the localization of caste politics in uttar pradesh after .../2441/5gpts2mlf09rtau5...the...

Institut d'études politiques de Paris

ECOLE DOCTORALE DE SCIENCES PO

Programme doctoral Science politique, mention Asie

Centre d’Etudes et de Recherches Internationales (CERI)

PhD in Political Science, specialization Asia

The localization of caste politics in Uttar Pradesh after

Mandal and Mandir

Reconfiguration of identity politics and party-elite linkages

Gilles Verniers

Under the supervision of Christophe Jaffrelot, Directeur de recherche, CNRS-CERI

Defended on December 16, 2016

Jury: Mrs Mukulika Banerjee, Associate Professor, London School of Economics and Political Science (rapporteure) Mr Christophe Jaffrelot, Directeur de recherche, CNRS-CERI Mr James Manor, emeritus Professor of Commonwealth studies and Senior Research fellow, Institute of Commonwealth Studies Mr Philip K. Oldenburg, Adjunct Associate Professor and Research Scholar, South Asia Institute, Columbia University Mrs Stéphanie Tawa Lama-Rewal, Directrice de recherche, CNRS-EHESS, CEIAS (rapporteure)

1

TableofContents

ListofFiguresandMaps................................................................................................................3ListofTables......................................................................................................................................4Acknowledgments............................................................................................................................6Chapter1.Introduction..................................................................................................................81.1.Statementoftheproblem...............................................................................................................91.2.Findingsandmainarguments.....................................................................................................141.3.ApproachestoElectionsStudies.................................................................................................19

Casestudiesandecologicalanalysis...............................................................................................................19Surveyandquantitativemethods....................................................................................................................22Theanthropologyofdemocracy.......................................................................................................................24

1.4.Methodology......................................................................................................................................271.5.Literaturereview.............................................................................................................................30

Theearlyliterature:partypoliticsandthefragilityofCongress’dominance..............................31Identities,parties’recompositionandthestudyofagrarianchange...............................................34Newresearchdirectionsinthemid-1990s..................................................................................................37TheanthropologicalcritiqueandcontributiontotheUttarPradeshpoliticsliterature.........46TheU.P.Literatureinthe2000s:thetripletropesofviolence,identity,andpatronage.........48

1.6.Overviewofdissertation...............................................................................................................56Part.I–DecodingUttarPradeshPolitics..............................................................................592.1.ThefourphasesofUttarPradesh’spartysystem:fromdominancetofragmentation........................................................................................................................................................................62

2.1.1.Congressdominationandtheriseoftheopposition...................................................................632.1.2.Partingofalliancesandparties’recompositioninthe1980s.................................................742.1.3.CommunalandCastepolarizationanddysfunctionalcoalitionpoliticsinthe1990s..842.1.4.Thelimitsofcastepolarization:Post-identitypolitics?.............................................................92

2.2.Sub-regionaltrajectories...........................................................................................................1002.2.1.UttarPradesh’ssub-regions................................................................................................................101

Chapter3.ElectoralpoliticsinUttarPradesh:Therulesofthegame......................1183.1.HaveUttarPradeshelectionsbecomemorecompetitive?.............................................119

3.1.1.TurnoutinUttarPradeshelections,1962-2012.........................................................................1213.1.2.Effectivenumberofpartiesandcandidates.................................................................................1283.1.3.Thedisproportionalityeffectoftheelectoralsystem..............................................................1323.1.4.Marginsofvictory....................................................................................................................................134

3.2.Politicalconstraints.....................................................................................................................1363.2.1.TheroleofIncumbency.........................................................................................................................1373.2.2.Theroleofturncoats..............................................................................................................................1523.2.3.Competitionfortheticket:Thelongroadtotheelection......................................................1553.2.4.Thesizeofthestablepoliticalclass.................................................................................................1613.2.5.Politicalstrongholds...............................................................................................................................163

3.3.Conclusion.......................................................................................................................................169Part.IIThechangingprofileofUttarPradeshStateLegislators.................................171Chapter4.TheheterogeneisationofUttarPradesh’spoliticalclass.........................1714.1.TheEvolutionofcasterepresentation..................................................................................174

4.1.1.Thefourphasesofcastegrouprepresentation..........................................................................1764.1.2.Uneventrajectoriesamongtheuppercastes...............................................................................1784.1.3.AstabilizedOBCrepresentation........................................................................................................1794.1.4.ThesteadyriseofMuslims’representation.................................................................................184

2

4.1.5.Havepartiesactuallybecomeinclusive?.......................................................................................1914.1.6.Casterepresentationincabinets:persistenceofbiases.........................................................197

4.2.Sub-regionalvariations..............................................................................................................2034.2.1.WheretheSavarnashavefallen.........................................................................................................2044.2.2.WheretheSavarnashaveresisted....................................................................................................2084.2.3.TheWest:Free-forallcompetition..................................................................................................2114.2.4.Explainingvariations..............................................................................................................................212

4.3.Beyondcaste...................................................................................................................................2144.3.1.Education.....................................................................................................................................................2154.3.2.Theproblemofoccupation..................................................................................................................2164.4.Conclusion.......................................................................................................................................................222

Chapter5.Whowieldspower?LocalperspectivesonUttarPradeshelectoralpolitics.............................................................................................................................................2245.1.Thesourcesofpoliticalpower.................................................................................................224

5.1.1.ThecaseofWesternU.P.......................................................................................................................2265.1.2.ThecaseofEasternU.P.:morecontinuitiesthanchange.........................................................249

5.2.Whatdoelectedrepresentativesdo?Ajobdescription..................................................2595.2.1.Sevaasapoliticalandsocialobligation..........................................................................................2625.2.2.Oncandidate’seffectiveness...............................................................................................................2655.2.3.Thehighcostofentryintopolitics...................................................................................................268

5.2.DonorDayavan?Divergenttrajectoriesinthecriminalizationofpolitics...............2705.3.1.“Wehaveturnedcorporate”:thereconfigurationofcriminalorganizationsinWesternUttarPradesh.........................................................................................................................................................2725.3.2.TheresilienceofruralgangstersintheEast................................................................................279

5.4.Conclusion.......................................................................................................................................290Chapter6-Interpartycomparison.......................................................................................2936.1.Candidatesselectionprocess....................................................................................................294

6.1.1.TheCongressParty:stickingwiththedeadwood......................................................................2966.1.2.TheBJP:withinthefamily....................................................................................................................3016.1.3.TheBahujanSamajParty:outsourcingcandidates...................................................................3056.1.4.TheSamajwadiParty:factionalismandlocalelitesintegration..........................................312

6.2.Implications....................................................................................................................................3196.2.1.Theimportanceofpartyorganizations..........................................................................................3206.2.2.Whyareregionalpartiesdominating?...........................................................................................3216.2.3.Differentiatedparty-elitelinkages....................................................................................................3246.2.4.WhydoestheGoondaRajtagsticktotheSPandnottheBSP?............................................3266.2.5.Consequencesforgovernance............................................................................................................3296.2.6.Consequencesfordemocratization..................................................................................................334

Conclusion.....................................................................................................................................337Bibliography.................................................................................................................................342

3

ListofFiguresandMaps2.12.22.32.42.52.62.73.13.23.33.43.53.63.73.83.93.103.113.123.133.143.154.14.24.34.44.54.64.74.84.94.104.114.124.13

Mainparties’voteshareintheUttarPradeshStateAssembly,1962-2012Mainparties’seatshareintheUttarPradeshStateAssembly,1962-2012ImpositionofPresident'sRuleinIndia,1949-2015StatedistributionofPresident'sRule,1949-2015Sub-RegionsofUttarPradeshTotalnumberofHeavy/Smallscaleindustrialunits(2012)EmploymentinHeavy/Smallscaleindustrialunits(2012)Distributionofconstituency-wiseturnoutinstateassemblyelectionsinUttarPradesh(1962-2012)UttarPradeshAssemblyElections2002VoterTurnoutUttarPradeshAssemblyElections2007VoterTurnoutUttarPradeshAssemblyElections2012VoterTurnoutSub-regionalaverageturnout(1962-2012)Gender-basedparticipationinUttarPradeshstateassemblyelections(1962-2012)Constituency-levelnumberofcandidatesinUttarPradeshstateassemblyelections(1962-2012)RelativeweightofcandidatestypesNumberofpartiescontestingandrepresentedintheU.P.assembly(1962-2012)Effectivenumberofcandidatesinconstituenciesovertime(1962-2012)Disproportionalityofseatsandvoteshareinthe2007StateAssemblyElectionsDistributionofconstituency-levelaveragewinningthresholds(1962-2012)Distributionofconstituency-levelmarginsofvictorypercentagesinU.P.assemblyelectionsIndependentcandidatesratioFirst-timerunningIndependentwinnerCasteandCommunitiesintheUttarPradeshVidhanSabha(1969-2012)RepresentationofmainUppercastejatis(1969-2012)RepresentationofmainOBCandintermediarycastesjatis(1969-1980)CasterepresentationamongMuslimMLAs(1993-2012)CasteandcommunitiesrepresentationinDoabCasteandcommunitiesrepresentationinEastUttarPradeshCasteandcommunitiesrepresentationinBundelkhandCasteandcommunitiesrepresentationinRohilkhandCasteandcommunitiesrepresentationinAvadhCasteandcommunitiesrepresentationintheNortheastCasteandcommunitiesrepresentationinUttarakhandCasteandcommunitiesrepresentationinWesternUttarPradesh

4

ListofTables

Table2.1Table2.2Table2.3Table2.4Table2.5Table2.6Table2.7Table2.8Table2.9Table2.10Table2.11Table2.12Table2.13Table2.14Table3.1Table3.2Table3.3Table3.4Table3.5Table3.6Table3.7Table3.8Table3.9Table3.10Table3.11Table3.12Table3.13Table3.14Table3.15Table3.16Table3.17Table3.18Table3.19

PerformanceofsocialistPartiesandtheirsuccessorsinUttarPradesh,1952-2012CasteGroupRepresentationamongSocialistParties,1952-1962Performanceofmainpartiesinthe1980UttarPradeshassemblyelectionsCasteGroupRepresentationamongSocialistParties,1980-1991VoteshareandSeatdifferenceinthe1989and1991UttarPradeshassemblyelectionsCasteandCommunity-basedpartypreferencesinUttarPradeshassemblyelections,1996-2012DalitvotingproportionsbypartyinUttarPradeshassemblyelections,2002-2012CasteandCommunity-basedpartypreferencesinUttarPradeshassemblyelections,2007-2012Registeredvoterspopulationpersub-regionSub-regionwisepopulationandshareofMuslimpopulation,perlocalitySub-regionwisepopulationandshareofSCpopulation,perlocalityOccurrenceofseatsheldthreetimesandabovebythesamecasteinfiveUttarPradeshassemblyelections1993-2012RegionalvoteshareperformanceofmainpartiesinUttarPradeshassemblyelections,1962-2012RegionalseatshareperformanceofmainpartiesinUttarPradeshassemblyelections,1962-2012DecadalaverageturnoutinUttarPradeshstateelectionsDecadalaverageturnoutgapbetweenreservedandgeneralseatsinUttarPradeshstateelectionsDecadalaveragegenderturnoutgapinUttarPradeshstateelectionsDecadalaveragenumberofcandidatesperconstituencyDecadalaverageENOPDecadaldistributionofaveragewinningthresholdsinU.P.assemblyelectionsDecadalaveragemarginofvictory(inpercent)inU.P.assemblyelectionsParty-wiseaveragemarginofvictory(inpercent),1989-2012Patternofproandanti-incumbencyinUttarPradeshstateelectionsPercentageofseatschanginghandsinUttarPradeshassemblyelections(1980-2007)Seatretentioncapacityofpartiesbetweenthe1996and2002assemblyelectionsinUttarPradeshSeatretentioncapacityofpartiesbetweenthe2002and2007assemblyelectionsinUttarPradeshSeatvolatilityinAmethiandRaeBareliLokSabhaconstituencies,2002-2012SampleofindividualincumbencycodingIndividualincumbencyinUttarPradeshstateelections(1951-2012)SamajwadiPartyre-runningincumbentMainpartiesre-runningincumbentsStrikerateofincumbentcandidatesandex-MLAsinUPassemblyelectionsPercentageofMLAswhochangedpartyaffiliationduringtheircareer,1952-1012

5

Table3.20Table3.21Table3.22Table3.23Table3.24Table4.1Table4.2Table4.3Table4.4Table4.5Table4.6Table4.7Table4.8Table4.9Table4.10Table4.11Table4.12Table4.13Table4.14Table4.15Table4.16Table4.17

TurncoatsfieldedbymainpartiesinUttarPradeshassemblyelectionsandtheirperformance(1980-2012)NumberofMLAs’contestsbeforeelectioninUttarPradeshstateassemblyelections(1952-2012)DecadalestimatesizeofthestablepoliticalclassParty-wisebreak-upofthestablepoliticalclass,1980-2012Decadalpartydistributionandratioof‘stronghold’seatsCastesandCommunitiesinUttarPradesh,1931(in%)CasteandcommunitycompositionoftheU.P.VidhanSabha,1969-2012(%)Muslimcandidatesnominatedbymainpartiesinthe2012StateElectionRepresentationofMuslimsintheUttarPradeshLegislativeAssembly,1962-2012SeatspossiblylostduetosplitMuslimvote(1996)SeatspossiblylostduetosplitMuslimvote(2002)SeatspossiblylostduetosplitMuslimvote(2007)Muslimvoters'partypreferencesinfourstateelectionsCasterepresentationamongamongBSPDalitMLAs(1989-2012)Representationofmajorcastegroupswithinmainparties(1989-2012)Castesandcommunitiesamongmainpartiescandidatesinthe2012U.P.StateelectionsCastesandcommunitiesinU.P.CabinetsPortfolioconcentrationinU.P.Cabinets,1980-2012EducationofU.P.MLAsProfessionofMLAsinthe2012U.P.Assembly,perpartyProfessionofMLAsinthe2012U.P.Assembly,percastegroupProfessionofMLAsinthe2012U.P.Assembly,persub-region

6

Acknowledgments

ThisdissertationhasbeensolonginthemakingthatIoweitscompletiontothepatienceofmyadvisor,ChristopheJaffrelot,whohasextendedunflinchingsupportandhasbeenaconstantsourceofnewideasandprojects. Ihavelearnedfromhiminmanymorewaysthanoneusuallyexpectsfromathesissupervisorandregardhimasamentor.

Throughalltheseyears,Ihavebenefitedfromthesupportandguidanceofmanypeople,allofwhomcannotpossiblybeacknowledgedhere.Iamparticularlygratefultoanumberofcolleaguesandfriendswhohavesharedtheirinsights,advice,criticismonthisprojecton countless occasions. In particular, Francesca Jensenius, Neelanjan Sircar, AdnanNasimullah, Susan Osterman, Vasundhara Sirnate and Rahul Verma have considerablyhelpedme,directlyandindirectly,throughformalandinformalconversations,todevelopmyabilitiesasaresearcheraswellasmyaptitudetoengagecriticallywithourdiscipline.

IamdeeplygratefultoPradeepChhibberforhostingmeonvariousoccasionsatBerkeleyandmoreparticularlyforincludingmeinsomeofhishisweeklycolloquiums;someofitsmembershavebecomeverydearfriends.I thank Karthika Sasikumar, Matija Cuk, Susan Osterman and Stu for their Californianhospitalityandfortheirfriendship.I owe a special debt to Philip Oldenburg, who kindly and patiently perused everyparagraphof thisdissertationandprovided –always in the timeliestmanner– criticalcomments and suggestions that helped me to improve the quality and depth of thisdissertation. His early encouragements in the phase of writing also contributed toattenuatingtheanxietytraditionallyattachedtothisexercise.

PaulBrassandhiscompanionSueopenedtheirhomeformeinAcme,WashingtonState.Paulalsoopenedthetreasuretroveofhispersonalarchivesandhisprodigiousmemory.IhavelearnedconsiderablyaboutUttarPradeshpolitics intheshortspanoftimeIspentwiththemattheirhomeandduringourboutoffieldworkinWesternU.P.

Partsofthechaptersandmaterialthatwentintothisdissertationhavebeenpresentedinvarious seminars in New Haven, New York, London, Paris, Berkeley, and Delhi. I amgratefulforthecommentsandcriticismsofparticipantsoftheseseminars–inparticularthoseoftheEECURInetwork.Thesemademethinkandrethinkmyarguments:KanchanChandra, Virginie Dutoya, Zoya Hasan, Oliver Heath, Pralay Kanungo, Sudipto Kaviraj,Sanjay Kumar, Satendra Kumar, Loraine Kennedy, Jim Manor, Partha Mukhopadhyay,Sudha Pai, Manisha Priyam, Stephanie Tawa-Lama Rewal, Louise Tillin, and AdamZiegfeld.IamespeciallygratefultoMukulikaBanerjeeandLuciaMicheluttiforopeningmyeyestothevalueandbenefitsoftheanthropologicalapproachtopoliticsanddemocracy.Also,IwouldnothavebeenabletospendelevenyearsinIndiaandgaintheexperienceIgatheredfromthegroundwithoutthesupportofinstitutionsandpeoplethatgavemetheopportunitytoexpandmyknowledgeandintellectualcuriosityfarbeyondtheboundariesofmydiscipline.

7

TheeightyearsIdedicatedtoSciencesPoandthedevelopmentofitscooperationinIndiaenabled me to travel across the country, develop friendships and make professionalcontacts that ultimately helped me acquire a broader and deeper understanding ofcontemporary India. Iamgrateful toFrancisVerillaudandAlessiaLefebure forofferingme the opportunity to move to Delhi in 2005, and for their constant support andencouragement.

The bulk of the research and fieldwork that went into this dissertation was fundedthroughtheCSH,withwhichIhavebeenaffiliatedforsevenyears,andthroughvariousresearchprojectsconnectedtotheCentre,notablytheEECURIproject.Iamgratefultoitssuccessivedirectors,VeroniqueDupontandBasudebChaudhuri,fortheirsupport.IoweaparticulardebttoMushirulHasanwho,throughhis intervention,enabledmetoremaininIndiaagainstvariousoddstheyearafterItaughtatJamia.Iamalsogratefultomy colleagues at Jamia and inparticular toAmeenaQazi,who sharedher contacts andconnectionsinEasternUttarPradesh.

More recently, I have also found considerable support and encouragement in my newhome, Ashoka University. Being associated with the birth of a new university is a rarechanceandaprivilege. Iamgrateful toPramathRajSinha forsharinghisvision for thecreation of a unique institution and for expressing his affection and care by applyinggentlepressureswhenIneededitthemost.

IamalsothankfultoAshoka’sVice-Chancellor,RudrangshuMukherjee,andtotheDeanofAcademic Affairs, Jonathan Gil Harris, for helping me complete this dissertation bygrantingmeanindispensablesabbaticalleave.TwoofmyformerstudentsatAshokahaveplayedacrucialroleinhelpingmesetuptheTrivediCentreforPoliticalData.VenkatPrasathandRajkamalSinghdedicatedtheirheartandsoultotheCentreandhaveoftenbeenthehaplessrecipientsofmylatenightideas.Overthepasttwoyears,IhavelearnedconsiderablyfromRajkamalaboutUttarPradeshpolitics, thanks to his ground experience and his extraordinarily acute sense ofobservation.IthankShivamVijformaintainingtheoldjournalist-academicsfeudaliveandforhavingalwaysbeengenerouswithhisinsightsfromthefield.InLucknow,HaiderAbbasRizvihasbeen a great fieldwork companion and I am grateful to him for introducing me to theintriguingalleysofLucknow.I was introduced to Lakhnawi tehzeeb by the late Ram Advani, whose bookshop inHazratganj has been an abode of peace nestling in the clamour of U.P.’s tumultuousenvironment.I am thankful to themany anonymous sources,members of Parliament and LegislativeAssembly, party workers and local journalists who shared their insights, anguish,ambitionsandexperiences.And finally, I expressmydeepest gratitude tomywife, Sharmila,whohas enduredmyprolonged absences, my anxieties and short temper throughout the completion of thiswork.IdonotknowifIcanthankherenough.

8

Chapter1.Introduction

On March 15 of the year 2012, Akhilesh Yadav, son of former Chief Minister and

SamajwadiParty founder andPresidentMulayamSinghYadav, becameUttarPradesh’s

twentieth Chief Minister, with a single majority of seats in the State Assembly. He

succeeded the leaderof theBahujanSamajParty,Mayawati,who fiveyearsearlieralso

came to power on her own, securing a narrowmajority of seats in the StateAssembly

with30.4%ofvoteshare.Thesetwoelectionswereremarkableforaseriesofreasons.

First,theymarkedtheendofalongperiodofgovernmentalinstabilityinUttarPradesh,

caused by the fragmentation of the electorate and the state’s party system, and by the

inabilityofpartiestoworktogetherinalliancesorcoalitions.Nopartyhadwonasingle

majorityofseatssince1985.

Second, these majorities were obtained through mobilization strategies that in

appearance and discourse transcended traditional caste affiliations or antagonisms,

through campaigns that focused on programmatic and general interest issues. This

contrasted again with the preceding decades, marked by deep caste antagonisms as

partiessoughttomobilizetheirrespectivecoresupportbases.

Third,thepartiesresponsiblefortheseinclusivestrategiesarepreciselythoseoftenheld

responsibleforthefragmentationofthepoliticalspace,throughnarrowcaste-basedparty

politicsandmobilization.

Fordecades, thestateofUttarPradeshhasbeensynonymouswithwhatmanyconsider

the ailments of India’s democracy: fragmentation, caste-entrenched politics, the

criminalization of the political class, poverty and violence. Many of the stereotypes

associatedwithIndianpoliticsatlargecomefromdepictionsofUttarPradeshpolitics,a

statethatweighsconsiderablyonnationalpolitics,owingtoitsdemographicstrength,its

representation in national assemblies1, and its historical role as the cradle of India’s

liberationmovement(Kudaisya2006).

1TheStateofUttarPradeshcurrentlysends80representativestotheLokSabha(lowerHouseofParliament),and31representativestotheRajyaSabha(theUpperHouse),.Eightoutofatotalof

9

Throughthe1990s, theStateofUttarPradeshembodiedthestoryofchronic instability

thatalsomarkednationalpolitics. In less thantenyears, fourstateelectionswereheld.

Eightgovernmentsunsuccessfullyattemptedtorulethestate,attimesincoalitionsoras

minority governments. President’s Rule had to be declared on three occasions due to

hungverdictsor the inabilityofpolitical leaders to formcoalitions,orgovern together.

The period was also marked by a context of social and political violence, economic

slowdown(Singh2009),a sharpreductionofpublicexpenditures (Kohli2012)and the

sustained fall of the state’s ranking in national socio-economic development indicators.

Uttar Pradesh ranks among the lowest states in India on all human development

indicators,beittheinfantandchildmortalityrate2,thesexratio3, illiteracy4,orpoverty

andpovertyreductionratio5(Mehrotra2011).Thepartiesinpowerandtheirleadersare

oftenseenasanaggravatingfactor,ifnotacausalfactor,ofthisdiresituation.

1.1.Statementoftheproblem

This dissertation proposes to examine the continuities and discontinuities in electoral

politicsinthestateofUttarPradesh,intheperiodthatfollowedwhatiscommonlycalled

theMandalandMandirphases.Theearly1990srepresentedaturningpointinNorthern

Indian politics, with the rise of backward political forces, riding on the demand for

reservations in public jobs and higher education institutions for the Other Backward

Classes(OBC),withtheriseofaDalitparty–theBahujanSamajParty–bornfromalow-

castecivilservants’union,andwiththeascensiontopoweroftheBharatiyaJanataParty

(BJP),aHindunationalistpartythatrodeonawaveofreligiousmobilizationandviolence,

whichculminatedinthedestructionoftheBabriMasjid inDecember1992, inAyodhya.

The period also corresponds with the beginning of the liberalization of the Indian

economy,withtheintroductionofthefirstwaveofeconomicreformsinJune1991.

fifteenPrimeMinistershavecomefromUttarPradesh.ThecurrentPrimeMinister,NarendraModi,thoughfromGujaratandhadalsowoninitsVadoraraconstituency,ultimatelyretainedVaranasi,inUttarPradesh,ashisconstituency.250perthousand,accordingtotheerstwhilePlanningCommissionofIndia,in2013.3888womenperthousandmen,accordingtotheerstwhilePlanningCommissionofIndia,in2011.470%in2011,fourpointsbelowthenationalaverage,witha20percentagegapbetweenmenandwomen.537.7percentofthepopulationwasestimatedtobeunderthepovertylinebythePlanningCommissionin2011.

10

ThisparticularmomentinIndia’shistoryalsocorrespondswiththeculminationofdeep-

rooted processes of political transformation, marked by the mobilization of lower and

backwardcastes,thefragmentationofthepartysystem,thedeclineoftheCongressParty

andtheriseofregionalpoliticalforces.Bothinacademicliteratureandpopulardiscourse,

anarrativeemerged,equatingtheriseofthebackwardswiththeriseofregionalparties,

andthedeclineoftheCongresswiththedeclineofthetraditionalupper-caste-dominated

social order (Hasan 1993, 1998, Jaffrelot 2003a). Thus, Uttar Pradesh has been and

remainsaprivilegedgroundforthosewhostudyprocessesofdemocratizationinIndia.

Thishasbeenlargelyunderstoodandanalysedasaprocessofpoliticalempowermentof

social groups through political representation. Groups that were previously excluded

from and/or under-represented in the political sphere gradually supported political

partiesoftheirownwiththeaimofobtainingbothafairshareofrepresentationaswell

asaccesstopublicresources.Inthisprocess–characterisedbyRajniKothariin1990as

the“greatsecularupsurge”(Kothari1990)andsomeyears later,byYogendraYadav,as

the “second democratic upsurge” (Yadav 2000) – caste has been seen as the favoured

vehicleofpoliticalmobilisation.

Today, twonarrativesdominate the characterizationofUttarPradeshpolitics.The first

one, essentially journalistic, consists on underscoring of the prevalence and

predominance of caste as a factor shaping electoral outcome, be it within parties’

strategies or voters’ motivations. This narrative is sustained by the continued

performanceof so-called caste-based regionalparties andby theprevalenceof caste in

the imaginary of electoral politics in this state. The anthropological literature on Uttar

Pradesh politics, in particular, insists on the prevalence of caste (Jeffrey 2001, 2002,

Michelutti2007).

Thesecondnarrative,ontheotherhand, focusesonthecapacityofdominantpartiesto

mobilizebeyondtheirtraditionalsupportbase–YadavsandMuslimsfortheSamajwadi

Party,Dalits fortheBSP,uppercastes fortheBJP–andgathersupportacrossthecaste

spectrum(GuptaandKumar2007,Pai2013,Verma2007b,Verma2014b).Thisisoften

seen and interpreted as a sign of maturing of the electorate, driven more by issues or

economic voting rather than by ascriptive identities. A close look at recent political

11

transformationsshowsthatthesituationismorecomplexthanitappearsandthatboth

narrativessufferfromseriouslimitations.

It is true that theparties thatarecurrentlydominating thepolitical sceneareprecisely

thosethathavesucceededinopeningtheirdoorstocandidatesbelongingtogroupsthey

mighthave initially opposed.TheBSP’s2007victory is largely credited to its ability to

attractasignificantportionof theBrahminvote. In2012, theSamajwadidistributed its

ticketsacrossthecastespectrum,atthecostofdilutingitsOBCbasebutwiththereward

of gaining a majority of seats in the state assembly. The parties that rose through the

processoftheelectorate’sfragmentationanddivisivecaste-basedelectoralstrategiesare

slowlytransformingthemselvesintocatch-allparties.

Asaconsequence,theoverallshareofOBCrepresentationintheStateAssemblyhasbeen

on thedeclineafterapeak in19936.Thedeclineof theuppercastes,noted inprevious

contributions(Hasan1998,Zerinini2009),hasstabilizedandtheshareofrepresentation

of upper-caste candidates and representatives within the so-called low-caste or

backward-castepartieshasbeenontherise.Inotherwords,thelinkbetweentheriseof

backwardclassesandtheriseoftheso-calledbackwardpartiesisnotsostraightforward.

Thisevolutiondoesnotmean,however,thatthetiesbetweencasteandelectoralpolitics

have ruptured. Extensive fieldwork conducted during and between two state elections

(2007 and 2012) revealed that caste remains a central variable in defining parties’

strategicchoice,lessintermsofpoliticaldiscoursethaninpoliticalpracticesatthetimeof

elections.Itremainsamajorfactorintheselectionofcandidatesandpartiesdotakeinto

account the localbalanceofpowerandtherelativenumbersofeachcastegroupas the

chiefamongothervariables.Theemphasisoncastebypartieshas,inturn,animpacton

voters’choice,sinceparties’decisionsdetermineorshapethe‘supply’ofcandidates,thus

creatingornegatingopportunitiesforvotersto‘votetheircaste’.

Whatmattersmorethanideologyorparties’identityinclinationsiscontext.Sub-regional

trends inthesocialcompositionoftheStateAssemblyrevealsubstantialvariationsthat

challengeanydiscourseorintellectualconstructionbasedonaggregatepoliticaltrends.In

6TheshareofrepresentationofOBCsintheStateAssemblydippedfrom35.4%oftheseatsin1993to25%in2012.

12

severalimportantsub-regionsofthestate,thedominationofuppercasteshasremained

fairly unchallenged even as the number of low-caste and backward-caste parties rose.

These sub-regions tend to be the areas that have experienced the least economic

transformationordiversificationoftheirruraleconomy,suchasCentralUP(Awadh),and

theNortheast.Othersub-regionsintheWest(WesternUP,Rohilkhand,Doab),thathave

seenmoreeconomicchange,havealsoexperiencedmorepronouncedpoliticalchurning

inrecentyears7.

This shows theneed to contextualize the relationshipbetweencasteandpolitics at the

right level,whichcan’tbeanaggregateone.A largepartof the literatureon identityor

caste politics focuses on the caste variable of MPs and MLAs alone, disconnected from

othervariablesconstitutingtheirsociologicalprofile.Thishasbeenitsmainlimitation.

The primary aim of this dissertation is to examine what has happened to caste – as a

vehicleofpoliticalmobilization–overthepasttwenty-fiveyears,andnotablyhowdoes

casterelatesto–andindeedcompetes–withothertropesofmobilizationsuchasreligion

andclass.Whatbecomesofbackwardpoliticswhenso-calledbackwardpartiesopentheir

doors to non-backward individuals and groups? Is the case for the newfound

inclusivenessofpartiescompelling,exaggerated,ordisingenuous?Shouldn’tweintegrate

the caste factor with other relevant variables, such as the economic background of

individualscontestingelections?

Backward politics and the often concomitant rise of regional parties is one of India’s

majorpost-Independencepolitical event, adeep-rootedprocessof fragmentationof the

electoratealongcasteandcommunitylinesthatsoughttoopposethedominationofthe

CongressPartyfromtheoutsideatboththestateandatthecentrallevels.Theliterature

ontheriseofregionalpartieshasfocusedessentiallyontheidentitydynamicsatwork–

castemobilization,contestationoftraditionalsocialordersandtraditionalelites–andon

theimpactofpre-liberalizationeconomictransformation,suchastheZamindariAbolition

7Thereareexamplesofcross-statecomparisons(Jenkins2004,Pai2000b)butfewhaveexaminedintra-state variations. This is important since the variations that may be observed within stateswhosepopulationsoftencomparetothesizeof largedemocraticcountriesmayarguablybesuchthatthenarrativeoftheirtrajectorystandsdefeated,orinneedofseriousnuancingoramending.

13

Act or the Green Revolution. Few have examined the transformation of politics after

MandalandMandirfromthevantagepointofpoliticalactors8.

In order to do so, I examine the evolving sociological profile of candidates and elected

representatives in recent elections. I also attempt to ‘connect’ the caste variable with

other socio-demographic and socio-economicvariables collected through fieldworkand

interviewsover the years, to examine the claim thatpolitics inUttarPradeshhas gone

‘beyondcaste’.

