Thumbnailjpg

The Handbook of Gangs

Wiley Handbooks in Criminology and Criminal Justice

Series Editor Charles F Wellford University of Maryland College Park

The handbooks in this series will be comprehensive academic reference works on leading topics in criminology and criminal justice

The Handbook of Law and SocietyEdited by Austin Sarat and Patricia Ewick

The Handbook of Juvenile Delinquency and Juvenile JusticeEdited by Marvin D Krohn and Jodi Lane

The Handbook of GangsEdited by Scott H Decker and David C Pyrooz

The Handbook of DevianceEdited by Erich Goode

The Handbook of Criminological TheoryEdited by Alex R Piquero

The Handbook of Drugs and SocietyEdited by Henry H Brownstein

The Handbook of Gangs

Edited by

Scott H Decker and David C Pyrooz

This edition first published 2015copy 2015 John Wiley amp Sons Inc

Registered OfficeJohn Wiley amp Sons Ltd The Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

Editorial Offices350 Main Street Malden MA 02148‐5020 USA9600 Garsington Road Oxford OX4 2DQ UKThe Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

For details of our global editorial offices for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at wwwwileycomwiley‐blackwell

The right of Scott H Decker and David C Pyrooz to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright Designs and Patents Act 1988

All rights reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise except as permitted by the UK Copyright Designs and Patents Act 1988 without the prior permission of the publisher

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books

Designations used by companies to distinguish their products are often claimed as trademarks All brand names and product names used in this book are trade names service marks trademarks or registered trademarks of their respective owners The publisher is not associated with any product or vendor mentioned in this book

Limit of LiabilityDisclaimer of Warranty While the publisher and authors have used their best efforts in preparing this book they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom If professional advice or other expert assistance is required the services of a competent professional should be sought

Library of Congress Cataloging‐in‐Publication Data

9781118726877 (hardback)

A catalogue record for this book is available from the British Library

Cover image copy Eky Studio Shutterstock

Set in 1012pt Minion by SPi Global Pondicherry India

1 2015

The Handbook of Gangs

Wiley Handbooks in Criminology and Criminal Justice

Series Editor Charles F Wellford University of Maryland College Park

The handbooks in this series will be comprehensive academic reference works on leading topics in criminology and criminal justice

The Handbook of Law and SocietyEdited by Austin Sarat and Patricia Ewick

The Handbook of Juvenile Delinquency and Juvenile JusticeEdited by Marvin D Krohn and Jodi Lane

The Handbook of GangsEdited by Scott H Decker and David C Pyrooz

The Handbook of DevianceEdited by Erich Goode

The Handbook of Criminological TheoryEdited by Alex R Piquero

The Handbook of Drugs and SocietyEdited by Henry H Brownstein

The Handbook of Gangs

Edited by

Scott H Decker and David C Pyrooz

This edition first published 2015copy 2015 John Wiley amp Sons Inc

Registered OfficeJohn Wiley amp Sons Ltd The Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

Editorial Offices350 Main Street Malden MA 02148‐5020 USA9600 Garsington Road Oxford OX4 2DQ UKThe Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

For details of our global editorial offices for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at wwwwileycomwiley‐blackwell

The right of Scott H Decker and David C Pyrooz to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright Designs and Patents Act 1988

All rights reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise except as permitted by the UK Copyright Designs and Patents Act 1988 without the prior permission of the publisher

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books

Designations used by companies to distinguish their products are often claimed as trademarks All brand names and product names used in this book are trade names service marks trademarks or registered trademarks of their respective owners The publisher is not associated with any product or vendor mentioned in this book

Limit of LiabilityDisclaimer of Warranty While the publisher and authors have used their best efforts in preparing this book they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom If professional advice or other expert assistance is required the services of a competent professional should be sought

Library of Congress Cataloging‐in‐Publication Data

9781118726877 (hardback)

A catalogue record for this book is available from the British Library

Cover image copy Eky Studio Shutterstock

Set in 1012pt Minion by SPi Global Pondicherry India

1 2015

Wiley Handbooks in Criminology and Criminal Justice

Series Editor Charles F Wellford University of Maryland College Park

The handbooks in this series will be comprehensive academic reference works on leading topics in criminology and criminal justice

The Handbook of Law and SocietyEdited by Austin Sarat and Patricia Ewick

The Handbook of Juvenile Delinquency and Juvenile JusticeEdited by Marvin D Krohn and Jodi Lane

The Handbook of GangsEdited by Scott H Decker and David C Pyrooz

The Handbook of DevianceEdited by Erich Goode

The Handbook of Criminological TheoryEdited by Alex R Piquero

The Handbook of Drugs and SocietyEdited by Henry H Brownstein

The Handbook of Gangs

Edited by

Scott H Decker and David C Pyrooz

This edition first published 2015copy 2015 John Wiley amp Sons Inc

Registered OfficeJohn Wiley amp Sons Ltd The Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

Editorial Offices350 Main Street Malden MA 02148‐5020 USA9600 Garsington Road Oxford OX4 2DQ UKThe Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

For details of our global editorial offices for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at wwwwileycomwiley‐blackwell

The right of Scott H Decker and David C Pyrooz to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright Designs and Patents Act 1988

All rights reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise except as permitted by the UK Copyright Designs and Patents Act 1988 without the prior permission of the publisher

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books

Designations used by companies to distinguish their products are often claimed as trademarks All brand names and product names used in this book are trade names service marks trademarks or registered trademarks of their respective owners The publisher is not associated with any product or vendor mentioned in this book

Limit of LiabilityDisclaimer of Warranty While the publisher and authors have used their best efforts in preparing this book they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom If professional advice or other expert assistance is required the services of a competent professional should be sought

Library of Congress Cataloging‐in‐Publication Data

9781118726877 (hardback)

A catalogue record for this book is available from the British Library

Cover image copy Eky Studio Shutterstock

Set in 1012pt Minion by SPi Global Pondicherry India

1 2015

The Handbook of Gangs

Edited by

Scott H Decker and David C Pyrooz

This edition first published 2015copy 2015 John Wiley amp Sons Inc

Registered OfficeJohn Wiley amp Sons Ltd The Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

Editorial Offices350 Main Street Malden MA 02148‐5020 USA9600 Garsington Road Oxford OX4 2DQ UKThe Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

For details of our global editorial offices for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at wwwwileycomwiley‐blackwell

The right of Scott H Decker and David C Pyrooz to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright Designs and Patents Act 1988

All rights reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise except as permitted by the UK Copyright Designs and Patents Act 1988 without the prior permission of the publisher

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books

Designations used by companies to distinguish their products are often claimed as trademarks All brand names and product names used in this book are trade names service marks trademarks or registered trademarks of their respective owners The publisher is not associated with any product or vendor mentioned in this book

Limit of LiabilityDisclaimer of Warranty While the publisher and authors have used their best efforts in preparing this book they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom If professional advice or other expert assistance is required the services of a competent professional should be sought

Library of Congress Cataloging‐in‐Publication Data

9781118726877 (hardback)

A catalogue record for this book is available from the British Library

Cover image copy Eky Studio Shutterstock

Set in 1012pt Minion by SPi Global Pondicherry India

1 2015

This edition first published 2015copy 2015 John Wiley amp Sons Inc

Registered OfficeJohn Wiley amp Sons Ltd The Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

Editorial Offices350 Main Street Malden MA 02148‐5020 USA9600 Garsington Road Oxford OX4 2DQ UKThe Atrium Southern Gate Chichester West Sussex PO19 8SQ UK

For details of our global editorial offices for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at wwwwileycomwiley‐blackwell

The right of Scott H Decker and David C Pyrooz to be identified as the authors of the editorial material in this work has been asserted in accordance with the UK Copyright Designs and Patents Act 1988

All rights reserved No part of this publication may be reproduced stored in a retrieval system or transmitted in any form or by any means electronic mechanical photocopying recording or otherwise except as permitted by the UK Copyright Designs and Patents Act 1988 without the prior permission of the publisher

Wiley also publishes its books in a variety of electronic formats Some content that appears in print may not be available in electronic books

Designations used by companies to distinguish their products are often claimed as trademarks All brand names and product names used in this book are trade names service marks trademarks or registered trademarks of their respective owners The publisher is not associated with any product or vendor mentioned in this book

Limit of LiabilityDisclaimer of Warranty While the publisher and authors have used their best efforts in preparing this book they make no representations or warranties with respect to the accuracy or completeness of the contents of this book and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose It is sold on the understanding that the publisher is not engaged in rendering professional services and neither the publisher nor the author shall be liable for damages arising herefrom If professional advice or other expert assistance is required the services of a competent professional should be sought

Library of Congress Cataloging‐in‐Publication Data

9781118726877 (hardback)

A catalogue record for this book is available from the British Library

Cover image copy Eky Studio Shutterstock

Set in 1012pt Minion by SPi Global Pondicherry India

1 2015

Notes on Contributors vii

1 Introduction 1Scott H Decker and David C Pyrooz

2 The Logic of Defining Gangs Revisited 7G David Curry

3 Little Gang Research Big Gang Research 28David C Pyrooz and Meghan M Mitchell

4 Documenting Gang Activity Intelligence Databases 59C Ronald Huff and Julie Barrows

5 Gang Membership in a Developmental and Life‐Course Perspective 78Beidi Dong Chris L Gibson and Marvin D Krohn

6 Neighborhoods and Street Gangs 98Andrew V Papachristos and Lorine A Hughes

7 Gangs and Social Learning Theory What We Know What We Need to Know and Why It Matters 118L Thomas Winfree Jr and Adrienne Freng

8 Social Psychology of Gangs An Intergroup Communication Perspective 136DaJung Woo Howard Giles Michael A Hogg and Liran Goldman

9 Social Network Analysis and Gangs 157Michael Sierra‐Arevalo and Andrew V Papachristos

10 Gangs Guns and Violence Synergistic Effects 178Arna L Carlock and Alan J Lizotte

11 Gangs and Drugs Connections Divergence and Culture 193Mark S Fleisher

12 Gender Sexuality and Gangs Re‐envisioning Diversity 208Vanessa R Panfil and Dana Peterson

Contents

vi Contents

13 Joining the Gang A Process of Supply and Demand 235James A Densley

14 Leaving the Gang A Review and Thoughts on Future Research 257Dena C Carson and J Michael Vecchio

15 Micro‐Level Processes of the Gang 276Jean M McGloin and Megan E Collins

16 Street Gangs Terrorists Drug Smugglers and Organized Crime Whatrsquos the Difference 294Scott H Decker and David C Pyrooz

17 Police Gang Units and Effective Gang Violence Reduction 309Anthony A Braga

18 Gangs in Correctional Institutions 328Shytierra Gaston and Beth M Huebner

19 Legislative Approaches to Addressing Gangs and Gang‐Related Crime 345Beth Bjerregaard

20 The Gang Resistance Education and Training (GREAT) Program An Evaluatorrsquos Perspective 369Finn‐Aage Esbensen

21 The OJJDP Comprehensive Gang Strategy The Comprehensive Gang Model 392Erika Gebo Brenda J Bond and Krystal S Campos

22 The Legacy of Malcolm W Klein 406Cheryl L Maxson

23 The Legacy of Irving A Spergel 424James C Howell

24 The Legacy of James F Short Jr 440Lorine A Hughes

25 The Legacy of Walter B Miller 458Richard K Moule Jr

26 Understanding Gangs in Contemporary Latin America 478Dennis Rodgers and Adam Baird

27 Understanding European Gangs 503Frank van Gemert and Frank M Weerman

28 European Responses to Gangs 520Rob Ralphs and Hannah Smithson

29 Gangs in African Asian and Australian Settings 538Angela Higginson and Kathryn Benier

Author Index 558Subject Index 572

Adam Baird PhD is Assistant Professor of International Peace Studies and Urban Governance at the UN University for Peace He studies gangs youth gender and mascu-linity and violence prevention and reduction in Central America and the Caribbean