Thisanalysisneedstobecontextualizedwiththeevolutionofparties’electoralstrategies

andrepresentationaloutcomes.Iexaminetheevolutionofvariouspoliticaltrendsatthe

stateandsub-regionallevels,onthebasisofauniquedatasetcombiningelectionresults

andsocio-demographicvariablesonelectedrepresentatives (from1962 to thepresent)

and on candidates (from 1991 to the present). From this empirical base, I study and

comparethetrajectoryofUttarPradesh’mainparties,thosewhohavedominatedorare

dominatingthestate’spoliticalsceneovertheperiodconsideredinthisproject.

Finally, the question of the transformation of the sociological profile of candidates and

elected representatives need to be contextualized at the level of local contests, i.e. the

constituencylevel,wheretheconstraintsofcompetitivelocalpoliticsweighthemoston

thecandidates’shoulders.

Thecombinationof thesebroadresearchdirectionswillenablemetodemonstrate that

the growing inclusive character of caste-based parties does not mean that caste has

receded,orceasedtomatter,asavehicleofpoliticalmobilization,butthattothecontrary,

inclusivenessisbuiltthroughthelocalizationofcastemobilizationstrategies,withparties

distributing tickets and candidates forging local alliances according to local caste

circumstances.

8OneofthemostrecentandremarkablecontributioninthissubjectisJeffreyWitsoe’sCasteVersusDevelopment,abouttheoperationsoflower-castepoliticsatthelocalandsub-regionallevelinBihar(Witsoe2013).

14

1.2.Findingsandmainarguments

Ifindthatstate-levelnarrativesoncaste–orhorizontalformsofcaste-basemobilization

– todayoperate less to thebenefitof localarrangements,negotiations,andtransactions

between groups and individuals embedded in specific socio-economic contexts. In the

process, castebecomes further entrenched in electoralpoliticsbut in a less transversal

manner.Ifindthatpartiesseektoforgelocalalliances–inwhichcasteplaysacrucialrole

–whilemobilizingacrossconstituenciesonthebasesofgeneralisttropescuttingacross

casteorsectariandivisions,suchasdevelopmentorabroaddefinitionofequityorsocial

justice.

Further, I also find that while caste remains deeply entrenched in local political

competition, itdoesnotoperateasan isolated factor.The fieldwork Iconductedduring

and between the 2007, 2009 and 2012 elections in various parts of the state and the

prosopographyofcandidatesovertheperiodrevealthatwhilecastecontinuestomatter

locally, it does so in connection with other socio-economic variables, such as the

inscriptionofpartyorganizationsandcandidatesinlocalsocio-economicnetworks.More

specifically,Ifindthatstate-basedpartiestendtorecruittheircandidatesamonggroups

andindividualswhocontrolorexertaninfluenceoverthelocalpoliticaleconomyoftheir

constituencies or on a larger scale, contributing to the integration of local political and

economicelites.

Thisisrevealedbythefactthatwhilethereisthroughtimeagradualheterogeneisationof

representationon thebasisofcasteover time– throughtheassertionofbackwardand

lower-caste groups – there is also a concomitant process of homogenization of

representationonthebasisofclasswithinthemaincontestingparties.Aspartiesturnto

local elites, they recruit more candidates hailing from a local business or industry

background,andlessfromprofessionsthatusedtobeover-representedintheAssembly,

suchas farmersor liberalprofessions.Furthermore, I find that the twodominantstate-

basedparties–theSamajwadiPartyandtheBahujanSamajParty-tendtorecruittheir

candidatesfromthesamesociologicalpooloflocalelitegroupsandindividuals.

Mymainexplanationforthesedevelopmentsistwofold.First,politicalcompetitioncomes

withasetofconstraints,orrules–bothformalandinformal–thatweighonpartiesand

15

candidates9. In order to be successful, candidates need resources, a strong party ticket

andthecapacitytomobilizenumbersbothwithinandoutsidetheircasteorcommunity.

Some of the constraints of political life in Uttar Pradesh tend to filter out aspiring

candidates at the entry-level - candidates who cannot afford the cost of entry into the

electoral fray–andat theexit– themany incumbentMLAswho fail tobeelected fora

secondorthirdterm,orarenotevengiventhechancetore-runinthefirstplace.

Thesecondexplanationisthatpartieshaveadaptedthemselvestotheseconstraints–by

picking “winnable” candidates fromcertainbackgroundsand, for instance,by requiring

candidatestofundtheirowncampaigns.Indoingso,theyhavecontributedtoincreasing

thesystemicconstraintsthatweighoncandidates.

In short, the constraints of electoral politics in Uttar Pradesh – a high cost of entry, a

competitivepoliticalarenaandshortpoliticallifeexpectancy–allservetoaffectonwho

aspirestocontest,whogetschosenbypartiestorun,whorunssuccessfully,andwhomay

last in politics for more than a term or two. Parties and the set of electoral rules and

practicesunderwhichtheyoperatecreatebothincentivesandadvantagesforcandidates

from certain socio-economic backgrounds, shaping in turn the sociological profile of

electedassemblies.

Theideathatcasteisenmeshedwitheconomicconsiderationsisnotanewidea.Scholars

of the Congress era have shown how Congress candidates tended to win thanks to a

combination of high-caste status and land ownership (Brass 1964b, 1980a, 1984b,

Weiner1967)10.As thestate’seconomychanged– inparticular theruraleconomy–so

didthesocialandeconomicbasesuponwhichpoliticalpowerrestsorfromwhichitcan

bederived.Landnolongermattersthewayitdid.Theinscriptionofcandidatesandlocal

party organizations within local networks that control or have influence over local

economic institutions is critical to their chances of success in the political arena. The

9Borrowed fromF.G.Bailey’s expression, fromStratagemsandSpoils, inwhichhe states that thearenaofpoliticalcompetitionisdefinedbyasetofrules–legal,customaryorconventional–thatweighonpoliticalactors,whomustabidebythoserulesinordertobecompetitive(Bailey1969).

10ThesetwoauthorsalsonotethatthereweremorefactorstotheCongress’supremacythanthesetwo factors. The legacy of being the party that led India to Independence and the leadership ofJawaharlalNehrualsoplayedmajorpartinthecontinuingsuccessoftheCongressParty.

16

changingprofileofmembersoftheLegislativeAssembly–moreheterogeneousinterms

of caste but more homogeneous in terms of class – is an indicator of these

transformations.

The story that unfolds is that after a phase of silent revolution, which saw a gradual

transfer of power from the upper caste elites to various subaltern groups (Jaffrelot

2003b),politicsinUttarPradeshisnowbeingdominatedbylocalsocio-economicelites,

endowed with the attributes that help winning elections. The main difference with the

pastisthatthesocialidentityoftheselocalsocio-economicelitestendstocutacrosscaste,

evenifweseearesurgenceofupper-casterepresentationinrecentyears.

This leadsme to a reflection on the ultimate aim or purpose of state electoral politics,

fromthevantagepointsofpoliticalactors.Itisusuallyassumedthatrepresentationisthe

aimofpoliticalmobilizations,thewilltoobtaina‘fairshare’ofrepresentation,accessto

public goods and political influence. The literature on Dalit mobilization in particular

stresses on the emancipatory nature of caste politicization, electoral mobilization and,

ultimately,theacquisitionofpower.

Butviewedfromthepoliticalplayers’perspective,thepicturechangesasthepurposeof

electoralcompetition tends tohave less todowithrepresentationandmore todowith

exertingcontroloverindividuals,groupsandterritories.Fromthevantagepointoflocal

elites, the aim of electoral politics is territorial control. Democratic participation,

representation, party politics are tools that enable them to retain, develop and defend

theirlocalstatus,socialpositionandprivileges.

Onecouldarguethatpoliticalinstitutionsareboundtobecapturedbysomeformofelite

–theCongresssystembeinganarchetypalexampleofthisphenomenon.Butthereareat

leasttwoimportantdifferencesinthecurrentconfiguration,comparedtothepast.

The first is that these ‘newelites’aredrawn from farmorediversegroups thanbefore.

Theyarenotconfinedtoaspecifictypeofcastes,evenifsomecongruencebetweensome

castesandthelocalelitesremains.

17

Theseconddifferenceisthatthesenewelitesarenotalignedwithspecificparties.Infact,

theyhavebeenknowntoshifttheirallegiancebetweenpartieseasilyandmakeforacut-

throatpoliticalstage.

StateelectionsinUttarPradesharehighlycompetitive.Thiscompetitivenessisnotonly

reflected in the alternating governments– no party has succeeded in winning two

consecutiveelectionssince1989–butalsothehighturnoverofrepresentatives ineach

election(anaverageof51%turnoveroverthesameperiod,withamarkedincreaseinthe

lasttwoelections)aswellasintensevyingwithinpartiesandsocialgroupsforpositions

of power. The increase in the number of aspiring candidates, the phenomenon of pre-

electoral competition, parties’ practices for candidate nomination, and, at times, the

auctioningofticketshavealsocreatedconditionsfortheemergenceofapoliticalmarket

whereinpartiespickcandidatesmainlyaccordingtotheircasteandeconomicprofiles.

Inthisstoryofpoliticaltransformation,twopartiesstandout:theSamajwadiParty(SP)

andtheBahujanSamajParty(BSP).FoundedinOctober1992,theSamajwadiPartyisthe

largest single party in Uttar Pradesh and heir to the state’s socialist tradition, whose

exemplarsareRamManoharLohiaandChaudharyCharanSingh.Underthestewardship

of its founder and leader,MulayamSinghYadav, theparty evolved fromabroad-based

socialist formation into a caste-based party known to represent and champion the

interestsofaparticulargroup,theYadavs.

TheSamajwadiPartyemergedintheearly1990sasthemainbeneficiaryoftheMandal

mobilization, succeeding in rallying its Yadav base and attracting the support of voters

whosoughttodislodgetheCongressPartyandotheranti-reservationpoliticalforces.At

the same time, it also emerged as the defender of the state’s Muslim minority. In the

contextoftheBabriMasjiddemolitionandtheriseoftheBJP,MuslimsturnedtotheSP

forprotection,whichnootherpartywaseitherwillingorabletoprovide.Theallianceof

backwardclassesandMuslimsprovedformidableandenabledthepartytorulethestate

onthreeoccasions,from1993to1995,from2002to2007,andsinceMarch2012tothe

present. As of March 2016, the Samajwadi Party has been in power for a total of nine

years11.

11Orabovetenyears,ifoneincludesthe1989JanataDalgovernment,ledbyMulayamSinghYadav.

18

TheSamajwadiPartyalsoembodies thebrandofmuscularpolitics characteristicof the

region.ItsleaderMulayamSinghYadavliterallystartedhiscareerfromthewrestlingpits

ofEtahdistrict12.Similartootherstate-basedpartiesinotherpartsofthecountry13,the

SP forged an image of itself as a party of action, capable of ‘getting things done’, even

throughviolentmeanswhennecessary.

In popular view, the Samajwadi Party embodies everything that ails the state of Uttar

Pradesh.Itisusuallythefirstpartyquotedinanyconversationaboutthecriminalization

of politics, as their rule is associated with disturbed law and order, arbitrariness,

unsavory political figures and proximity with criminal elements. Their rough political

style,withatastefortherusticandlocalidioms,iseasilyshunnedbytheurbaneliteswho

see in the Samajwadi Party an incarnation of their nightmare of a polity dominated by

plebeians.

TheotherpartythatstandsoutistheBahujanSamajParty,apartycreatedbyKanshiRam

in 1984 and meant to be the vehicle and instrument of political empowerment of the

state’smostunderprivilegedsocialgroups,theDalits.Thepartygrewbyconsolidatingits

supportamongDalitsthroughafierycaste-basedrhetoricandthedenunciationofsocial

injustices. It scaled the zenith of power by forging alliances with parties and including

candidateswhobelongedtotheverygroupsitdenounced,atleast, initsinitialphaseof

ascension.TheBSPnowrecruitsitscandidateswithinthesameelitepoolofitsmainrival,

theSamajwadiParty,thoughwithsomemarkeddifferences.

Both parties have benefited froma series of transformations that have occurred in the

post-liberalization period: the diversification of the rural economy, the penetration of

marketforcesinruralareas,urbanization,thedevelopmentoflocalindustries,aswellas

some amount of social mobility among subaltern groups (Kapur et al. 2010). These

transformationshavenotonlycontributedtotheemergenceofnewelitesbuthavealso

transformed the economic base fromwhichpolitical power canbe derived. Thosewho

controllocaleconomiccapitalcanpotentiallyderivepoliticalcapitalfromit.Thesuccess

oftheSamajwadiPartyliesprincipallyinitsabilitytoco-optsuchindividualsandgroups

12HewasinductedintopoliticsbyNatthuSingh,acloseaideofChaudharyCharanSingh,whowaslookingforastrongmancapableofreinvigoratingtheSamyuktaSocialistPartyorganization.13See(Hansen2001)foraremarkablestudyofthedevelopmentoftheShivSenainMumbai.

19

who derive political influence from their social position and economic assets. These

individualsandgroupsoftenseek toexpandthesocialandeconomiccontrol theyexert

over certain territories through participation in democratic politics. By selecting their

candidates among them, the Samajwadi Party and the Bahujan Samaj Party have

contributed over the past two decades to a process of integrating local political and

economicelites.

Over thepast twenty-fiveyears,bothnationalparties– theCongressand theBharatiya

Janata Party (BJP) – have markedly declined. The Congress ceased to be a relevant

political force fromthemid-1990sonwardsandtheBJPhas fallenbackon itserstwhile

urbanstrongholds,afteraperiodofstrengthinthe1990s.Bothpartieshavesufferedin

the 2000s from a disconnect with the new elites of the state, and have retained a

pronouncedupper-castebiasintheirorganization.

1.3.ApproachestoElectionsStudies

ThereisalongtraditionofstudiesofelectionsandelectoralpoliticsinIndia14.Insteadof

presenting a comprehensive overview of the evolution of the discipline, I will briefly

present the main approaches that have been used to study electoral politics in India,

namely: the case study method, surveys, quantitative methods, mixed methods and

politicalanthropology.

Casestudiesandecologicalanalysis

Earlier studies of Indian elections originally consisted of local field studies, mostly

conductedbyAmerican scholars and scholars from theUniversityofDelhi.Thesewere

local-level accounts of elections, or the contextualization of general or parliamentary

electionsonthescaleofalocalityoraconstituency.Fifteenoftheseoriginalfieldstudies,

conductedduringthe1967GeneralElectionsandthe1971Parliamentaryelections,were

compiledin2007byA.M.Shah(SrinivasandShah2007).Thesefieldstudiesfocusedon

14Summariesof election studies canbe foundwithKondo (2007),Palshikar (2007), Lama-Rewal(2009)andKumarandRai(2013).AbroaderviewontheintellectualhistoryofthestudyofIndianpoliticscanbefoundintheOxfordCompaniontoIndianPolitics(RudolphandRudolph2010).

20

narratives on the conduct of elections and on the socio-political history of these

constituencies, social divisions and factional politics, underlining the necessity to

understandthebroadercontextinwhichelectionstakeplace.Theoutlookwasverymuch

sociologicalandanthropological,relyingoninterviewsandparticipantobservation.

This approach was criticized for its localism and for its failure to provide broad

explanationstopoliticaltransformations(Palshikar2007).Fromthere,theliteraturetook

two distinct directions. Some scholars pursued the case study approach by giving it a

comparativeturnwhileothersturnedtosurveymethods.

MyronWeinerdevelopedacasestudymethodologyinhisworkontheCongressPartyin

fivedistricts(Weiner1967),inwhichhesoughttounderstandwhatmotivatedpeopleto

jointheCongressPartyafterIndependenceandhowthepartyorganizeditselftosustain

its domination. His comparative fieldwork enabled him to unravel the clientelistic /

patronage nature of the relationship between the party and voters, and underline the

criticalnecessity for thepartytoaddress itsconstituents’grievancesandassist themin

the settlement of local disputes. He pursued this method in his account of the 1977

GeneralElection(Weiner1978),basedonfieldworkandinterviewsconductedinvarious

cities across the country. Following his example, many would subsequently publish

single-electionmonographs(Kumar1997).

Thatliteratureonelectionswasenrichedbyaseriesofclassicmonographsonpartiesat

thenationallevel,asalsomonographsonparticularstates:ontheCongress(Brass1964a,

Kochanek 1968), the Jan Sangh (Baxter 1971, Jhangiani 1967), the socialist parties

(BurgerandUniversityofCaliforniaBerkeley.CenterforSouthandSoutheastAsiaStudies.

1969),theCommunistparties(FieldandFranda1974,Franda1971)andtheSwatantra

Party (Erdman 1967). These contributions focused on parties’ ideologies, internal

organizationandelectoralstrategies.

Between1974and1977,MyronWeinerandJohnOsgoodFieldco-editedfourvolumeson

electoralpoliticsinIndianstates(Barnett1975,FieldandFranda1974,FieldandWeiner

1977,1975), theoutcomeofan IndianElectionDataProject thathadstartedunder the

impetus of Myron Weiner at MIT in 1968. These volumes included contributions on a

seriesoftransversalquestions,suchastherelationshipbetweenelectoralbehaviourand

21

someaspectsofmodernization,suchastheGreenRevolution,theimpactofmigrationand

urbanization.Theyalso includedmonographicstudiesof leftparties inWestBengal,on

theperformanceofwomencandidatesinstateelections,comparisonsofelectoralpolitics

between various kinds of backward areas, in former directly administered colonial

territories and former princely states. Other essays focused on the study of regional

variationsofpolitical trends, theroleofvariouscleavages–religious, caste, ideological,

urban-rural, factional alignments – in the shaping of political preferences and electoral

outcomes.Thesestudiessoughttocombineelectionresultswithcensusandothersocio-

economicdata.Assuch,theydefinedtheresearchagendaoncontemporaryIndianpolitics

fordecades.Theguidingprinciplebehindthesestudieswastheimportanceof“proximity

variables”, such as ethnic concentrations, topography, the presence or absence of non-

farmeconomicactivitiesorindustrialization.

ThoughsteeredfromtheUnitedStates,thesestudieswerecoordinatedbyvariousIndian

universitydepartments (notably theUniversityofRajasthan,under IqbalNarain)15,and

aroundthenewlycreatedCentrefortheStudyofDevelopingSocieties(CSDS),underthe

direction of Rajni Kothari, who himself contributed landmark studies on the Congress

parties, the role of caste in politics, and the particular political trajectory of India’s

democracy(Kothari1970b,a).OneofKothari’scontributionswastogroundhisanalysis

inempiricalevidence–ascarceresourceinhistime–coupledwiththedeparturefrom

themorenormativeMarxistapproachthatwasinvogueinthatperiod.

One of the contributors of the Indian Election Data project, Paul Brass, developed and

formalizedthecasestudymethodbyaddingwhathecalledthe“ecologicalanalysis”,ora

systematic study of the correlation between electoral demographic and socio-economic

data with political phenomena, such as turnout, party performance and voters’

preferences(Banerjee,2009:20).Themeritof thisapproachwasthat itwasmindfulof

contexts and studied socio-political variables in connection with each other, and not

separately.Italsohadtheadvantageofnotrelyingonasinglesourceofinformationbut

tocomparevarioussourcesandnotediscrepanciesinofficialdata.Thisrecommendation

wouldproveessential toanyrelevantstudyof theroleof caste inpolitics, for instance.

BrasspursuedthismethodinsubsequentstudiesofNorthIndianpolitics.Hiscollectionof

15IthankPhilipK.Oldenburgforremindingmeoftheimportanceofthesedepartments.

22

essays in the 1980s (Brass 1984a, 1985) include several constituency or locality-level

studiesor electoralpolitics.Brass justifies the selectionof constituencies as illustrating

“differentaspectsofthemainsocialconflictsthathavebeenprominentinUPpolitics”.The

selection is thus made on the basis of exemplarity and not randomness. The same

principleguidedhisworkon communal riots, inwhichhe compared the trajectoriesof

thecitiesofAligarhandMeerut(Brass2004).Hislaterworkwouldfocusmoreonevents,

incidentsandindividualportraits,suchastheportraitofalocalBJPheroinKanpur,oran

incident of rape in Daphnala, as starting points for deeper reflections on political

authorityandviolence(Brass1997b).

Anotherproponentof the ecologicalmethodwasHarryBlair, a scholarworkingon the

stateofBihar. Inhisbookonelectoralpolitics in IndiaandBangladesh,Blair randomly

selected constituencies and conducted interviewswith voters, political actors and local

bureaucrats,underliningamongotherfindingstheimpactofthepresenceorabsenceof

membersoflocaldominantcastesatthepollingstation(Blair1979).Hewasalsopartof

thescholarswhousedadvancedstatisticsforthefirsttimetostudythesecorrelations16.

Yet,thenecessitytograsppoliticalprocessesonalargerscaleledscholarstodevelopthe

surveymethod.

Surveyandquantitativemethods

Political surveys had been in existence in India since the 1950s. Eric da Costa, an

economist turned journalist, is credited with having conducted the earliest opinion

surveys in India. He founded the Indian Institute of Public Opinion (IIPO) in 1956,

modeledaftertheAmericanInstituteofPublicOpinion,createdbyGallupatPrincetonin

1935 (Kumar and Rai 2013). There were other contributions that emanated from the

private sectorbut the first scientific election surveyswere conductedby theCSDS.The

firstNationalElectionSurvey(NES)wasconductedin1967byRajniKothari,Ramashray

Roy and Bashiruddin Ahmed. Roy and Kothari had both studied at the University of

Michigan, where they learned survey methods. They sought to apply these methods in

16Another examplewouldbeBruceBuenodeMesquita’sdissertationon strategiesof oppositionpartiesincoalitiongovernmentsatthestatelevelinIndia.

23

Indiaupontheirreturn.TheaimoftheNESwastomapandmeasurethevotingbehaviour,

opinionandattitudesofIndianvotersandexplainelectoraloutcomes17.

SanjayKumarandPraveenRaiexplainthemeritsandthelimitationsofthesefirstsurveys,

whichwereconductedforthe1971and1980elections,whichweretheninterruptedfora

periodofoverfifteenyears18.

Theresurgenceofsurveyinthe1990stookplaceinitiallyoutsidethewallsofacademic

institutions. Another economist turned journalist, Prannoy Roy, trained in the United

Kingdom, sought to apply survey methods in India and back his media coverage of

electionswithdata,startingwiththe1984elections19.AlongwithDavidButler,Roywas

inchargeofapopularregularpublicationbasedonelectionsurveyscalled“Indiadecides”

(Butler,Lahiri,andRoy1995),whichhelaterconvertedintoatelevisionformat,afterthe

establishmentofthenewschannelNDTV,in1998(TawaLama-Rewal2009).Atthesame

time,theCSDSreviveditsdataunit,underthedirectionofYogendraYadav.TheNational

ElectionSurveyserieswasrestoredandtheCSDSbuiltanation-widenetworkofscholars

andpartners–Lokniti–tocovereverygeneralandstateelection.Thefindingsof these

surveyswouldeventuallymaketheirwayintopopularacademicpublications,suchasthe

EconomicandPoliticalWeekly(EPW),andotherjournals.Theirmonopolisticpositionin

thisfieldmadethemthequasi-soleproviderofsurveydatatoscholars,inparticulardata

scientists.NESdatahaveprovidedtheempiricalbackboneofmanystudiesconductedon

elections,notably thebookseriesco-editedbyRamashrayRoyandPaulWallace,books

on caste and politics such as the Rise of the Plebeians (Jaffrelot and Kumar 2009), or

Christophe Jaffrelot’s Silent Revolution (Jaffrelot 2003b). Lokniti scholars and their

colleagues would contribute to numerous volumes on elections and state politics

(Chatterjee 1997, Gould and Ganguly 1993, Palshikar, Suri, and Yadav 2014, Vora and

Palshikar2004,WallaceandRoy2003).

In recent years, the discipline, particularly outside India, has followed the quantitative

turn that political science in general has taken. Statistical regressions and natural

experimentshavebecomecommontoolsandmethodsusedtoexploreaspectsofelectoral

17KumarandRai,ibid.,p.21.18Ibid.19See(Oldenburg1988).

24

politics,particularlyworkconductedwithanevaluativepurpose,onthelearningeffectof

women’s reservations (Bhavnani 2009), the effects of caste bias on state governance

(Acharya,Roemer,andSomanathan2015)orthestudyoftheincumbencyeffectinstate

elections (Uppal 2009, 2011, Uppal and Baskaran 2014). In the Indian context, these

contributionsareoftenmadebyeconomiststrainedinthesemethodologies,contraryto

politicalscientistswhohavetendedtoremainalooffromquantitativemethods.

Theanthropologyofdemocracy

Lastly, therehas recentlybeena return to theethnographicmethod,under the labelof

anthropologyofdemocracy.With theexceptionof figures likeHaroldGould,F.G.Bailey

and Adrian Mayer, anthropologists have traditionally stayed away from the study of

electionsanddemocraticprocesses,consideredtobetheturfofpoliticalscientists.Fora

long time, therewasadivisionof labourbetweenpolitical scientists,whostudied large

political processes, electoral outcomes and institutions, sociologists, who studied

politically assertive and marginalized social groups, and anthropologists, who

concentratedoneveryday localpolitics, issuesofviolenceorpatronage,anddemocratic

rituals20.

ThereusedtobeatraditionofanthropologicalworkonthepoliticalinSouthAsia.Oneof

itspioneeringfigures,BritishanthropologistFrederickGeorgeBailey,producedabodyof

work on politics based on his extensive fieldwork conducted in Bisipara, a village in

Orissa, and in various other parts of this state in the late 1950s. His best-known

contributionsareananalysisof the rules,both formaland informal, regulatingpolitical

competition(Bailey1969),theneedofkeepingenemiesinpolitics(Bailey1998),andthe

inevitable association of deceit and moral breakouts with political leadership (Bailey

1988,1991).Thecommonthreadofhisworkhasbeentoconsiderpoliticalpracticesas

theyare,disconnectedfrommoralornormativeconsiderations.

Yet, for all its value, the profile of the discipline dipped considerably in the following

decade, tobe revived in theearly2000s (Spencer2007).A2002essaywrittenby Julia

20Therewereagainexceptions,intheformofcross-disciplinarystudiesofpolitics,suchasthetwovolumeseditedbyFrancineFrankelandM.S.A.Rao(Frankel,Frankel,andRao1990,FrankelandRao1989),orpoliticalworkthatwassociologicallyinscribedinurban(Oldenburg1974)andruralsettings(Retzlaff1962,1959).

25

Paley, from the University of Pennsylvania, in the Annual Review of Anthropology

providesareviewofthefieldaswellasservesasamanifestofortherenewedinterestof

anthropologists in democratic processes and institutions (Paley 2002). This approach,

accordingtoPaley,

“…bring[s]tothestudyofdemocracyanexaminationof localmeanings,circulating

discourses,multiple contestations and changing forms of power that is rare in the

scholarlyliteratureondemocratictransitions,whichhaslargelyfocusedonpolitical

institutionsandformalregimeshifts”21.

Anthropologistsunderscorethatthemeaningofconceptssuchasdemocracy,equalityor

citizenship is contextual and that consequently, these contextualizedmeaningsmustbe

investigated,payingparticularattention to the languageused todescribe thesenotions.

Beyond the study of meaning, anthropological approaches focus on practices and their

intersectionwithmeaning,or,inPaley’swords,whatisdonewithmeaning(Paley2008)–

(Paley’semphasis).

Anothercommonunderstandingofthisapproachconsistsofsayingthatanthropologists

explorethe‘why’ofpolitics,whilepoliticalscientistsandpollstersfocusonthe‘what’of

politics.Whydo Indians vote?Whydopoorer voters tend to votemore than themore

affluentones?Whydopeoplevoteknowinglyfortaintedcandidates?Theytendtobeless

interested in formal aspects of democracyor the explanationof electoral outcome, and

moreinterestedinmattersofsubstanceofpolitics.

Mukulika Banerjee’s “Why India Votes?” is a landmark contribution to the study of

democracyinSouthAsia,notsomuchforheranthropologicaloutlookbutforthefactthat

she and her collaborators convincingly reconcile the local gaze of the ethnographic

methodwiththenecessityofcomparison,tobuildageneralargumentonwhypeoplevote

in the first place. Banerjee avoids some of the common limitations and traps of her

discipline – localism, resistance to comparison – by combining several methods:

ethnographic, comparative and quantitative. The methodology debate, opposing

quantitativeandqualitativemethods,oftenleadstosterileparochialconfrontationsorto

21Paley,ibid.,p.1.

26

celebratory yet unspecified calls for methodological unity. ‘Why India votes?’ for once

givestheconvincingdemonstrationthatnotonlycanvariedmethodscoexistwithinthe

frameworkofparticular researchbut that they canactuallyproduceanoutcome larger

than their individual contributions, mutually enriching the material collected and its

analysis.

One collection particularly stands out for its contribution to the field. Routledge’s

collection Exploring the Political in South Asia, edited by Mukulika Banerjee has so far

produced ten volumes on caste and politics (Michelutti 2008, Still 2014), state politics

(RaghavanandManor2009), crime (Sanchez2016), and themeaningsandpracticesof

power and influence in South Asia (Price and Ruud 2010). It also includes the Rise of

Plebeians,inthecontinuationofwhichthisdissertationislocated.

What I retain from this body of work is the necessity to focus on practices, on what

politicalactorsdo,andnotontheverificationofwhethertheiractsconformtowhattheir

formalstatusmandatesthemtodointhefirstplace.Forthepurposesofthisdissertation,

itwasimperativetoconsiderthe‘actualjobdescription’ofpoliticians22,definedbothby

politicians’self-perceptionoftheirjobdutiesandrequirementsandbyvoters’notionsof

what their representatives should be doing, rather than seek to verify whether their

actions correspond with or fit the mould of their formal institutional mandate. Studies

that aim at establishing whether practices conform to formal norms are bound to

disappointwiththeirnearlysystematicnegativeresponsesorirrelevantfindings.

It is also necessary to contextualize political practices at their most relevant level of

observation,whichtendstobelocal.Anelectionsignifiesmorethantheactofchoosinga

representative.Therearematerialand intangibleconsiderationsatstake, related to the

balance of power between individuals and groups. Local political competition is

embedded within a context of transforming local configurations of domination and

subordination–orpowerrelations–betweengroups.

The third element of interest is that it is possible to ‘scale up’ ethnographic findings

throughcomparisonormulti-sitefieldwork.Alargepartofourfieldworkhasconsistedof

22IborrowthisexpressionfromaconversationwithPhilipOldenburg.

27

observingandaskingquestionsonthedistributionoflocalpowerandpoliticalinfluence

indifferentlocations,andtocomparenotesonvariedconfigurationsofthedistributionof

powerandinfluence.

SomeofthemostrelevantandinterestingcontributionstoIndianelectoralpoliticshave

been based on mixed methodologies, at the crossroads of quantitative and qualitative

approaches. Survey data is used as an empirical back-up for more in-depth studies of

aspectsofelectoralpolitics,suchasthemeaningoftheactofvoting(AhujaandChhibber

2010), the interconnections between caste and class in electoral behaviour (Jaffrelot

2015a),thepoliticalbehaviourofminorities(Heath,Verniers,andKumar2015),therole

of gender in political participation (Deshpande 2004) and, of course, the large body of

literatureontheexplanationofelectoraloutcomeingeneralandassemblyelections.

1.4.Methodology

This dissertation employs a mix of methodological approaches., consisting in using

empiricaldataanddescriptivestatisticstodrawacontextofpoliticalaction–whatIcall

thesystemicrulesofpoliticalengagements–aswellastostudytheevolutionofthestate

assembly’smembers’profile.

Thecontentofthisdissertationalsodrawsfromextensivefieldworkconductedoversix

years across the state of Uttar Pradesh, between 2007 and 2012. The first exploratory

fieldworkwasconductedinLucknowandvariousconstituenciesinEasternUttarPradesh

duringthemonthsofMarchandApril2007,beforeandafterthestate14thstateAssembly

election.Subsequently,regularvisitstoLucknow,thestate’scapitalenabledmetobuild

theempiricalbaseofmuchoftheanalysesconductedinthisdissertation.Otherroundsof

fieldwork were conducted before and after the 2009 General elections and around the

2012 StateElections, across constituencies inWesternUttar Pradesh andCentralUttar

Pradesh.

The method used has been essentially based on semi-directive interviews with local

politicalleaders,candidates,partycampaigners,academicsandlocalpoliticalobservers.I

havetried,throughmyfieldwork,topayattentiontocontext,demography,socialrivalries,

28

factionalism,aswell as thepoliticaleconomyof constituencies.Extensivesemi-directed

interviewsalsoprovedfarmoreeffectiveandricherthanpre-establishedquestionnaires,

orevenfollowingafixedpro-forma.

The fieldwork also aimed at locatingpoliticianswithin their local contexts, identify the

typesofnetworks towhich theybelongand fromwhich theydraw their resourcesand

influence.Fieldworkconductedinconstituencieswasguidedbythetwobroadquestions,

‘who exerts political influence here and how is power distributed?’; and how has this

evolvedthroughtime?

The empirical base of this dissertation is composed of building three datasets, two of

which are unique. The first dataset consists of the digitization and the expansion of

publicly available ECI (Election Commission of India) reports, which provide fairly

detailed information on General and State election results. These reports have been

digitized23, cleansed and expanded with the addition of new variables, such as sub-

regions,thematchingofAssemblyandParliamentaryseatsandsomesocio-demographic

variablesdrawnfromtheIndiandecadalcensusforthemostrecentyear.

The second dataset pertains to the sociological profile of Members of the Legislative

Assembly(from1962to2012)andcandidates(comprehensivelyfortheyears2007and

2012).Variousvariables,suchasoccupation,education,individualinformationoncareer

and family background,were collected through theWho’swho, or biographical notices

publishedbytheUttarPradeshVidhanSabhalibrary.Thesevolumeshavebeentranslated

anddigitized.ThisWho’sWhodatawasverified,andaugmentedwithcastedatathrough

fieldwork and interviews with candidates and representatives, local journalists, party

workers and other political observers. Interviews were conducted in individual

residences, party offices, on the campaign trail and at times in vehicles, train stations,

dhabasandallkindsofhotels,rangingfromdodgytofancy.

Therehasbeenarigorousefforttoincludecandidates’dataintothedatasetsandanalysis

as far as possible. Any study of elected representatives should include unsuccessful

contestantsaswell,sincelittlecanbesaidofoneifitisnotcomparedwiththeother.Most

23Francesca Jensenius provided a reformatted ECI data until the year 2007. Data cleaning andadditionofnewvariableswasdonebytheauthor.

29

studiesonrepresentationhavefocusedonelectedrepresentativesandnotoncandidates,

essentiallydue to the lackofdataon the latter.Understandinghowpartieswinor lose

electionsrequiresthatweunderstandwhocontestsinthefirstplaceandunderwhatkind

ofconstraints.

The third dataset undertakes the coding of individual political trajectories, that is, the

attributionofauniqueidentificationnumberforeverycandidatehavingcontestedState

Assemblyelections(73,480entriesfrom1951to2012).Namesof individualcandidates

have been matched, both manually and with the help of a fuzzy name-matching script.

This enabled the coding of individual career trajectories (how many times listed

individualscontestedandtheresult),thestatusofcontestants(re-runningorincumbent

candidates,ex-MLAsormaidencontestants),of‘turncoat’candidates(candidatesshifting

party affiliations between two elections) and of ‘migrating’ candidates (candidates

contestingfromdifferentconstituenciesthroughtime).Thisdatasetprovidesthemeasure

ofindividualincumbency,orthecapacityofanelectedrepresentativetobere-electedor

electedmorethanonce,whichisanindicatorofelectoralvolatility.

Thisdissertationrestsessentiallyon theprosopographyofpolitical actors– candidates

andelectedrepresentatives.Thedatasetsassembled for thisdissertationareessentially

meant to provide an empirical bedrock to what is essentially a qualitative analysis of

politicalpractices.

Icontendthatthedataitselfdoesnotcontainanswerstothemanyquestionsithelpsto

raiseand formulate,andthatqualitative fieldworkandtheobservationof localpolitical

practicescanandshouldbecontextualizedandtestedagainstthebackdropofempirical

data,bothintimeandspace.

This dissertation builds on previous contributions of scholars who worked on the

sociologicalprofileofrepresentativesinIndia(JaffrelotandKumar2009,Jayal2006). It

aimsnotonlyto‘update’thesestudiesbyincorporatingdataonrecentelections,butalso

toexpandtheseapproachesbyincludingnewvariablesandbycontextualizingthisdata

throughcomparativelocalizedqualitativefieldwork,thereforedrawinganewframework

ofanalysisforthestudyofpoliticalrepresentation.

30

BeforeIgetintocontext,itisnecessarytoevokethequestionsthathavebeenraisedby

scholars about politics in Uttar Pradesh and examine some of the responses that they

haveoffered.

1.5.Literaturereview

Reviewing the literature on Uttar Pradesh politics is not an easy task since it is not

cohesively organized under a regional denomination, the way Punjab studies or Tamil

studies are, for example. This is largely due to the fact that Uttar Pradesh, despite its

centrality in Indian politics and the public’s imagination, is rarely seen as a cohesive

regionalentity.Thestatedoesnothaveaspecific languageof itsown. It isdivided into

groups and communities that often find more commonalities with their cross-border

counterpartsthananattachmentwiththestateasawhole.

ItisrevealingthatUttarPradeshdoesnothaveregionalistparties,butregional,orstate-

basedpartiesthatdonotarticulateaparticularnotionofregionalidentity.Rather,these

parties seek to identifywithparticular segmentsof the state’spopulation, or refer to a

broader, national register, as is the case for the Congress and the BJP. Further, the

denomination of “heartland”, often used to describe Uttar Pradesh, also lends a vague

charactertothisspace,aswellasageographicalconnotationthathasnotbeenconducive

totheemergenceofabindingoroverarchingregionalidentity24.

Besides, thepaucityofpublicuniversities throughout the statehasmeant that regional

academic productions, in the forms of journals such as the Uttar Pradesh Journal of

Political Science,or theUttarPradesh Journalof Social ScienceResearch,have failed to

makeamarkbeyondtheboundariesofthestateanditsregionaluniversities25.Thishas

not alwaysbeen the case.The stateofUttarPradeshused tohaveuniversities of great

standinginLucknow,Allahabad,VaranasiorAgra26.Theseinstitutionshavesufferedfrom

24ThislackofacohesiveregionalidentityforUPisillustratedbythefactthattheSamajwadiParty’sbranchinMumbaipresentsitselfasthepartyoftheUttarBhartiyans(NorthIndians),anacceptionthatgenerallycoverstheNorthIndianhindiphonesphere.25AnotableexceptiontothissadstateofaffairsisProf.A.K.Verma,ProfessorofPolitcalScienceatChristCollege,Kanpur,andstatecoordinatoroftheLoknitinetwork.26In2013,T.N.MadanpublishedahistoryofSociologyat theUniversityofLucknow(ReddyandHaragopal1985).

31

decades of under-funding, politicization and general neglect, a dire situation well

described in Craig Jeffries’ books on youth and education in Western Uttar Pradesh

(Bakshi,Chawla,andShah2015)27.

What we find instead are various periods and streams of scholarly work based on

fieldworkconductedinUttarPradesh.Thepresentationoftheliteraturethatfollowsdoes

not aim to be comprehensive but to offer a diachronic analysis of some the main

questions that have been raised (mostly) by political scientists as well as to give an

overview of the answers they have proposed and the debates that these answers have

generated. This exercise is also limited to the post-Independence literature, which

correspondswith the date of the emergence of political science as a discipline in India

(RudolphandRudolph2010,561).

Theearlyliterature:partypoliticsandthefragilityofCongress’dominance

The early literature on U.P. politics tends to focus on state-specific political questions,

suchasfactionalismwithintheCongressParty(Brass1965,1984a,1985),andtheriseof

oppositionparties (Baxter 1971,Burger 1969). In 1969,RalphC.Meyer completed the

first prosopographical study of Uttar Pradesh’s political elite, gathering data on the

sociological profile of U.P. MLAs from 1952 to 1962. Most of these early contributions

either focus on parties and state-level politics, with some exceptions, such as Ralph H.

Retzlaff,whostudieddecisionmakingprocessesinaWesternU.P.village(Retzlaff1959,

1962),orRobertS.Robinwhowroteonelite formation in threePanchayatelections in

UttarPradesh,beforeandafterIndependence(Robins1967).HaroldGould’searlywork

oncastepoliticswasbasedonfieldworkconductedinUttarPradesh,notablyinLucknow

andFaizabad(Gould1963,1969).

Givenitsdemographicandpoliticalimportance,UttarPradeshalsooccupiesalargespace

inliteraturedealingwithnationalpoliticsandnationalpoliticalprocesses.Oneexampleis

thestudyofMuslimpoliticalelitesbyTheodoreP.WrightJr.,inwhichU.P.Muslimsfigure

prominently (Wright Jr. 1964, 1966), and he highlights the elite character of Muslim

27TheRudolphshadalreadydescribed,in1969,howuniversitieslocatedintheHeartlandsufferedfromlessfundingandattentionthantheirearliercounterpartslocatedoncoasts,orintheRimland(Kapuretal.2010).

32

representatives,who,religionapart,sharemoresocialcharacteristicswiththeirelected

Hinducounterpartsthanwiththeirownbrethren(WrightJr.1964,267).

The aforementioned state-specific studies were written on the basis of fieldwork

conductedmostlyintheearly1960s,atatimewhentheCongressdominationwasatits

peak. All of them, however, point to some contradictions or vulnerabilities in this

dominantsystem.InhisstudyoffactionalpoliticsintheCongress,Brassanalyzeshowthe

Congress domination depends on its embeddedness in local economic and social

environments, particularly by controlling local political and economic institutions

through theco-optationof locallydominantruralcommunities28.Healsouncovershow

theCongressdominationrestsonlocalallianceofdominantgroups,ratherthanonelite

capturefromasingleuppercaste(i.e.theBrahmins).Hefurthershowsthatthecapacity

oftheCongresspartytoretainpoweralsodependsonitsabilitytomaintainthecohesion

ofitslocalbranches,ortodealwiththedisintegrativeimpactoffactionalism29.Heargues

that internal factionalismismoresignificantthaninter-partycompetitionforexplaining

variationsintheCongressvote.

InhercomparativestudyoftheCongressPartyinU.P.anditsnascentopposition,Burger

identifies five sources of the Congress’ vulnerability: ideological clarity and purpose,

personnel in organization, the difficulties of governing, particularly in relation to the

bureaucracy, the inability to meet voters’ expectations, and the party’s growing

identification with specific elite groups 30 . While she recognizes that the social

composition of the Congress leadership remains frozen in the social groups that were

initially mobilized (traditional upper castes elites), she also points at the ability of the

partytoopenitsdoorstoothergroupslocallywhentheoldconfigurationledthemtolose

seats31.

By followingthetrajectoryandperformanceof the JanaSangh inmunicipalelections in

thelate1950s,BaxtershowshowquicklytheCongressstartedlosinggroundtotheHindu

right in urban seats, notably in the 1959 elections in which it lost all the major cities,

28Op.cit.,p.229.29Ibid.,p.239.30Op.Cit.,p.264-265.31Op.Cit.,p.271.

33

includingLucknow32.Heshowstheprocessthroughwhichaformercoresupportbaseof

the Congress Party – the upper castes – split their votes between parties according to

theirlocalinterests.

In his study of the profile of the U.P. Assembly, Meyer perceives early changes in the

economic profile of elected representatives, warning in his conclusion that if the

interlocking of economic and political power could bring some stability to the political

system,itcouldalsowellbetothedetrimentofthepoorandatthecostofadysfunctional

economicgrowth,duetotheaversionoftheseelitestoredistributingwealth(Meyer1969,

350).

Finally,Robin’sstudiesofPanchayatelectionsfrom1946to1961indicatethat“statusasa

traditionalleader,beingliteratebelowthehighschoollevel,enjoyinghighcastestatus,and

havinganon-agriculturaloccupationareallpositiverecruitmentfactorsatthenominating

level”33,andthatpoliticalconflictsaresortedduringthephaseofnominationandnotat

the time of the election (which sees a large number of Panchayat leaders elected

unopposed),whichshowshowlocaldemocraticprocessescanbesubjecttoelitecapture.

Mostauthorsofthisperiodemphasizetheimportanceofpopularparticipationandsocial

mobilization, and explain how party politics hampers the transformative aspirations of

these movements. Meyer in particular underlines how the socialist parties, especially,

remainbiasedtowardstheuppercasteintheirorganization.

There is also an interest, post-Independence, in some of the transformative policies

implementedbythefirststategovernments,notably landreforms(Mayer1958,Metcalf

1967, Neale 1962). Many of these contributions were made by former or active civil

servants,andsometimesbypoliticiansthemselves–atrendthatwouldlaterdisappear34.

Mostofthesestudies,conductedeitheratthevillage leveloratthedistrict level,would

32Op.Cit.,p.180-181.33Ibid.,p.29.34CharanSingh’swritingonagriculture, landreformsandsocialrelationsareacase inpoint.See(Singh1947,1959,1964).AnotherusefulsourceisRamManoharLohia’scompletewritings,editedbyMastramKapoor(LohiaandKapoor2011).

34

pinpoint the crucial role of local democratic institutions and processes in the

implementationofpublicschemesandeconomicpolicies.

Identities,parties’recompositionandthestudyofagrarianchange

Paul Brass remains the main figure of studies on U.P. politics through the 1970s and

1980s.Thecrystallizationofgroup-basedpoliticaldemandsintheperiodinspiredBrass

to explore the role of language and religion in the formation of group identities. In his

1974book,Brassproposesaconstructivistreadingofidentityformationbyunderscoring

theroleofpoliticalelites,whomobilizeonthebasisofselectedsymbolsofgroupidentity

(Brass1974). Comparing the trajectories of threepoliticalmovements inNorth India –

theMaithilimovementinBihar,thepoliticaldifferentiationofMuslimsinUttarPradesh,

and the formation of a Sikh Punjabi linguistic state in Punjab – Brass examines the

conditionsunderwhich linguisticandreligiouspoliticalmovementssucceedor fail,and

posits thecentral roleof smallelites in the formationof collective identities.Successful

movements are those in which political elites effectively mobilize on the basis of a

particular symbol (language or religion)while subsuming other symbols (caste, region,

state),underit.

Thus,politicalpartiesandsocialmovementsarenotmerelytheextensionofparochialor

communalpopulardemands,but theyalsoshapethesedemandsbyreinforcingdivisive

social cleavages35.Theyarecentral to the formationandchannelingofgroup identities.

This is important in the context of a polity where political elites claim to attend to

‘primordial’needsofpre-existingpoliticalcommunities36.

In the following decade, Brass shifts his attention from the Congress to the socialist

opposition,whichleadshimtoexploretheconnectionsbetweenagrarianchangeandthe

emergence of a new divided political family, grounded in the small and middle-

landowningpeasantclasses(Brass1984a,1985).Thetwovolumespublishedin1984and

1985, Castes, Factions and Party in Indian Politics, contain his contributions on the

interconnection between caste mobilization, inter and intra-party dynamics, and broad

35Op.Cit.,p.41.36SeethepoliticizationoftheYadavsin(Michelutti2008).

35

socio-economic change in Uttar Pradesh over twenty years. While the first volume

containsanumberofcomparativestudiesonelectoraland legislativepoliticsaswellas

nationalpolitics,mostof the twovolumes’ contributionsarebasedon thecaseofUttar

Pradesh, where he traces the roots of political instability in parties’ lack of internal

discipline, theweakness of their organizations outside the legislature, the opportunism

and careerism of political leaders, and the lack of institutionalized support in the

electorate (Brass 1984a). Here again we find traces of his political elite-focused

constructivistapproach,aconstantreminderoftheprimacyofpoliticsintheconstruction

ofpoliticalandsocialmovements.

Invariouschapters,heunderlinesthespecificityof Indianelectoralpolitics,notablythe

decoupling of the principles guiding national or state-level politics, where parties and

leaders in power pursue a modernist developmentalist agenda, and local politics, that

remains embedded in local structures of authorities. While the state and the national

governmentseektoreformthecountry,localcongresscadresaremorepreoccupiedwith

theprotectionoftheirinterests,throughlandcontrol,accesstolocalresources,andvoice

concernsaboutinter-casteandinter-communalrelationsthattheleadershipoftheparty

wouldwishaway(Brass1984a,5).Thus,thelinkagesbetweenlocalandstatepoliticsare

markedbycontradictions,whicheventuallytranslateintofactionalismanddivisions.

WhileBrassstressesmoreonpartypoliticsthansocialtransformations,heacknowledges

thattheriseandfallofpartiesisalsoconnectedtodeepersocialmovements.Chaudhary

Charan Singh’s ascension to power in 1967 and 1970 – both mandates aborted by

President’sRule–signalednotonlythepossibilityofconsolidationofthesocialistcamp,

but also the consolidation of the middle peasantry, which had undergone profound

changes(Brass1980a,b).

Thesechangeswere in largepart consequencesof landreformsand transformationsof

theruraleconomy.Inthisperiod,variousauthorsexaminedtheimpactoftheabolitionof

theZamindarisystemandoftheintroductionofthetenancyreformsandlandceilinglaws

(Hasan 1989, Metcalf 1967, Neale 1970, Oldenburg 1987, Pai 1986). Sudha Pai in

particular studiedagrarian relations in fourdistricts inEasternU.P. in theearly1980s.

Sheexaminedtheconditionoftraditionaluppercastelandlords,thestructureandtypes

oftenancyandlandsize,thepositionofthelandlessclassandtheissueofbondedlabour.

36

Her choice of terrain was not conventional since most of the earlier (and subsequent)

studies on agrarian transformations had been conducted in areas that were relatively

more prosperous and that had been exposed to the Green Revolution (the Western

regions). In her first book, she proposes a typology of landowners, breaking down the

broadandvaguecategoryoffarmerintofourmainagrarianclasses:biglandowners,both

self-cultivators and rentiers; medium landowners, both self-cultivators and share-

croppers; petty land-owners; and landless peasants37. Her fieldwork revealed that the

classofbiglandownerssucceededinpreservingmuchoftheirpastassets,despitevarious

wavesoflandreform,whichmeantthattheycouldmaintainmuchoftheirpastinfluence.

Caste divisions among the middle peasantry and landless farmers acted as a hurdle to

theirorganizationasaclass38.

Aninterestingfeatureofherworkisherreluctancetomakesweepingcategorizationsof

socialtransformations.Whilesheinitiallyattemptedtoseewhetherfeudalismcontinued

to prevail over expanding capitalistic modes of rural production, the diversity of

situations and the over-determination of local contexts over caste and class relations

preventedherfromprovidingastraightanswertothequestion39.Shealsoacknowledged

thatcasteconfigurationsvariedliterallyfromvillagetovillage.

In1989,ZoyaHasanaddressedsimilarquestionsinthecontextofWesternUttarPradesh

(Hasan1989). Sheproceeds to abroadhistorical overviewof the evolutionof agrarian

relations,fromthemid-nineteenthcenturytoIndependence,andthenexaminesthesocial

and political impact of land reforms and agrarian change in the post-Independence

context.Insodoing,shearticulatesmoreexplicitlythanPaithepoliticalconsequencesof

agrariantransformations,thatistosay,theconsolidationofanewclassofrichpeasants

externaltotheCongress-affiliatedtraditionalpatronagenetworks.Shealsodescribeshow

dominantlocalgroupssucceedinconcentratingthegainsfromthe‘newagrarianstrategy’

of the 1960s and 1970s, “while the conditionof the small peasantsworsenedbecause of

theirinabilitytomeettheincreasingcostsofproduction”40.

37Ibid.,p.119.38Ibid.,p.131.39Ibid.,p.128.40Op.Cit.,p.164.

37

InWesternU.P.,agrarianreformsandtheGreenRevolutioncontributedtotheemergence

of a politicized elite segment of the peasantry,who gradually gained control over local

democratic institutions through class-based mobilization and money power, a fact

observable in other Indian states, such as Maharashtra, Punjab, Karnataka or Tamil

Nadu41. InNorthernIndia,therichandmiddlepeasantrywashistoricallyanti-Congress,

asitwasconsideredtobedominatedbyandbiasedtowardstheuppercastes42.

Thisanti-CongressismtranslatedintoamassivesupportforCharanSingh’sBharatiyaLok

Dal,apartyovertlydedicated to their representationand thedefenseof their interests.

This consolidation behind the Lok Dal was initially a Western UP phenomenon. As Pai

shows, therich landowningclasses inEasternUPcutacross theuppercastesandsome

dominantOBCs.

Variousarticlesandcontributionsenrichthis literaturewithcasestudies,whichtendto

documenthowreformsandpoliciesgetsubvertedoncetheypassthroughthefilterofthe

localcontextsofsocialdomination(Singh1974,Singh1976,Subas1984).Inasimilarvein,

Oldenburg, in his study of the impact of the land consolidation policy of the 1960s

(chakbandi),analyzeshowtheimplementationofruraltransformationschemesinducesa

cultureandpracticesofcorruption,particularlythroughtheemergenceofaprofessional

classofintermediaries(Oldenburg1987)43.

The politics of the 1960s and 1970s is well summarized in a richly documented

contributiontoIqbalNarain’svolumeonstatepoliticsinIndia,bySaraswatiSrivastava,a

lecturerinPoliticsatBenaresHinduUniversity(Srivastava1976).

Newresearchdirectionsinthemid-1990s

41SeeAnthony Carter (Carter 1974) for a study of elite politics in ruralMaharashtra or PranabBardhan(Bardhan1982),forabroaderoverview.42Ibid.,p.165.43Oneof theearliercontributionsonthesubjectof landconsolidationwasmadeby JosephElder(Elder1962).

38

Thelate1980sandearly1990sconstitutedamajorturningpointinU.P.politics,withthe

declineoftheCongress,theriseoftheBJPandtheHindurightandthepoliticalassertion

ofDalitsthroughtheBSP,andofsegmentsofthebackwardclassesthroughtheJanataDal

and, post-1993, the Samajwadi Party. Electoral competition intensifies and turns more

violent,asthestatesinksintopoliticalinstabilityandfinancialcrisis.TheliteratureonU.P.

politicsreflectsthesedevelopmentsandbranchesout infourmaindirectionsorfieldof

studies: identity politics and the rise of the lower castes, the rise of the Hindu right,

political violence (Brass 1997b, 2003, 2006,Wilkinson2006,Wilkinson2013), and the

studyofpatronageandclientelism(Chandra2000,2004a,c).

A number of authors contributed to several of these academic streams, which clearly

overlappedwitheachother.The1990scontributionswouldinitiallyadhereonlytofacts

and events, and would give way in the 2000s to several exercises in formalizing the

explanations put forward into more general arguments about Indian politics. In other

words,theeventsoftheearly1990sgeneratedawholesetofnewquestionsthatwould

framethesubsequentstudyofU.P.politics.

Lowercastepolitics

TheriseofDalitpoliticsinthelate1980sandearly1990sspurredarenewedinterestfor

the study of caste politics and in particular the rise of lower caste parties. Various

historicalaccountsareavailableregardingtheriseoftheBSP(Duncan1997,1999,Hasan

1998,Jaffrelot2003b,Mendelsohn1993,MendelsohnandVicziany1998,Pai1997,1999).

There is little to findon the travailsof theRepublicanPartyof India (RPI),Ambedkar’s

formationwhichcontestedanumberofseatsinUttarPradeshinthelate1960s(theywon

10seatsinthe1967elections,mostlyinDoabandWesternUttarPradesh)44.

Theseaccountsdonotdifferverymuchfromeachotherandofferapicturethatfitwell

withPaulBrass’constructivistapproach,astheydescribehowasmallDalitbureaucratic

eliteproceededtocreateamovementandapartythatcontributedtothepoliticizationof

44AngelaBurger,inherstudyoftheCongressoppositioninUttarPradesh,doesnotmentiontheRPI.

39

vastsegmentsofthestate’sDalitpopulation.Manyofthesecontributionsalsounderline

thefactthatthegainstherebyobtainedhavebeenlargelysymbolicandthattheBSPdid

little to structurally alter the state of exploitation and exclusion most Dalits are still

reelingunder.

IanDuncanforexampleunderlinesthattheBSP’srisehasbeenessentiallybasedoncaste

appeal,andnotoncampaignsbasedonmaterialissuesaffectingDalits(Duncan1999).As

aresult,whiletheinsistenceonsocialoppressionandexclusionhelpedtheBSPtoforgea

Dalitpolitical identity, it alsoprevented it frombuildingabroadelectoral supportbase

amongtheruralpoor,beyonditsDalitcore.

Others have been less severe with the BSP and have studied in detail how the party

proceededtodivertthestatemachinerytoworkinfavouroftheDalitsoncetheywerein

power through a politics of transfers and public job nominations (Jaffrelot 2003b) or

through a clientelistic diversion of public resources towards the Scheduled Castes

(Chandra2004c).InTheSilentRevolution,Jaffrelotdescribesindetailthetransformation

ofabackwardcastes’civilservicesunionintoapoliticalparty.BothChandraandJaffrelot

insist on the importance of reservations in the forming of a small but politicized and

mobilized middle-class educated elite among the Dalits. Jaffrelot and I would later on

detail how the party expanded its base by opening its doors to non-Dalit groups,

discardingthenotionoftheBSPasanexclusiveDalitparty(Jaffrelot2010a,Jaffrelotand

Verniers2012).

TheriseoftheHindunationalistmovement

Similarly, the riseof theBJPput to the fore thequestionofethno-religiousnationalism

andculturalmobilization.TheAyodhyamovement–orthequestforthere-appropriation

of a mosque allegedly built on the birthplace of Ram – has been widely covered and

analyzed as the culmination of a long process of development of the Hindu nationalist

movement (Hansen1999,Hansenand Jaffrelot2001,Hasan1994,1998, Jaffrelot1996,

Parikh 1993, Zavos 2000). These contributions examine the emergence, the ideological

foundations as well as the changing support base of the Hindu nationalist movement.

40

Zavos,Hansen,andJaffrelot inparticularofferthreedifferentreadingsofthesourcesof

themovement,whichcomplementratherthancontradicteachother.

“[Jaffrelot]explainsthecurrentdominantpositionoftheHindunationalistsastheresultof

decades of systematic […] organizational work and imaginative political strategies”

(Hansen1999,4).ZavosexaminesthehistoricalandideologicalfoundationsoftheHindu

nationalistmovementinthelate19thandearly20thcenturiesandbaseshisanalysisona

more historical and cultural base. Hansen focuses the bulk of his analysis on the re-

emergenceoftheBJPandaffiliateorganizations inthe1980s.HansenandJaffrelotthen

analyze thepoliticsof theBJP in the1990sand thepolitical consequencesof theBabri

Masjid demolition. Similarly to Zavos, Hansen’s main argument locates the origins of

Hindunationalismnotspecificallyinthepoliticalorreligiousdomainbutintherealmof

public culture45. All three recognize that the purpose or aim of the Hindu nationalist

movement is to effectuate social change, to “transform Indian public culture into a

sovereign, disciplined national culture rooted inwhat is claimed to be a superior ancient

Hindupast,andtoimposeacorporatistanddisciplinedsocialorganizationuponsociety”46.

Politicalpower–andthereforetheBJP–isseenasaninstrumenttoachievethesegoals,

andnotasanendinitself.

Theexperienceof theBJP’spower inUttarPradeshandat theCentre in the late1990s

and early 2000s would see the party adapt its posture and try to adopt a more

“mainstreamfigure”,notably throughthevernacularizationof itsnationalisticdiscourse

(Narayan2006),ashift fromovert tocovert formsofreligiousmobilizations(VanDyke

1997), and a shift of its political discourse towards more consensual themes, such as

governanceanddevelopment(Zerinini-Brotel1998,AdeneyandSáez2005).

The literatureonHindunationalism isnotU.P.-specificbutgiven the importanceof the

stateinnationalpoliticsandthefactthatmanyofthemajoreventsthathavemarkedits

historyhaveoccurredthereexplainwhythetwolevelsofanalysis–nationalandregional

–areoftentreatedasinterchangeable.

45Op.Cit.,p.4.46Ibid.,p.4.

41

BackwardClassesPolitics

Backward-classpoliticsconstitutesthethirdpartofthepoliticaltriptychthatdefinesU.P.

politicsinthe1990s.Thisliteratureisdividedintofourmainthemes:thetransformation

ofkisanpoliticsintocastepolitics,quotapoliticsandtheMandalaffair,partypoliticsand

theriseofOBCrepresentationintheAssembly,andthe‘backwardization’ofotherparties.

The first theme refers to a period of political turmoil that saw the socialist family – or

JanataParivar–dislocateintovariouspoliticalformationsbasedondistinctsocialbases.

ThesplitoftheJanataDalintotheSPandtheRLDintheearly1990ssignaledtheendof

kisan politics – understood as a politics of the representation of a broad spectrum of

backward castes, spanning from the Scheduled Castes to the dominantOBCs – and the

adventofcaste-basedmobilizations.

Hereagain, Jaffrelotprovides themostcomprehensiveoverview,detailingthehistorical

andsociologicalprocesses that led to thepoliticizationofbackwardcastes initiallyasa

broad social category, defined essentially in terms of class and occupation (Kisan) to

caste-basedformsofpoliticalmobilization,aroundtheissueofquotas(Jaffrelot2000a,b,

2003b).WhilehisunitofanalysisistheHindibelt(NorthIndianHindi-speakingstates),

UttarPradeshoccupiesacentralplaceinhisanalysis.Hedemonstrateshowquotapolitics,

which initially emerged from the farmer’s movement, initially complemented Kisan

politics but at the turn of the 1990s unleashed a process of caste polarization that

substitutedcasteasavehicleofpoliticalmobilizationtothebroadcategoryofbackward,

orKisan47.

The caste politics literature of the 1990s stresses particularly on party and caste

alignmentsthroughthearticulationofcasteandpartyidentitiesandthedefenceofcaste-

basedinterests.Littleattentionispaidtotheheterogeneityofcastes–evencastegroups.

Inrecentyearshowever,variousscholarshavebroughtelementsofcastedifferentiation

totheanalysis,usingdiversemethodsandreachingvariousinterpretations.

Zerinini looksat thedifferentiatedtrajectoriesofcastewithincastegroupsusing jatias

theunitof analysis. Shehas shownempiricallyhowspecificdominantOBCcasteshave

47Op.Cit.,p.343.

42

benefitedfromthebackwardmovementandhowlowerOBCsremainunder-represented

intheStateAssembly.(Zerinini2009).

In an original contribution to the caste literature in Uttar Pradesh, Sunit Singh, an

Allahabad-based scholar, looks at castes as internally stratified entities, in which a

nucleuscomposedofprosperousanddominantmembersholdskeypositionswithintheir

caste network (Singh, 2002: 179). The members of this nucleus are comparatively

privileged in terms of education, share in services, business, landholding, etc. The

remainingmembersofthecastearesituatedatdifferentpointsondifferentorbitsaround

thenucleus, thedistancebeing the reflection of their relative economic strength. Singh

conductedasurveyandcreatedfocusgroupstoestimatethesize,orthelevelofinfluence

concentrated among broad caste groups.He finds that both SC andOBC castes tend to

have a small nucleus anda largeperiphery, indicating a strong concentrationofpower

within these castes. This is consistent with the literature on the creamy layer, which

indicates that reservations have contributed to the emergence of a small elite within

targetedgroups,whotendtoseizethelion’sshareofthebenefitsofquotasandpolitical

influence.