Julie Barrows PhD is Adjunct Professor of Sociology at the University of Minnesota Her dissertation examined the formation network structure and effectiveness of gang task forces In addition to conducting gang research she has worked in the criminal justice system for many years and currently serves as a federal agent

Kathryn Benier is a PhD Candidate in the School of Social Science at the University of Queensland and a Research Assistant at the Institute for Social Science Research Her dis-sertation focuses on the community factors influencing the incidence of hate crime in Australia

Beth Bjerregaard PhD is a Professor and Chair in the Department of Criminal Justice amp Criminology at the University of North Carolina at Charlotte She received her PhD in Criminal Justice from the State University of New York at Albany Her research interests include capital punishment gang membership and gang delinquency

Brenda J Bond PhD is an Associate Professor of Public Service at Suffolk University Her work focuses on the management and performance of police organizations as well as cross‐agency collaboration She has examined these issues in a book co‐edited with Erika Gebo Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) and has recently published a paper on crime outcomes associated with effective group work in policing

Anthony A Braga PhD is the Don M Gottfredson Professor of Evidence‐Based Criminology in the School of Criminal Justice at Rutgers University and a Senior Research Fellow in the Program in Criminal Justice Policy and Management at Harvard University His research involves addressing illegal access to firearms reducing gang violence and controlling crime hot spots

Notes on Contributors

viii Notes on Contributors

Krystal S Campos is a recent graduate of Suffolk Universityrsquos Dual Masterrsquos Degree Program Crime and Justice Studies and Mental Health Counseling She has been a research assistant on studies focusing on youths who are considered high‐risk for gang and criminal involvement and the effects of youths living in high‐risk environments

Arna L Carlock is a PhD Candidate in the School of Criminal Justice at the University at Albany and an analyst for the Rochester Youth Development Study Her primary research interests include anticipated early death and gang violence stability and desistance

Dena C Carson PhD is an Assistant Professor in the School of Public and Environmental Affairs at Indiana University Purdue University Indianapolis Her general research interests include youth violence victimization gangs and delinquent peer groups Her recent pub-lications have appeared in Youth amp Society Journal of Criminal Justice and Youth Violence amp Juvenile Justice

Megan E Collins is a PhD Student in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include policing practices gang violence and criminal justice policy

G David Curry PhD is Professor Emeritus in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis He earned his PhD in sociology from the University of Chicago in 1976 His research addresses issues dealing with gangs and delinquency military service hate crime and domestic violence

Scott H Decker PhD is Foundation Professor in the School of Criminology and Criminal Justice at Arizona State University His main research interests are in the areas of gangs vio-lence criminal justice policy and the offenderrsquos perspective He is a Fellow in the American Society of Criminology and the Academy of Criminal Justice Sciences

James A Densley PhD is Assistant Professor of Criminal Justice at Metropolitan State University Minnesota He earned his doctorate in sociology from the University of Oxford Densleyrsquos teaching and research interests include street gangs criminal networks violence and theoretical criminology He is the author of How Gangs Work An Ethnography of Youth Violence (2013)

Beidi Dong is a doctoral candidate in the Department of Sociology and Criminology amp Law at the University of Florida His research interests include developmental and life-course criminology youth gangs and violence crime and place as well as juvenile delinquency and justice in a comparative sense

Finn‐Aage Esbensen PhD is the E Desmond Lee Professor of Youth Crime and Violence and also serves as the Chair of the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis

Mark S Fleisher PhD a cultural and linguistic anthropologist is a Research Professor at the Jack Joseph and Morton Mandel School of Applied Social Sciences Case Western Reserve University He has written Warehousing Violence (1989) Beggars and Thieves (1995) Crime and Employment (2004) Dead End Kids (1998) The Myth of Prison Rape (2009) and Few Escape (2015)

Notes on Contributors ix

Adrienne Freng PhD is currently a Full Professor in the Department of Criminal Justice at the University of Wyoming Her research interests include juvenile delinquency gangs and race and crime issues including specifically how they relate to American Indian populations She has written and co‐authored numerous articles based on these topics as well as co‐authored Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010)

Shytierra Gaston MS is a doctoral candidate in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis and a National Science Foundation graduate research fellow Her research areas focus on corrections prisoner reentry families of offenders and race crime and justice

Erika Gebo PhD is an Associate Professor of Sociology at Suffolk University Her areas of interest include youth violence and the implementation and evaluation of crime policies and programs She recently co‐edited the book Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) with Brenda J Bond

Chris L Gibson PhD is a Research Foundation Professor and Associate Professor of Criminology at the University of Florida He also holds affiliate appointments in the College of Medicine Institute for Child Health Policy and the College of Lawrsquos Criminal Justice Center He recently co‐edited with Marvin D Krohn the book Handbook of Life‐Course Criminology Emerging Trends and Directions for Future Research (2013)

Howard Giles PhD is Professor of Communication at the University of California Santa Barbara He is founding Editor of the Journal of Language and Social Psychology and the Journal of Asian Pacific Communication and was past President of the International Communication Association and the International Association of Language and Social Psychology

Liran Goldman is a recent PhD graduate from the Department of Behavioral and Organizational Sciences at Claremont Graduate University in Los Angeles She is a recipient of a Department of Homeland Security Fellowship as well as a START (Study of Terrorism and Responses to Terrorism) Research Award

Angela Higginson PhD is a Lecturer in Criminology at the School of Social Science at the University of Queensland Her research evaluates policing and community crime control interventions Recent work includes systematic reviews of the predictors and prevention of youth gangs in low‐ and middle‐income countries

Michael A Hogg PhD is Professor of Social Psychology at Claremont Graduate University in Los Angeles He is founding Editor of Group Processes and Intergroup Relations a former Associate Editor of the Journal of Experimental Social Psychology and past President of the Society of Experimental Social Psychology

James C Howell PhD is a Senior Research Associate with the National Gang Center in Tallahassee FL where he has worked for 19 years Dr Howell enjoys gang research and is active in helping states and localities address gang and juvenile delinquency problems using evidenced‐based services and programs

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

vi Contents

13 Joining the Gang A Process of Supply and Demand 235James A Densley

14 Leaving the Gang A Review and Thoughts on Future Research 257Dena C Carson and J Michael Vecchio

15 Micro‐Level Processes of the Gang 276Jean M McGloin and Megan E Collins

16 Street Gangs Terrorists Drug Smugglers and Organized Crime Whatrsquos the Difference 294Scott H Decker and David C Pyrooz

17 Police Gang Units and Effective Gang Violence Reduction 309Anthony A Braga

18 Gangs in Correctional Institutions 328Shytierra Gaston and Beth M Huebner

19 Legislative Approaches to Addressing Gangs and Gang‐Related Crime 345Beth Bjerregaard

20 The Gang Resistance Education and Training (GREAT) Program An Evaluatorrsquos Perspective 369Finn‐Aage Esbensen

21 The OJJDP Comprehensive Gang Strategy The Comprehensive Gang Model 392Erika Gebo Brenda J Bond and Krystal S Campos

22 The Legacy of Malcolm W Klein 406Cheryl L Maxson

23 The Legacy of Irving A Spergel 424James C Howell

24 The Legacy of James F Short Jr 440Lorine A Hughes

25 The Legacy of Walter B Miller 458Richard K Moule Jr

26 Understanding Gangs in Contemporary Latin America 478Dennis Rodgers and Adam Baird

27 Understanding European Gangs 503Frank van Gemert and Frank M Weerman

28 European Responses to Gangs 520Rob Ralphs and Hannah Smithson

29 Gangs in African Asian and Australian Settings 538Angela Higginson and Kathryn Benier

Author Index 558Subject Index 572

Adam Baird PhD is Assistant Professor of International Peace Studies and Urban Governance at the UN University for Peace He studies gangs youth gender and mascu-linity and violence prevention and reduction in Central America and the Caribbean

Julie Barrows PhD is Adjunct Professor of Sociology at the University of Minnesota Her dissertation examined the formation network structure and effectiveness of gang task forces In addition to conducting gang research she has worked in the criminal justice system for many years and currently serves as a federal agent

Kathryn Benier is a PhD Candidate in the School of Social Science at the University of Queensland and a Research Assistant at the Institute for Social Science Research Her dis-sertation focuses on the community factors influencing the incidence of hate crime in Australia

Beth Bjerregaard PhD is a Professor and Chair in the Department of Criminal Justice amp Criminology at the University of North Carolina at Charlotte She received her PhD in Criminal Justice from the State University of New York at Albany Her research interests include capital punishment gang membership and gang delinquency

Brenda J Bond PhD is an Associate Professor of Public Service at Suffolk University Her work focuses on the management and performance of police organizations as well as cross‐agency collaboration She has examined these issues in a book co‐edited with Erika Gebo Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) and has recently published a paper on crime outcomes associated with effective group work in policing

Anthony A Braga PhD is the Don M Gottfredson Professor of Evidence‐Based Criminology in the School of Criminal Justice at Rutgers University and a Senior Research Fellow in the Program in Criminal Justice Policy and Management at Harvard University His research involves addressing illegal access to firearms reducing gang violence and controlling crime hot spots

Notes on Contributors

viii Notes on Contributors

Krystal S Campos is a recent graduate of Suffolk Universityrsquos Dual Masterrsquos Degree Program Crime and Justice Studies and Mental Health Counseling She has been a research assistant on studies focusing on youths who are considered high‐risk for gang and criminal involvement and the effects of youths living in high‐risk environments

Arna L Carlock is a PhD Candidate in the School of Criminal Justice at the University at Albany and an analyst for the Rochester Youth Development Study Her primary research interests include anticipated early death and gang violence stability and desistance

Dena C Carson PhD is an Assistant Professor in the School of Public and Environmental Affairs at Indiana University Purdue University Indianapolis Her general research interests include youth violence victimization gangs and delinquent peer groups Her recent pub-lications have appeared in Youth amp Society Journal of Criminal Justice and Youth Violence amp Juvenile Justice

Megan E Collins is a PhD Student in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include policing practices gang violence and criminal justice policy

G David Curry PhD is Professor Emeritus in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis He earned his PhD in sociology from the University of Chicago in 1976 His research addresses issues dealing with gangs and delinquency military service hate crime and domestic violence

Scott H Decker PhD is Foundation Professor in the School of Criminology and Criminal Justice at Arizona State University His main research interests are in the areas of gangs vio-lence criminal justice policy and the offenderrsquos perspective He is a Fellow in the American Society of Criminology and the Academy of Criminal Justice Sciences

James A Densley PhD is Assistant Professor of Criminal Justice at Metropolitan State University Minnesota He earned his doctorate in sociology from the University of Oxford Densleyrsquos teaching and research interests include street gangs criminal networks violence and theoretical criminology He is the author of How Gangs Work An Ethnography of Youth Violence (2013)