Following the advent of simple majorities and the rise of ‘rainbow coalitions’, some

authorscontestthecentralityofcasteinelectoralmobilizationprocesses.Paidevelopeda

moderateviewonwhatshecalls“post-identity”politics,oraprocessofcomplexification

of thedeterminantsofvote, inwhichcastecontinues toplaya rolebutalongsideother

factors(Pai2013).Others,lessnuanced,havecometostatethatcasteplaysaminorrole

in electoral choice and the determination of electoral outcomes, in the face of the

multiplicity of intervening factors and the local heterogeneity of social groups (Gupta

2016,GuptaandKumar2007).

Finally,morerecently, JaffrelothasusedCSDSsurveydatatobreakdowncastesupport

forpartybyclass, andobservedapositive relationshipbetween theupper segmentsof

thebackwardcastesandtheBJP.Thiswouldindicatethatinrecentyears,atleast,aclass

cleavagecutsacrosscastedivisionsandaffectshowvoterscasttheirvotes.

The literature of the 1990s culminated in one of the last efforts to provide a cohesive

overviewofthemajorpoliticalprocessesatworkduringthiscrucialdecade(Hasan1998).

43

In“QuestforPower”ZoyaHasanhassoughttodescribeandanalyzehowanewpolitical

order has emerged from the mobilization of various forms of identities, how the

intertwiningof religiousandcastemobilizationscreatedapolitical spacedominatedby

theHindurightandstate-basedparties,precipitatingthedeclineoftheCongressParty.In

order todoso, she juxtaposes fourchaptersdealingwith thedeclineof theCongress,a

reformulation of the political consequences of agrarian change, backward-class

mobilizations and the struggle around Ayodhya. These four streams of political

transformationarerevealingofthreedeeptransformationsofthepoliticalorderinUttar

Pradesh: increasing inter-group conflicts over control of government, a growing

disjuncture between increasing political equality and persistent socio-economic

inequalitiesand the legitimizationand institutionalizationof identities–bothcasteand

religious–asthebasesforpoliticalmobilization48.Thebookhoweverstopstheshortof

offering a comprehensive overview of political changes in UP, since it barely mentions

Dalitpolitics.

The academic streams that I just described have been complemented by the electoral

analysis produced by various scholars, who have chronicled and analyzed successive

elections,generally focusingonparty strategiesandelectoralbehaviour.Post-1996, the

CSDS-Loknitisurveysenrichedthesecontributionswithsystematicsurveydata(Parmar

1996,Shankar1996,Amaresh1997,Anirudh1997,Brass1997a,Duncan1997,Pai1998b,

a)49.

One of the first limitations of this literature is that it tends to overstate the reality (or

simplifythecomplexity)ofcaste-partyalignments,easilyassumingtheequationbetween

parties’ proclaimed identities and the social composition of their electoral base. Zoya

Hasan’sQuest forPower is an illustration of this limitation. Available CSDS survey data

reveals that fewcastesvotecohesively foraspecificparty (seechapter2)and that few

partiessurvivewiththesupportofasinglesocialgroup(seechapter4).

48Op.Cit.,p.235-236.49 The main contributor to this literature in the 2000s is A.K. Verma, the resident LokniticorrespondantandU.P.scholar,basedatChristCollege,Kanpur.Seeamongothers(Verma2002a,2003,2004b,2005a,2006,2007b,a,2012b,2014a).

44

Asecondlimitationisthatmostoftheseparty-centriccontributionstendtobebasedona

macro-analysisofstatepolitics.Fewhavepaidattentiontointernaldisparities–notably

thespatial.Thereisalsolittlediscussionabouthowthesetransformationstranslateinto

local political practices. The role of crime and violence in the assertion of political

dominanceatthelocallevelisrarelymentionedinthisliterature50.

ThepoliticaleconomyofUttarPradesh

Butthemainlimitationofthe‘MandalandMandir’literaturesistheabsenceofthethird

major factor of social and political change: the market. Barring a few exceptions, the

literatureonbackwardpoliticsandreligiousmobilizations tends to focusessentiallyon

political and social factors – the role of political actors and parties – and inter-caste

dynamics. Contrary to the previous literature that examined the relations between

economic transformations and social and political change 51 , this literature almost

completelydiscardseconomicfactorsfromtheanalysis.

Although there is debate over whether the post-1991 liberalization policies have

benefited the poor in India’s backward states52, it is generally agreed that recent

economictransformationsandthetrajectoryofIndia’sgrowthhaveincreasedinequalities

(Thorat and Dubey 2012), as well as regional and intra-regional disparities

(Suryanarayana2009,Chakraborty2010,Dubey2010,ThoratandDubey2012,Singhet

al.2014,Bakshi,Chawla, andShah2015).Liberalizationhas sharpened inequalitiesnot

just between but also within groups. These inequalities – particularly within dominant

groups,suchastheJats inHaryanaorthePatidars inGujarat–havebeenthesourceof

muchofthecaste-basedviolenceinNorthernIndiainrecentyears.

Therehavebeenanumberofempiricallygroundedcontributionsbypoliticaleconomists,

political scientists and anthropologists shedding light on the connections between the

economy(includingland),casteandpolitics.

50SeeLuciaMichelutti’sreviewofJaffrelot’sSilentRevolution(Michelutti2004).51OnecanmentionFrancineFrankel’sworkonthepoliticalimpactofagrarianmodernizationandtheGreenRevolutioninparticular(Frankel1969,1971).52Foranoptimisticaccountontheeffectoftheimprovementofmaterialconditionsoncaste-baseddiscriminations,see(Kapuretal.2010)

45

Lieten and Srivastava conducted fieldwork throughout the 1990s in three districts –

Muzaffarnagar(West),RaeBareli(Awadh)andJaunpur,intheEast(LietenandSrivastava

1999). They discuss variations in types of agriculture, land ownership patterns, asset

status, educational andoccupational statusbetween castes.As far as landownership is

concerned, theymaketwoobservations.Thefirst is thatthe leading land-owningcastes

are those that traditionallyheld superior rights in these regions (Jats inMuzaffarnagar,

BrahminsandThakursinJaunpur,upper-casteHindusandMuslimsinRaeBareli)53.But

they also observe a great deal of inter-panchayat variations, due to recent social and

economictransformations.Similarly,theyobservethatsocialdominanceremainslargely

perceivedby their respondents in termsof castedynamics54.They conclude thatat the

local (Panchayat) levels, patternsof dominance are resilient55, and alsonote thatwhen

castedominanceischallenged,thedominantgroupsarelikelytotoresorttoviolenceto

maintaintheirposition.

Jens Lerche conducted fieldwork in 1993 in villages in Muzaffarnagar and Jaunpur

districts(Lerche1999).He,too,showsthetransformationoflabourrelations–duetothe

developmentof agricultural technology (tubewells andmechanization)–have loosened

theexploitativerelationofdependencythattiedlandlesslabourerstolargelyupper-caste

ordominantOBCs landowners56.Heconnects theriseof theBSPto thepoliticalagency

that rural landless labourers have gained from breaking the chains of economic

dependency(Lerche1999,213).

Despite these changes, these scholars, and others, have also illustrated how dominant

groupshavemaintainedorexpandedtheirinfluencebycontrollinglocalstateinstitutions

(Jeffrey2003,LindbergandMadsen2003,PaiandSingh1997,Singh1992).

While these contributions link economic transformations to social andpolitical change,

othershaveexaminedthe impactofpolitical transformationsondevelopmentalpolicies

and outcomes. One general argument is that focus on caste-based politics and policies

53Op.Cit.,pp.110-111.54Ibid.,p.131.55Ibid.,p.139.56Op.Cit.,p.190.

46

designed to please specific caste segments of the electorate (or implemented to that

effect)haveledsuccessivegovernmentstoneglectsomeofthebasicfunctionsofthestate

welfaremachinery,leadingtolowinvestmentsinthehealthandeducationsector,andto

a general deterioration of public services (Jeffery, Jeffrey, and Lerche 2014). The 2010

Planning Commission Report recognized that centrally sponsored schemes had not

substantiallyalleviatedpovertyinUttarPradesh(PlanningCommission2010).

In 2007, Sudha Pai co-edited a volume that sought to take stock of the more recent

transformationsthathadtakenplaceinUttarPradeshwithregardtoidentitypoliticsand

political mobilizations, as well as matters of governance and macro-economic reforms

(Pai2007).Whilethevolumecontainsusefulcontributionsonpartiesandidentitypolitics,

aswell as thedepressedmacro-economic trajectoryof the state, it doesnotbind these

questionstogetherandfallsshortofofferinganewanalyticalframeworkthataddresses

the interconnections between social, political and economic change in the post-

liberalizationperiod.

There are other ‘holes’ in the literature, such as the study of public institutions or the

political sociology of the bureaucracy. Akhil Gupta did propose a reading of the

functioningofthestate(Gupta2012), inwhichheassertsthatthecentralfeatureofthe

functioningofthelocalbureaucracyisarbitrariness,ratherthancorruption.

There isalsoa literatureonPanchayatiRaj,whichIamnotquotinghere,whichusually

focusesonaspectsofdeliveryoroutcomes,butrarelyonthepoliticalprocessesaffecting

theircompositionandfunctioning.

TheanthropologicalcritiqueandcontributiontotheUttarPradeshpoliticsliterature

Whilethereisarelativeconsensusonthedescriptionofpoliticaleventsandprocessesin

U.P.,therearedivergencesontheirsignificanceandinterpretation.TheriseoftheBSPin

particularhasgivenwaytothehopethatintheviolentpoliticalcontextoftheearly1990s,

there was at least one political party dedicated to the emancipation of India’s poorest

citizens.Theriseofbackwardclassesisalsoassociatedwithaperiodthatsawincreased

participationfromthelowersegmentsoftheelectorate,whatYogendraYadavtermedas

47

the ‘second democratic upsurge’ (Yadav 2000). Many scholars at the turn of the

Millennium agreed that Indian politics was becoming more inclusive and participatory

(Sheth1996,Weiner1997,Yadav1996a,b,2000).

In recent years, political anthropologists and sociologists have questioned the

interpretative frameworkofpoliticalchange inU.P. inherited fromthe1990s literature.

Theseauthorshaveraiseddoubts,particularlyabouttheemancipatorypotentialoflower-

caste politics by conducting local ethnographies of the processes through which these

‘new’ low-caste politicians actually emerged (Jeffrey 2001, 2002, Jeffrey, Jeffery, and

Jeffery 2008a, Kumar 2014, 2016). They have criticized the interpretation of political

scientistswhoseeintheriseoftheBSPboththesignalandthevectorofdeepgrassroot-

level lower-caste political assertion, through access to local resources, powernetworks

andculturalemancipation(Jeffrey,Jeffery,andJeffery2008a,1371-75).Whiletheydonot

deny their politicization nor the importance of symbolic victories, they point out that

political change in thepost-liberalization context has “tended to bolster thepositionof

dominantsectionsofsociety”(Jeffrey,Jeffery,andJeffery2008a,1371).

IntheirstudyofnewDalitpoliticiansinBijnordistrict, JefferyandJeffreyshowthatthe

riseoftheBSPhascontributedtotheemergenceofanewclassofskillfulDalitpoliticians,

although“theriseoflower-castepartieshasnotresultedinincreasedleverageandpolitical

powerforDalitsontheground”57,ashasbeenthecaseinSouthIndia.Theyalsonotethat

theriseoftheOBCsmostlybenefitedtheiruppersectionsaswellastheJats,whocould

furtherenhancetheircontroloverlandholdingandtheirinfluenceoverlocalgovernment

bodies,throughtheiralignmentwiththeSamajwadiParty58.

They have asserted further that the emergence of ‘backward politicians’ cannot be

interpreted in isolation as the by-product of deep and organized caste-based social

movement, or be seen as the spearheadof emancipatory transformativepolitics. These

new politicians emerge from a political milieu and context that are deeply embedded

withinlocalsocio-economiccontexts,whicharemarkedbyharshcompetitiveness,crime,

violenceandcorruption,aswellastheabsenceofprospectsfortheyouthbesidespolitics

(Jeffrey2010b,Jeffrey,Jeffery,andJeffery2008b).

57Op.cit.,p.1390.58Ibid.,p.1369.

48

Corruption and violence emerge as two registers widely used by dominant groups to

preserve their social status and control over local territories, as well as to counter the

assertion of subaltern groups. Craig Jeffrey showed how many rich farmers in rural

Meerutdistrictmaintaintheirpositionbycolludingwithlocalstateofficials(Jeffrey2002).

Earlieron,hemadeanimportantcontributionwhenhearguedthatthereproductionof

socialinequalitiesremainedmediatedbycasteandthephenomenonofcastedominance,

inconnectionwithotheraxesofpower,suchasclass(Jeffrey2001).

What emerges from this literature is a portrayal of a polity undergoing deep

transformationsbutnotnecessarilyheadedinasingledirection.Inhisethnographyofthe

ruraleconomy,SatendraKumarshowshowthediversificationoftheruraleconomyhas

leadtoamorehorizontalpoliticalcompetition,ortheemergenceofahighlycompetitive

environmentbetweencontendingcastegroups(Kumar2014,2016).

It is to be noted that these contributions are almost exclusively based on fieldwork

conducted inWesternU.P., indistrictsadjacent toDelhi,or inadjacentRohilkhand, two

sub-regionscomparativelymoredynamicandprosperousthantherestofthestate.More

comparative ethnographies across the state are needed if one were to generalize their

conclusionsatthestatelevel.

The U.P. Literature in the 2000s: the triple tropes of violence, identity, andpatronage.

Most of the literature described so far consists in descriptive and analytical account of

political processes and transformations. Barring the contributions of Paul Brass on the

analysisofethnicviolence(seebelow),fewhaveattemptedtolinktheirempiricalworkto

larger theoretical issues. Post-2000, a series of scholars have used the work they

conductedinUttarPradeshoronUttarPradeshtobuildgeneralargumentsabouteither

Indianpoliticsand/orbroadertheoreticalacademicdebates,ontheissuesofviolenceand

crime,identitypoliticsandpatronage.

Politicalviolenceandthecriminalizationofpolitics.

49

TheissueofviolenceinU.P.politicsisn’tanewdebate,eventhoughRudolphnotesthatit

took a long time for political scientists to address the issue of post-Partition political

violence (Rudolph and Rudolph 2010, 571). The state of Uttar Pradesh has seen many

episodesofcommunalviolenceafterIndependence,particularlyinthe1980s,whencities

suchasMeerut,AligarhandMoradabad,were regularlyhitby large-scalepogromsand

riots59. In the late 1980s, theRam Janmabhoomi issue accentuated communal tensions

and violence across the state, which peaked in the aftermath of the Babri Masjid

demolition.

In the early 2000s, three authors – Wilkinson, Varshney and Brass – proposed three

different interpretations on the causes of communal violence. Wilkinson details the

electoral incentives that lead parties in power to prevent the spread of communal

violence or incite it, stating that communal violence is not an issue of state capacity

(Wilkinson2004,2005,Wilkinson2006).Varshneyarguesthatcommunalviolencecanbe

preventedwheninter-religiouslocalassociativetiespre-exist,quashingcommunalsparks

beforetheyflareintoconflagrations(Varshney1996,2001).

WhileWilkinsonandVarshneyuseacommondatasetcompilingtheoccurrenceofriotsin

India (Varshney and Wilkinson 2006), Brass bases his analysis on extensive fieldwork

conducted inAligarh andMeerut, over aperiodof twenty years.The three authors are

concernedchieflywithinter-religiousviolenceoccurringincitiesinthespecificcontextof

politicalcompetition.

In The Production of Hindu-Muslim Violence in India (Brass 2003), based on extensive

fieldworkconductedinthecityofAligarh,Brassformalizessomeoftheargumentsthathe

haddevelopedinhisearlierethnographyofriotcasesinUttarPradesh(Brass1997b).He

seekstounderstand,amongothers,whyriotspersist,whataccountsforvariationsintime

and space, and who stands to gain from routinized communal violence. The planning

required to organize communal riots and the institutionalization and banalization of

communal violence in every day life are part of what he calls an institutionalized riot

system (IRS), which sets a context propitious to the activation of social tensions into

violenceinperiodsofmobilizationoratthetimeofelections(Brass2004,4839).

59Foradetailedaccount anddescriptionof communal riots inpost-IndependenceUttarPradesh,see(GalonnierandGraff2013).

50

While sparkling a vivid intellectual debate, these contributions have their limitations,

particularlyinthecaseofVarshneyandWilkinson,whobasetheiranalysisonacommon

datasetcodinginstancesofcommunalriots(onlywheninvolvingfatalities),tracedfroma

single source (the national edition of the English daily The Times of India)60. Brass in

particularhascriticizedVarshney’sargumentonthegroundsthatinter-religiouscivicties

cannotstandinthefaceofthepoliticalwilltocreatecommunalviolence(Brass2004).

BrassandWilkinsonbothagreeonthepoliticaloriginofcommunalviolence.Whiletheir

respectiveworksfocusoncities,thereisalsoplentyofevidencepointingtothefactthat

ruralareasarealsosubjected to frequentcommunaleruptions.BadriNarayanrefers to

thiswhenhedescribeswhathereferstoasa‘phenomenonofsmallriots’(Narayan2014),

or the occurrence of episodes of small-scale communal violence at short and repeated

intervals,whichnurtureaclimateofconstanttensionbetweenreligiouscommunities.In

2014, police data compiled for an Indian Express investigation showed that in the ten

weeks that followed the 2014 General Elections, 605 communal incidents took place,

mostlyinruralareas.Two-thirdsoftheseincidentstookplaceinoraroundconstituencies

tabledtogoforabye-election,aftertheelectionoftheirMLAstotheLokSabha(Suresh

2014a).Theinvestigationfurtherrevealedthatthetriggersofviolencewereoftenbanal

neighborhoodissuesinflatedatthebehestoflocalpoliticians(Suresh2014b).

Thisliteraturehowevertendstofocusonaparticulartypeofviolence–inter-religious–

atthecostofscrutinyonotherformsofroutinizedviolence,notablycaste-based.Thereis

plentyofevidencethatU.P.’sdailylifeismarkedbypervasivesocialtensionsandviolence,

domesticorcaste-based.MyownfieldworkinLucknowwasfrequentlyinterruptedbythe

impositionofcurfewsinsituationsofstreetviolencethatseldomattractednationalmedia

attention.During the fieldwork conducted inWesternU.P. in the fall of 2012, I seldom

encounteredvillagesthatdidnothavearecenthistoryofviolenceandmurder,linkedto

family disputes, caste feuds or conflicts over land. Factional politics within dominant

communitiesorbetween(andattimeswithin)dominantlocalpoliticalfamiliesalsolead

tofrequentclashesandactsofviolence.

60Thisexplainswhytheirdatalistessentiallyriotsoccuringincities.

51

Socialandeconomicchangeaffectsthebalanceofpowerbetweencastes.Theassertionof

lowercastesisoftenlocallymetwithviolence,perpetratedbymembersofdominantlocal

groupsthatseektomaintaintheirdominance(Jodhka2015).

Further,thequestionoftiesbetweenpoliticsandviolenceis furthercomplicatedbythe

social legitimacy that a number of criminal politicians enjoy. Two recent contributions

haveunderlined thedepthof thenexusbetweenviolenceandelectoralpoliticsand the

sociallegitimacyattachedtoit.

In his 2012 dissertation and in prior publications, Milan Vaishnav has laid down the

motivesthatpushvotersandpartiestochoosecandidateswithapubliclyknowncriminal

record(Vaishnav2011,2012).Votersmayrationallychoosetosupportsuchcandidates

astheyareperceivedtobemorecredibleprotectorsofgroup-basedinterests.Partiesare

also incentivised to field ‘tainted’ candidatesbecause theresources theypossess–both

financial and criminal – give them a competitive advantage over ‘clean’ or ‘cleaner’

opponents.

Usingamoreanthropologicalstandpoint,Micheluttidescribesandanalyseshowtheself-

presentationofpoliticiansandtheircampaigningstyleisimbuedwithviolentmasculine

references,andwhysomesegmentsoftheelectoraterespondpositivelytothetropesof

‘muscularpolitics’(Michelutti2010,2014).

Party-voterslinkages,orthetropesofpatronageandclientelism

Thesecond themeof thepost-2000 literaturedealswithparty-voters linkages.Muchof

theliteratureonthatsubjectdefinestherelationbetweenparties,politiciansandvoters

as transactional, that is to say based on the exchange of commoditized goods between

holdersofpositionsofpowerandvoters,whohavetheirsupporttoofferinexchangeof

thosegoods,orthepromiseofthedistributionofthosegoods.Theliteratureonpatronage

andclientelismshowstheformsthesetransactionshavetakenandtheconditionsunder

52

which patronage or clientelism operate (Brass 1964b, Chandra 2004c, 2009, Cox and

McCubbins1986,KitscheltandWilkinson2007,Piliavsky2014)61.

DuringthefirsttwodecadesafterIndependence,theCongress’dominationrestedonthe

systemicpatronageoflocalnotablesandontheorganizationofexchangeofvotesagainst

benefitswithintheframeworkoflocaltraditionalformsofauthority(Singh2002).Brass

calledpatronage,alongside factionalismandadministration, the ‘substance’ofCongress

power(Brass,1964:212).

WiththeCongress’dominationchallenged,patronageandclientelismincreasedduetothe

higher competitivenessof theparty system (Kitschelt andWilkinson2007). Candidates

and parties compete for votes on the basis of their ability to meet the demands and

expectationsofvoters,whichinturncreatesacontextofcompetitivepatronage,inwhich

parties and candidates attempt to outdo each other with the promise of or the actual

redistribution of resources. The greater availability of resources – public and private –

after liberalization increased further the scope for patronage. As Wilkinson puts it, the

costofclientelismisincreasing62.

In the Indian context, clientelistic relations are often organized, or mediated, through

caste.Castesworkasinterestgroupsandthusofferindividualstheopportunitytoweigh

collectivelyintheirnegotiationwithpowermongerswhoprefertoengagewithcollective

entities,suchascastesorcasteblocks,ratherthanaheterogeneousgroupofindividuals.

InWhyEthnicPartiesSucceed,KanchanChandraargues that, inpatronagedemocracies,

local clientelistic networks prevail over policies or other possible determinants of

electoralbehaviour,sincevotersareunderinformationconstraints(Chandra2004b).The

elusivenessof ‘state-leddevelopment’alsoencouragesvoters to turn to localpatronage

networks rather than expect benefits from state policies. Chandra also underlines that

61These authors use the terms “clientelism” and “patronage” interchangeably. It designates aparticularmodeofexchange,orcontractualrelationbetweenvotersandpoliticians,characterizedby the personalization of the relation (direct exchanges), the exchange of particular goods(resources versus electoral support) and dependent from control mechanisms. For a completedefinition, see (Kitschelt and Wilkinson 2007, 7-23). Patronage is usually contrasted with“programmatic politics”, which are supposed to be framed regardless of specific individualinterests.62Op.Cit.,p.112.

53

elections in Indiahavebecomesubjected toauctionsor tradeofpublic servicesagainst

the support of segments of the electorate (Chandra 2004a). Voters thus form their

politicalchoicesaccordingtothematerialbenefits theyexpect toderive fromtheactof

voting(Kothari1964,1974)63.

ThedepictionofIndiaasa‘patronagedemocracy’hasbeencriticizedonseveralaccounts,

notablyforitsgeneralized‘blanket’character,andforitsexaggerationoftheeffectiveness

ofpatronagenetworksinsecuringvotes.

Aclientelisticrelationimpliestheexistenceofadirectexchange,oradirectrelationship

between the patron and the client (Kitschelt and Wilkinson 2007, 10). Ahuja and

Chhibber point that politicians cannot possibly reach every voter in their constituency

(Ahuja and Chhibber 2010). They also find that different categories of voters entertain

differentiatedexpectationsfromtheirrepresentativesandthat“asubstantialchunkofthe

voters in Indiavotewithoutany tangibleexpectation in termsof improvedaccess tostate

servicesorprivatebenefit”64.ThissupportsMukulikaBanerjee’sfindingthatpoorvoters

participatemoreinelectionsbecauseoftheirattachmenttotheactofvoting,beforeeven

thequestionofchoicearises.Forpoorruralvoters,theactofvotingrepresentsavaluable

andtangibleexperienceofpoliticalandsocialequality,aconsiderationthatprevailsover

material benefits (Banerjee 2014). Much of the literature on patronage democracy

assumes that “there is a uniformity in the relationship of the State to its citizens and,

furthermore, that all citizens view their relationship with the State similarly” (Banerjee

2014,2).

Similarly, many groups find themselves excluded de facto from patronage networks,

particularly, the lower castesandminorities.There is alsoamyriadof small, dispersed

castes that do not weigh much politically and therefore do not appear in the

configurationsofvarioussocialalliancesdesignedbypartiesandcandidates.

63She, however, moderates this view by stating that the increase of the welfare state capacitieshavereducedthecentralityandroleofintermediariesinprovidingaccesstosocialbenefits.Accesstothesesocialbenefits,however,remainsunequalandsubjecttovariousformsofcorruptionandextortionpractices.64op.cit..p.6.

54

Furthermore, critiquesof thepatronagedemocracy argumentunderline thatpoliticians

arenotthesolemediatorsbetweenvotersandtheStateandthatmostvotershavetodeal

directly with the local administration in their day-to-day life. As such, politicians’

credibility is low and so is the faith in their capacity to deliver on their promises

(Banerjee2014,6).

Otherscholarshavearguedthatthetraditionalpatronagelinkagesdonotoperateasthey

used to in a liberalized economy, which has created more avenues for upward social

mobility (Jenkins 2005, Manor 2010). Jenkins and Manor point out that maintaining

patronagenetworksdoesnotpreventincumbentsfromlosingelections.Structurally,itis

probablynotpossible tomeeteveryrequestorexpectation inconstituencies thatcount

over a million inhabitants on an average. As Bailey noted in the 1950s in Orissa,

candidatesare individualswhohave limitedmeansat theirdisposalwithwhich togain

theirends(Bailey,2001:35).

Othercritiquespointat thecontradictionbetween thenecessityof redistributinggoods

andthe imperativeofpreyingonone’sconstituency inordertoraisesufficient fundsto

enterandlastinpolitics.

Thesecritiquesdonotdenytheexistenceofpatronagenetworksassuch,butcastdoubts

overtheireffectivenessindeterminingelectoraloutcomesoreventheindividualfatesof

politicians.Insuchanenvironmentoframpantrivalry,thereislittle,ifnotnocontrolon

howindividualvoterswillbehaveinsidethepollingbooth.AsAhujaandChhibberputit:

“Patronagenetworksdoexist,buttheconsumersoftheservicesofsuchnetworksarelimited

innumber”65.

Recent contributions underline the fact that patronage linkages have the effect of

maintaining existing patterns of social domination, despite democratization. In an

ethnographicarticleonthe2009StateelectionsinAndhraPradesh,CarolynElliottnotes

thatuppercasteshavebeenabletoretaintheirpoliticalinfluencethroughtheclientelistic

redistribution of welfare and patronage benefits to marginalized segments of the

electorate(Elliott2011).Similarly,TariqThachildemonstrateshowtheBJPmadeinroads

65Ibid.,p.6.

55

among subaltern groups in Jharkhand and Chhattisgarh by organizing and operating

social services targeted at these groups (Thachil 2014). The expansion of clientelistic

networks is also away for parties to extend their support baseby focusing on specific

populationsbeyondtheircoresupportbase.

From the perspective of politicians, forging and maintaining patronage networks is an

essentialmeans tobuildandretainelectoral support. Inhis surveyofMLAsacross five

states, Chopra finds that 47%ofU.P.MLAsdefine theirmain role as attending to their

constituentsandworkingon thedevelopmentof their constituency,beforeworking for

theirparty (21%)orworkingon thedevelopmentof thestateasawhole (22%).Nota

singleMLAsurveyed(67intotal)mentionedAssemblywork(Chopra,1996:151).

Finally, the notion of clientelism usually implies a relationship that is not only

transactionalbutalsoasymmetrical.Patronsandclientsdonotstandonafootofequality.

Evidence from the ground often points to the contrary. Local groups of voters tend to

negotiate with several candidates wooing them for electoral support, in a sort of

competitive bidding game. In his survey of 408 villages across Rajasthan and Madhya

Pradesh, Anirudh Krishna finds that the spread of education, the increased political

competitionandtheexpansionofstateruralexpenditurehavecontributedtodisconnect

intermediation from closed kinship ties (Krishna 2003). As a consequence, patronage

networksarefarfrombeingstable.Infact,theyarequitefluid.

IagreewithChhibberandAhuja’sassessmentofthelimitedimpactofdirectclientelistic

networks (particularly in the case of Lok Sabha elections, in which voters per

constituencyarecountedinmillions).ButIwouldaddthatclientelisticnetworkscontinue

to be of primordial importance for candidates and elected representatives, since they

represent the main way through which they can relate with their constituents. My

experienceinU.P. indicatesthat,whilepatronagenetworksmaynotbesufficienttowin

an election, they are nonetheless a necessary condition to be competitive. In a way,

buildingpatronagenetworks–anddevelopingacapacityofredistribution–ispartofthe

pricethatanaspiringcandidatemustpayinordertobeabletowinaseat.

These three tropes – violence, identity and patronage do not exhaust the range of

questions that have been asked, or that can be asked aboutUttar Pradesh politics. But

56

theyconstitutethethreemainacademicframeworksunderwhichU.P.statepoliticshas

beenstudied.

Theseframeworksneedtobere-visitedorupdated.Muchoftherecentliteratureisstill

focusedonthepoliticsofthe1990s,orusesthepoliticsofthe1990sasagridtoanalyze

contemporary UP politics. Few have attempted to formulate or adapt their analytical

frameworktothemorerecentperiod,orpaidattentiontosub-regionalvariations.

1.6.Overviewofdissertation

Thisdissertationhasbeendivided into six chapters. In this first introductorychapter, I

have described the main arguments that will unfold in the subsequent chapters, and

providedadescriptionofthemethodsfollowedtoinvestigatetheunderlyingquestions.

Chapter2providesapoliticalandpoliticaleconomycontexttothequestionsraisedinthis

dissertation. It sketches in broad terms the main lines of transformation of the state’s

politics, of its party system, of the trajectory of its main political actors, while paying

attention other broad transformations in the state’s political economy. This chapter

focusesonsub-regionalvariations,athemethatcontinuesinsubsequentparts.

Chapter 3 deals with the evolution of the profession of politics and the changing

conditionsofelectoralcompetitioninUttarPradesh.Themainargumentinthissectionis

that the rules of political engagement – to borrow F.G. Bailey’s expression – have a

filtering impactonwhoaspirestocontestandwhogets towin.Someof theserulesare

inducedbytheelectoralsystemitself,bytheoverallcompetitivenessofelectoralpolitics

and by some specific features of electoral competition, party politics, and by voters’

expectations and behaviour. They create a universe of constraints and pressures that

weigh on candidates and parties. Thereafter, it examines how the literature on Uttar

Pradeshpoliticshassoughttomakesenseofthesetransformations,byreviewingsomeof

the major questions that have been raised and some of the answers that have been

advancedbyscholars.

57

Chapter 4 makes a quantitative examination of the evolution of the sociological

composition of the state assembly. This chapter aims at criticizing aggregative head

counting approaches, opting, instead, to dissect data into various levels of observation:

jati-wise,party-wise,candidate-wise,aswellassub-regionalvariations.It looksatother

availabledataonthesocio-economicprofileofMLAs,notablyeducationandoccupation,

toconcludeonthemeritsandlimitationsofthisapproach.

Chapter5aimsatcontextualizingthequestionofthechangingprofileofMLAsatamore

relevantlevelofobservation.Anyworkontheroleofcasteinpoliticsshouldfocusona

qualitative assessment of how caste exerts and manifests itself in the most tangible

manner,which is locally. In this dissertation, local essentiallymeans constituency-level

observation,althoughnecessarilyconductedinvariouslocationswithinconstituencies.