Beidi Dong is a doctoral candidate in the Department of Sociology and Criminology amp Law at the University of Florida His research interests include developmental and life-course criminology youth gangs and violence crime and place as well as juvenile delinquency and justice in a comparative sense

Finn‐Aage Esbensen PhD is the E Desmond Lee Professor of Youth Crime and Violence and also serves as the Chair of the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis

Mark S Fleisher PhD a cultural and linguistic anthropologist is a Research Professor at the Jack Joseph and Morton Mandel School of Applied Social Sciences Case Western Reserve University He has written Warehousing Violence (1989) Beggars and Thieves (1995) Crime and Employment (2004) Dead End Kids (1998) The Myth of Prison Rape (2009) and Few Escape (2015)

Notes on Contributors ix

Adrienne Freng PhD is currently a Full Professor in the Department of Criminal Justice at the University of Wyoming Her research interests include juvenile delinquency gangs and race and crime issues including specifically how they relate to American Indian populations She has written and co‐authored numerous articles based on these topics as well as co‐authored Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010)

Shytierra Gaston MS is a doctoral candidate in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis and a National Science Foundation graduate research fellow Her research areas focus on corrections prisoner reentry families of offenders and race crime and justice

Erika Gebo PhD is an Associate Professor of Sociology at Suffolk University Her areas of interest include youth violence and the implementation and evaluation of crime policies and programs She recently co‐edited the book Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) with Brenda J Bond

Chris L Gibson PhD is a Research Foundation Professor and Associate Professor of Criminology at the University of Florida He also holds affiliate appointments in the College of Medicine Institute for Child Health Policy and the College of Lawrsquos Criminal Justice Center He recently co‐edited with Marvin D Krohn the book Handbook of Life‐Course Criminology Emerging Trends and Directions for Future Research (2013)

Howard Giles PhD is Professor of Communication at the University of California Santa Barbara He is founding Editor of the Journal of Language and Social Psychology and the Journal of Asian Pacific Communication and was past President of the International Communication Association and the International Association of Language and Social Psychology

Liran Goldman is a recent PhD graduate from the Department of Behavioral and Organizational Sciences at Claremont Graduate University in Los Angeles She is a recipient of a Department of Homeland Security Fellowship as well as a START (Study of Terrorism and Responses to Terrorism) Research Award

Angela Higginson PhD is a Lecturer in Criminology at the School of Social Science at the University of Queensland Her research evaluates policing and community crime control interventions Recent work includes systematic reviews of the predictors and prevention of youth gangs in low‐ and middle‐income countries

Michael A Hogg PhD is Professor of Social Psychology at Claremont Graduate University in Los Angeles He is founding Editor of Group Processes and Intergroup Relations a former Associate Editor of the Journal of Experimental Social Psychology and past President of the Society of Experimental Social Psychology

James C Howell PhD is a Senior Research Associate with the National Gang Center in Tallahassee FL where he has worked for 19 years Dr Howell enjoys gang research and is active in helping states and localities address gang and juvenile delinquency problems using evidenced‐based services and programs

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

Adam Baird PhD is Assistant Professor of International Peace Studies and Urban Governance at the UN University for Peace He studies gangs youth gender and mascu-linity and violence prevention and reduction in Central America and the Caribbean

Julie Barrows PhD is Adjunct Professor of Sociology at the University of Minnesota Her dissertation examined the formation network structure and effectiveness of gang task forces In addition to conducting gang research she has worked in the criminal justice system for many years and currently serves as a federal agent

Kathryn Benier is a PhD Candidate in the School of Social Science at the University of Queensland and a Research Assistant at the Institute for Social Science Research Her dis-sertation focuses on the community factors influencing the incidence of hate crime in Australia

Beth Bjerregaard PhD is a Professor and Chair in the Department of Criminal Justice amp Criminology at the University of North Carolina at Charlotte She received her PhD in Criminal Justice from the State University of New York at Albany Her research interests include capital punishment gang membership and gang delinquency

Brenda J Bond PhD is an Associate Professor of Public Service at Suffolk University Her work focuses on the management and performance of police organizations as well as cross‐agency collaboration She has examined these issues in a book co‐edited with Erika Gebo Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) and has recently published a paper on crime outcomes associated with effective group work in policing

Anthony A Braga PhD is the Don M Gottfredson Professor of Evidence‐Based Criminology in the School of Criminal Justice at Rutgers University and a Senior Research Fellow in the Program in Criminal Justice Policy and Management at Harvard University His research involves addressing illegal access to firearms reducing gang violence and controlling crime hot spots

Notes on Contributors

viii Notes on Contributors

Krystal S Campos is a recent graduate of Suffolk Universityrsquos Dual Masterrsquos Degree Program Crime and Justice Studies and Mental Health Counseling She has been a research assistant on studies focusing on youths who are considered high‐risk for gang and criminal involvement and the effects of youths living in high‐risk environments

Arna L Carlock is a PhD Candidate in the School of Criminal Justice at the University at Albany and an analyst for the Rochester Youth Development Study Her primary research interests include anticipated early death and gang violence stability and desistance

Dena C Carson PhD is an Assistant Professor in the School of Public and Environmental Affairs at Indiana University Purdue University Indianapolis Her general research interests include youth violence victimization gangs and delinquent peer groups Her recent pub-lications have appeared in Youth amp Society Journal of Criminal Justice and Youth Violence amp Juvenile Justice

Megan E Collins is a PhD Student in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include policing practices gang violence and criminal justice policy

G David Curry PhD is Professor Emeritus in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis He earned his PhD in sociology from the University of Chicago in 1976 His research addresses issues dealing with gangs and delinquency military service hate crime and domestic violence

Scott H Decker PhD is Foundation Professor in the School of Criminology and Criminal Justice at Arizona State University His main research interests are in the areas of gangs vio-lence criminal justice policy and the offenderrsquos perspective He is a Fellow in the American Society of Criminology and the Academy of Criminal Justice Sciences

James A Densley PhD is Assistant Professor of Criminal Justice at Metropolitan State University Minnesota He earned his doctorate in sociology from the University of Oxford Densleyrsquos teaching and research interests include street gangs criminal networks violence and theoretical criminology He is the author of How Gangs Work An Ethnography of Youth Violence (2013)

Beidi Dong is a doctoral candidate in the Department of Sociology and Criminology amp Law at the University of Florida His research interests include developmental and life-course criminology youth gangs and violence crime and place as well as juvenile delinquency and justice in a comparative sense

Finn‐Aage Esbensen PhD is the E Desmond Lee Professor of Youth Crime and Violence and also serves as the Chair of the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis

Mark S Fleisher PhD a cultural and linguistic anthropologist is a Research Professor at the Jack Joseph and Morton Mandel School of Applied Social Sciences Case Western Reserve University He has written Warehousing Violence (1989) Beggars and Thieves (1995) Crime and Employment (2004) Dead End Kids (1998) The Myth of Prison Rape (2009) and Few Escape (2015)

Notes on Contributors ix

Adrienne Freng PhD is currently a Full Professor in the Department of Criminal Justice at the University of Wyoming Her research interests include juvenile delinquency gangs and race and crime issues including specifically how they relate to American Indian populations She has written and co‐authored numerous articles based on these topics as well as co‐authored Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010)

Shytierra Gaston MS is a doctoral candidate in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis and a National Science Foundation graduate research fellow Her research areas focus on corrections prisoner reentry families of offenders and race crime and justice

Erika Gebo PhD is an Associate Professor of Sociology at Suffolk University Her areas of interest include youth violence and the implementation and evaluation of crime policies and programs She recently co‐edited the book Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) with Brenda J Bond

Chris L Gibson PhD is a Research Foundation Professor and Associate Professor of Criminology at the University of Florida He also holds affiliate appointments in the College of Medicine Institute for Child Health Policy and the College of Lawrsquos Criminal Justice Center He recently co‐edited with Marvin D Krohn the book Handbook of Life‐Course Criminology Emerging Trends and Directions for Future Research (2013)

Howard Giles PhD is Professor of Communication at the University of California Santa Barbara He is founding Editor of the Journal of Language and Social Psychology and the Journal of Asian Pacific Communication and was past President of the International Communication Association and the International Association of Language and Social Psychology

Liran Goldman is a recent PhD graduate from the Department of Behavioral and Organizational Sciences at Claremont Graduate University in Los Angeles She is a recipient of a Department of Homeland Security Fellowship as well as a START (Study of Terrorism and Responses to Terrorism) Research Award

Angela Higginson PhD is a Lecturer in Criminology at the School of Social Science at the University of Queensland Her research evaluates policing and community crime control interventions Recent work includes systematic reviews of the predictors and prevention of youth gangs in low‐ and middle‐income countries

Michael A Hogg PhD is Professor of Social Psychology at Claremont Graduate University in Los Angeles He is founding Editor of Group Processes and Intergroup Relations a former Associate Editor of the Journal of Experimental Social Psychology and past President of the Society of Experimental Social Psychology

James C Howell PhD is a Senior Research Associate with the National Gang Center in Tallahassee FL where he has worked for 19 years Dr Howell enjoys gang research and is active in helping states and localities address gang and juvenile delinquency problems using evidenced‐based services and programs

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

viii Notes on Contributors

Krystal S Campos is a recent graduate of Suffolk Universityrsquos Dual Masterrsquos Degree Program Crime and Justice Studies and Mental Health Counseling She has been a research assistant on studies focusing on youths who are considered high‐risk for gang and criminal involvement and the effects of youths living in high‐risk environments

Arna L Carlock is a PhD Candidate in the School of Criminal Justice at the University at Albany and an analyst for the Rochester Youth Development Study Her primary research interests include anticipated early death and gang violence stability and desistance

Dena C Carson PhD is an Assistant Professor in the School of Public and Environmental Affairs at Indiana University Purdue University Indianapolis Her general research interests include youth violence victimization gangs and delinquent peer groups Her recent pub-lications have appeared in Youth amp Society Journal of Criminal Justice and Youth Violence amp Juvenile Justice

Megan E Collins is a PhD Student in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include policing practices gang violence and criminal justice policy

G David Curry PhD is Professor Emeritus in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis He earned his PhD in sociology from the University of Chicago in 1976 His research addresses issues dealing with gangs and delinquency military service hate crime and domestic violence

Scott H Decker PhD is Foundation Professor in the School of Criminology and Criminal Justice at Arizona State University His main research interests are in the areas of gangs vio-lence criminal justice policy and the offenderrsquos perspective He is a Fellow in the American Society of Criminology and the Academy of Criminal Justice Sciences

James A Densley PhD is Assistant Professor of Criminal Justice at Metropolitan State University Minnesota He earned his doctorate in sociology from the University of Oxford Densleyrsquos teaching and research interests include street gangs criminal networks violence and theoretical criminology He is the author of How Gangs Work An Ethnography of Youth Violence (2013)

Beidi Dong is a doctoral candidate in the Department of Sociology and Criminology amp Law at the University of Florida His research interests include developmental and life-course criminology youth gangs and violence crime and place as well as juvenile delinquency and justice in a comparative sense

Finn‐Aage Esbensen PhD is the E Desmond Lee Professor of Youth Crime and Violence and also serves as the Chair of the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis

Mark S Fleisher PhD a cultural and linguistic anthropologist is a Research Professor at the Jack Joseph and Morton Mandel School of Applied Social Sciences Case Western Reserve University He has written Warehousing Violence (1989) Beggars and Thieves (1995) Crime and Employment (2004) Dead End Kids (1998) The Myth of Prison Rape (2009) and Few Escape (2015)

Notes on Contributors ix

Adrienne Freng PhD is currently a Full Professor in the Department of Criminal Justice at the University of Wyoming Her research interests include juvenile delinquency gangs and race and crime issues including specifically how they relate to American Indian populations She has written and co‐authored numerous articles based on these topics as well as co‐authored Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010)

Shytierra Gaston MS is a doctoral candidate in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis and a National Science Foundation graduate research fellow Her research areas focus on corrections prisoner reentry families of offenders and race crime and justice

Erika Gebo PhD is an Associate Professor of Sociology at Suffolk University Her areas of interest include youth violence and the implementation and evaluation of crime policies and programs She recently co‐edited the book Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) with Brenda J Bond

Chris L Gibson PhD is a Research Foundation Professor and Associate Professor of Criminology at the University of Florida He also holds affiliate appointments in the College of Medicine Institute for Child Health Policy and the College of Lawrsquos Criminal Justice Center He recently co‐edited with Marvin D Krohn the book Handbook of Life‐Course Criminology Emerging Trends and Directions for Future Research (2013)

Howard Giles PhD is Professor of Communication at the University of California Santa Barbara He is founding Editor of the Journal of Language and Social Psychology and the Journal of Asian Pacific Communication and was past President of the International Communication Association and the International Association of Language and Social Psychology

Liran Goldman is a recent PhD graduate from the Department of Behavioral and Organizational Sciences at Claremont Graduate University in Los Angeles She is a recipient of a Department of Homeland Security Fellowship as well as a START (Study of Terrorism and Responses to Terrorism) Research Award

Angela Higginson PhD is a Lecturer in Criminology at the School of Social Science at the University of Queensland Her research evaluates policing and community crime control interventions Recent work includes systematic reviews of the predictors and prevention of youth gangs in low‐ and middle‐income countries

Michael A Hogg PhD is Professor of Social Psychology at Claremont Graduate University in Los Angeles He is founding Editor of Group Processes and Intergroup Relations a former Associate Editor of the Journal of Experimental Social Psychology and past President of the Society of Experimental Social Psychology

James C Howell PhD is a Senior Research Associate with the National Gang Center in Tallahassee FL where he has worked for 19 years Dr Howell enjoys gang research and is active in helping states and localities address gang and juvenile delinquency problems using evidenced‐based services and programs

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

Notes on Contributors ix

Adrienne Freng PhD is currently a Full Professor in the Department of Criminal Justice at the University of Wyoming Her research interests include juvenile delinquency gangs and race and crime issues including specifically how they relate to American Indian populations She has written and co‐authored numerous articles based on these topics as well as co‐authored Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010)

Shytierra Gaston MS is a doctoral candidate in the Department of Criminology and Criminal Justice at the University of Missouri‐St Louis and a National Science Foundation graduate research fellow Her research areas focus on corrections prisoner reentry families of offenders and race crime and justice

Erika Gebo PhD is an Associate Professor of Sociology at Suffolk University Her areas of interest include youth violence and the implementation and evaluation of crime policies and programs She recently co‐edited the book Looking Beyond Suppression Community Strategies to Reduce Gang Violence (2012) with Brenda J Bond

Chris L Gibson PhD is a Research Foundation Professor and Associate Professor of Criminology at the University of Florida He also holds affiliate appointments in the College of Medicine Institute for Child Health Policy and the College of Lawrsquos Criminal Justice Center He recently co‐edited with Marvin D Krohn the book Handbook of Life‐Course Criminology Emerging Trends and Directions for Future Research (2013)

Howard Giles PhD is Professor of Communication at the University of California Santa Barbara He is founding Editor of the Journal of Language and Social Psychology and the Journal of Asian Pacific Communication and was past President of the International Communication Association and the International Association of Language and Social Psychology

Liran Goldman is a recent PhD graduate from the Department of Behavioral and Organizational Sciences at Claremont Graduate University in Los Angeles She is a recipient of a Department of Homeland Security Fellowship as well as a START (Study of Terrorism and Responses to Terrorism) Research Award

Angela Higginson PhD is a Lecturer in Criminology at the School of Social Science at the University of Queensland Her research evaluates policing and community crime control interventions Recent work includes systematic reviews of the predictors and prevention of youth gangs in low‐ and middle‐income countries

Michael A Hogg PhD is Professor of Social Psychology at Claremont Graduate University in Los Angeles He is founding Editor of Group Processes and Intergroup Relations a former Associate Editor of the Journal of Experimental Social Psychology and past President of the Society of Experimental Social Psychology

James C Howell PhD is a Senior Research Associate with the National Gang Center in Tallahassee FL where he has worked for 19 years Dr Howell enjoys gang research and is active in helping states and localities address gang and juvenile delinquency problems using evidenced‐based services and programs

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

x Notes on Contributors

Beth M Huebner PhD is an Associate Professor at the University of Missouri‐St Louis She received her PhD from the Michigan State University in Criminal Justice Her current research interests include prisoner reentry criminal justice decision making and public policy Her current research explores the efficacy of current sex offender policy and the long‐term patterns of recidivism among serious violent offenders

C Ronald Huff PhD is Professor Emeritus at both the University of California Irvine and the Ohio State University His current research focuses on gangs wrongful convictions of innocent persons and public policy He is a Fellow and past President of the American Society of Criminology

Lorine A Hughes PhD is Associate Professor in the School of Criminology and Criminal Justice at the University of Nebraska at Omaha The last of Professor Shortrsquos graduate stu-dents she has spent the past few years digitizing and reanalyzing Short and Strodtbeckrsquos data using modern methods Resulting publications (with Short) appear in Criminology and the Journal of Quantitative Criminology

Marvin D Krohn PhD is at the University of Florida He is interested in life course approaches His work on the Rochester Youth Development Study has led to numerous research articles and a co‐authored book Gangs and Delinquency in Developmental Perspective which was the American Society of Criminologyrsquos recipient of the 2003 Hindelang Award for Outstanding Scholarship He was recently named a Fellow of the ASC

Alan J Lizotte is Dean and Professor of Criminal Justice at the University at Albany and a co‐principal investigator of the Rochester Youth Development Study His research interests include guns and gun control correlates and causes of delinquency over the life course and policy health issues for clients of the criminal justice system

Cheryl L Maxson PhD is a Professor in the Department of Criminology Law and Society at the University of Californiarsquos Irvine campus Her research addresses street gangs status offenders youth violence juvenile justice legislation and community treatment of juvenile offenders

Jean M McGloin PhD is an Associate Professor in the Department of Criminology and Criminal Justice at the University of Maryland Her research interests include peer influence co‐offending and offending specialization Her work has been published in Criminology the Journal of Research in Crime and Delinquency and the Journal of Quantitative Criminology

Meghan M Mitchell MS is a Doctoral Student at Sam Houston State University Her inter-ests include the intersection of social justice issues (race class gender) and crime along with gangs and school‐based delinquency

Richard K Moule Jr MS is a Doctoral Student in the School of Criminology and Criminal Justice at Arizona State University His research interests include gangs and deviant net-works developmentallife‐course criminology and the intersection of technology and criminological theory He is the archivist for the Walter B Miller Library

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

Notes on Contributors xi

Vanessa R Panfil PhD is a Post‐Doctoral Associate in the School of Criminal Justice at Rutgers University She studies how intersections of gender and sexuality shape individ-ualsrsquo experiences with gangs crime victimization and the criminal and juvenile justice systems She co‐edited with Dana Peterson Handbook of LGBT Communities Crime and Justice (2014)

Andrew V Papachristos PhD is an Associate Professor of Sociology Public Health and Law at Yale University His research applies the growing field of social network analysis to understanding patterns of crime victimization in US cities His writing has appeared in Foreign Policy The American Journal of Sociology The Annals of the American Academy of Social and Political Science The American Journal of Public Health The Journal of Urban Health and Criminology amp Public Policy among other outlets

Dana Peterson PhD is an Associate Professor in the School of Criminal Justice at the University at Albany She studies youth gangs violence and how sex and gender structure these Publications include Youth Violence Sex and Race Differences in Offending Victimization and Gang Membership (2010 with Esbensen Taylor and Freng) and Handbook of LGBT Communities Crime and Justice (2014 with Panfil)

David C Pyrooz PhD is an Assistant Professor of Sociology at the University of Colorado Boulder He studies gangs and deviant networks violence and developmental and life-course criminology He is the co-author with G David Curry and Scott Decker of Confronting Gangs Crime and Community (3rd edition 2013) and the recipient of the 2015 Academy New Scholar Award from the Academy of Criminal Justice Sciences

Rob Ralphs PhD is a Senior Lecturer in Criminology at the Manchester Metropolitan University United Kingdom His current research focuses on UK gang policy including his recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

Dennis Rodgers PhD is Professor of Urban Social and Political Research at the University of Glasgow A social anthropologist by training he works on issues relating to urban development conflict and violence in Nicaragua Argentina and India His most recent publication is the volume Global Gangs Street Violence across the World co‐edited with Jennifer Hazen (2014)

Michael Sierra‐Arevalo MA is a Doctoral Student in the Department of Sociology at Yale University and an affiliate fellow at the Institution for Social and Policy Studies (ISPS) His research focuses on gangs the police urban violence and the causes and effects of legal cynicism

Hannah Smithson PhD is a Reader in Criminology in the Department of Sociology and Criminology at Manchester Metropolitan University United Kingdom Her current research focuses on UK gang policy including her recent co‐authored article ldquoUsed and Abused The Problematic Usage of Gang Terminology in the United Kingdom and Its Implications for Ethnic Minority Youthrdquo in the British Journal of Criminology

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

xii Notes on Contributors

Frank van Gemert PhD is Assistant Professor at the Department of Criminal Law and Criminology at VU University Amsterdam As a qualitative researcher he prefers to collect data as an ethnographer and more recently as a biographer His work is on gangs drug dealers homicide squatters and more generally cultural criminology

J Michael Vecchio PhD is an Assistant Professor in the Department of Criminal Justice and Criminology at Loyola University Chicago His research interests include youth vio-lence and victimization youth gangs and responses to victimization His most recent pub-lished work appeared in Deviant Behavior

Frank M Weerman PhD is a senior researcher at the Netherlands Institute for the Study of Crime and Law Enforcement (NSCR) His publications focus on the explanation of juvenile delinquency on co‐offending and youth gangs and on quantitative analyses of social networks delinquent peers and peer‐related activities

L Thomas Winfree Jr PhD is Visiting Professor in Arizona State Universityrsquos School of Criminology and Criminal Justice He has co‐authored multiple editions of five textbooks including Understanding Crime Essentials of Criminological Theory (3rd edition 2009) he has also co‐edited two anthologies including Social Learning Theories of Crime (2012) Winfree is co‐author of dozens of theory‐based articles

DaJung (DJ) Woo (MA Kansas State University) is a PhD student in the Communication Department at the University of California Santa Barbara Her research projects examine the processes and dynamics in which individuals become socialized and assimilated into groups organizations and occupational roles through communication

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

The Handbook of Gangs is a comprehensive volume that presents the current state of knowledge about gangs Each of the chapters is written by a leading expert in the world on their chosen topic and stands alone as an insightful review of what is known about that topic But taken together the chapters provide us with a broad understanding of the foundation for gang research the theories and methods used to study gang behaviors and processes the many forms of gangs the correlates of gangs and gang membership gang prevention and intervention programs and gang experts Indeed we believe that one of the strengths of the book is the series of interrelationships among the chapters While the topics of each chapter are different they are related through themes problems theories and methods