Chapter6aims toexplainwhyasimilarbroadcontextofsocial,political,andeconomic

change has resulted in varied trajectories for different parties. This has been done by

comparingtheperformanceofthefourmainpartiesoverthepasttwentytotwenty-five

yearsandbycomparing theirevolution, in termsoforganization,electoralstrategyand

relationwith thenewelitesof thestates.Parties in India–especiallyregionalparties–

are reputed tobeweakorganizationsheadedby strong individuals.Many insist on the

organizational, institutionalweaknessesofparties,onthehyper-centralizationofpower

within their organizations, on the lack of internal democracy (Chhibber, Jensenius, and

Suryanarayan2014).Whilethisislargelytrue,regionalpartiesalsotendtoexertpower

and influence outside the realm of their formal organizations, developing and at times

controllingformalandinformaltieswithnetworksofindividuals,groupsandinstitutions

who in their own right and domains exert some form of influence. Parties must be

analyzed as part of larger formal and informal networks, which cover institutions,

political familiesorthe localdominationofparticulargroupsoverparticularterritories.

The fieldwork has been attentive to how parties build their networks, both local and

supra-local.

Thesecondpartofchapter6servesasconclusionandoffersananalyticalframeworkthat

characterizesstateelectoralpolitics,itstransformations,andwhatelectoralandpartisan

practicesrevealaboutthepolityatlarge.Inparticular,itreflectsonthechangingplaceof

58

caste in state electoral politics and on the meaning of these transformations for the

emancipatorypromiseofbackward-classmobilization.

The days of horizontal mobilization of backward castes – excluding Dalits – may be

behind us for the time being, owing to the localization and marketization of electoral

politicsandtheevolutionsofparties’electoralstrategies,butitdoesnotmeanthatcaste

has disappeared altogether as a vehicle ofmobilization. In fact, caste gets reconfigured

oncemoreby theworkingsof electoralpolitics, in the sense that it is subsumedunder

local political economic contexts. In this sense, it confirms Sudipto Kaviraj’s statement

that elections in India have "constantly reconfirmed ordinary people's community

orientationinsteadofunderminingit"66.Thelocalentrenchmentofcastepoliticsaffectsin

turn how parties and candidates conduct themselves, frame strategies, and ultimately

approachtheelectorate.

66SeeS.Kaviraj'scontributiontoF.Frankel,R.Bhargava,Z.HasanandB.Arora,TransformingIndia.SocialandPoliticalDynamicsofDemocracy: "Democracy and Inequality",Delhi,OxfordUniversityPress,2000,pp.89-119.

59

Part.I–DecodingUttarPradeshPolitics

The state of Uttar Pradesh occupies a particular place in national politics, owing to its

demography (199.8 million inhabitants, 16.5 per cent of India’s population)67and the

place it occupies in the national political imaginary. Eight of India’s fourteen Prime

Ministerhavecome fromthis state68,whichcurrently sends80parliamentarians to the

LokSabha(14.7percentofthetotalseatsoftheLowerHouse)and31totheRajyaSabha

(12.6percentoftheseatsoftheUpperHouse).

Beyond its demographic size and political weight, Uttar Pradesh holds particular

importancesincethemainfaultlinesofIndianpoliticsmanifestthemselvesinthisstate,

often in explicit andexacerbated forms.The state’s recentpoliticshasbeenprofoundly

markedbyreligiousandlower-castemobilizations,whichhaveledtotheriseoftheBJP

andstate-basedparties.It isinUttarPradeshthataDalitwomanbecameChiefMinister

on fouroccasions, at theheadof India’smost successfulDalitparty, theBahujanSamaj

Party.

SomeofthemajorpoliticaleventsthathavetakenplaceinUttarPradeshhavehadnation-

widerepercussions,suchastheuppercasteanti-reservationsprotestsintheearly1990s

orthedestructionoftheBabriMasjidatAyodhya,on6December1992.

The state is often imagined as a “heartland”, either on a linguistic (Hindi and Urdu),

religiousorhistoricalbasis.Thestate is also the siteofmanyeventsand locations that

havemarkedIndia’s liberation’sgest, fromthe1857SepoyMutinytotheIndependence

movement.ItholdsparticularsignificanceforMuslimsaswellasforpracticingHindus69.

67Source: Census 2011. Uttar Pradesh’s population grew by 20.23 per cent since the previousCensus,in2001.68Asnotedearlier, the currentPrimeMinister,NarendraModi, chose to retain theVaranasi seat,havingalsosuccessfullycontestedfromVadodara,Gujarat.69InRegion,Nation, ‘Heartland’, Gyanesh Kudaisya deconstructs this notion of UP as a heartlandanddisintguishesfivewaysinwhichthestateasbeendefinedandimaginedasIndia’sheartland:aBritish colonial definition of the ‘model province’, as the site of the freedom struggle by thenationalistmovement,asapowerhouseofMuslimpolitics,untilthePartition,asaHinduheartlandorbattlegroundforHindutvaforces,andfinallyastherepositoryofa ‘compositeculture,anotherpost-colonialconstructionintunewiththeCongressprojectofnationalcommunitybuilding(Bailey2001).

60

ForMuslims, there isa concentrationofmanysitesofnational importance in thestate:

the Aligarh Muslim University (AMU), the Darul Uloom Islamic School of Deoband, the

NadwaSeminaryinLucknow,theShibliCollegeofAzamgarh,theGhaziMiyanDargahin

Bahraichandthewell-knownBabriMasjidinAyodhya.Thenotionofheartlandthatcan

be criticizedasanex-antehistorical andpolitical constructionbut it keepsnonetheless

saliencyinpublicimaginaries(Jha2010).

Insocio-economicterms,thestateofUttarPradeshranksatthebottomofmostindicators

of thehumandevelopment indexand is known, alongsideBihar, asoneof India’smost

backwardstates.Itspopulationismainlyrural–77.7percentinthe2011Census),with

highconcentrationofpoverty–29.4percentof thepopulationwasofficiallybelowthe

povertylinein201170.In2011,theliteracyratestoodat67.68percent71,morethansix

per cent lower than the national average. In the period of 2006-2010, the average life

expectancy was 62.7 years, nearly four years below the national average, and nearly

twelve years lower than India’s best performing state, Kerala. Gender disparities are

prevalent in most of these indicators. Twenty percentage points separate men and

womeninliteracy.Amongchildrenbelowsix,therewere902girlsforathousandboysin

201172.

Despite the prevalence of poverty, the state has undergone deep economic

transformations over the past twenty-five years, particularly its rural economy. The

state’s GDP ranks second in the country, at 6.8 lakh crores rupees73, but its per capita

70Againstanationalaverageat25.7percent.ThenumberofBPLpersonsinruralandurbanareasisrespectively30.41percentand26.06percent.Source:LokSabhaUnstarredQuestionNo.289,dated11.07.2014.ThemeasurementofpovertyinIndiaissubjecttointensedebate.Foranuancedyetcriticalassessmentofpovertymeasurementmethods,see(Kudaisya2006).71For the UP population above seven years of age, and with a ten per cent gap between ruralliteracy(65.46percent)andurban literacy(75.14percent).Maleandfemale literacyalsovariesgreatly(77.28percentagainst57.18percent).Thegendergapisnearly23%inruralareas(76.33percentagainst53.65percent)and11percentinurbanareas(80.45percentagainst69.22percent.SeeCensus2011).72The overall sex ratio has improved since the previous census, but not among the youngestpopulation,whichisasourceofconcern.Thesexratioismoreskewedinurbanareas(885)thaninruralareas(918),forthisagebracket.Overall,thestateranks26thinIndia.73About90billionEuros.

61

incomewasonly29,785rupees in2011, ranked31st in India, justaheadof the stateof

Bihar74.

Although UP’s rural economy has become diversified, the process has not taken place

uniformlyacrosstheterritory.Intra-regionaldisparitieshavegrown,particularlyduetoa

fasterpaceofgrowthanddevelopmentinWesternUttarPradeshandpartsofDoab,the

state’smosturbanizedsub-regions.Industrializationhasstagnatedoverthepastdecade,

growth being pulled by the tertiary sector, agro-business and construction-related

activities.

Thestate’spublic life isconsideredasbeingvastlycriminalized,markedby lawlessness

andimpunityforperpetratorsofviolence.Partiesareseenasshelteringperpetratorsof

violence,whothemselvesoftencontestelections(Vaishnav2011).

Thestate’spoliticsalsoremainsdominatedbycaste,notthe least inpopular imaginary.

Alongside its neighbour, Bihar, with whom it shares many of the abovementioned

features, Uttar Pradesh’s politics is perhaps more explicitly shaped by caste and caste

mobilizationsthananyotherstate,wherethecastecalculustendstobeconcealedbehind

ideologicalorregionalistdiscourses75.

Inter-partycompetitionhasbeenandremainsthemeetingpointandpointofconflictfor

socio-economic transformations. The fate of national parties often plays out in Uttar

Pradesh.Itsregionalpartiesareimportantforthebalanceofnationalcoalitions,although

itstwomainregionalparties,theSamajwadiPartyandtheBahujanSamajParty,havenot

participatedingovernmentcoalitionsattheCentre76. Inotherwords,UttarPradeshhas

been and remains an epicenter formany of the phenomena that havemarked national

politics.Thisisnottosaythatthesephenomenadidnotmanifestthemselveselsewhere,

butthattheyareallfoundinthisparticularstate,ofteninexacerbatedforms.

74Calculatedatconstantcost.Source:ReserveBankofIndia.75PrernaSinghhasarguedthatthelackofregionalidentityandthepoliticizationofsocialdivisionsin U.P. have contributed to poor outcomes in matters of social development and policies (Singh2015).Othershaveargued that the identificationof theU.P.with thenation led to theneglectofstate-leveldevelopmentalissues(Zerinini-Brotel1998).76WiththeexceptionofthesoicalistsintheJanataPartycoalitionin1977.

62

Chapter2–UttarPradeshpolitics:ahistoricaloverview

ThissectiontracesthefourphasesoftransformationofthepartysysteminUttarPradesh

and gives an account of the intra- and inter-party dynamics that have contributed to

changesintheconfigurationofpoliticalcompetitioninUttarPradesh.Thesecondpartof

thischapterfurtherdetailsthespatialdimensionofthesetransformations.

2.1.ThefourphasesofUttarPradesh’spartysystem:fromdominancetofragmentation

TheevolutionofthepartysysteminUttarPradeshisclassicallydividedintofourphases.

The first phase is a period of the Congress’ domination of a fragmented multi-partisan

scene, fromthe firstelections in1951 to themid-1960s, followedbya secondphaseof

that sees the rise of opposition forces. This phase, which also witnessed the Congress

beingsplitandcoalitiongovernmentsemerginginvariousstates,particularlyinthe1967

elections,culminatedintheimpositionoftheEmergencyandtheinstallationofthefirst

non-CongressstategovernmentinUttarPradesh,in1977.

Thethirdphase,throughthe1980s,seesthereturnoftheCongresstopower,andtherise

ofthreeparties: theBJP, theBSPandvarioussocialist formations.Thisphasepavedthe

wayforafourthphase,thatofoffour-partdivisionofthepartysystemthroughthe1990s,

a phase further marked by caste and religious polarization. The fourth phase is also a

phase of chronic governmental instability, as the distribution of votes among the four

partiespreventedanyofthemfromformingagovernmentontheirown,andtheintense

inter-party rivalry and social polarization that mark electoral competition doomed all

coalitionstofailure.

The situation stabilizes in the 2000s, as the national parties decline and the regional

parties shed their traditional upper castes antagonism to embrace more inclusive

discoursesandelectoralstrategies.

63

2.1.1.Congressdominationandtheriseoftheopposition77

The Congress Party ruled Uttar Pradesh virtually unchallenged in the state’s first two

elections,in1951and1957.Itwon90percentoftheseatswith47.9percentofthevotes

in1951,and66.5percentoftheseatswith42.4percentofthevotesin1957,inamulti-

partisan and fragmented party system. In the 1960s, it succeeded in obtaining single

majoritiesofseatswitharoundathirdofthevotes78.ThesuccessoftheCongressParty

lay in its capacity to incorporate political competitionwithin its ranks, and to draw its

supportfromtraditionallydominantuppercastegroupsaswellasfromthemarginalized

segments of the population – the lower castes and theMuslims.What connected these

groups to the party was its ability to include the local traditional notabilities into

patronage networks and use them in turn to mobilize poorer voters (Kochanek 1966,

1968,Kothari1964,1967).

The second phase starts in the mid-1960s, when various socio-political oppositional

movementsemergeandstartcoordinatingtheireffortsinordertochallengetheCongress’

dominant position (Brass 1980a, b, 1981, Duncan 1988). The opposition remained

polarizedthroughthe1960sbetweenvariousbrandsofsocialistpartiesthathademerged

asbreakawayfactionsoftheCongressinthelate1950sandearly1960s;theJanaSangh,a

HindunationalistpartythathademergedfromtheRSS(Baxter1971);andtheRepublican

Partyof India,abranchofAmbedkar’sMaharashtra-basedRPI,whichhada short-lived

existence inNorthern India.TheCommunistsalsohadsomepresence inUttarPradesh,

particularlyintheeast,indistrictsadjacenttoBihar,aswellasinBundelkhand.

Therewereotherpoliticalformationsintheopposition,whoinitiallydidwellbutquickly

faded the moment large alliances started to emerge. Among those parties was the

Swatantra Party, founded in 1959 by C. Rajagopalachari, a figure of the liberation

77I have drawn heavily in this section from Paul Brass’s various meticulous accounts of theseevents(Himanshu2010).78Thisdissertationpresentsstateelectionresultsstartingfrom1962,andnot1952.Thereasonforthis is that the first two elections followed a dual/triple-member constituency system, in whichreserved seats for Scheduled Castes and for Scheduled Tribe candidates were added to existingseats,ratherthanattributedseparately.IntheElectionCommission’sstatisticalreports,theresultsobtainedbythesevariouscandidatesinsimilarseatsareundifferentiated,whichmakesthetaskofbuildingelectoraltimelineverydifficult.Onthequestionofdual/triple-memberconstituencies,see(Brass1980a,b,1984a,1985,2011).

64

movement,India’slastGovernor-GeneralandaformerCongressman.Itwasconceivedas

aright-wing,ornon-leftistalternativetotheCongressParty.TheSwatantraPartyleaders

tookonthemselvesthetasktomergevariousanti-Congressforceswithinitsfold–such

as S.K.D. Paliwal’s Independent Progressive Legislature Party (IPLP), a group of

independentlegislatorsstandingbetweentheLeftandtheHinduright–andtoadvocate

foroppositionunity(Erdman1967).Theycouldnot,however,breakthestrangleholdof

factionalism and rival ambitions that divided the socialist camp. Subsequently, its

numbersdwindledaftertwogoodperformancesinthe1962and1967elections79.

Thedysfunctionalriseofthesocialists

The socialists represented the main opposition to the Congress in the 1960s and early

1970s, but their divisions prevented them from effectively challenging the Congress’

supremacy in the state assembly. They remained divided until the founding of the

Bharatiya Lok Dal by Charan Singh in 1974. Prior to that, various streams of socialist

partiescompetedagainsttheCongressbutinascatteredmanner.Therehadbeenvarious

attempts at merging the various socialist factions in the state, but they failed due to

ineffectiveleadershipandthesuccessoftheCongressPartyinco-optingsocialistcadres.

The Uttar Pradesh socialists also belonged to national formations, which were divided

nationally, on doctrinal matters and over disagreements on alliance strategies80. These

nationallinesoffracturestranslatedintostate-levelrifts.

ThesocialistsinNorthernIndiahadinitiallyemergedasafactionoftheCongressParty–

the Congress Socialist Party – created by Jay Prakash Narayan and Narendra Deva in

Patna in 1934. Its aims were to mobilize the lower peasantry against landlordism, to

promoteequitablelandreforms(Jaffrelot2003b,253)aswellasthesocializationofbasic

industries.ItbrokeawayfromtheCongressin1948,yearofthecreationoftheSocialist

79TheSwatantrawasalsocrippledbythefactthatitrapidlylostitstwomainfigures.S.K.Paliwallefttheparty–asdidRajaRaghavendraPratapSinghofManakpurin1964.80By then the quest for socialist unity was already on the wane. The first breakaway from theCongresstookplacein1948,ledbyAshokMehtaandRamManoharLohia.TheyformedtheKisanMazdoorPrajaParty(KMPP)threeyearslater.Subsequenttoitsdefeatin1952,theKMPPmergedwiththreeothersocialistformationstoformthePrajaSocialistParty,whichLohiaquicklyleft.ForanaccountofthemotivesbehindthissplitandthecontextoftheSSP’semergence,see(Schoenfeld1965).

65

Party(SP).TheSPbecamethePrajaSocialistParty(PSP)in1954,afteramergerwiththe

KisanMazdoorPrajaParty(KMPP)ofJ.B.Kripalani,anotherdisgruntledex-Congressman.

ThePSPwas ledby itsGeneralSecretary,RamManoharLohia,oneof therising figures

among the socialists leaders.Within twoyears, thePSPsplit twice, the first timeat the

initiative of Ram Manohar Lohia himself who, alongside Madhu Limaye, left the PSP in

1955 to form a new Socialist Party81. The second split took place in 1962, after the

expulsionofAshokMehta,whowasheldresponsibleforthelargenumberofdefectionsof

socialistcadrestotheCongress.

Thedefeatof the socialists in the1962elections led their leaders towork towards the

unification of their camps. The rise of the Jana Sangh and a new alliance between the

Congress and the Communists left them with no choice but to join hands (Schoenfeld

1965).

They formed the Samyukta (“United”) Socialist Party (SSP) in June 196482. The new

alliance proved weak, being devoid of common minimum views on leadership and

electoralstrategy.TheSSPsplitwithinayear,overtheissueofLohia’sleadershipofthe

party, itssystematicanti-Congresspositionandofitsreadinesstoallywithanyonewho

wouldhelphimdefeattheCongress(includingtheJanaSangh).Thosewhohadopposed

LohialefttheSSPtoreformthePSP,leavingtheSSPasthemainsocialistformationinthe

state83.LohiaremainedattheheadoftheSSPuntilhisdeathin1967.

The dismal performance of the PSP and the SSP in the 1969 state elections (they won

threeand thirty-theeseats respectively)and in the1971LokSabhaelections (inwhich

81There had been incidents in Kerala, leading a PSP government to shoot at protesters in ademonstration.LohiaandLimayealsoopposedtheideaofanelectoralalliancewiththeCongress,which was mooted by Ashok Mehta following the Awadi declaration of the Congress, which setsocialism as the goal of political transformation in India, and the bases of a platform to bringtogethertheCongressandthesocialists(McMillan2000,Singer2012).82TheSamyuktaSocialistParty(‘UnitedSocialistParty’)wasformedin1964throughasplitinthePraja Socialist Party. Both formations were reunited as the Socialist Party in 1972. GeorgeFernandesledthepartyforthreeyears.83Brass,ibid.,161.

67

Further,afreshsplitintheCongressPartyprovidednewopportunitiesforalliances.The

Congress(O)84,wasformedafterIndiraGandhi’screationoftheCongress(R)85,following

herownexpulsionfromtheCongressinNovember1969.ItwaslednationallybyKamaraj

andMorarjiDesai,andinUttarPradeshbyC.B.Gupta,aformerChiefMinister.

OnAugust9,1971,anewSocialistPartywasformed,withthemergerofthePSP,theSSP

andthreeothersplintergroups,undertheleadershipofKarpuriThakur,asocialistfigure

from Bihar and former chairman of the SSP, and with Madhu Dandavate, ex-PSP, as

GeneralSecretary.

Onceagain,thealliancefailed.Bothpartieswereinternallydividedovertheopportunity

to allywith theCongress (O),whichwas ledby themore conservative elements of the

formerlyunitedCongressParty.Internalrivalriesalsotookprecedenceoverthepledgeof

unity.Withinayear,RajNaraincreatedhisownpartyinUttarPradeshafterhavingbeen

temporarily expelled over a Rajya Sabha ticket nomination86. So did Karpuri Thakur

himself,in1972,inalastbidtofostersocialistunity.

84(O)for‘Organization’.85(R)standsfor‘requisitionist’.86SeeBrass,ibid.,pp.163-167.

68

Table2.1PerformanceofsocialistpartiesandtheirsuccessorsinUttarPradesh,1952-201287Yearofelection

Party Contested Won Voteshare(%)

Voteshareinseatswherecontested(%)

1952 SOP 349 20 12.03 13.631957 KMPP 268 1 5.7 8.23 PSP 261 44 14.47 22.791962 SOC 273 24 8.21 12.74 PSP 288 38 11.52 17.341967 PSP 167 11 4.09 10.7 SSP 254 44 9.97 16.861969 PSP 92 3 1.72 7.89 SSP 258 33 7.82 12.861974 BKD 396 106 21.22 22.53 SOP 226 5 2.9 5.41977 JNP* 422 352 47.76 48.041980 JNP 2 0 0.01 2.06 JNP(JP) 239 4 2.89 5.16 JNP(SC)** 399 59 21.51 22.7 JNP(SR)*** 302 4 4.17 5.78 LKD 4 0 0.27 24.461985 JP 311 20 5.6 7.6 LD 385 84 21.43 23.561989 JD 356 208 29.71 35.27 JNP(JP) 119 1 0.74 2.66 LKD(B) 204 2 1.19 2.481991 JD 374 92 18.84 21.05 JP 399 34 12.52 13.13 LKD 107 0 0.35 1.381993 JD 377 27 12.33 13.76 JP 298 1 0.52 0.74 SP 256 109 17.94 29.481996 JD 54 7 2.56 19.99 JP 31 0 0.13 1.8 SP 281 110 21.8 32.122002 JP 23 1 0.27 4.51 RLD 38 14 2.48 26.82 SP 390 143 25.37 26.272007 RLD 254 10 3.7 5.76 SP 393 97 25.43 26.072012 RLD 46 9 2.33 20.05 SP 401 224 29.13 29.29Source:AdaptedfromECIdata.*CandidatesfromtheJanaSanghandvariousCongressbreakawayfactionsincluded.**CharanSingh'sfaction.***RajNarain'sfaction.

87MinorfactionsandJanataParivarpartiesfromotherstateshavebeenignored.

69

ConsolidationbehindCharanSingh’sBKD

Thesemain streamsof the socialist parties finally joined the umbrella of theBharatiya

Kranti Dal (BKD), established by Chaudhary Charan Singh, in 1974. Charan Singh, a

former Minister in various cabinets of C.B. Gupta, was a Jat leader and political

spokesperson of the middle peasantry88. A minister in the Cabinets of three Congress

Chief Ministers, he defected from the party in 1967 to form his own formation – the

BharatiyaKrantiDal89–aswellasthefirstnon-CongressgovernmentinUttarPradesh,a

broadcoalitionofoppositionpartiesinwhichtheJanaSanghrepresented42percentof

thestrengthintheAssembly(Kashyap1969,144)90.TheSamyuktaVidhayakDal(SVDor

‘UnitedLegislativeParty’)governmentlastedamereelevenmonths,tornbyitsmembers’

rivalambitionsandcontradictorypolicyviews.CharanSinghresignedinFebruary1968

and President’s Rule was imposed for a full year. The SVD experiment was indeed a

‘coalitionofopposites’asitwaspopularlytermed,butitwasalsoalandmarkeventwith

nationalrepercussions.IftheCongresscouldbedefeatedinIndia’slargeststate–andone

ofitshistoricstrongholds–itcouldalsobedefeatedinnationalpolitics.

C.B.Guptamadeacomeback in the1969stateelections,missing thesimplemajorityof

seatsbyabare0.35percentofthevotes91.Thesocialistsreachedatotalof31.3percent

ofthevoteshare,barelytwopointsbehindtheCongress,butstillindispersedorder.The

BKD became the second party, with 21.3 per cent of the votes and 98 seats. The Jana

88CharanSinghhadacareerintheCongressPartythatspannedover45years.Hismainpoliticalfeats were the introduction of major agrarian reforms, pushing towards a commercializedarchitecture.Hewasinstrumentalintheintroductionandimplementationofthe1952ZamindariAbolitionAct,anambitiouslandreformwhich,althoughpartiallyflawed,didleadtosubstantialland redistribution and the empowerment of the lower and middle peasantry. He is alsoremembered for the Chakbandi (land consolidation), a policy that helped landowners toconsolidatetheirpropertybyswappingdiscontinuousholdings.Onthe lifeofCharanSingh,seePaulBrass’definitivethree-volumebiography(Brass1984a,160).89InitiallycalledtheJanaCongress.90Inthe1967stateelections,theCongressfellshortofamajorityforthefirsttime,with46.8percentof the seats, for32.2per centof thevotes.TheoutgoingChiefMinister,C.B.Gupta, ralliedsupportfromoutsidethepartybutwasthwartedafternineteendaysbyCharanSingh’sdefection,whotooktheopportunitytoseizepowerattheheadofanalternativecoalition.TheSVDincludedtheJanaSangh,theSSP,CharanSingh’sBKD,thetwocommunistparties,thePSP,theSwatantraParty,theRepublicanPartyand22independentMLAs.91He formed his government with the support of a few independent MLAs and the SwatantraParty,whichwasreducedtofivemembersinthestateassembly.

70

Sangh,whichwasthesecondpartyintheAssemblyin1967,lost3.7percentofthevotes,

whichtranslatedintoalossofnearlyhalftheirseats(from98to49).

Bruised by his defeat, Charan Singh concentrated his efforts – with little success – on

buildinganationalcoalition,andexpandinghisparty’sreachbeyondthebordersofUttar

Pradesh.C.B.Gupta’sgovernmentfelleightdaysaheadofitsfirstanniversary,duetothe

splitbetweentheCongress(O)andtheCongress(R).C.B.Guptalost139MLAsandhalfof

hiscabinettotheCongress(R),underthe leadershipofKamlapatiTripathi.C.B.Gupta’s

decision to remainwith theoldCongress costhimhisChiefMinistershipaswell ashis

political career.With thesupportof IndiraGandhi,CharanSinghwassworn inasChief

Ministerforthesecondtime,inApril1970.

The relationship betweenCharan Singh and IndiraGandhi grew sour, as the latter had

envisaged a merger of the BKD with the Congress, and the substitution of Tripathi by

CharanSinghasChiefMinister(Brass,2014,58-75).TheCongresswithdrewitssupport,

allCongressministersresigned,leavingCharanSinghinasituationsimilartotheonein

1967, when he himself defected from the government. President’s Rule followed for a

periodof17days.TheCongress,however,didnothavethestrengthintheAssemblyto

formagovernmentandhadtomakewayforanewformulationoftheSVDgovernment,

basedthistimeonanalliancebetweenCharanSinghandC.B.Gupta.Unabletodecideon

whoamongthemshouldlead,theypickedathirdweakcandidate,TribhuvanaNarayana

Singh,whowasanoutsidertobothformations.TheSVDcoalitionwascompletedwiththe

supportoftheJanaSangh,theSwatantraPartyandtheSSP,butcollapsedonceagainafter

thelandslidevictoryoftheCongressinthe1971Generalelections(theCongresswon73

outof85seatsinUttarPradeshalone).TheBKDreturnedjusttheoneseat.

After initiallycontemplatingatripartitealliancewiththeJanaSanghandtheSwatantra,

Charan Singh’s BKD entered into an alliance with the SSP as well as with the Muslim

Majslis,aMuslimmicro-partyledbyDrAbdulJaleelFaridi,amedicaldoctorfromawell-

to-doMuslimfamilyofLucknow.TheBKDthenmergedwiththeSwatantraParty,ahead

of the1974assemblyelections.CharanSingh’sparty turned into theBharatiyaLokDal

71

(BLD)andmergedwithRajNarain’sSSP,aswellaswithafewothermicro-parties92.The

BLD’scorewasorganizedaroundCharanSingh,whohedtheWestoftheStatethroughhis

followingamong the Jats, anda triumvirateof socialists –RajNarain,ChandrajitYadav

andChandraShekhar–whowerebasedintheEast93.

Despitethisconsolidationofoppositionparties,theCongresswonthe1974electionsby

the skin of its teeth (50.7 per cent of the seats for 31.9 per cent of the votes). H.N.

Bahuguna,whohadbeenappointedChiefMinisterafterTripathi’sforcedresignation,was

reconfirmedinhisposition.

TheBLDemergedas the secondpolitical force in the state,withCharanSingh’s former

BKDoccupyingthemajorityofthespacewithinthisnewalliance.UnliketheSVDcoalition,

the parties that joined the BLD did so by exhaustion, after being reduced to tatters in

successivepolls.Theoppositionlandscaperemaineddividedbetweenthreeclusters:the

socialists, the Jana Sangh, and theCongress (O),whichwas reduced to ten seats in the

1974elections.

UnityinEmergency

Indira Gandhi provided them the cause and the opportunity to form a new alliance by

declaringastateofEmergency,onthe25thofJune1975.Thethreeoppositionformations

joinedhandstodefeatIndiraGandhiinthe1977elections,underthecommonbannerof

the Janata Party94. The alliance won 82.8% of the seats with a combined vote share of

92Including the Utkal Congress, the Uttar Pradesh branch of Biju Patnaik’s dissident CongressfactioninOrissa.93Raj Narain contested from Rajgarh in 1974, and lost. Chandrajit Yadav was a Lok Sabha MPfromAzamgarhandaformerCongressGeneralSecretaryandMinisterforSteelandMinesintheIndiraGandhiCabinet.ChandraShekharwasaLohiaitebasedinBallia,aformerPSPcadrewhoserved intheRajyaSabhabetween1962and1967.He joinedtheCongress in themiddleofhisterm,in1964,butlaterwentontoopposeIndiraGandhi,whichearnedhimajailtermduringtheEmergency.HewouldbecomethePresidentoftheJanataPartyin1977andbrieflyserveasPrimeMinister,aftertheresignationofV.P.Singh.94TheywerealsojoinedbyCongressforDemocracy(CFD),anotherCongresssplintergroupledby Jagjivan Ram, H.N. Bahuguna, the former Orissa Chief Minister Nandini Satpathy, amongothers.TheCFDmergedwith the JanataPartyonMay5, 1977. Subsequently, in1979, Jagjivan

72

47.76percent,while theCongress lostnearly80percentof its seats (from215 to47)

whilemaintainingitsoverallvoteshare(31.9percentagainst32.3percent).

The1977electionsmarktheendofthesecondphaseofthepartysystem,whichisalong

andcomplexphaseofrecompositionoftheanti-Congressopposition.Theconsolidationof

thesocialistcampprovedtobelonganddifficult,markedbymanyfailuresandachronic

incapacity toactuallybuilda socialistplatform.ThePSPand theSSP failed to forge the

allianceofbackwardclassestheirideologyhadpushedtobuild.Ineffectiveleadershipand

permanentfactionalismcrippledtheireffortsatworkingtogether.Nationaldivisionsalso

permeated into state politics, complicating matters further. At the ground level,

factionalism and the political personnel’s lack of ideology, or its willingness to trade

ideology for opportunism,meant that the ranksof the socialist partieswere frequently

depleted,andtheyhadtorenewtheircandidatesduringalmosteveryelection95.

Besides,thesociologicalcompositionofthesocialistparties–predominantlyuppercastes

andamongthempredominantlyBrahmins–ultimatelypreventedthemfromconnecting

withthebasestheysoughttomobilize.ThemountingrejectionoftheCongressinthelate

1960sand1970s,whichcamefromthebackwardclasses,wasarejectionof theupper-

castedominatedsocialorder.Clearly,theywereonthewrongside96.

Lastly, Indira Gandhi pre-empted their ideology by taking a socialist and populist turn

afterthe1969split,thuscuttingthegroundfromundertheirfeet,(Fickett1973).

Rambroketieswiththealliance,takingalongwithhimanumberofhissupporters.HefoundedtheCongress(J)–‘J’forJagjivan–whichretainedasmallpresenceandthendisappeared.95SubhashKashyapestimatesthatthePSPlosttwo-thirdsofitscadrestotheCongressby1964,andanotherhalfbeforethe1971LokSabhaelections(Brass2011).96This was particularly true about the leadership. Ram Manohar Lohia was a Bania, AcharyaNarendraDevaKhatri,JayprakashNarayanaKayasth,andsoforth.LewisP.FicketJr.estimatesthat 75 per cent of the PSP national party leadership was from the upper caste (50 per centBrahmin).Thepartyorganizationcontainedonly12.5percentof lowercastes, amere fivepercentofDalits,andalmostnoMuslims(Kashyap1969,295-296).

73

Table2.2CasteGroupRepresentationamongSocialistParties,1952-1962

1952 1957 1962

UC 13(48.15%) 32(48.48%) 30(52.63%)

OBC 5(18.52%) 23(34.85%) 19(33.33%)

SC 8(29.63%) 9(13.64%) 6(10.53%)

Muslims 1(3.7%) 2(3.03%) 2(3.51%)

N 27(100%)

76(100%)

57(100%)

*1957and1962combineboththeSocialistPartyandthePSP97.Source:Adaptedfrom(Meyer1969,175-178).

Thepartythatemergedvictorious–CharanSingh’sBKD–happenedtobethepartythat

hadtheleastupper-castebias.CharanSinghhadformedtheBKDasafrontformiddleand

backwardcastes,orasafrontofmiddleandsmalllandowners,whowerelargelyexcluded

from the Congress’ social coalition of upper castes, Dalits and minorities. The BKD

articulatedabroaddiscourseofsocialjustice,championingtheinterestsofthesmalland

middle peasantry (Brass 1980a)98. Although there were limits to the BLD’s alliance of

backwardclassesasCharanSinghnevermademuchofanoverturetowardstheDalits,it

wouldremainforthesecondthenthemainpoliticalforceinUttarPradeshforalongtime.

TheBKD’s experience, however,would be short lived. Tensionswith the Jana Sangh at

both the central and state levels led to divisions among the socialists, precipitating the

departureofCharanSingh–thenUnionHomeMinister–onApril9,1978.Thesocialist

blocagainsplitintovariousparties,themaintwobeingledbyCharanSinghandanother

byRajNarain,afterhisownexpulsionfromtheUnionCabinetongroundsofindiscipline99.

97Anexaminationofthesocialistcandidatesinthe1962electionsrevealsthattheratioofticketsdistributed to upper caste candidates was slightly below 48%. A quarter of the tickets weredistributedamongOBCs,20%amongSCcandidatesandaround7%amongMuslimcandidates.98Although itwasverymuchanUttarPradeshparty, theBKDwasestablishedas an ‘All-India’party, in a meeting in Delhi on April 9, 1967, which several Chief Ministers and non-Congressparty leaders attended. Its first chairmanwasMahamayaPrasad Sinha, ChiefMinister ofBihar(Brass,2014,37).99RajNarainhadopenly criticized twoofhis colleagues from theCabinet, L.K.Advani andAtalBihariVajpayee,forhavingparticipatedinanRSSrallyinDelhi.

74

A few months later, the Central government lost its majority and Morarji Desai had to

tenderhisresignationonJuly15,1978,makingwayforCharanSinghtotakeover,with

theinitialoutsidesupportofCongress.UnabletoobtainamajorityinParliament,he,too,

resigned on August 20 of the same year. The divisions and the collapse of the Janata

coalitionattheCentrepavedthewayfortheresurgenceoftheCongressattheCentreand

in the state. In the 1980 UP state elections, Charan Singh could save only 59 seats100.

Throughoutthe1980s,asPaulBrassnoted,'theonlyelectorallysignificantstruggleinUP

as a whole has become that between the Congress and the Lok Dal' (Brass 1986, 664).

[Brass,1986:665].

2.1.2.Partingofalliancesandparties’recompositioninthe1980s

The third phase starts soon after the restoration of democracy by the Janata coalition.

This phase was marked by the return of the Congress Party and the decaying of its

organization(Pai2000b), therebirthof the JanaSanghas theBharatiya JanataParty in

1980, the reconfigurationof the socialist camp,which led toa temporary consolidation

aroundCharanSinghbeforeanewphaseofdivisionappearedafterhisdeath.The1980s

werealsomarkedby thebirthofanewpoliticalparty, theBahujanSamajParty,which

succeededinmobilizingasectionoftheDalitsvotersbyprovidingthemwithapartyof

theirown,forthefirsttimesincethevirtualdemiseoftheRPIinthelate1960s.

The Janata Party experiment, to useMadhu Limaye’s expression, did not last long. The

failureofthesocialists,thesanghisandtheex-Congresstomaintaintheircoalitionatthe

Centre, the resurgence of Indira Gandhi on the national stage and the tenure of a

lacklustreChiefMinisterinUttarPradesh101pavedthewayforatriumphalreturnofthe

Congressasearlyas1980.Attheheadofafracturedparty, itwasinUttarPradeshthat

shewonherparty’smostthumpingvictoryintwenty-twoyears,bagging72.7percentof

100TheCongress(I)won309seatsoutof425,with37.65%ofvoteshare.101Ram Naresh Yadav was designated Chief Minister of Uttar Pradesh in 1977. A provinciallawyer and a political novice, he had owed his nomination to being the least threateningconsensuscandidateoftheJanatacoalition(Kohli1987,197).HewasquicklyreplacedbyBanarsiDas,afterayearandahalfinoffice.

75

theseatswith37.65percentof thevotes.Onceagain,a fragmentedoppositionenabled

theCongressPartytoconvertaminorityofvotesintoamajorityofseats.

Thesocialists,whoalsoemergedbitteranddividedfromtheJanataexperiment,brokeup

intofourvariantsoftheJanataParty102.Theleaderofitsmaincomponent,CharanSingh,

hadbeensworn inasPrimeMinister in July1979,at theheadofabrittlecoalitionthat

dependedontheoutsidesupportoftheCongressParty.Thatsupportdidnotlast,forcing

Charan Singh to resign and call for fresh elections after a month in office. The tearing

downof the Janatacoalition, themishandlingof IndiraGandhi’sarrestprior to this,her

re-electioninaby-electionin1978103,thedeathofJayprakashNarayaninOctober1979,

preparedthegroundforIndiraGandhi’striumphalreturninthe1980Generalelections.

ThevariouscomponentsoftheJanatacoalitioncontestedontheirown,undercuttingeach

otheragainsttheircommonenemy.Atthenationallevel,theCongresswon353seatswith

42.7percentof thepopularvote.TheLokDal, renamed in theseelections JanataParty

(Secular–CharanSingh),won41seats (essentially inWesternUttarPradesh,Haryana,

andPunjab)with9.4percentofthevotes,tenmoreseatsthantheresidualJanataParty,

whichhadtwicethenumberofvotersthanCharanSingh’sfaction.

102TheJNP(JP)wasabreakawayfactionthathadsomepresenceintheEastandNorth-East,theJNP(SC), ‘SC’ for ‘Secular-CharanSingh’suppportedCharanSingh, the JNP(SR), ‘SR’ for ‘Secular-Raj Narain’, under the leadership of Raj Narain, and a residual JNP who only contested in twoseatsandwonnone.TherewasalsoaLokDalthatcontestedthe1980election.Itwasafour-manoperation,whichquicklydisappearedandhadnorelationwithCharanSingh’sLokDal.103Shehadwonaby-electioninChikmaglur,inKarnataka,inNovember1978,thankstoCongressMPD.B.ChandreGowda,whovacatedhisseatsoshecouldcontest.ShewondespitethesplitoftheCongress(R)intotheCongress(I)–forIndira–andtheCongress(U)–forDevarajUrs,thethenKarnatakaChiefMinister,ayearearlier.

76

Table2.3Performanceofmainpartiesinthe1980UttarPradeshassemblyelections Contested Won Forfeited

DepositVoteshare

Voteshareinseatscontested

BJP 400 11 283 10.76% 11.50%

CPI 155 7 134 3.55% 9.75%

CPM 18 0 14 0.47% 10.36%

INC(I) 424 309 10 37.65% 37.76%

INC(U) 339 13 290 6.38% 7.99%

JNP 2 0 2 0.01% 2.06%

JNP(JP) 239 4 219 2.89% 5.16%

JNP(SC) 399 59 156 21.51% 22.70%

JNP(SR) 302 4 276 4.17% 5.78%

LKD 4 0 0 0.27% 24.46%

Others 60 1 58 0.45%

Ind. 2267 17 2212 11.87% 12.01%

Total 425 100%

Source:AdaptedfromECIreports.

Theyears that followed the1980electionswereyearsof the re-compositionof parties

andtheemergenceofnewpoliticalforces.WhiletheCongressdominatedthedecade,its

organizationsuffered fromthecentralizedcontrolexertedby theHighCommandof the

party. The 1980s were a period of organizational decay for the Congress (Pai 2000b),

undertheauthoritarianrulesofIndiraGandhiattheCentre,ofV.P.SinghinUttarPradesh

between1980and1982, andunder themismanagementofweakchiefministers in the

secondhalfofthedecade(Stone1988)104.

Duringherlastyearsinpower,IndiraGandhihadmadeastrategytoappointweakchief

ministers in Congress-ruled states, and to dismiss chief ministers in states ruled by

opposition parties, through the imposition of President’s Rule (Art. 356 of the

Constitution)105.In1980alone,President’srulewasdeclaredontendifferentoccasions.

104ThedismissalofV.P.Singh in July1982andhisreplacementbytheSpeaker,SripatiMisra,amandevoidofanyfollowingofhisown,markedthebeginningof theabovementionedphaseofweak Congress chief ministers. Typically, the announcement of his appointment was made inDelhiandhecouldnotchoosemostmembersofhisCabinet(Fickett1973).105With10 impositions,UttarPradesh ranks fourth among states in India, precededbyPunjabandJammu&Kashmir,twostateswithahistoryofinsurrectionandcivilunrest,andPuducherry.

78

the Congress nonetheless succeeded in retaining power, owing to the weaknesses and

divisionsamongitsopponents.

In1980,V.P.SinghbecameChiefMinister106.Histenurewasmarkedbyasignificantrise

of criminal violence andbydeadly communal violence in the state,whichhe sought to

address with a firm response. So much so that the unleashing of the PAC (Provincial

ArmedConstabulary),asortofspecialpoliceforceknownfortheirbrutality,fuelledcivil

unrest insteadofcalmingdownthesituation(Chawla2014).Attimes, theytookpart in

riots themselves or ‘looked the other way’ while riots were taking place, notably in

Meerut,Aligarh(Brass2004)andMoradabad(Engineer1984).Thesituationdeteriorated

tothepointthatV.P.Singh’sownbrother,C.S.P.Singh,ajudgeoftheAllahabadHighCourt,

wasmurderedalongwithhisnine-year-oldsonbydacoitsinMarch1982,whilereturning

fromahuntingparty.Barelyamonthlater,themassacrebydistinctgangsoftenYadavs

inKanpurdistrictandsixDalitsinMainpuriledhimtoresign(Mustafa1995,52).Indira

GandhitookadvantageofV.P.Singh’sresignationtoappointaseriesofweakandpliable

chiefministers–SripatiMishra,N.D.TiwariandVirBahadurSingh,successively.

AftertheEmergency,theBLDre-emergedastheCongress’strongestopponent.Theother

components of the Janata Parivar – the Congress (O) and the SSP – disappeared while

most of their members were absorbed in the BLD. On the eve of the 1984 elections,

CharanSinghformedyetanotherparty,theDalitMazdoorKisanParty(DMKP),througha

mergerwithH.N. Bahuguna’s Socialist Front and theNational Socialist Party. Its newly

appointedheadof theUPunitwasamannamedMulayamSinghYadav,aYadav leader

fromEtawah,whomCharanSinghhadrecruited forhismobilizationandorganisational

abilities in the late1960s.Thepartywas launchedon the20thofOctober1984.Eleven

days later, Indira Gandhi was assassinated by her bodyguards. The DMKP was swept

awaybythepro-CongresswavethatfollowedIndiraGandhi’sdeath.CharanSinghcould

onlysavehisseatinBaghpat,andanotherone,inEtahdistrict,inthegeneralelections.

106V.P. Singh was born in 1931 in a Rajput zamindari family, formerly ruling the Kingdom ofManda.Hewonhisfirstelectionin1969inSoraonandthenproceededtowinaLokSabhaseatinPhulpur, JawaharlalNehru’sold constituency.Hewould serve five terms in theLokSabha, twofrom Allahabad and two from Fatehpur. He was appointed Deputy Minister for Commerce inIndira Gandhi’s Cabinet and served as Minister for Commerce during the Emergency. For apoliticalbiographyofV.P.Singh,see(Stone1988).

79

PowergrabintheLokDal

Themid-1980swereaperiodoftransitionfortheLokDal.Sincetheearly1980s,Charan

Singhhadgraduallybeenwithdrawingfromactivepolitics.Hehadentrustedthetaskof

leading theanti-V.P. Singhagitationcampaign toMulayamandmadehimLeaderof the

OppositionintheVidhanSabha.In1985,MulayamSinghtookovertheorganizationand

ledtheLokDal’scampaign,winning84seatswith21.43%ofvoteshare.

On January 10, 1987, Ajit Singh dislodged Mulayam from the post of Leader of the

Opposition, with the support of the Congress. Mulayam retaliated by creating the

KrantikariMorcha,anewparliamentaryalliancegatheringtheJanataPartywiththetwo

communistpartiesandthreeminorformations107.

CharanSinghpassedawayonMay29,1987.Meredaysafterhisdemise,theLokDalsplit

intotheLokDal(A),afactionledbyhissonAjit,andtheLokDal(B),ledbyH.N.Bahuguna.

AftersidingbrieflywithBahuguna,MulayamSinghlefttoformtheJanataDal(JD).TheJD

underwentsomeupheavalsinitiallybut,ultimately,Mulayamprevailedoveralltheother

splintergroupsoftheLokDal.HisfactioncommandedthelargestshareoftheLokDal’s

political base among the leading middle status agricultural castes in UP, the Yadavs

figuringpredominantlyamongthem(Brass2011).ThecoresupportersofAjitSingh–the

Jats–weregeographicallyconfinedtoafewdistrictsofWesternUP.Moreover,Mulayam

Singh’s superior manegerial skills also helped him to retain much of the Lok Dal’s

organizationalstrength.

OnSeptember1987,MulayamlaunchedfromthetownofAkbarpurtheKrantiRath(the

“RevolutionProcession”),aprocessioninwhichhedemandedtheimplementationofthe

Mandalreport,aswellaspromotedseveralpro-farmermeasures,suchastheindexation

ofagriculturewages to theprice index, a loanwaiver for farmers.TheKrantiRathwas

meanttobeademonstrationofstrengthtotheCongressaswellastoitsrivalsfromthe

Lok Dal. It was also meant to be an opportunity to rally the Yadavs to his cause. The

processionstarted fromAkbarpurandthenproceededthroughthe threeparliamentary

segmentsofKannauj,MainpuriandEtawah,threeYadavstrongholdsandconstituencies

107TheJanwadiParty,theSanjayVicharManchandtheCongress(J)).

80

that Mulayam Singh Yadav and his family would subsequently hold. The rathended in

Jaswantnagar,whereMulayamhadwonhisfirstelection,in1967108.

WhileMulayamtouredtheplainsofDoab,theLokDalbecameembroiledinabattle for

leadership at the national level, the Devi Lal faction of Haryana clashing with H.N.

Bahuguna, thennationalPresidentof theParty.Bahuguna’sweakenedposition inUttar

Pradesh helped Devi Lal to prevail. He supported the decision to merge the Lok Dal –

includingitsU.P.branchandMulayam’sJanataParty(thencalledSamajwadiJanataParty)

withV.P.Singh’s JanMorcha,ananti-corruption/anti-RajivGandhimovementthathad

become the base for the formation of a large nation-wide coalition of regional parties,

which would lead to the formation of the National Front government in 1989109. The

mergerofthevariousfactionsoftheLokDalwiththeJanMorchaon11October1988led

to the creation of the JanataDal. Themerger gaveMulayam the opportunity to prevail

overhisrivalsinUttarPradesh,andtobecomechiefministerafterthe1989election.The

Congress lost11percentofvoteshareand65percentof itsseats.The tworemaining

splinterfactionsoftheLokDal–theJNP(JP)andtheLokDal(B)–obtainedoneandtwo

seatsrespectively,withlessthantwopercentofthevotescombined.

TherebirthoftheJanaSangh

The Jana Sangh, which had joined the Janata Party coalition, refounded itself as the

BharatiyaJanataPartyinApril1980.Itcontestedthe1980assemblyelectionsunderthis

name and obtained a punishing result. Compared to the 1974 elections, the rightwing

party’svotesharedroppedby7percent, reducing it to11seats in theassembly. Itgot

108A second rath followed the first one, this time on Ajit Singh’s own turf in Western UttarPradesh. This was meant to symbolically capture the legacy of Charan Singh and divide thesupportbaseofAjitSinghinhisownstronghold.109AfterhisresignationasDefenceMinisterin1991,V.P.Singhcreateda‘people’splatform’–theJan Morcha – alongside Arif Mohammad Khan and Arun Nehru, aimed at campaigning againstRajivGandhiinparticularandagainstcorruptioningeneral.This‘non-political’platformhadtheverypolitical objective to gather theopposition. It enjoyed thenon-Congress stalwarts such asJyotiBasuinWestBengal,RamakrishnaHegdeinKarnataka,andevenL.K.Advani,fromtheBJP.OnthecreationanddevelopmentoftheJanMorcha,see(Mustafa1995).OntheformationoftheNationalFront,see(Mustafa1995,89-103).

81

through the1985and1989electionswithabouthalfof the supportbase that ithad in

1967.

UndertheimpetusofAtalBihariVajpayee,thesaffronpartyattemptedtodevelopamore

moderateapproach,broadenitssupportbaseandjustifytherupturewiththeJanaSangh

legacy(Hansen1999,158).While thestrategyworked inassemblyelections inMadhya

Pradesh,GujaratandRajasthan–wheretheRSSnetworkswerestrong– it failed in the

‘cowbelt’statesofBiharandUttarPradesh110.

InUttarPradesh,theyremainedbelowthebaroftwelvepercentofthepopularvotefor

threeconsecutiveelections.Asaresult,towardstheendofthedecade,thepartyadopted

a more aggressive mobilization strategy based on religious appeal as an attempt to

consolidateitssupportbaseinitiallyamongtheuppercastes(Hansen1999,Jaffrelot1996,

2010c,Zavos,Wyatt,andHewitt2004).

ItalsoviolentlyopposedthecampaigntoimplementtherecommendationsoftheMandal

Commission’sreport–effectivelyexpandingthequotaregimeinpublicemploymentand

educational institutions by 27%, by including the OBCs – and led a vast mobilization

campaign for reclaiming of the supposed birthplace of Ram, which culminated in the

destructionoftheBabriMasjidinAyodhya,onthe6thofDecember1992.

EmergenceoftheBahujanSamajParty

Finally,aroundthesameperiod,anewpoliticalforcedeveloped.TheBahujanSamajParty

(BSP), created by Kanshi Ram in 1984, emerged from the ranks of the Backward and

Minority Castes Employees Federation (BAMCEF), a union consisting mostly of Dalit

government employees, and more specifically from the Dalit Shoshit Samaj Sangharsh

Samiti(DS-4)111.BAMCEF’sagitationwingwascreatedbyKanshiRamin1982inorderto

mobilizeDalitsthroughcycleralliesandmassmobilizationcampaigns(Jaoul2010).

After leaving the Congress, Dalit voters never really had an attractive alternative to

support. The BJP was consolidating its upper-caste vote base and did not pay much

110Hansen,ibid.,p.158.111‘CommitteeforthestruggleoftheDalitsandexploitedcommunities’

82

attention to lower castes– at least initially.The socialists and theCharanSingh faction

nevercaredmuchforDalits,despitetheiregalitarianrhetoric.MostDalitsworkedasdaily

wage farm labourers for the landowning castes that formed the core of the socialists’

supportbase.Theirinterestswereoftenatodds.

TheBSPemergedasthealternativeforDalits,whocouldfinallysupportapartyoftheir

own112.TheBSPgrewbymobilizingacoreofDalitvoters.Itgathered9.4percentofthe

votes in its first participation in a state election, earning it 13 seats. Later on, the BSP

wouldseek toexpand itsbase toother lowercastesandminorities, andgainpowerby

forginggovernmentalalliances.

FromKisanpoliticstocaste-basedpolitics

The1980s are a keydecade tounderstand contemporarypolitical transformations and

dynamics.Atthenationallevel,thisdecadesawthepeakofvariousfarmers’movements

andthatoftheirpoliticalcounterparts,aswellasthebeginningoftheirdownfall,dueto

theirpoliticalandideologicaldivisions(BentallandCorbridge1996,Frankel,Frankel,and

Rao1990,FrankelandRao1989,Varshney1995).InUttarPradesh,asinotherNorthern

Indianstates, thesedivisionscausedarapidsuccessionofmergersandsplitsofparties,

leadingtoareconfigurationofthepartysystemaswellasanevolutionofthewaypolitical

interestscrystallizeandtranslateintopoliticalmobilization.Politicalleadersandfactions

competingforpowerwithintheJanataParivargraduallyturnedtocasteandidentityas

the main trope for political mobilization, rather than appealing to a broad category of

backwardclassesorKisanidentity.

In reality, no party is mobilized solely on the basis of one factor; rather, they seek to

appealtovotersthroughvariouspoliticalandsemanticregisters.AsBruceGrahamnoted,

politicalinterestsinUttarPradeshalwaysexpressedthemselvesonavarietyofregisters:

land interests, class interests, caste interests, religious interests and some sectional

interests (Graham 1993, 189). This variety of interests was reflected in the factional

112TheRepublicanPartyofIndiawaslargelylimitedtoWesternUttarPradesh,anditscadreweretooclosetotheCongresstodevelopadistinctidentity(Ruparelia2015,Chapter5).

83

compositionof the JanataParivar,whichwasmadeofvariouscomponentsanimatedby

divergentandoftencontradictinginterests.

Splitsandmergers in thesocialist camp traditionally tookplaceonaccountofpersonal

rivalries between leaders, fed by divergences on matters of strategy and ideology.

Somewhat abstract doctrinal debates often led to rifts between leaders, rifts that

translatedintosplits(Fickett1973,829).Duringthe1980s,splitstookplaceontheissue

of party control and power as well as caste113. Charan Singh’s Lok Dal split not only

between contending political figures, but between contending political figures that

represented – and where supported by – different castes: the Jats with Ajit Singh, the

YadavswithMulayam.Thenewgenerationofpolitical leaderswithintheJanataParivar

foughtforpowerwithinthestateandwithintheirpoliticalfamily,bankingonthesupport

of their castes, articulating a discourse of social justice and equity around specific

ascriptive identities, leaving aside in practice the socialists’ ambition to rally the

backward classes against the upper castes. The resurgence of the debate over the

extensionofthereservationsystemtoOtherBackwardClassesinthe1980sservedasa

crystallizerforcaste-basedmobilizations114.Asaresult,theupper-castebiasthatmarked

thesociologicalcompositionofthesocialistformationsofthe1960sand1970swanedto

givewaytoahigherrepresentationofOBCs.

Thesetransformationsmadewayforthefourthphaseoftheoverhaulofthepartysystem

in the 1990s, a new political configuration that emerged through the juxtaposition of

religion-based and caste-based mobilizations (Hasan 1993, 1998, Jaffrelot 2003b, Pai

2000b).

Three parties - the BJP, the BSP and the SP – surged while the Congress collapsed,

electorally,andorganizationally.

113WhatremainedcommonbetweenthetwoperiodswastheoverdeterminationofthedivisionsbetweenthevariousfactionsandcomponentsoftheJanataDalatthenationallevel.114ForadetailedhistoryofreservationsinIndiaandacomprehensivecontemporaryassessment,see(Pai1994,309).

84

Table2.4CasteGroupRepresentationamongSocialistParties,1980-1991

1980 1985 1989 1991

UC 16(25.4%) 28(26.92%) 75(35.89%) 31(24.60%)

OBC 24(38.10%) 33(31.73%) 67(32.06%) 51(40.48%)

SC 11(17.46%) 29(27.88%) 52(24.88%) 29(23.02%)

Muslims 9(14.29%) 14(13.46%) 13(6.22%) 14(11.11%)

Unidentified 3(4.76%) 2(0.96%)

N 43(100%)

104(100%)

209(100%)

125(100%)

Source:Author’sfieldwork.

The1990swouldbeadecadeofstructuralinstability115,markedbytheinabilityofparties

towinmajoritiesontheirown.Thisforcedthemtoenterintoaregimeofdysfunctional

coalitions, which would plunge the state of Uttar Pradesh into chronic instability and

recurrentperiodsofPresident’sRule.

2.1.3.CommunalandCastepolarizationanddysfunctionalcoalitionpoliticsinthe1990s

Threepartieshavebeeninpower,eitherontheirownorincoalitions,intheperiodthat

followedthedestructionoftheBabriMasjid.TheBJPruledthestatecontinuouslyforfour

years and 169 days, with three different chief ministers116. The Bahujan Samaj Party

(BSP)ruledforsevenyearsandsixteendaysunderfourMayawati-ledgovernments(two

incoalitionwiththeBJP,onewiththeSamajwadiParty).

Before the2012StateAssembly elections, the SamajwadiParty (SP) ruled the state for

fiveyearsandseventy-threedays,underMulayamSinghYadav.Duringthesameperiod,

115Duringthatdecade,onlytheBJPsucceededtowinamajorityofseats,in1991.116KalyanSingh(21September1997–12November1999),RamPrakashGupta(12November1999–28October2000)andRajnathSingh(28October2000-8March2002)

85

President’sRulewasdeclaredonthreeoccasions,foratotalperiodoftwoyearsandtwo

hundredandeightdays.

ThebirthoftheSamajwadiParty

TheSamajwadiPartywasformallycreatedonthe4thofOctober1992.Itemergedasthe

victorious faction from the fratricidal struggle between the claimants for the political

successionof Charan Singh after his death in 1987. Its leader,MulayamSinghYadav, a

prominentpoliticalfigurebothamonghiscaste–theYadavs–andthestate’ssocialists,

succeeded in capturing Charan Singh’s legacy and built a party dominated by the elite

segmentsofthestate’sbackwardclasses117.

ThepartyclaimstheduallegacyofRamManoharLohia–asideologicalfoundingfather–

and Chaudhary Charan Singh – as tutelary political figure. Both had advocated that a

socialistpoliticalformationshouldaimatforminganallianceoftheruralmiddleandlow

peasantry, alongside Dalits and Muslims, to be able to defeat a Congress Party largely

dominatedby theuppercastes. InLohia’sview,suchanalliancewasalsoanallianceof

caste and class, an opportunity to practise political equality among backward and

marginalizedgroups,preparingtheadventofamorejustsociety.

That alliance, as seen, would prove difficult to set up due to feuds within the various

factionsandbranchesofthesocialistmovementaswellasantagonismbetweenupwardly

mobile and assertive segments of the OBCs and other lower peasant castes. The more

prosperousJatsinWesternUP,forinstance,andthelowerpeasantcastesofEasternUP

couldnot setupacommonplatform(Verma2004a,1509).Besides,neitherwere these

117Many of the small parties or residual parties from earlier coalitions – such as the BhartiyaKisanKamgarParty(BKKP), launchedbyAjitSinghinSeptember1996,withthesupportoftheJatleaderMahendraSinghTikait,theJanataDal,theLokDal(Bahuguna)–graduallydisappearedor became localized phenomenons, indicating the increased polarization of voters’ preferencesbetweenthefourmainparties.Somelocalpartieshavesubsisted,suchasAjitSingh’sRashtriyaLok Dal in Western Uttar Pradesh. New local parties have carved out a modicum of space forthemselves,suchastheApnaDal,straddlingtheborderbetweenlowerDoabandEasternUttarPradesh.These local parties subsist on accountof thedemographic concentrationof their coresupportbase– Jats fortheRLD,LodhsfortheApnaDal–asupportthathasbeenerodingoverrecentyears.

86

twogroupsfundamentallycommittedtobuildingallianceswithDalits,whichlimitedthe

expansionoftheirelectoralbases.

Mulayamhademergedas thewinner in the1989elections.However,hewas inaweak

position. First, he prevailed over Ajit Singh for the leadership of the Janata Dal state

LegislatureParty–apreludetohisappointmentasChiefMinister–byathinmargin118.

Asaresult,hisownmajoritystillcontainedelementsfaithfultohisrival–56ofthe206

JanataDalMLAswerereportedlydevotedtoAjitSingh,whofelthehadbeencheatedof

the chief ministership. They could topple the government down at any time. To make

matters worse, Ajit Singh used his proximity to the Jat leader Mahendra Singh Tikait,

PresidentoftheBharatiyaKisanUnion(BKU),afarmer’suniondominatedbytheJatsin

WesternU.P.,tostiragitationagainstthegovernment,andleadarebellionofJanataDal

MLAsfromWesternUttarPradesh(Fickett1993,92).

ThesecondmajorchallengethatMulayam’sgovernmentfacedwastheintensificationof

the Hindu nationalists’ campaign for the Ram Temple. Mulayam found himself at odds

withhisownally,V.P.Singh,thenPrimeMinister,whosegovernmentdependedfromthe

BJP’ssupport.AturningpointinMulayamSingh’scareertookplaceonOctober30,1990,

when he ordered the police to fire at kar sevaks who had gathered at Ayodhya,

threatening to destroy the Babri mosque. More violence took place on November 2.

Sixteensevaksdiedandmanywere injured.Thatdecisioncosthimtheelection in1991

butearnedhimthestaunchsupportofMuslimvoters,whowouldprovidehimasecond

coresupportbaseinsubsequentelections.Mulayamhadalreadymadeanameforhimself

in the early 1990s among the state’s largest minority, by opposing V.P. Singh’s violent

repressionofcriminality,whichcausedmanycasualtiesamongMuslims119.TheAyodhya

incidentgavehimastatusofprotectorofminoritiesthatnooneelsecouldclaimatthat

timeinUttarPradesh120.

118DilipAwasthi,journalistwithIndiaToday,estimatedthemargintobeof11votes,outof212legislators(Galanter1984,McMillan2005).119ThePAChasbeenaccusedofcondoning, ifnothelping,withtheMoradabadriotsof1980, inwhichhundredsofMuslimsdied.120MulayamSinghYadav isnottheonlypolitical figurewhoemergedasa‘muslimprotector’ inthis troubledperiod. InBihar,LalooPrasadYadav’sshot to famewhenhearrestedL.K.Advani,preventinghimandhisRathYatratoenterU.P.

87

Following the October 30 incident, the BJP withdrew its support to V.P. Singh at the

Centre, precipitating his downfall. Following that, the national Janata Party split and in

Uttar Pradesh, Mulayam followed Chandra Shekhar’s faction. The latter renamed his

splinter group the Samajwadi Janata Party, before he becamePrimeMinister. After the

split, Mulayam’s government became dependent from the Congress’ support, the same

wayChandraShekhar’sminoritygovernmentdependedfromtheCongressattheCentre.

HoundedbyinternaldivisionsandweakenedbythefallofthegovernmentattheCentre,

MulayamtenderedhisresignationonApril4,1991.HewenttothepollundertheJanata

Partybanner,whileafactionledbyV.P.SinghcontestedasJanataDal.Thecampaignwas

markedbyviolence(crudebombswerehurledinfrontofMulayamSingh’sresidenceon

April24,injuringnumberofhissecuritystaff),byanoutbreakofcommunalincidents121,

andbythedeathofRajivGandhi,duringthecampaignfortheGeneralelections.

In the1991General elections that followed, theCongress failed toobtainedamajority.

TheBJPemergedasthesecondpartywith120seatsand20percentofthepopularvote.

V.P.Singh’sJanataDalwasreducedto59seats(31inBiharand22inUttarPradesh)and

Mulayam’sJanataPartytofive122.

InUttarPradesh,theBJPgainednearlytwentypercentofvoteshareandobtainedforthe

first timea singlemajority in theAssembly (221seats), leading to the formationof the

firstKalyanSinghgovernment.TheCongress’seatsharewascutbyhalf(from94to46).