The goal of this volume is to provide a definitive reference for professionals working in the field of gang prevention or suppression researchers and students What we know about gangs has become interdisciplinary and internationalized in the course of the past 20 years and our book reflects both of those trends We believe that this volume pulls together the most contemporary reviews on the key topics in the understanding of gangs and the responses to gangs To accomplish this we have engaged the best scholars in their area many of whom have chosen to work with an emerging scholar Each chapter provides a critical review of what is known about the topic as well as the insights of the authors about the topic In this way the book will be grounded in the current knowledge about the specific topics but also provides new material that reflects the knowledge of the leading minds in the field While there is a strong orientation toward sociological criminology in the book we believe that the chapters reflect a broad approach in both method and theory continuing to expand the boundaries of gang research

The book is organized into seven sections

1 Laying the Foundation for Understanding Gangs (Chapters 2ndash4)2 Theories of Gangs (Chapters 5ndash10)3 Gang Correlates (Chapters 11ndash12)4 Gang Processes (Chapters 13ndash16)

IntroductionScott H Decker and David C Pyrooz

1

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

2 Chapter 1

5 Responding to Gangs (Chapters 17ndash21)6 Pioneers in Gang Research (Chapters 22ndash25)7 Gangs in International Context (Chapters 26ndash29)

There are many unique features to the book and we use this introduction to walk the reader through those features One aspect of the book that we are particularly proud of is the personal touch in many of the chapters We encouraged the authors to give us insights into their personal knowledge about the research and researchers in their topic area We think this makes the chapters more interesting to the reader This is an especially important feature of this book as many of the authors are intimately involved with the production of knowledge in the area of their chapter know other individuals working on contemporary research in the area and have a strong appreciation for the history of work in that area For example Andrew Papachristos is among the leading network theorists and methodoshylogists in the social sciences He has teamed with Michael Sierra‐Arevalo to produce an outstanding chapter (9) on the use of network analysis and theory with gangs which inteshygrates network theory with their ongoing network analysis Finn Aage‐Esbensen wrote a chapter (20) on the Gang Resistance Education and Training (GREAT) project that is insightful for what it tells us both about GREAT from an evaluatorrsquos perspective as well as about evaluation methods Similarly Anthony Braga ranks among the very best police scholars in the world and has written an excellent chapter (17) on the role of the police in responding to gangs particularly focused deterrence approaches We believe that the other chapters reflect a comparable level of prominence among the authors Indeed the names of many of these authors are synonymous with the topics of their chapters

Over the past two decades there has been a virtual explosion of gang research Indeed the third chapter of this book written by David Pyrooz and Meghan Mitchell traces the history of gang research and documents this explosion in great detail what they term the transition from ldquolittlerdquo to ldquobigrdquo gang research A key theme in the history of gang research is the definition of a gang a theme that David Curry revisits in his chapter (2) nearly two decades after his work on the logic of defining gangs and measuring gang activities In their chapter (4) on documenting gang activity Ronald Huff and Julie Barrows show us not only why the definition and measurement of gangs is so important but also the essential features of how gang intelligence is gathered and recorded Indeed the explosion of research on gangs has occurred in tandem to responses to gangs which rely on the very records carefully described by Huff and Barrows

Research on gangs has grown in theoretical sophistication as well in methodological rigor Theoretical developments have followed Shortrsquos tripartite focus on macro micro and individual theories and levels of explanation Each of these levels of explanation is represented in this volume This includes new perspectives such as social psychological approaches to the study of gangs (DaJung Woo Howard Giles Michael Hogg and Liran Goldman chapter 8) emerging perspectives such as the growing attention to gang membership in developmental and life course perspective (Beidi Dong Chris Gibson and Marvin Krohn chapter 5) and longstanding perspectives such as social learning theory (Thomas Winfree and Adrienne Freng chapter 7) Further the correlates of gangs and gang membership are documented in chapter 10 on violence and synergy between gangs and guns (Arna Carlock and Alan Lizotte) chapter 11 on the connections between gangs drugs and culture (Mark Fleisher) and chapter 12 that rethinks gender and gangs (Vanessa Panfil and Dana Peterson) At the macro‐level Andrew Papachristos and Lorine Hughes (chapter 6) provide a chapter on neighborhoods and gangs that is not anchored

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

Introduction 3

to any single theoretical perspective which is often contentious instead viewing gangs as ldquoindependent variablesrdquo and ldquodependent variablesrdquo as causes and consequences of neighborhood social organization and problems

Perhaps most importantly the book also pays explicit attention to explanations of gang‐related processes and behaviors This remains the least studied yet perhaps most important of the remaining items on the gang research agenda Chapters 13 and 14 tackle the imporshytant processes of how people enter and exit from gangs James Densley presents readers with a theory of joining gangs that is rooted in a signaling perspective something that offers great value to gang research Dena Carson and Michael Vecchio alternatively review the current state of the evidence on disengaging from gangs a burgeoning area of gang research True to Shortrsquos calls for micro‐level gang research Jean McGloin and Megan Collins expertly outline in chapter 15 the micro‐level intra‐ and inter‐gang processes theorized to elevate levels of crime and violence among gangs and gang members These processes are believed to make gangs qualitatively different which is why in chapter 16 we (the editors) ask what are the structural organizational and social features that gangs uniquely possess compared to other criminal and extremist groups

Units of government (cities states and the federal government) have worked to develop responses to gangs While some of these responses have emphasized suppression to the exclusion of other approaches there is a growing emphasis on prevention and intervention programs For example the Centers for Disease Control and the National Institute of Justice have collaborated to produce a volume titled Changing Course that reviews the primary areas of responding to gangs from an integrated public health and criminal justice perspecshytive Many of its chapters explicitly acknowledge the efforts to control the spread of gangs and the harm that gang membership causes through increased criminal involvement This handbook captures these trends in several chapters most notably Shytierra Gaston and Beth Huebnerrsquos chapter on gangs and corrections (18) Beth Bjerregaardrsquos chapter on gang legislation (19) and Erika Gebo Brenda Bond and Krystal Camposrsquos chapter (21) on the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Comprehensive Strategy

The spread of gangs in the United States and the worldwide growth in the use of the Internet has meant that gangs have ldquogone globalrdquo American‐style youth gangs can now be found in a growing number of countries around the world While there are some core simshyilarities across these gangs in many cases they are quite distinctive reflecting religious cultural historical and institutional differences of the countries Frank van Gemert and Frank Weerman (chapter 27) examine trends patterns and correlates of contemporary European gangs Their chapter is nicely complemented by Rob Ralphs and Hannah Smithsonrsquos review (chapter 28) of gang intervention strategies in Europe While much gang research has crossed the Atlantic to Europe Dennis Rodgers and Adam Baird (chapter 26) identify a growing body of research in Latin American countries and Angela Higginson and Kathryn Benier (chapter 29) close the book with a review of what is known about gangs in Asia Africa and Australia These chapters are explicitly comparative and contemporary

But we believe that the most unique contribution found in the book comes in the four chapters that examine the work of four individual gang researchers who have made contrishybutions to this literature since the middle classic era of gang research The four individuals whose contributions are highlighted in separate chapters advancing the understanding of gangs are Malcolm Klein Irving Spergel James F Short Jr and Walter Miller Each of these individuals is responsible for developing a pillar of the foundation of gang research We term these the ldquolegacyrdquo chapters Each chapter examining their involvement in gang research has elements of autobiography historiography and philosophy of science in addition to a

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

4 Chapter 1

review of their work These chapters were written by individuals who have worked with or have been intimately involved with the work of the four pioneers Lorine Hughes who wrote the chapter about James F Short Jr was one of his doctoral students and has worked with him and the data from his Chicago project Cheryl Maxson wrote the chapter about Malcolm Klein Cheryl and ldquoMacrdquo have worked together for over 20 years and no one knows Macrsquos work better than Cheryl James C Howell worked closely with the late Irving Spergel to develop the ldquoSpergel Modelrdquo which became the basis for nearly $100 million of federal gang intervention efforts Richard Moule wrote the chapter about Walter Miller It is fair to say no one has read more of Walterrsquos work than Richard and while they never met he understands Millerrsquos work better than anyone in the field today We hope you like reading them as much as we have

It is appropriate to ask what the next steps are for gang research and policy What are the questions that need to be answered that have not been resolved or not even been asked We see four broad areas that are ripe for the attention of young scholars

First it is important to address the gaps in our current knowledge about gangs This includes a number of salient topics such as the role of technology in gang behavior and the spread of gang behavior The digital empowerment assumption holds that the Internet and social media offer key advantages to gangs and other criminal networks yet it might also prove to be a detriment What is needed is an understanding of if and how these new techshynologies influence gang behavior Also so much of what is known about gangs is derived from a criminological perspective yet the very mechanisms leading to criminal behavior are not exclusive to this behavior but general to a variety of non‐criminal consequences as well Therefore it is necessary to learn about the implications ndash negative and positive ndash of gangs for other socializing institutions

Second we know all too little about group process in gangs despite Shortrsquos work from nearly 50 years ago exhorting us to better understand group process This concern is closely tied to the level of measurement brought to understand gangs This is an issue that is far more than an arcane devotion to method and measurement indeed it is at the core of what we need to know about gangs Short urges us to consider whether we are studying and responding to individuals (gang members) groups (gangs) or structural aspects of society (class neighborhood variables or the like) Despite his influential work on this topic in the 1960s and 1980s there is still inadequate attention to these issues

Third we lack an understanding of how gangs compare to other groups involved in crime The growth of interest in terrorist and extremist groups organized crime groups drug smugglers and money launderers (to name but a few) is illustrative of this interest Do these groups share common organizational structural or group process variables Are these groups wholly distinct and different from each other Are there overlaps between these groups that draw our attention to a common understanding and shared responses Do individuals belong to multiple gangs or transition from one form of crime organization (such as gangs) to another (such as organized crime or terror groups) We have a long way to go in pursuing these questions and much of the progress will likely be made outside of gang research

Fourth and in a related point we need to import models and methods from other discishyplines Scholars and policymakers working toward a better understanding and response to organized crime have made valuable use of trial transcripts as sources of data pointing the way toward more innovative sources of information about the topics we wish to study Those who study terrorists and terrorism use data from social media such as Twitter Facebook and email to better understand and respond to such groups The world of gang

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

Introduction 5

scholarship has been slow to utilize such data and many have been resistant to the use of new forms of data This is indeed ironic since most gang members are at the peak ages for the use of social media the late teens