V.P.Singh’sJanataDalcamethirdwith92seatsand18.8percentofthevote.Mulayam’s

factionobtainedonly34seats.

Parties’ splits tend to create havoc with local organization. District organizations get

dissolvedandoftensplitonthebasesofcastefactions.Theprocessofpolarizationamong

backwards(includingtheJats)inWesternU.P.canbetracedtothatperiod.

121Theelectionscouldnot takeplace insixconstituencies,all inWesternUttarPradesh,due tothecommunalviolence:Agota,Hastinapur,Kharkauda,Kithore,Meerut,andMeerutCantonment.122FourinUttarPradesh,includingChandraShekhar,andH.D.DeveGowda,inKarnataka.

88

Table2.5VoteshareandSeatdifferenceinthe1989and1991UttarPradeshassemblyelections

Voteshare Seats

1989 1991 Swing 1989 1991 Swing

BJP 11.61 31.45 19.84 57 221 164

INC 27.9 17.32 -10.58 94 46 -48

BSP 9.41 9.44 0.03 13 12 -1

JD 29.71 18.84 -10.87 208 92 -116

JP 12.52 34

Ind. 15.46 7.44 -8.02 40 7 -33

Others 5.91 2.99 -2.91 13 7 -6

100 425 419

Source:adaptedfromECIreports.

ThetenureofKalyanSinghwasmarkedbyvariouskindsofviolentagitationmovements.

Ononehand,thecommunalviolencethathadbroughthimtopowerdidnotsubside.In

fact, it intensified. For one, theHindumilitants saw the advent of aBJP government in

Lucknowasanopportunitytoreachtheirgoal–thebuildingofaRamtempleinAyodhya

–andsecondly,thestakesofthemovementwentfarbeyondthescopeofstatepolitics.As

Zoya Hasan points, “the demolition of the Babri Masjid was the culmination of a mass

movementintendedtopromotereligiouslygroundednationalism”.(Hasan1998,189).

Themovementalsomarkedanewstage in thecastepolarizationof theelectorate.The

BJPconsolidateditssupportamongtheuppercastethroughitsoppositiontoreservations

andthroughthetropesofreligiousnationalism.Thesetropesalsoenabledittodentinto

the OBC vote, by mobilizing lower OBC voters such as Lodhs, Kurmis, Sainis and

Kushwahas, who were in frequent local conflict of interests with the Yadavs. The

distribution of tickets to lower OBC candidates and the projection of Kalyan Singh – a

Lodh – as figurehead of the party served as bait and incentives for the lower OBC to

supporttheBJP.OtherOBCfigureswithintheparty,suchasUmaBharti(alsoaLodh)and

VinayKatiyar(aKurmi),alsocontributedtotheexpansionofthevotebaseoftheBJP.

89

Inthemidstofallthis,MulayamdistancedhimselffromhisallyChandraShekhar,onthe

basisofadisagreementovertheMandalagitation.ARajputfromBallia,ChandraShekhar

wastoodependentfromtheuppercastevotetoforcefullysupporttheimplementationof

the Mandal Report’s recommendation. On the 29 September 1992, the SJP legislature

groups splits in two.MulayamabandonedChandra Shekhar to formhis ownparty, the

Samajwadi Party123. The SP was launched in Lucknow, in presence of several socialist

leadersfromacrossthecountry.Soonafter,heannouncesanalliancewiththeBSPforthe

upcomingstateelections.Itdevelopedastrategyofdistancingvis-à-vistheBJP,inorder

to consolidate its support among Muslims, and sought to re-create Charan Singh’s old

AJGAR social alliance (Ahirs, Jats, Gujjars and Rajputs) through targeted caste

mobilization(Pai1994,302).

The1990s:Asuccessionofdysfunctionalcoalitions

Throughthe1990s,therelationshipbetweenpartiesandpartyleaderswascharacterized

by intense acrimony, both between and within alliances. Electoral outcomes were so

fragmentedthatpartieshadtodependfromdefectionsfromrivalpartiestosecureshort

andunstablemajorities124.

The State Assembly of Uttar Pradesh was dissolved the day the Babri Masjid fell, as

President’s rule was immediately declared. A year later, in December 1993, the

SamajwadiParty and theBahujan SamajParty came together in a coalition, preventing

theBJP’sreturntopower.MulayamandKanshiRamhadholdregulartalkseversincethe

formerlosthisChiefMinistership.Thisalliancewasconsonantwiththesocialistobjective

ofcreatingabroadallianceofbackwards,spanningfromhedominantOBCstotheDalits.

The 1993 verdict came in as a surprise. The BJP’s momentum with the Ayodhya

movement came to a halt, as the party lost a significant number of seats (although

maintainedastablevoteshare).TheCongressregistereditsworsteverperformance,with

123Since he had lost the Prime Ministership, in April 1991, Chandra Shekhar lost most of hisinfluence.Mulayamdepartedfromaspentforce.124ThefourBJPgovernmentsbetweenJune1991andMarch2002dependedupondefectorsfromtheCongressandotherparties(Awasthi1989).

90

28seatsintheassembly.TherivalcomponentsoftheJanataParivarwerealsoroutedas

many of its members defected to other parties, notably to Mulayam’s newly created

SamajwadiParty.Theywouldneverrecoverfromthedefeat, leavingtheentirespaceof

theJanataParivartotheSP,minusthefewpocketsfaithfultoAjitSingh’sRLD.

The BSP and the SP were the only met gainers of these elections, although the BJP

remainedthefirstparty,with177seats.TheBSPmultiplieditsnumberofseatsbymore

than five (from 11 to 67) with a slight increase in vote share, and the SP tripled its

presenceintheassembly(from34to109)witha5.4positiveswing125.

However,thecoalitionquicklycameunderstrainascasesofatrocitiescommittedagainst

Dalitsshotupthroughthestate.BSPofficeholdersandleadersfeltunderminedbytheir

SP counterparts and the relationship between Mulayam Singh Yadav and Mayawati,

consideredbythelatterasapoliticalnovice,grewestranged.Mayawatibrokethealliance

in May 1995, by forming a counter-coalition with the BJP, who offered her the post of

ChiefMinister126. IntheBSP’sview,thealliancewiththeBJPwasameantoreconstruct

the erstwhile successful Congress’ coalition of extremes. The BSP would gather the

supportoftheDalitsandtheMuslims,distributeticketsacrossthelowerOBCs,whilethe

BJPconsolidatedtheuppercastevote(Pai2009).

ThearrangementwiththeBJPlastedonly137days,theBJPpullingoutitssupportahead

ofthe1996LokSabhaelections.AfteranotheryearlongperiodofPresident’sRule,both

partiesbrokeredanewalliance,underthepromisethattheChiefMinisterialpostwould

rotate between the two parties on a six-monthly basis127. It eventually did but the

coalition fell once again, on account of thebad relationshipbetweenMayawati and the

125Onthedetailofthe1993electionssee(Duncan1997).126Mulayam Singh Yadav attempted to counter the breach of alliance by attempting to breakdown theBSP, coercingor luringsomeof itsMLAs todefect inhis favour.Hesucceeded togetfifteenBSPlegislatorstodefectbutfellshortofthenumberoftwenty-threerequired(athirdoftheBSP’sstrengthintheAssembly)toavoidtheanti-defectionlaw.OnJune2,fiveadditionalBSPlegislators were forcibly removed from their guest house, where Mayawati was consulting herranks, and coerced into joining a breakaway faction of the BSP led by Raj Bahadur, who hadsignedapledgeofallegiancetoMulayam.127TheagreementalsoincludedparityintheCabinet,theSpeakerpostfortheBJPandatoppartyleadersbipartisanpaneltomonitorthecoalition.

91

newly re-installed Chief Minister Kalyan Singh. The latter succeeded to maintain its

government by engineering defections from various parties, including the BSP, the

CongressandtheJanataDal.HeandthethenBJPPresidentRajnathSinghusedtheircaste

appeal to attract legislators from opposition parties, rewarding them with ministerial

berths. Factionalism within the BJP ultimately led to the dismissal of Kalyan Singh in

November1999,andhis replacementbyanageingRamPrakashGupta,himselfquickly

replacedbyRajnathSingh.

The 2002 State elections produced once again a hung Assembly, the Samajwadi Party

cameout aheadwith143 seats, theBahujanSamajParty secondwith98 seats and the

Bhartiya Janata Party third with 88 seats128. The incapacity of the SP to find coalition

partnersoutsidetheBSPandtheBJPledthesetwopartiestotieupforathirdtime.The

rallying of Ajit Singh’s Rashtriya Lok Dal and of a number of small parties and

independentsenabledthemtoobtainamajorityintheLowerHouse.Onceagain,thetwo

partners grew apart and the Mayawati-led government fell in August 2003, when the

Rashtriya Lok Dal withdrew its support to the Mayawati led government, with the

blessingsoftheBJP.

AjitSingh,thenalsoaUnionMinister inthesecondNationalDemocraticAlliance(NDA)

governmentattheCentre,shiftedhissupporttotheSamajwadiParty,whothuswrestled

powerbycobblingtogetheracoalitionofsmallparties,theremainsoftheCongressParty,

seven independents and thirteen defectors from the BSP129. The third Mulayam Singh

Yadav’sgovernmentlasteduntiltheendofthetermandwassucceeded,forthefirsttime

since1991,byasinglepartymajoritygovernment,ledbyMayawati.

128TheBJPhadapre-pollalliancewitheightparties:RashtriyaLokDal (AjitSingh), JanataDal-United,LokJanshaktiParty,theSamataParty,ManekaGandhi’sShaktiDal,AmarmaniTripathi’sLokTantrikCongress,theLokParivartanParty(RKChaudhariandBerkhooRamVerma)andtheKisanMazdoorBahujanSamajParty(ChaudhariNarendraSingh).TheBJPalsosupportedthreeindependent candidates – Raghuraj Pratap Singh, alias Raja Bhaiya (Kunda), Ram Nath Saroj(Bihar)andPappuJaiswal(Pipraich).InPai(2002b).

129The Samajwadi Party formed the government in 2003 with 142 seats, with the support ofCongress (16 seats), theRashtriya LokDal (14 seats), theRashtriyaKranti Party (2 seats), theCPI-CPM (2 seats), smaller parties and Independents (19) and 13 defectors from the BahujanSamajParty,foratotalof208seats.

92

2.1.4.Thelimitsofcastepolarization:Post-identitypolitics?

Inter-partyacrimonyandintra-partyfactionalismwerenottheonlytworeasonsbehind

theinabilityofpartiestoformstablegovernments.Thefour-partdivisionoftheelectoral

scene also stemmed from the fact that each party devised electoral strategies that

targetedspecificandfundamentallynarrowsegmentsoftheelectorate,attheexclusionof

themajorityofvoterswhodidnotbelongtothosesegments.Torenderisinasimplistic

manner, the BJP targeted the upper castes and the lower backward castes through its

religious mobilization, conducted on the ground by the RSS. The BSP sought to

consolidateitssupportamongDalitsbyantagonizingtheuppercastesandtheOBCs.The

SamajwadipartysoughttoachieveasimilargoalbyfavoringitsYadavbase,whilewooing

MuslimswhohadlenttheirsupporttoMulayamaftertheAyodhyaincident.TheCongress,

deprivedofacoresupportbaseofitsown,furtherdeclined.

Inreality,electoralbehaviourwasmorecomplexanddidnotfollowthissimplisticfour-

partdivisionof thepolitical space.Noneof theseaforementionedsocial categoriesvote

foranyspecificpartyenbloc.Parties’supportbasesamongvariousgroupskeepchanging

overtimeandoverspace(seeTable2.6andsection2.2.1).Somecastesdoclusteraround

specificparties,providingthemwithacoresupportbase.Buteventhesecorecaste-party

alignmentsarequitefluid.TheBJPsucceededinconsolidatingthesupportfromtheupper

caste intheearly1990sbutthatsupportquicklyeroded,tothebenefitofotherparties.

TheeffectofthereligiousappealoftheBJP’scampaigndidnotlastanduppercastesvotes

gotrapidlydividedamongvariousparties.

ItisoftensaidthatuppercastesinU.P.donotbelongtoanypartyandvotestrategically

accordingtolocalconfigurations,withtheaimofmaximizingtheirrepresentationinthe

Assembly. It isparticularlytruefortheThakurs,whothroughthe2000ssplit theirvote

moreandmorebetweentheBJPandtheSP.Itisalsonecessarytokeepinmindthatonly

afewcastesparticipateintheselargecaste-partyalignments.Mostcastesareeithertoo

small or too geographically dispersed to constitute effective blocs of voters, beyond

specificconstituenciesordistricts.Ifweaddtheknownuppercastestotheofficialstate

listofOBCandSCcastes,andifwecomparethatlisttothecastecompositionofthestate

93

assemblyovertime,onecanseethataround40percentofcastesinUttarPradeshnever

hadanyrepresentationinthestateassembly.

Table2.6Casteandcommunity-basedpartypreferencesinUttarPradeshassemblyelections,

1996-2012130 Congress BJP 1996 2002 2007 2012 1996 2002 2007 2012Brahmins 4.00% 26.50% 18.70% 12.90% 70.70% 48.70% 39.10% 37.80%Rajputs 4.50% 9.70% 9.40% 13.30% 74.80% 45.70% 43.50% 29.40%Vaishyas 3.70% 12.10% 9.80% 20.90% 81.10% 48.30% 44.60% 41.80%OthersUcs 2.30% 22.40% 12.00% 12.90% 78% 44.80% 39.30% 16.90%Jats 6.50% 2.70% 10.60% 31.80% 9.10% 21.10% 6.80%Yadav 6.60% 4.70% 4.10% 4.30% 5.40% 5.20% 3.90% 9.30%Kurmis/Koeris 6.20% 41.40% OtherOBCs 5.70% 7.10% 7.90% 12.60% 47.30% 27.10% 19.00% 18.90%Jatavs 14.30% 4.00% 2.10% 4.50% 5.30% 2.00% 3.00% 4.70%OtherSCs 12.50% 9.00% 5.10% 13.60% 7.90% 11.40% 10.50% 8.20%STs 9.10% 54.50% Muslims 12.20% 10.00% 14.10% 18.00% 1.90% 1.70% 2.40% 6.60%Others 6.30% 4.30% 12.40% 9.10% 36.80% 13.00% 13.70% 15.60%Total 8.10% 32.10% 20.10% 22.40% 17.30% 8.80% 19/7% 22.80%

BSP SP 1996 2002 2007 2012 1996 2002 2007 2012Brahmins 3.70% 6.00% 16.70% 19.00% 5.50% 2.60% 10.30% 18.80%Rajputs 5.30% 4.80% 11.90% 13.60% 1.90% 8.10% 20.60% 25.50%Vaishyas 4.90% 3.40% 13.60% 14.90% 3.70% 15.50% 12.00% 11.90%OthersUcs 1.20% 5.20% 12.00% 17.10% 5.80% 13.80% 14.90% 14.10%Jats 0.90% 12.30% 15.90% 4.50% 11.10% 6.80%Yadav 3.30% 5.20% 7.70% 10.70% 60.70% 70.80% 72.50% 65.80%Kurmis/Koeris 11.70% 31.40% OtherOBCs 14.50% 19.20% 26.50% 18.80% 15.90% 17.60% 19.60% 29.30%Jatavs 64.90% 78.50% 84.80% 61.90% 6.00% 1.60% 3.50% 14.70%OtherSCs 60.60% 55.10% 55.00% 47.90% 9.90% 14.40% 13.60% 18.50%STs 9.10% Muslims 12.30% 9.70% 17.60% 30.40% 48.00% 53.00% 47.70% 39.40%Others 20.10% 13.00% 29.20% 23.40% 14.60% 13.00% 23.00% 30.60%Total 24.70% 24.00% 8.60% 17.00% 30.40% 25.40% 18.60% 11.60%Source:CSDS/LoknitiNESData.

130DatabasedonsurveysconductedbyLokniti.TheNforthefoursurveysare,chronologically:5592,2058,9530,and6614.

94

Thenotionofcoresupportbasehowevershouldnotbediscarded.Partiesarticulatetheir

discourse and policies around the specific interests of their core support base – in the

formofsubsidiesorloanwaiversforfarmers,housingschemesforScheduledCastes,and

soforth.Further,thecontroloflocaldemocraticinstitutions,suchasdistrictPanchayats

(orZillaParishads)orcooperatives,isoftenexertedthroughlocalcaste-basedpatronage

networks, in which the alignments between specific castes and specific parties often

determinehowpublicresourcesareallocated.

As a consequence, the focalization of parties on specific groups limited their ability to

expandtheirsupportbaseacrossgroups,asthemobilizationofacoresupportgroupon

thebasisof identityor religiousaffiliationoftenhas theeffectofalienating thesupport

fromothergroups.

Thistrendwouldcontinuethroughthe1990stowardstheendofwhichpartiesstartedto

changetheir toneand implement lessexclusiveelectoralstrategies.Until then,electoral

strategiesofpartiesaimedatconsolidatingacoresupportgroupandseekrepresentation

fromothergroupsinareaswheretheircoregroupwasn’tdemographicallystrongenough.

Inthefourthphase,partiesseektotransferthevotesoftheircoresupportbasetowards

candidatesbelonging toother groups, inorder to createwinning local social coalitions,

includinginconstituencieswheretheircoresupportbaseisstrong.

The systematic failure of coalitions led the two regional parties to broaden their social

base by wooing each other’s bases and by distributing tickets across castes and

communities (seechapter4).Theychanged theirdiscourse fromcaste-baseappeal toa

more generalist discourse on social justice, calling on a broader definition of the

backwards,inthecaseoftheSamajwadiParty,ortheredefinitionoftheBSPasasarvajan

party(“fortheentiresociety”),ratherthanabahujan(“majority”)party–thatisforthe

lowercastes(Jaffrelot2010a,Pai2009).Thiswasatwo-prongedstrategy.Thefirstaspect

was to develop a catch-all discourse, focused on development and social justice. The

seconddimension,crucial,wastodistributecandidatureticketsacrosscastesinorderto

matchthecatch-allambitionwithpractice.

95

However,contrarytopopularbelief,thisdistributionofticketsdidnotnecessarilyfollow

a particular pre-determined calculation of inter-caste balance. Instead, parties focus on

the local demographic and sociological features of constituencies to determine whom

candidatefromwhichgroupshouldgettheticket.

The logic consists in seeking the best possible local combination of castes, in order to

maximize the party’s chance of winning. This system draws on the notion of

transferability of vote bank, or the transfer of the votes from the core supporters of a

partyinfavorofacandidatefromanothercaste.TheBSP’scalculation,forinstance,isthat

with an average of 20 per cent of Dalit voters across constituencies, it needs to seek

candidatewhocandeliveratleasttwentyotherpercentofthevoteswhich,addedtothe

local Dalits support, will guarantee victory (see Section 2.3.2 for more details on this

arithmetic). Thus, the choice of candidate is guided by local considerations and

pragmatism,andnotsimplybyabroadnewly-foundinclusiveideology.

Thisfactprovidesthebasefortheargumentthatcastepoliticshasbecomemorelocalized

in the post-Mandal and post-Mandir years, against the popular perception that parties

have suddenly become less caste-minded because they suddenly have stopped talking

aboutcastepublicly.The2007assemblyelectionsareacaseinpoint.

The2007elections:aturningpoint

The2007electionsmarkedaturningpointinUttarPradeshpoliticsastheBSPsucceeded

inwinning a singlemajority of seats (206out of 403),with30.5%of the vote share, a

positiveswingof7.3%comparedto2002.TheBJPvotesharedecreasedby3.12percent,

to17percent.TheSPandCongressbothmaintainedtheirvoteshareat25.4and8.6per

centrespectively.

TheBSPsucceededforthefirsttimetoattractasubstantialpartoftheuppercastevote,

having fielded a large number of upper caste candidates. Between 2002 and 2007,

Mayawati doubled the number of tickets distributed to upper caste candidates,

particularly in favor of Brahmins and Vaishyas (see Chapter 3). The Samajwadi Party

adoptedavotebaseexpansionstrategy too,by fieldinga largenumberofnon-Brahmin

96

candidates, notablyBanias andRajputs. Bothparties in sumattempted to expand their

basebywooingthesocialcategoriesthattheyhadinitiallyopposed131.

Why did this strategy work better for the BSP than the SP? The first reason usually

invokedisthatthecohesionofthecoresupportbaseoftheBSP–theDalits–islarge(21

per cent of the population) and stronger than in other parties. In the 2002 assembly

elections,72.2percentofDalitsvotedfortheBSP(SeeTable2.7).Thatnumberreduced

significantlyin2007butremainedabove50%withinthisbroadsocialcategory132.

Table2.7DalitvotingproportionsbypartyinUttarPradeshassemblyelections,2002-2012UttarPradesh 2002 2007 2012INC 4.20% 7.40% 8.60%BJP 6.40% 5.00% 5.90%BSP 72.20% 52.00% 57.70%SP 7.60% 27.60% 14.30%Others 9.50% 8.00% 13.50%

Source:NESData.QuotedfromGorringe,JefferyandWaghmore(2016).WhilethecoresupportgroupoftheSP,theYadavs,isindeedsmaller(8.7percentofthe

population,accordingtothe1931Census),itisnolesscohesivethantheBSP’ssupport

amongtheDalits.Itisinfactmorecohesivein2012.

Whatthedataabouttheselasttwoelectionsindicateisthatthecoresupportofthesetwo

partiesactuallyerodesbetween2007and2012,aprocessthathadstartedearlierforthe

BSP.Atthesametimebothpartieshaveincreasedtheirsupportamongtheuppercaste,

moresoamongtheBrahminsfortheBSP,andamongtheRajputsfortheSP.Bothin2007

and2012,thebulkoftheuppercastesremainedwiththeBJP133.

131This strategy is what A.K. Verma, a political scientists from Kanpur, called the ‘reverseosmosis’and‘sandwichcoalition’strategies(Verma2002b).132Dalits, like any other caste group, do not form a cohesive entity. We know through Loknitisurveys that various groups among Dalits support the BSP differentialy. In the 2009 generalelections,86percentofJatavsvotedfortheBSP,against64percentforPasisand61percentforotherDalits(Verma2007b,a).133In2007,theBJPreceived44percentoftheBrahminvoteand46percentoftheRajputvote.Thesenumbersdecreasedto38percentand29percentrespectivelyin2012.Verma,op.cit.

97

Table2.8CasteandCommunity-basedpartypreferencesinUttarPradeshassemblyelections,

2007-2012 BSP SP

Jati 2007 2012 2007 2012Brahmins 16 19 10 19Rajput 12 14 20 26Vaishya 14 15 12 12OtherUpperCastes 10 16 17 15Jats 7 10 8 7Yadav 7 11 72 66Kurmi/Koeri 16 19 17 35OtherOBC 30 19 20 26Jatav 86 62 4 15Balmiki 71 42 2 9Pasi/Pano 53 57 16 24OtherSC 58 45 16 18Muslim 17 20 45 39Others 30 23 23 31

Source:CSDS-LoknitiSurveyData.QuotedinVerma(2009).Figuresinpercentandroundedoff.

Thesecondreason,morecogent, isthattherewerelesscontradictionsintyingapartof

theuppercastevotetotheDalitvotethanwithOBCvoters.Inmanypartsofthestate,the

social,economicandpolitical competition tends to takeplaceamong theupperand the

backward castes, in particular the dominant OBCs, who still form the bulk of the SP’s

organization. The induction of Rajput and Bania figures in the 2007 elections created

tensionsandresentmentwithintheranksoftheSP.AseriesofSamajwadiPartyfigures

left the party in protest or refrained their campaign enthusiasm. As we shall see in

Chapter four, the local branches of the Samajwadi Party are mingled with local elite

groups.Mostof its candidatesaredrawn from its localorganization.While for theBSP,

there is a strict division of labour between the party organization, predominantly

composed with Dalit members, and the candidates who are recruited outside the

organization134.Although this duality created tensions after the elections, therewasno

contradiction in bringing in the ranks of the party elements who did not share the

emancipatoryidealsoftheparty’sbase.FortheSP,inductingoutsidersalmostinevitably

generateconflictsofinterests.

134Thisappliestogeneralseats.

98

Athirdreason,morecircumstantial, thatexplainedthegapinperformancebetweenthe

SPandtheBSP,isthesimplefactthattheSPwastheincumbentpartyandthatsince1985,

noparty inUttarPradeshhassucceeded inbeingre-elected.Besides, thepartysuffered

fromapoorimageduetothedeteriorationoflawinorderundertheirrule.Thespectreof

thegoondaRaj(criminals’rule)hauntedthepartyanditscampaign.Itdidnothelpthatin

ordertowootheuppercaste,theSPchoseascampaignmascotssomeoftheworstpublic

criminalsthestatehasproducedinrecenttimes.ItgaveatickettoAmarmaniTripathi,a

Brahmin leader from the East and to one of U.P. most archetypal criminal politician,

RaghurajPratapSingh(aliasRajaBhaiya),aThakurfigurefromKunda(North-East).

That being said, the Samajwadi Party resisted well despite these advantages, since it

maintained itsvote share.TheSPwon97seatsand trailed in167seats (including123

againstaBSPcandidate).Intheseelections,BSPandSPcandidatesoccupiedthefirsttwo

slotsin200seats,outof403.

TheBSPcampaignsucceededbydistributingticketsacrosscastelinesandbyappealingto

a broader base of voters through a generalist discourse focused on the promise of

development and caste inclusion, but it was also greatly helped by a series of

circumstances.

ThefirstonewasthatbothCongressandtheBJPwerestillreelingfromtheirdefeatinthe

2002 elections and in the 2005 by-elections. This made the election a two-horse race,

instead of the three-corner contest of 2002 (Pai 2009). The BSP in 2007 reaped the

benefitofashift inelectoralstrategy initiated twodecadesearlierbyKanshiRam,who

hadalreadystartedtoopenthegatesofthepartytonon-Dalitcandidates.Thedifference

thenwastheBJPwasamuchstrongercontenderfortheuppercastevote,whoshunned

theBSPonaccountofitsexplicitlyanti-uppercasterhetoric.Thesecondcircumstanceis

thattheSamajwadiParty,aswesaw,sufferedfromanincumbentdisadvantage.

Moreimportantly,theBSPbenefitedfromthelowturnout(46%),whichloweredthebar

ofabsolutevotesnecessarytoconvertaminorityofvotesintoamajorityofseats.Many

fieldaccountsdescribedhowBJP,CongressandevenSPsupportersdidnotvotethatyear,

99

notexpectingtheirpartytodowell.Inaway,theBrahminswhodidnotvotein2007also

helpedtheBSP,alongsidethosewhovotedforitscandidates.

The2007BSPvictory,followedbytheSPvictoryfiveyearslater,ledmanycommentators

to pronounce the end of caste as a factor in Uttar Pradesh politics. They harbored the

illusion that the BSP had shed its caste-mindedness and saw in the rainbow coalition

gatheredbyMayawatithesignthatcastecalculationsandnarrowstrategieshadfailed,to

thebenefitofanewly-foundsenseofgeneralinterest(GuptaandKumar2007).

Sudha Pai describes perfectly to what extent the party went to attract Brahmins, by

organizing Brahmins jodo Sammelans (Brahmin enrolment conventions) 135 , using

brahminical rituals such as Vedic hymns or the blowing of conches, and by organizing

other caste-specific events across the state (Pai 2009). The very fact that the BSP

addressedcaste-basedralliesanddevisedcaste-basedtacticsandstrategiesindicatethat

castewasverymuchattheheartofitsstrategy,evenifabsentfromitsinclusivediscourse.

Gupta and Kumar are right however in their critique that caste alone does not explain

electoral outcomes. Other scholars, more nuanced, have noted the gradual decline of

ascriptiveidentities,thepresenceofnewalignmentsofpartiesandvoters,andthegreater

appeal among voters of transversal issues such as security, law and order, and

developmentandtheincreasedsaliencyofclassdiversificationofcastegroups,ortherole

ofintra-castedisparities(Pai2013,Singh2014,Sridharan2014,Jaffrelot2015a).

The 2012 successful Samajwadi Party campaign seemed to confirm that trend. The SP

campaignedonthethemeofinclusivedevelopment,socialharmony,withafocusonthe

youth.ThesonofMulayamSinghYadav,Akhilesh,ledthecampaignasthenewfigurehead

oftheparty,projectinganimageofyouthfulchange.Itscampaignslogan–UmmedkiCycle

(the cycle of hope – the bicycle also being the party’s symbol) – was an inclusive one.

Televised advertisement showcased a social rainbow encompassing farmers, students,

urban professionals, housewives and women professionals. The party also organized a

massiveroadshow–aKrantiRath–reminiscentofMulayam’s1987campaign,inwhich

135TheseconventionsexcludedDalits’participation.

100

Akhileshheld5to6meetingsaday,addressingaudiencesinthehundredsofthousandon

adailybasis.

Butdespite the inclusive toneof the campaignand theassertions that candidateswere

selected on the basis of competence and clean records, the distribution of tickets

remained essentially based on the local reading of caste configurations. Instead of

attempting to forge transversal social alliances, the SP emulated the BSP strategy

consisting in pickingup candidates according to local circumstances. This resulted in a

morediversifiedprofileofcandidates(seesection4.1.3).

Parties nowadays tend to lead parallel campaigns. At the macro level, they develop an

inclusive discourse stressing on mobilization tropes that have a catch-all appeal –

development, social justice and providential leadership. At the local level, the electoral

competition remains grounded into the local configurations of competition among

contendingsocialgroups,oftenorganizedalongcastelines. Negatingdifferencesonthe

public stage while activating these differences locally has become the formula of a

successfulelection,aformulathattheBJPwouldadoptinUttarPradeshandelsewherein

the 2014 general election. In both cases, the traditional tropes of mobilization do not

disappear but are rather concealed under a generalist inclusive discourse adopted by

parties.

2.2.Sub-regionaltrajectories

Thedescriptionmadeso farofpolitical trendsanddynamicshasbeenmade ingeneral

terms,consideringthestateasawhole,regardlessofsub-regionalvariations.Itisobvious

however that the transformationsdescribedandanalyzedbeforeaswell as thegrowth

anddeclineofpoliticalpartiesdidnottakeplaceuniformlyacrosstheterritory.Thereare

necessarily spatial variations, which are important to consider. To begin with, political

competition takes place in a variety of contexts that include demographic variations,

differentiated,variedeconomictrajectoriesandlocalizedsocialtransformations.Mostof

theliteratureonelectoralpoliticsconsiderseitherthestateortheconstituencyasunitof

analysis, as if therewasnothingelse inbetween,oras if somesingleconstituencycase

wasrepresentativeofthewhole.

101

The State of Uttar Pradesh is divided into various sub-regions that represent cohesive

historical,socialterritorialentitieswithinthestatetopeopleandthereforevoters.Some

of these sub-regions carry specific sub-regional identities. In fact, State boundaries cut

across, or artificially divide, a number of ethno-cultural regions. The sub-region of

Bundelkhand belongs to a larger ensemble located in Northern Madhya Pradesh. The

regionofPoorvanchal, intheEast,hasmuchtosharewith itsneighborBhojpur,partof

thestateofBihar.TheJat-dominatedregionsofWesternU.P.andHaryanahavealsomuch

in common, somuchso thatapartywas created in the1960s tounite the “Jatland”on

both sides of the Yamuna River136. Finally, the Shravasti and Balrampur districts also

belongtoalargesocio-geographicalensemble–theTerai–aplainregionthatstretches

fromHimachalPradeshtoBangladesh,throughNepal,UttarPradeshandNorthernBihar.

Forourpurpose,IstaywithintheboundariesofthestateofUttarPradesh,andfindthat

dividingaggregatepoliticaldataatasub-regionallevelrevealsimportantvariationsthat

challengethenarrativesandexplanationsthatscholarsandcommentatorsputforwardto

makesenseofthestate’spolitics.