As we move ahead in gang research there is a need to include work from the past in gang research as a foundation to build on but not to be so tied to prior research that we cannot move ahead to new paradigms Lorine Hughes has done an excellent job using the Short and Strodtbeck data from Chicago gangs in the 1960s Andrew Papachristos has ldquodiscovshyeredrdquo data from the Chicago Crime Commission about the Capone family activities in Chicago in the 1920s Decker and Moule are working with the data from the Boston Special Youth Project in the 1950s and Moule is analyzing the Gluecksrsquo gang data It appears that the Cambridge‐Somerville data is now being used Of course Sampson and Laub have led the way by making effective use of the Gluecksrsquo delinquency data from the Lyman School in the 1940s In addition to the utility of data sets from earlier times gang scholars need to preserve the memories work and records of earlier researchers This includes the legacy of individual scholars such as Short Klein Miller and Spergel as well as countless research projects The US Justice Department has funded tens of millions of dollars of gang research and evaluation Perhaps the single best leveraged investment comes from evaluations of the GREAT project Finn‐Aage Esbensen has led multiple evaluations of GREAT programs While the evaluation focus has been strong and the methodology has been top notch the data has provided some of the most important basic and theoretical findings about gangs Similarly the Causes and Correlates funding has led to important findings in basic and applied knowledge about gangs Leaders from both of these projects have contribshyuted chapters (5 10 and 20) to this volume The importance of this work lies in the very strong research designs that allow them to track changes in behavior over time and thus speak of the causes of gang behavior

It is also true that our knowledge about gangs needs expansion beyond its current geographic limits This includes looking past the United States not only to Europe where thanks to Malcolm Klein Cheryl Maxson Hans‐Juumlrgen Kerner and others there is a set of burgeoning research projects but also to Africa South and Central America Asia and Australia The boundaries of gang research do not just include countries they also include the scholars involved in the enterprise Gang research has been conducted largely at a small number of American universities The boundaries of gang research need to be expanded to include new methods new locations new groups and new scholars We know very little about gangs in rural and suburban areas a situation that needs to be addressed Questions about race ethnicity gender sexual preference family violence and countless other topics need to be included in the study of gangs By asking questions about these correlates of gangs in a cross‐cultural context we can illustrate the gang context in other countries as well as provide a contrast to the situation in the United States

The study of gangs needs to avoid a fixed image of the topic Too often common stereoshytypes are reified and accepted as if they are true In many instances this is a consequence of being swayed too much by the ldquoofficial versionrdquo of what gangs are The ldquoofficial versionrdquo of gangs has always been different from what researchers find from what programs report and what gang members have to say about their gangs This book balances the ldquocommon wisdomrdquo about gangs with what objective research has found

How should one read this book We think that there is a sequential ordering to the chapshyters and that the sections of the book reflect common themes We have also provided links between the various chapters That said a theory or correlate chapter could be paired with an intervention chapter to contrast how well they intersect For example do the risk factors

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

6 Chapter 1

or correlates of gang involvement identified by a theory get addressed by an intervention This would be a way to determine whether interventions are addressing known causes of gangs or if researchers and practitioners are talking ldquopastrdquo each other The chapters on gangs outside of the United States are useful to compare to those on gangs within the United States

We brought these authors and their chapters together to produce a comprehensive volume of knowledge about gangs and gang members The chapters are written by the leading authorities in the world on their given topic It is our goal that this book will be used by those who seek to understand gangs and to craft changes to improve the lives of gang members and those affected by their behavior We hope you find that the book accomplishes these goals

Finally it is important that we acknowledge those working ldquobehind the scenesrdquo who made this book come together We thank Chantal Fahmy Rick Moule Matthias Woeckener and Jun Wu each of whom spent an extraordinary amount of time and care poring over the chapters The good folks at Wiley Blackwell notably Linsay Bourgeous Haze Humbert Julia Kirk Allison Kostka Breanna Locke and Julia Teweles were excelshylent to work with and ensured the success of this project across each stage of the process Our colleagues at Arizona State University and Sam Houston State University including Gaylene Armstrong Cassia Spohn and Vince Webb among others are deserving of recshyognition for supporting this project over the last three years Most importantly we are grateful to our families Thank you to JoAnn Sara Laura and Elizabeth And thank you to Natty Cyrus and Adalyn A final thank you is in order to each of the contributors for participating in this project

Sadly our friend and colleague Dave Curry passed away on April 26 2015 We are very proud to include chapter 2 his last published work in this volume Dave was a pivotal figure in the study of gangs and juvenile justice His work was always motivated by and focused on improving the lives of young people We miss you man

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Handbook of Gangs First Edition Edited by Scott H Decker and David C Pyrooz copy 2015 John Wiley amp Sons Inc Published 2015 by John Wiley amp Sons Inc

Irsquove had it with gangs(Klein 1971)

Irsquove had it with gang definitionsWhile I was in the words of Katz and Jackson‐Jacobs (2004) busily feeding the ldquomoral panicrdquo over gang problems in the United States I never felt comfortable reporting the number of gangs Of all the statistics reported about gangs the number of gangs is the least reliable and meaningful because of the difficulty in consistently distinguishing sets cliques and subgroups within gangs Furthermore a gang is rarely the target of an intervention Over the last 20 years I have become increasingly aware that gang violence was costing and disrupting lives dispro-portionately the lives of minority and poverty youth On the basis of that awareness I have spent my career trying to provide reliable estimates of the magnitude of gang problems because as Huff (1990) notes both undercounts (denial) and overcounts (over‐identification) are coun-terproductive in controlling gang crime But let us face it the number of gang members and even the number of gang homicides are just not as scary as ldquo29000 gangs threaten USrdquo

A definition of a gang however is an essential element of a survey instrument or interview The questions (for individual self‐reports) ldquoHave you ever been a gang memberrdquo as well as (for law enforcement agencies) ldquoDoes your jurisdiction have gangsrdquo are central to building knowledge about gangs and gang members While I confess to not caring exactly what lexical or nominal definition for gangs is used I insist on using the same definitions as much as possible Consistent definitions are what makes it social science instead of social philosophy

Gang problem prevalence is linked to how gangs are defined and what unit of analysis ndash gang gang members or gang crimes ndash is used The guiding focus in this chapter is how researchers use definitions of gangs in their research This chapter draws from my prior work on the logic of gang definitions (Ball and Curry 1995) and officially recorded delinquency and gang membership (Curry 2000) The importance of operational definitions over lexical definitions is stressed throughout this work as the appropriate direction for gang research The Appendix at the end of the chapter explains the methods of definition that are referenced in the chapter

The Logic of Defining Gangs RevisitedG David Curry

2

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

8 Chapter 2

This chapter tackles three specific tasks First I trace the evolution of definitions and the measurement of gangs and gang membership Here I start with Thrasher and end with the Eurogang definition the contemporary ldquoacceptedrdquo definition of a gang despite its discontents An important part of this section is the definitional debate over including involvement in crime in the definition of gang membership I then compare official and self‐reported gang data The utility of research on officially recorded delinquency and gang involvement is defended Finally I move to an examination of gang homicide the gang crime we know most about

The Evolution of Definition and Measurement of Gangs

Always Begin with Thrasher

In his gang course Irving Spergel encouraged his students to always begin learning and thinking about gangs with Frederic Thrasher I agree that all gang discussions should begin with Thrasher (1927) One of the first graduate students in the department of sociology at the University of Chicago Thrasherrsquos ldquoNrdquo of 1313 gangs has been attributed to graduate school culture (Geis and Dodge 2002 Katz and Jackson‐Jacobs 2004) rather than anything to do with gangs Supposedly no one including Thrasher ever counted the 1313 gangs The 1313 was a street address of some repute known to University of Chicago graduate students and its inclusion in a dissertation title was part of a gambling pool

No definition of gang will ever be so elegant or as aesthetically pleasing as Thrasherrsquos tripartite definition (Howell 2012) According to Ball and Curry (1995) Thrasher was influenced by a desire to move beyond the earlier Darwinist definition of the gang offered by Puffer (1912) by producing a sociological definition In its full form Thrasherrsquos definition is as follows ldquoAn interstitial group formed spontaneously and then integrated through conflictrdquo (Thrasher 1927 57)

Thrasher (1927) found his gangs in the interstices of the city Technically interstices are the empty spaces within the physical and social structure of the city not being used by other elements of city life and are available for the gang and its activity It is no surprise that the first space that Thrasher turns over to the gang is the ldquoplaygroundrdquo ndash that brutal milieu of unsupervised childhood activity Thrasherrsquos gangs emerge spontaneously as playgroups The third definitional attribute of Thrasherrsquos gang is that gangs become organized through processes of conflict Interstitial spontaneity and conflict ndash sheer intellectual poetry

As stated above my primary focus in this chapter is how researchers use definitions of gangs in their research That includes Thrasher Thrasher (1927) began with three crucial concepts and worked outward to create a complete catalogue for everything that might be called a gang in early twentieth‐century Chicago Thrasher is the prototypical taxonomist as he moves from schoolyard to organized crime to corrupt politicians as well as a new generation of civic organizations and young menrsquos athletic clubs So long as Thrasherrsquos gangs are interstitial emerge spontaneously and evolve through conflict with other groups over time they fit his definition and they fit into his entertaining volume Gang activity is characterized by Thrasher as filled with imagination romance and adventure

Ball and Curry (1995) point out that Thrasherrsquos definition is ldquodurablerdquo and a ldquosynthetic definition by descriptionrdquo However Ball and Curry also suggest that Thrasherrsquos defini-tion contains some subcategories of definition not so acceptable to ldquothe logic of definitionrdquo Thrasherrsquos (1927 46) identification of the gang as an ldquointerstitial grouprdquo reflects the ldquosynthetic method of definitionrdquo in two ways Thrasher locates his gang in two transitional

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Logic of Defining Gangs Revisited 9

zones one psychological and one sociological The maturation developmental phase between childhood and adulthood is massively populated by social science psychology social work and education theory and research Thrasher went one step further in his synthetic definition to add a socio‐geographic category full of a large portion of our population whose social organizational structure is not well understood those who live in the transitional zone between the central business district and more stable residential areas

Ball and Curry (1995) have one small corner of Hell reserved for ldquocausal synthetic definitionsrdquo and they cast part of Thrasherrsquos definition into that crevice Ball and Curry (1995 237) attest that Thrasher (1927) is less consumed by Darwinism than Pufferrsquos (1912) earlier characterization of gangs but ldquopart of Thrasherrsquos classic definitionrdquo represents the same kind of causal definition as Puffer It is that part about terming the gang as ldquoa group that is formed spontaneously then integrated through conflictrdquo On causal synthetic definitions Ball and Curry (1995 237) recognized that correlational definitions are acceptable and useful if they are clarified causal definitions ldquoshould be confined to axiomatic systems such as geometry where tautology is proof of consistency rather than a source of error and should be strictly avoided in any field wishing to develop through empirical researchrdquo Thrasher did many services for gang research his definition of a gang however durable and poetic was not one of them

Definitions from Gang Researchrsquos Past and Their Discontents

While there has been no definition as durable as Thrasherrsquos (1927) there are several prior to the Eurogang definition that merit recognition The first enduring post‐Thrasher defini-tion of a gang was offered by Klein (1971 13 Klein and Maxson 2014) Maybe more than Thrasherrsquos definition Kleinrsquos (1971) definition is durable and widely known The definition as described by Klein and Maxson (2014) is based on Kleinrsquos study of five large clusters of gangs in Los Angeles and characterized as based on a ldquosocial‐psychological frameworkrdquo This definition reads

[a juvenile gang is] any denotable group of youngsters who (a) are generally perceived as a distinct aggregation by others in their neighborhood (b) recognize themselves as a denotable group (almost invariably with a group name) and (c) have been involved in a sufficient number of delinquent incidents to call forth a consistent negative response from neighborhood residents andor enforcement agencies (Klein 1971 13)

Bursik and Grasmick (1993) had a little fun with Kleinrsquos (1971) definition by pointing out that a certain deviant category of college fraternities fit Kleinrsquos definition of a gang fairly snuggly Klein (1995) abandoned this definition in the face of what he has called so much controversy over this or any other definition

Perhaps no one has been more troubled by the absence of a post‐Thrasher (1927) defini-tion of a gang than Walter Miller (see chapter 25 in this volume) Miller lamented ldquoAt no time has there been anything close to consensus on what a youth gang might berdquo (1975 115) According to Ball and Curry (1995 237) Millerrsquos (1958) first definition of a gang was a causal synthetic definition Millerrsquos (1975) effort to reach a consensus definition was to ask a national survey of youth service agency workers police officers community outreach workers judges criminal justice planners probation officers prosecutors public defenders educators city council members state legislators ex‐convicts and past and present

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

10 Chapter 2

members of gangs for their definition of the term Although the result was a list of 1400 different characteristics 85 agreed on six items defining a youth gang as

a self‐formed association of peers bound together by mutual interests with identifiable lead-ership well‐developed lines of authority and other organizational features who act in concert to achieve a specific purpose or purposes which generally include the conduct of illegal activity and control over a particular territory facility or type of enterprise (Miller 1975 121)

Klein and Maxson (1989 205) regard ldquovotingrdquo on a definition as ldquodiscouragingrdquo or as a ldquopopularity pollrdquo (Klein and Maxson 2006) Ball and Curry (1995) maintain that voting on a lexical definition is ldquoappropriaterdquo so long as the population is specifically defined but voting on a stipulative definition is a ldquopoorrdquo method when the subject of the definition is ldquohighly emotional and different segments of society seem to be approaching the subject from different perspectives with different purposesrdquo As for me I ldquovoterdquo that Miller is ldquoappropriaterdquo with his voting population selection and his stipulative definition

Post‐Thrasher (1927) critiques of gang research in the United States did not really begin until Joan Moorersquos (1988) preface to John Hagedornrsquos People and Folks Gangs Crime and the Underclass in a Rustbelt City Hagedorn also contributed a chapter to C Ronald Huff rsquos (1990) reader Gangs in America that reviewed the general state of theory and methods in gang research Hagedorn got some of the other authorsrsquo chapters before publication and was able to get in some blistering jibes in the published edition where he expanded his bibliog-raphy of ldquobadrdquo gang research that he had begun in People and Folks The most recent expanded critique of gang research has been offered by Jack Katz and Curtis Jackson‐Jacobs (2004) where they condemn among others the work of Hagedorn Since aggressive criti-cism is comparatively rare in gang research (except maybe in anonymous reviews) and since Katz and Jackson‐Jacobs (2004) deal with definitional issues I devote some detailed attention to their work

Katz and Jackson‐Jacobs (2004) divide gang research into three fundamental stages (1) the naturalistic which escaped criticism (2) the ldquowindow framerdquo phase and (3) the ignorance of causal issues It is easy to understand their preference for the naturalistic as well as their disdain for (2) and (3) which they viewed as having major ldquodefectsrdquo and ldquoperversionsrdquo Thrasher (1927) alone is stage (1) and hence beyond criticism Katz and Jackson‐Jacobs list three exceptions to the dearth of good research post‐Thrasher Two of them are not really comparable to Thrasher while one is One is William Sandersrsquos (1994) analysis of police reports and his field notes from police ride‐alongs in a study of evolving gang problems in an emerging gang city The other is R Lincoln Keiserrsquos (1969) The Vicelords Warriors of the Streets an ethnography of the Chicago gang of that name The third is Scott H Decker and Barrik Van Winklersquos (1996) Life in the Gang Family Friends and Violence which is very likely the kind of book Thrasher would have written about gangs in St Louis in the 1990s

What they label stage (2) and condemn was also condemned by Hagedorn in his language as ldquotoo much theory too little factsrdquo For Katz and Jackson‐Jacobs (2004) stage (2) is an absence of data and presence of secret agendas of theories too often concealing ready‐made program designs Most typical and downright evil were Richard A Cloward and Lloyd Ohlin (1960) Katz and Jackson‐Jacobs (2004 96) maintain that Cloward and Ohlinrsquos publication and its associated programs provided the structure and policies that shaped many of the programs that became Johnsonrsquos War on Poverty as well as adding a plethora of awards to the authorsrsquo shelves Katz and Jackson‐Jacobs wrote

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Logic of Defining Gangs Revisited 11

criminologists have seen through the gang using the gang as a window onto phenomena which are treated as far more important than documenting the everyday realities of gang members on their own turf and in their own terms Like politicians and journalists who shape popular culture gang criminologists have been preoccupied with the gang as etonym icon or index (2004 94)

Who Says the Art of Invoking Metaphor Has Been LostThat Cloward and Ohlin (1960) apply Mertonrsquos model of anomie to gangs is the stretch taken by Katz and Jackson‐Jacobs (2004 102) and is the weakest theoretical stretch since the word ldquogangsrdquo was added to the subtitle of Delinquency and Opportunity I feel that Katz and Jackson‐Jacobs are extremely soft in dealing with Cloward and Ohlin I ask them to show me their ldquocommunity of readersrdquo (2004 106) for Delinquency and Opportunity A Theory of Delinquent Gangs That community does not exist as best as I can tell especially among gang researchers themselves

Their handling of the gang as a unit of analysis is part of Katz and Jackson‐Jacobsrsquos (2004) theoretical frustration ldquoGangrdquo is used interchangeably by Katz and Jackson‐Jacobs as a collection of individuals falling under some definition of a gang and as a social phenomenon crossing generations and urban geography The third issue of Katz and Jackson‐Jacobs is that they insist repeatedly on the need to establish an empirical relation-ship between gangs and violent crime that fits within the definition of the gang and does not imply deviant behavior As a central theme this is second only to the theme with which they begin their chapter where they critique the absence of analytic utility of official record data for gang research One concern is unnecessary and the other is wrong ldquoThe gangrdquo as they describe it as social phenomenon and social movement does not require an established link to crime unless the researcher assigns it one in an operational definition I feel that offi-cial records data is a valuable dialectical source of what we have to learn about gangs and will address that argument below For me solutions for our nationrsquos gang problem must involve law enforcement the courts and especially corrections Their policies and actions in response to gangs are a part of the gang problem and have to be a part of any solution

Tautology and the ldquoCausalrdquo Question

To use crime in the definition of a gang or not to use crime That is the question In my discussion of the ldquocausal questionrdquo I explore the relationship between gangs or ldquotherdquo gang and violence (or delinquency) without being anchored to a specific stipulative definition of a gang I am quite aware that definitions of the gang that include negative behavior such as illegal activity deviance or crime will ultimately tell us very little about any relationship between gangs and such behavior (which I return to below) It is a tautology with a capital T For the most part though it is important to know there are important quests to perform other than finding the holy grail of the link between gangs and violence through definition There are a number of conditions that make any search difficult If we keep specific gangs in the questions models are going to be hierarchical requiring transitions between micro‐ and macro‐levels of social processes (Coleman 1990 Matsueda 2013) Behaviors linked between different points in time will underlie almost every aspect of gang behavior requiring dynamic longitudinal models and analyses something recently proposed by Decker Melde and Pyrooz (2013) that is not easy to achieve

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

12 Chapter 2

There is a tradition of gang definition that is exceptional in that it avoids any inclusion of deviance or law‐breaking in its criteria of a gang For James F Short Jr

Gangs are groups whose members meet together with some regularity over time on the basis of group‐defined criteria of membership and group‐defined organizational characteristics that is gangs are non‐adult sponsored self‐determining groups that demonstrate a continuity over time (1996 5)

Shortrsquos definition avoids any risk of tautology in examining the relationship between gangs and law‐violating behavior A number of more contemporary definitions follow Shortrsquos lead Similar approaches to defining gangs without referencing law‐breaking behavior are offered by a number of ethnographic researchers For Brotherton (1997) gangs are a social construction of forces outside the gang Conquergood (1997) speaks of gangs as a particular form of literacy in symbols associated with street life1 Garot (2010) suggests that gang involvement is just one form of ldquoperformancerdquo available to students who resist their social status in the authority struc-ture of our society All of these approaches based on definitions of gangs without an element of law‐breaking leave empirical researchers with categorizations of gangs that are too broad for focused analysis a problem that we will revisit below with the Eurogang definition

Katz and Jackson‐Jacobs (2004) contend that we will understand little about the essence of gang behavior until we cast it in the literature and theoretical language of social movements For me this is one other reason to work patiently on unraveling the relationship between gangs and violence rather than waste time looking for a lexical definition to solve the problem without empiricism The causal question whether we are studying the individual behaviors of gang members the collective behaviors of gangs or the criminal rates of gang neighborhoods is not going to be possible to address without an understanding of the dimension of time

According to Katz and Jackson‐Jacobs (2004) the causal question refers to the lack of ldquoa good basis for thinking gangs cause crimerdquo They correctly realize that embedding measures of negative behavior into an operational definition of a gang traps us in a tautology They are not the first James F Short Jr (1990) has always counseled that delinquency or crime should not be an element of how we measure gang involvement This was a mistake I made in my efforts with Irving Spergel to create scales of gang involvement a couple of decades ago (Curry and Spergel 1992) Fighting was such a routine boyhood behavior for Irving and me that we honestly did not think of it as criminal behavior We corrected our efforts to create models of gang involvement with passing time in response to the gnawing criticism of reviewers It was when the longitudinal gang and delinquency studies appeared in several large ndash though not particularly gang‐troubled ndash cities that I concluded that answers to causal questions about gangs and delinquency might ultimately be obtainable (Krohn and Thornberry 2008) This has played out at the individual level not as gangs as groups (see chapter 3 in this volume) And because of that this research has escaped the tautology critique because criminal involvement is not a prerequisite for gang membership at least in the measures of gang membership advocated by Decker et al (2014b) Esbensen et al (2001) and Thornberry et al (2003)

The importance of time in any question of the relationship between gangs and crime has led researchers to turn inevitably to life course approaches to criminology delinquency and gang involvement Foremost in the life transition approach to understanding gangs and delinquency is the outstanding work of Terence Thornberry and his colleagues (1993) Thornberry suggests three models of how gang participation and criminal behavior could result in relationships between gangs and crime These models were not presented as proof

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Logic of Defining Gangs Revisited 13

of a positive relationship but as potential structural relationships among selected variables bridging the macro‐level of gang and community and the micro‐level of gang membership The three models are labeled ldquoselectionrdquo ldquoenhancementrdquo and ldquofacilitationrdquo In the selection model potential gang members offend more than their peers the process of selecting fellow delinquents as friends and the opposing rejection by non‐delinquent peers leads to the delinquent youths being isolated together In selection gang members are just as delinquent before and after joining the gang Gang members are in a phase of their lives where their offending is high whether they are in a gang or not In the facilitation model gangs ldquofacilitaterdquo or provide opportunities and encouragement for delinquency and crime associated with the group context and dynamics of the gang Enhancement is a blending of the selection and facilitation models Only under facilitation or enhancement is there a positive relation-ship between gangs and crime That does not mean that the relationship is assumed It means that it needs figuring out No causal definition will serve our needs In fact the logic of definition warns that it will not work Empirical analysis is needed

While I do not think that there is overwhelming agreement with Thornberry et alrsquos (1993) facilitation model Krohn and Thornberryrsquos (2008) and Curry Decker and Pyroozrsquos (2014) reviews of the studies explicitly testing these models show that there is more support for facilitation than selection models However it appears that there is even more support for the blended model the enhancement model

Even if an enhancement model best reflects the social reality it means that the ldquocausalrdquo debate as a dichotomy (yes or no) can be put to rest in favor of the factors that explain or moderate this relationship Indeed the vast majority of this research has been conducted in a handful of sites and among youth which reflects a much different reality than law enforcement records which show that the gang problem is concentrated among adults not juveniles (Curry 2000) Research that focuses only on juveniles misses out on a very large portion of the population of gang members Pyrooz (2014) showed that 40 of those with a history of gang membership were gang members in adulthood using national longitudinal survey data Of the six trajectories of gang membership that he mapped using group‐based trajectory modeling three of the trajectories he identified (labeled ldquoadult onsetrdquo and ldquoearlyrdquo and ldquolate persistentrdquo) crossed the line into adulthood and make it certain any prolonged official report records would look more and more like those of adult gang offenders

In just 20 years of research we have progressed from speculating about how gang mem-bersrsquo delinquency and gang careers might overlap with each other to deriving the most common membership trajectories in a national‐level sample of gang members Knowing that distinctly different chronological patterns of gang involvement exist moves us beyond potentially baffling questions about what population those of us who are studying gangs (Curry 2000 Curry et al 2014 Decker et al 2013) are actually studying to knowing the diversity of combinations available It is my contention that researchers are making progress in answering the so‐called ldquocausal questionrdquo of deciphering the association between gangs and delinquency with better methods and data rather than a definition of gang that will fit the philosophical needs of gang researchers

The Eurogang Takeover

The Eurogang definition is ldquoA street gang is any durable street‐oriented youth group whose involvement in illegal activity is part of its group identityrdquo (Klein and Maxson 2006 4) It is rather straightforward The definition was introduced in Klein (2001) in The Eurogang

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

14 Chapter 2

Paradox the first volume of the Eurogang program of research (see chapter 22) As Klein and Maxson pointed out there are defining components of gangs and then there are descriptors They claim that the Eurogang definition focuses only on the defining components

Many feel that the Eurogang definition is ldquocommonly usedrdquo Four edited Eurogang volumes of research is proof that it is And this definition is now commonly seen and sometimes it is even the operational definition of gangs or gang membership in journal articles According to Klein and Maxson ldquoThis is the consensus nominal definition agreed to by a consortium of more than 100 American and European researchers and policy makers from more than a dozen nations meeting in a series of eight workshops between 1997 and 2005rdquo (2006 4)

I attended two of those workshops and I do not remember signing off on that definition but there are other things that I do not remember from those workshops Based on his logical analysis of the definition itself and what he knows about street gangs and the spirit of international unity Klein (2014) recommends that from now on we adopt the Eurogang definition for future research Klein feels such an agreement in the past would have avoided the fractured nature of 60 years of research and theory on youth gang research I endorse the Eurogang definition but it is only one of the definitions that I have endorsed in my career and feel comfortable endorsing in this chapter I was a student of Irving Spergel (1989 13) who noted that gang definitions ldquohave varied over time according to the percep-tion and interests of the definer academic fashions and the changing social reality of the gangrdquo But as anyone with even a brief familiarity with gang research knows consensus with gang definitions is rare and the Eurogang definition is no exception

Aldridge Medina‐Ariz and Ralphs (2012) identify problems with the Eurogang defi-nition especially for British gangs They dispute the part of the definition that describes gangs as street‐oriented since many gangs avoid the streets so as not to be subjected to police harassment or drive‐by shootings which are not uncommon in their study area that they refer to as Research City They argue that some gangs are street‐oriented but some are not They also take issue with the part of the Eurogang definition that states that gangs have illegal activity as part of their identity They describe three groups that fit the Eurogang definition in their identity being based on illegal activity but that no one in their commu-nities or the authors consider to be ldquogangsrdquo These include what Americans would call stoners ravers and a group that takes drugs and goes to night clubs All of these are kinds of groups that Thrasher (1927) describes as gangs but Klein (2014) has frequently excluded from his own research on gangs Aldridge et al (2012) turn from ethnographic analysis to survey data to make their case against the Eurogang definition Specifically they examine the Offending Crime and Justice Survey in the United Kingdom and survey data from Madrid From their analysis of the survey data the authors argue that the Eurogang defini-tion condition of the group being involved in illegal behavior is probably not a valid measure of gang involvement Aldridge et al consider that a gang is a group that is willing to engage in violence and suggest that gang researchers studying in an international context would develop a better understanding of that aspect of gang behavior

Apparently the United Kingdom is a country in which there is little agreed‐upon legal or academic definition of gangs Shute and Medina (2014) begin their critique of the response to gangs in Britain with the description of an imaginary beast from a childrenrsquos book called The Gruffalo They argue that in its response to gangs Britain has created its own Gruffalo in gangs and gang crime This threat is used to invoke ldquomoral panic and justifying greater police powers in socially marginal communitiesrdquo The authors point out that in an official report from the Home Office the term ldquogangrdquo is used 266 times without ever providing an

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

The Logic of Defining Gangs Revisited 15

evidence‐based operational definition of what a gang is In the same vein as Shute and Medina Smithson Ralphs and Williams (2013) reference a think tank report accusing an actual defini-tion of a gang by the Home Office as distorting the gang problem resulting in ldquogroups of Black and Asian men (being) seen as gangs criminalized and then dealt with on this basis by the policerdquo In their study of ldquoNorthvillerdquo these authors found no youths who identified themselves as belonging to any organized group They found the same denial of gangs in interviewing members of the community including elderly residents who testified that their communities have no gangs or gang problems Still some of the authors referenced above signed on to a doc-ument known as the Manchester Gang Response Network (date unknown) urging that gang responses in the United Kingdom use an empirically grounded gang definition in responding to gangs While some of them have found fault with the Eurogang definition they view the Eurogang definition as a useful ldquostarting pointrdquo in the United Kingdom But they also feel it could lead to labeling minorities and net‐widening Finally the Manchester Network recom-mends ldquothe use of Eurogang measuresrdquo for all respondents in the childrenrsquos supplement of the Crime Survey of England and Wales as ldquothe best way forwardrdquo

Matsuda Esbensen and Carson (2012) compare the Eurogang definition with two other widely used methods of identifying gang members using one of the Gang Resistance Education and Training (GREAT) samples of 3162 students selected from 31 schools and 195 classrooms The authors used three definitional methods of identifying gang members self‐nomination the Eurogang definition and friends in the gang This last definitional category has been treated as ldquogang involvementrdquo as opposed to gang membership by some authors (Curry Decker and Egley 2002) The results were quite interesting The Eurogang definition identified the largest percentage of the sample as gang members (68) followed by friends in a gang (55) and self‐nomination (48) The definitional overlap of all three definitional methods was only 9 The overlap between the Eurogang definition and self‐nomination was only 13 and the overlap between gangs as friends and the Eurogang definition was 17 Since I am a ldquoself‐nominationrdquo sympathizer but ldquoapproverrdquo of the Eurogang definition I am most discouraged by the small overlap between self‐nomination and the Eurogang definition regardless of how well each definition independently identifies high‐risk youth Between this evidence and that of the critiques of the UK contingent it would suggest that the Eurogang definition is worthy of continued empirical investigation so that the empirical consensus on the definition is on par with the researcher consensus on the definition

Official Records and Gangs Gang Membership and Gang Violence

Dual Realities of Gangs The Distrust of Official Records

Consistently research has found a positive correlation between self‐reported gang mem-bership and self‐reported delinquency (Curry et al 2002 Esbensen and Huizinga 1993 Thornberry et al 1993) Similarly the relationship between self‐reported delinquency and officially recorded delinquency (Elliott Huizinga and Moore 1987 Hawkins et al 1998) has been explored systematically In Curry (2000) official records for a five‐year period on a population of at‐risk youth were used to examine the relationship between survey self‐report measures of gang involvement and delinquency in early adolescence and subsequent officially recorded delinquency and police‐identified gang involvement

Understanding the relationship between survey measures of gang involvement and delinquency and official records on gang involvement and delinquency has implications for

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence

16 Chapter 2

both theory and policy A correlation between gang involvement by younger offenders and subsequent officially recorded delinquent offenses can be interpreted to support a model of a single unified gang problem (trajectories with multiple geometric characteristics) varying across age ethnicity and community context Based upon what gang researchers have assumed so far some younger marginally involved gang youth become older ldquohard corerdquo gang members identified by the juvenile and criminal justice systems Others drop out of the gang and diminish involvement in delinquency (Decker Pyrooz and Moule 2014a Sweeten Pyrooz and Piquero 2013 Thornberry et al 1993) Still other youth might become involved in gangs at a later age (Pyrooz 2014) If the ldquoenhancementrdquo or ldquofacilitationrdquo models described by Thornberry et al (1993) is correct most youth would become involved in gangs prior to becoming involved in greater levels of delinquency This image of a compre-hensive unified gang problem in a community has framed contemporary strategies of responding to gang crime problems that link prevention intervention and suppression strategies under comprehensive programs Programs might need an age‐specific focus if uniform or different age trajectories for gang membership appear to be more ubiquitous than researchers (including me) have assumed (Pyrooz 2014)

The alternative possibility faced by researchers (Curry 2000 Katz and Jackson‐Jacobs 2004) just ten years ago that there might be no relationship between self‐reported gang involvement and officially recorded delinquency would have suggested that there are actually two parallel gang problems in US communities ndash one involving younger self‐identified gang members and their comparatively heavy involvement in minor juvenile offending and another population of older police‐identified gang members and their involvement in more serious offenses tracked by law enforcement agencies If there were two gang problems the first based on the behavior of juveniles would be the one studied by field researchers (Decker and Van Winkle 1996 Miller 2000) and survey researchers (Esbensen and Huizinga 1993 Thornberry et al 1993) The second gang problem based on the recording of law enforcement would be the one described by analyses of official records on gangs (Block and Block 1993 Maxson and Klein 1990) If this were true the disturbing reality would be that the former would have little bearing on the latter That is to the extent that gang research is based on juveniles it would contribute little knowledge to the policies and practices of the criminal justice system

The law enforcement perspective of gangs has been more subject to skepticism and criticism by gang researchers than self‐reported gang membership and observation by field observers Hagedorn (1988) notably calls those who use official data ldquocourthouse criminol-ogistsrdquo Thrasher (1927) observed that the relationship between police and gangs was confounded by enmity connivance and even corruption Since Thrasher other field researchers have similarly emphasized the role of local politics in shaping gang intelligence and record‐keeping on gangs in law enforcement agencies (Klein 1995 Spergel 1995) The National Youth Gang Survey (1999) cautions readers repeatedly that law enforcement reporting on gangs is subject to local political considerations in which gang problems can be exaggerated or denied

Survey researchers in examining the relationship between gang involvement and delinquency in the 1990s have used self‐report measures simply because these are the only measures available Fagan (1990 190) supported his use of a self‐report question on gang membership as preferable to official definitions ldquoOfficial (police and social agency) definitions and rosters of gang membership often vary by city and agencyrdquo Fagan noted ldquomore often reflecting the organization of social control agencies than empirical realities about gang mem-bership or gangsrdquo In Esbensen and Huizingarsquos view ldquoOfficial data provide rather subjective assessments of gang behaviorrdquo (1993 566) These are not words that inspire confidence