2.2.1.UttarPradesh’ssub-regions

Therearemultiplepossibleways todivide the territory.Geographicallyspeaking, there

are three broad sub-regions in Uttar Pradesh – The plain area of the Gangetic basin,

where thebulkof thepopulation lives, theNorthernMountains and the Southernhills.

Agronomists divide the territory into eight agro-climactic zones137. In her study on the

correlationbetweenelectoralandeconomicvariables inU.P.,FrancineFrankelusesfive

136TheVishalHaryanaParty(the«GreaterHaryanaParty)wascreatedbyRaoBirendraSingh,aCongress defector, in 1967. Singh became the first Chief Minister of Haryana that year, to bequicklyunseatedandreplacedbyaCongressChiefMinister,BansiLal.Singh,whohadbeenoneoftheleadingvoicefortheseparationoftheHindi-speakingdistrictsfromPunjab,mergedhispartywiththeCongressin1978.137The Western Plain Zone, Mid-western Plain Zone, South-western Plain Zone, Central PlainZone,BundelkhandZone,North-easternPlainZone,EasternPlainZoneandVindyanZone.

102

ecological regions (Frankel 1977, 153-156)138. Sudha Pai uses an economic criteria to

divide the territory into five sub-regions: The Northern Hills (currently Uttarakhand),

WesternU.P.,centralU.P.,EasternU.P.,andBundelkhand(Pai1993,29).Administratively

speaking, the state of Uttar Pradesh is divided into 71 districts, clustered into 18

divisions139.ThelatePlanningCommissionusedtodividethestateintofoursub-entities–

Bundelkhand, Central, Eastern and Western Uttar Pradesh – each region being further

dividedintoanumberofcircles,aggregatingthreetosevendistricts(India2007).Finally,

partieshavetheirownwaystolookatthestate’sterritory.TheSPandtheBJPbothdivide

itinto10circles,basedontheirreadingofcastedemographics.TheBSPdividesthestate

intofourentitiesbasedontheprojectionofthestatecarvedintosmallerstates140.

I choose to divide the territory into seven sub-regions (or eight, if one includes

Uttarakhandbefore itscreationasaseparatestate)141, followinganomenclature that is

widelyreferredtopopularly,orpartofcommonparlanceinthestate142.Theseterritories

roughlycorrespondtocohesivesocio-politicalhistoricalentities,althoughtheboundaries

of these territories are necessarily shifting through time. They are distinct in their

histories,demographicsandsocio-economictrajectories.

138TheHimalayanRegion,TheWest,CentralandEastPlain(thatistheGangeticbasin),andtheSouthernHillsandPlateauRegion.139Saharanpur, Moradabad, Bareilly, Lucknow, Devipatan, Basti, Gorakhpur, Meerut, Aligarh,Agra,Kanpur,Faizabad,Azamgarh,Jhansi,Chitrakoot,Allahabad,VaranasiandMirzapur.140ThebifurcationplanincludesHaritPradeshcomprising22districtsinWesternUttarPradesh(another version is named Braj Pradesh or Paschimanchal, and includes parts of NorthernMadhyaPradesh);Awadh,orCentralUttarPradesh,twicethesizeofBelgiumandfivetimesitspopulation;Poorvanchal,comprisingtheEasternandNorth-easterndistricts;andBundelkhand,clubbingthetwopartsofthathistoricalregioncurrentlylyingacrossSouthernU.P.andMadhyaPradesh.141So does Ralph C. Meyer in his 1969 study of the sociological profile of U.P. legislators. Hedivides the territory into 8 entities: Mountains, Northwestern (Rohilkhand and Western U.P.),Western (around Agra), Central, North-Central (Pilibhit, Kheri and Bahraich districts), NearEastern,FarEasternandSouthernU.P.142Although technically, Western Uttar Pradesh should be considered as Upper Doab, it makessensetorefertoitasaseparatesocio-politicalentity.

103

In2012,thedistributionofthenumberofregisteredvotersacrossthesefoursub-regions

isasfollows:

Table2.9Registeredvoterspopulationpersub-region

Sub-region Registeredvoters

AVADH 24.36%BUNDELKHAND 5.00%

DOAB 21.51%EAST 15.58%

NORTH-EAST 10.37%

ROHILKHAND 12.31%WEST 10.88%

Total 100.00%

Source:Jensenius,F.,Verniers,G.IndianStateAssemblyElectionandCandidates

Data(1962-Present)The largest sub-region, Avadh, comprises 100 assembly seats and a quarter of the

electorate. This sub-region contains some of the richest districts in the state – in

104

particular the state’s capital, Lucknow – and some of its poorest, such as Shrawasti,

BahraichandBalrampur(seeAnnexure3).Thenameofthatsub-regioncomesfromthe

eponymprincely state, ledbya Shiadynasty from theearly SixteenthCenturyuntil its

annexationbytheBritishin1856.

Thesecondlargestsub-region,Doab,isalongtractoflandthatstretchesfromtheWest

totheEast,followingthebasinoftwoconfluentrivers–theGangesandtheYamuna.Both

rivers joins at Sangham, in Allahabad, which marks the Eastern boundary of that sub-

region.Historically,Doabwasdividedintothreeareas:UpperDoab,comprisingpartsof

Uttarakhand,WesternUttarPradeshandDelhi,CentralorMiddleDoab,aportionofthe

sub-regionthatstretchesfromAligarhandAgratotheSoutherndistrictsofEtawahand

Kannauj.Andfinally,LowerDoab,whichcomprisesthedistrictsstretchingfromKanpur

toAllahabad.Inthisdissertation,DoabreferstoCentralandLowerDoabandcontains89

seats.

Thethirdsub-region,Rohilkhand,drawsitsnamedfromtheRohillas,acommunityfrom

Afghanistanwho founded a Pashtun state in these areas in the 17th Century. This sub-

regionislocatedintheupperGangesalluvialplains,southofUttarakhandandNepal.Itis

the cradle of several Muslim-rules princely states, such as the Rampur State (whose

current heir is an MLA). This sub-region comprises 52 constituencies, many of them

containing a large number of Muslim voters, who represent 35.31 per cent of the

populationinthatregion(against19.26percentatthestatelevel).

The East is divided into two sub-entities – the East (61 seats) and the North East (40

seats). They are also often jointly referred to as Poorvanchal and are nestled between

Nepal intheNorth,AwadhontheWest, thestateofBihartotheEastandtheregionof

Bagelkhand in the South, bordering Madhya Pradesh. The main cities are Allahabad,

Varanasi in the East, and Gorakhpur in the North East. The predominant language in

theseareasisBhojpuri.

105

Table2.10Sub-regionwisepopulationandshareofMuslimpopulation,perlocality

Sub-Region Totalpopulation

Rural Urban TotalMuslims(%)

TotalRuralMuslims

TotalUrbanMuslims

MuslimsinRuralareas%

MuslimsinUrbanareas%

Avadh 51489825 85.26% 14.74% 19.40% 74.48% 25.52% 16.95% 33.58%

Bundelkhand 9681552 77.33% 22.67% 7.17% 47.61% 52.39% 4.41% 16.57%

Doab 42849809 71.03% 28.97% 12.43% 44.99% 55.01% 7.87% 23.60%

East 29741466 84.71% 15.29% 11.69% 58.87% 41.13% 8.13% 31.45%

North-east 18065568 90.34% 9.66% 16.57% 87.09% 12.91% 15.98% 22.15%

Rohilkhand 25798559 74.01% 25.99% 35.31% 60.18% 39.82% 28.71% 54.11%

West 22185562 58.10% 41.90% 31.08% 55.89% 44.11% 29.90% 32.72%

Total 199812341 77.73% 22.27% 19.26% 62.77% 37.23% 15.55% 32.20%

Source:AdaptedfromCensus2011.

The Eastern districts have historically lagged behind the rest of the state in terms of

economicdevelopment.WithBundelkhand,theyaretheleastindustrialized.Mostofthe

people living below the poverty line are concentrated in the East, where per capita

income is also far below the state average143. Most of the rural economy rests on

subsistenceagriculture,withlittlecropdiversification.Thefewindustriesthatremained

in the East – notably textiles in and around Varanasi, or carpets around Mirzapur and

Bhadohi (once one of Asia’s largest hand-woven carpets production centers) have

considerablysufferedinthe lastdecade,duetocheaperChineseimports,highercostof

rawmaterials(suchasthewoolyarn)andtothefalloftheRupee(Bhatt2013)144.

Before Independence, these two sub-regions followed the zamindari system of land

revenue system (zamindar literally means ‘land-holder’), a system that relied on big

landlordsbelongingtolocalhereditarydominantfamilies,mostofthemuppercastes145,

whoexertedtheirpowerandinfluenceonbehalfofahigherformofpoliticalauthority–

the colonial state before 1947, and the Congress party, for a time after Independence,

143PlanningCommissionofIndia,UttarPradeshDevelomentReport,op.cit.144Carpet weaving started in this region in the 16th Century. The Bhadohi carpets are nowmanufactured across nine districts in Eastern U.P., employing around 3.2 million people,including2.2millionruralartisans.145ThethreemaindominantuppercasteinEasternUttarPradesharetheBrahmins,theKayasthsandtheBhumihars.

108

Theseventhsub-region isWesternUttarPradesh,acomparativelysmaller tractof land

that borders the capital, Delhi, and stretches North towards the Himalayan range and

Uttarakhand,alongsidetheEastbankof theYamuna.Owingto itsgeographic location–

near the capital and bordering Haryana, Western U.P. is the most urbanized and

industrializedsub-region151.Muchofthewealthof thestate isconcentratedin its forty-

fourconstituencies,whiletherestofthestateiscomparativelylowincome.

Socio-economically speaking,WesternUttarPradesh shares some important traitswith

the neighboring state of Haryana. The land tenure regime historically follows the

bhaiacharasystem,aregimemarkedbyawidedistributionoflandamongself-cultivating

landowningcommunitieswholargelybelongtothemiddlepeasantcastes.Inthecaseof

HaryanaandWesternUttarPradesh,theJatsoccupyadominantposition.InWesternU.P.,

other groups, such as the Gujjars (both Hindus and Muslims), Tyagis and Rajputs also

practice this formof land tenure system.Compared to the zamindari systemandother

relatedagrarianregimes, land ismoreequallydistributedwithinparticular landowning

groups, which does not preclude the exclusion of lower castes and Dalits, who largely

workaslabourersonlandthattheydonotown(Friese1990,136).

The bhaichara system also prevails in Bundelkhand, where the OBCs have historically

ownedmostoftheagricultureland.ItisnotentirelyabsentintheEastbutismuchrarer

thanthezamindarisystem(Stokes1975,516).

ThegrowththathasoccurredinseveraldistrictsofWesternUttarPradesh–particularly

thoseadjacenttoDelhi,ismorelinkedtothegrowthofthenationalcapitalthantorestof

the State. Arguably, if the recent economic transformations of areas such as Noida,

Ghaziabad, and the commercial and industrial belt on the road to Meerut were made

possibleinpartbyimportantpolicychangesattheStatelevel,theybenefitedmorefrom

their proximity from Delhi and specific policy treatment that did not concerned other

partsoftheState(specificlandtenureregulation,politicsofdispensation).

151TheeightdistrictsofWesternUttarPradeshconcentratemorethanhalfoftheState’sindustry.SeePlanningCommissionofIndia,UttarPradeshHumanDevelopmentReport,2006.

109

Beyondtherealestate,industrialandIThubsofGhaziabadandNoida,alargepartofthe

industry in Western Uttar Pradesh is linked to agriculture, notably food processing.

WesternU.P.isthefoodandsugarbasketofthestate(India2007,32)152.

These variations, very briefly sketched here, are important to keep in mind since they

produce varied socio-economic contexts for electoral competition. The land tenure

regime, the hold of certain groups over economic resources, the diversification of the

rural economy and the opportunities offered by urbanization have all political

consequences.

Sub-regionalpartytrajectories

TheJanaSanghinitiallyemergedasapoliticalforceintheurbansegmentsofAwadhand

throughoutRohilkhand,twosub-regionswhereformerrulersandbiglandlords–mostly

uppercastes–hadretainedmuchoftheirinfluencepost-Independence.Itthendeveloped

a presence in Bundelkhand and in the Eastern district, two areas adjacent to Madhya

Pradesh. The Jana Sangh’s baseswere nestled among the upper segments of the urban

electorate–tradingcommunities,businessmen–aswellamongthelargelandownersin

theruralareas(Brass1984a,Burger1969).Towardsthelate1960s,theJanaSangmade

inroads into rural constituencies, by building support among the middle class

agriculturist. Italsoexpandeditsholdorurbanseatsbygarneringthesupportofurban

professionals,andalsodevelopedastrongcadreoffulltimepartyworkers(Pai1993,55-

56).

Afteritsre-foundationin1980,thepartyregaineditsstrengthfirst inBundelkhandand

Avadh. By 1991, the BJP had spread across the eight sub-regions of the state, scoring

above 30 per cent vote shares, barring the East, where it lagged at 25 per cent.

Subsequently,itregistereditshighestscoresinUttaranchal,carvedasaseparatestatein

2000.Itsdeclineinthe2000swasmoremarkedinAvadh,whereitlost10pointsofvote

share between 2002 and 2012, Bundelkhand, where it lost 11 per cent of vote share

152WesternU.P.produces45percentofthestate’sgrainproductionand60percentof itssugarproduction.

110

between2002and2007.In2012,theBJPissituatedbelowthe20percentbarofaverage

voteshareacrosstheeightsub-regions.

TheperformancesoftheCongresshavebeenfairlyuniformacrosstheterritoryovertime.

In the 1960s, the party was strongly ahead across all sub-regions, with a stronger

presence (measured invote share) inUttarakhandandAwadh. Itsperiodof vote share

consolidationinthe1980salsotookplaceacrossthestate,andsodiditscollapse,which

startedtobemeasuredinthe1989elections.TheCongressloses13percentofvoteshare

acrosstheeightsub-regions,thatyear.In1996,itlosesafurther7.6percentofvoteshare

again across the territory. In fact, the Congress resisted comparatively better in

Uttarakhand,where itmaintainednearly30%voteshareuntil1993.But itdidcollapse

there too, before the region was carved out as a separate state. The other sub-region

whereCongress comparativelyperformsbetter isBundelkhand,where it grew from11

percentin2002to18.6percentin2012.

Historically, the socialists had a weak presence in Uttarakhand, Rohilkhand, and in

Bundelkhand. They initially emerged in the West, where the alliance between Jats,

Muslimsandother landedbackwardgroupssuchas theGujjarsand theAhirs (Yadavs)

formedawidebasesupportingCharanSinghandtheBKD.Inthe1969elections,theBKD

received 34.45 per cent of the votes in that region against 21.3 per cent in the state

overall.Thepartymade inroads in theEast in themid-1970s,after itsalliancewith the

SSP(Duncan1988,41).ItgrewfurtherincentralU.P.butretainedtheWestandDoabas

theirtwopoliticalbases.

In 1991, the split of the Lok Dal caused the Mulayam faction to drop to 6.2 per cent,

against47.7percenttwoyearsearlier.Sincethen,theSPneverquiterecoveredthespace

the socialistsoccupied in that region.Thepolarizationbetween theMulayam factionof

the LokDal and theAjit Singh faction, backedby the Jats, explainswhy the Samajwadi

scoresparticularly low in thesubregion in theearlyandmid1990s.Also,both theBJP

andtheBSPprogressedthereinthe1990s,byconsolidatingrespectivelytheuppercaste

and theDalit vote, and by denting into the support of lowerOBCs for theBJP, and the

Muslimvote,fortheBSP.

111

Today,theSPtendstobeweakerinU.P.’sborderdistrictsandstrongerinthehinterland.

Avadhcounts38constituenciesconsistentlyheldbyanuppercasteMLA(20byRajputs

alone) and Doab counts 14 seats where Yadav candidates keep getting re-elected over

time.TheBJPholdsmostoftheurbanuppercastestrongholdswhiletheSPdominatesin

theruralandsemi-urbancastestrongholds.

Table2.12Occurrenceofseatsheldthreetimesandabovebythesamecaste

infiveUttarPradeshassemblyelections1993-2012 Avadh Bundelkhand Doab East North-East Rohilkhand WestBrahmins 8 0 4 2 4 0 2Rajput 20 2 6 3 11 6 4Bania 3 0 1 0 0 4 1Bhumihar 6 0 0 3 0 0 0Kayasth 1 0 1 1 0 0 0Khatri 0 0 1 0 0 1 0Jats 0 0 3 0 0 2 4Yadav 4 1 14 7 0 2 1Kurmis 6 0 4 2 1 1 0Gujjar 0 0 0 0 0 0 6Lodh 0 1 1 1 0 0 0Muslims 9 0 5 6 2 12 2Source:Author’scalculations.

The BSP initially registered strong performances in Bundelkhand and in Poorvanchal,

some of the poorest areas with comparatively larger SC populations. The party had a

weakpresence intherestof thestate(between3to8percentofvoteshare),until the

1996 election, in which its alliance with the SP made it gain 13 points and gave him

inroadsintomanyareaswhereithadanegligiblepresenceearlier.Itsperformancessince

havebeenfairlyuniformacrosstheterritory,withacomparativelyhighervoteshare in

Bundelkhand.Italsolostuniformlyacrossthestatein2012,exceptintheWest,wherethe

combinationofsupportfromDalitsandMuslimsvotersenabledittoremainaheadofthe

SP.

Conclusion

Itisoftentemptingtolookatpoliticaltransformationsastheextensionintothepolitical

domain of social changes or movements affecting the balance of power between

112

politicized groups. In the case of Uttar Pradesh, much of the political change has been

drivenbypartypolitics,thatistosaybydynamicsmarkingtheirrelationstootherparties

aswellastheirinternaldivisions.Thus,forlong,theriseofthesocialistswascrippledby

theirinternaldivisions–bothstrategicandideological.ThecollapseoftheCongresswas

alsoacollapseofitsorganization,thepartylosingmanyofitscadresandmilitantstorival

formations.

The early scholarship on U.P. politics focused on factionalism within parties – in the

Congress first (Brass 1965), then the socialists (Brass 1984c, a, 1985, Burger 1969,

Duncan1988)andtheHinduright(Baxter1971,Burger1969).Thistraitwouldcontinue

todefinepartypoliticsinastatewhere,uptothisdayandwiththenotableexceptionof

the BSP, parties remain fairly weak organizations, riddled by internal rivalries and by

theirinabilitytogoverninalliancewithoneanother.

The1980swereaperiodof reconfigurationofpolitical formationsasparties started to

identify themselveswith specific segmentsof thevoters, largelydefinedon casteoron

religion,andmobilizedthesesegmentsofvoters inwaysthatsought toalignsocialand

politicalantagonisms.Theresultwastheproductionofanintenselyagonisticandviolent

politicalscene,fragmentedinwaysthatmadeitimpossibleforanypartytogovernonits

own.Thebitternessofinter-partycompetitiontrumpedtheresponsibilitytogovernand

madeit impossibleforpartiestoformsustainablealliances.The1990sexperimentwith

coalition politics in U.P. has been a disaster, which contributed to the state’s economic

morass,atatimewhenIndiastartedtoopenitseconomy.

ThetransformationofelectoralstrategiesoftheBSPintheearly1990sandthenoftheSP

towardstheendof thedecadebroughtanendtothatphaseofstructural instability.By

broadening their social bases through local caste-based alliances and through the

adoption of a generalist development-oriented discourse, regional parties succeeded to

wintwoabsolutemajorities,in2007and2012.

Thisdoesnotmean that thedeep antagonisms thatmarked electoral competitionhave

receded.NordoesitmeanthattheformsofpoliticalmobilizationassociatedwithMandal

andMandirhavedisappearedorfaded.Theyhaveinsteadbeenreconfiguredthroughthe

113

localization of parties’ electoral strategies and concealed behind the generalist tone of

theirpublicexpression.

These successive transformations have also had consequences on the evolution of the

sociological profile of the political class, as we shall see in chapter 4. They also affect

politics as a profession, as we shall see in the next chapter, by creating and imposing

constraints on both parties and individuals contesting elections. In order to capture

partiesandpoliticians’behaviour,itisimportanttoconsidertheconditionsofexerciseof

the political profession, and how the constraints that weigh on political actors impact

boththeirrecruitmentandtheirbehaviour.

114

Tabl

e2.

13Regionalvoteshareperformanceofm

ainpartiesinUttarPradeshassemblyelections,1962-2012

Region

1962

1967

1969

1974

1977

1980

1985

1989

1991

1993

1996

2002

2007

2012

BJS/BJP

Avad

h22

.20%

26

.07%

23

.74%

22

.48%

14.1

1%

11.8

0%

13.8

3%

32.7

2%

33.3

9%

31.7

8%

22.3

1%

14.1

4%

11.9

4%

Bu

ndel

khan

d17

.08%

32

.50%

22

.22%

26

.48%

13.2

0%

16.1

6%

24.3

6%

30.0

9%

29.7

6%

31.5

2%

24.0

4%

13.1

7%

18.9

0%

Do

ab

14.1

0%

18.8

4%

15.6

7%

17.2

2%

11

.32%

9.

28%

11

.47%

30

.30%

33

.85%

33

.26%

20

.18%

18

.78%

15

.41%

East

15

.37%

22

.20%

17

.51%

13

.07%

7.34

%

5.50

%

8.61

%

24.5

9%

29.2

6%

30.0

0%

18.7

0%

13.4

0%

11.1

0%

No

rth-

East

14

.78%

22

.40%

17

.68%

16

.40%

9.85

%

10.8

2%

11.7

7%

31.3

7%

31.8

0%

29.4

8%

20.1

0%

19.7

8%

17.5

0%

Ro

hilk

hand

19

.93%

23

.05%

17

.82%

15

.71%

11.5

4%

12.4

4%

13.5

0%

35.5

6%

33.7

6%

31.6

8%

17.7

4%

20.2

7%

18.8

3%

Ut

tara

khan

d9.

23%

14

.39%

19

.27%

13

.20%

8.98

%

6.14

%

6.05

%

37.7

9%

36.4

9%

41.4

2%

Wes

t10

.43%

13

.13%

9.

95%

11

.02%

6.61

%

6.71

%

5.77

%

32.8

9%

38.3

8%

36.5

3%

17.5

6%

20.0

6%

17.0

8%

Al

l16

.46%

21

.67%

17

.93%

17

.12%

10.7

6%

9.73

%

11.6

1%

31.4

5%

33.3

0%

32.5

2%

20.0

9%

16.9

7%

14.9

8%

INC

Avad

h37

.43%

33

.97%

38

.17%

34

.30%

32

.80%

40

.02%

42

.89%

33

.68%

19

.91%

16

.11%

8.

91%

9.

54%

10

.75%

13

.65%

Bund

elkh

and

37.8

9%

34.6

0%

37.9

9%

32.4

9%

26.5

0%

43.0

8%

43.8

3%

28.7

8%

20.6

4%

19.8

0%

12.8

9%

11.0

0%

13.2

8%

18.5

8%

Do

ab

31.4

3%

30.8

4%

31.8

5%

32.4

3%

30.7

4%

39.0

2%

40.0

5%

28.9

9%

14.9

1%

14.7

0%

7.99

%

8.18

%

7.86

%

9.36

%

Ea

st

37.7

9%

33.2

1%

32.9

7%

28.7

9%

31.4

5%

36.2

7%

33.0

9%

22.6

0%

13.7

4%

9.69

%

4.91

%

6.70

%

5.36

%

8.21

%

No

rth-

East

39

.67%

34

.27%

32

.97%

31

.85%

31

.10%

30

.22%

36

.55%

18

.65%

17

.21%

13

.60%

6.

63%

8.

84%

8.

79%

11

.37%

Rohi

lkha

nd

31.3

9%

27.4

2%

32.1

6%

27.5

5%

29.6

0%

32.3

1%

37.6

0%

23.6

1%

16.8

4%

12.2

0%

9.37

%

10.4

2%

8.51

%

10.4

0%

Ut

tara

khan

d50

.36%

37

.56%

40

.21%

42

.62%

35

.59%

41

.89%

55

.05%

38

.34%

28

.01%

29

.88%

13

.21%

W

est

40.2

4%

31.0

0%

27.9

8%

34.5

1%

38.0

4%

40.6

7%

38.4

2%

28.5

3%

15.3

9%

17.0

3%

8.57

%

9.92

%

7.56

%

14.1

1%

Al

l36

.33%

32

.20%

33

.69%

32

.29%

31

.94%

37

.65%

39

.73%

27

.90%

17

.32%

15

.08%

8.

35%

8.

96%

8.

61%

11

.59%

BSP

Avad

h

8.40

%

8.70

%

6.85

%

20.1

2%

22.5

5%

31.2

6%

25.4

0%

Bu

ndel

khan

d

15.5

7%

20.3

1%

24.6

1%

22.7

7%

28.0

1%

33.2

4%

26.2

0%

Do

ab

6.

43%

8.

99%

9.

72%

21

.76%

23

.06%

32

.38%

27

.19%

East

15.2

4%

16.4

5%

27.6

9%

23.0

7%

25.4

8%

31.6

6%

26.9

6%

No

rth-

East

11.8

4%

10.4

2%

13.3

1%

19.4

2%

22.1

4%

25.9

9%

23.0

3%

Ro

hilk

hand

10.8

1%

6.24

%

3.26

%

18.0

6%

19.9

8%

27.0

6%

22.3

8%

Ut

tara

khan

d

5.24

%

4.37

%

5.69

%

10.1

4%

Wes

t

5.03

%

2.76

%

4.89

%

14.7

8%

22.8

3%

29.5

4%

28.6

6%

To

tal

9.

41%

9.

44%

11

.12%

19

.64%

23

.06%

30

.43%

25

.80%

Source

:Ada

pted

from

ECI

Rep

orts

.

115

Tabl

e2.

13RegionalvoteshareperformanceofmainpartiesinUttarPradeshassemblyelections,1962-2012(continued)

Region

1962

1967

1969

1974

19771980

1985

1989

1991

1993

1996

2002

2007

2012

SP*

Avad

h7.

97%

10

.00%

15

.14%

15

.65%

17.5

9%

16.8

6%

23.9

1%

15.5

1%

27.3

1%

25.2

7%

28.4

5%

28.6

8%

32.8

9%

Bu

ndel

khan

d16

.86%

4.

84%

14

.47%

9.

99%

16.3

1%

13.0

2%

13.1

8%

8.94

%

3.68

%

16.7

5%

23.1

0%

26.1

3%

25.3

0%

Do

ab

12.2

7%

13.9

9%

23.5

9%

22.7

0%

24

.27%

27

.06%

36

.42%

16

.61%

22

.13%

23

.75%

23

.71%

24

.74%

29

.30%

East

13

.58%

12

.98%

17

.67%

29

.06%

20.9

2%

21.1

3%

27.1

2%

13.2

1%

11.3

8%

23.8

0%

26.1

5%

27.7

5%

31.9

9%

No

rth-

East

12

.72%

13

.27%

19

.91%

19

.74%

21.6

5%

15.0

6%

27.5

9%

8.70

%

13.5

9%

17.8

9%

26.2

3%

25.2

1%

27.4

8%

Ro

hilk

hand

11

.62%

4.

24%

26

.52%

20

.71%

20.4

9%

17.4

9%

27.0

5%

12.0

0%

22.6

5%

24.1

5%

24.9

9%

25.1

9%

28.4

1%

Ut

tara

khan

d15

.14%

2.

16%

12

.41%

8.

78%

14.0

9%

15.5

2%

21.3

0%

3.70

%

5.07

%

9.09

%

Wes

t10

.28%

6.

85%

34

.45%

30

.31%

32.0

6%

35.4

8%

47.7

0%

6.18

%

7.77

%

15.3

1%

21.0

0%

16.8

4%

20.0

1%

Al

l11

.52%

9.

97%

21

.29%

21

.22%

21.5

1%

21.3

5%

29.7

1%

12.5

2%

17.9

4%

21.8

0%

25.3

7%

25.4

3%

28.9

9%

1977

JNP

Avad

h46

.05%

Bu

ndel

khan

d44

.46%

Do

ab

51.8

7%

East

46

.32%

No

rth-

East

51

.27%

Ro

hilk

hand

15

.37%

Ut

tara

khan

d48

.19%

W

est

51.5

9%

All

47.7

6%

Source

:Ada

pted

from

ECI

Rep

orts

.

*Inc

lude

sthe

mai

nbr

anch

oft

heso

cial

istf

amily

:SSP

,PSP

,BKD

,BLD

,JNP

(SC)

,LKD

,JD,

JP,S

P.

116

Tabl

e2.

14Regionalseatshareperformanceofm

ainpartiesinUttarPradeshassemblyelections,1962-2012

Region

1962

1967

1969

1974

1977

1980

1985

1989

1991

1993

1996

2002

2007

2012

BJS/BJP

Avad

h19

.05%

35

.64%

13

.86%

21

.78%

4.95

%

1.98

%

16.8

3%

56.4

4%

40.5

9%

32.6

7%

26.7

3%

9.90

%

6.06

%

Bu

ndel

khan

d23

.81%

52

.38%

14

.29%

28

.57%

4.76

%

9.52

%

23.8

1%

52.3

8%

33.3

3%

33.3

3%

28.5

7%

0.00

%

15.7

9%

Do

ab

9.78

%

18.6

8%

9.89

%

10.1

1%

1.

11%

5.

56%

14

.44%

46

.67%

42

.70%

42

.22%

19

.57%

15

.22%

10

.11%

East

14

.75%

28

.33%

15

.00%

15

.00%

1.67

%

8.33

%

21.6

7%

70.0

0%

63.3

3%

63.3

3%

30.0

0%

23.3

3%

15.0

0%

No

rth-

East

4.

88%

17

.50%

5.

00%

13

.51%

2.70

%

8.11

%

10.8

1%

62.1

6%

43.2

4%

35.1

4%

21.6

2%

18.9

2%

17.5

0%

Ro

hilk

hand

14

.00%

24

.00%

12

.00%

21

.15%

5.77

%

7.69

%

17.3

1%

67.3

1%

46.1

5%

40.3

8%

20.0

0%

16.0

0%

15.3

8%

Ut

tara

khan

d0.

00%

4.

55%

9.

09%

4.

55%

81

.82%

59

.09%

77

.27%

W

est

2.50

%

10.0

0%

7.50

%

7.14

%

7.

14%

38

.89%

64

.29%

51

.22%

19

.05%

16

.67%

20

.45%

All

11.4

0%

23.0

6%

11.5

3%

14.3

9%

2.

59%

3.

76%

13

.41%

52

.74%

41

.75%

41

.04%

21

.84%

12

.66%

11

.66%

INC

Avad

h56

.19%

47

.52%

58

.42%

63

.37%

13

.86%

78

.22%

68

.32%

36

.63%

21

.78%

3.

96%

8.

91%

3.

96%

8.

91%

6.

06%

Bund

elkh

and

57.1

4%

38.1

0%

71.4

3%

42.8

6%

9.52

%

80.9

5%

71.4

3%

23.8

1%

19.0

5%

9.52

%

14.2

9%

4.76

%

14.2

9%

21.0

5%

Do

ab

50.0

0%

42.8

6%

46.1

5%

44.9

4%

4.44

%

73.3

3%

58.8

9%

13.3

3%

3.33

%

4.49

%

4.44

%

8.70

%

4.35

%

3.37

%

Ea

st

54.1

0%

53.3

3%

50.0

0%

38.3

3%

20.0

0%

83.3

3%

48.3

3%

15.0

0%

6.67

%

3.33

%

6.67

%

1.67

%

5.

00%

Nort

h-Ea

st

56.1

0%

52.5

0%

52.5

0%

51.3

5%

2.70

%

51.3

5%

62.1

6%

10.8

1%

13.5

1%

8.11

%

10.8

1%

13.5

1%

10.8

1%

12.5

0%

Ro

hilk

hand

56

.00%

42

.00%

38

.00%

34

.62%

17

.31%

63

.46%

69

.23%

17

.31%

7.

69%

5.

77%

13

.46%

6.

00%

2.

00%

3.

85%

Utta

rakh

and

90.0

0%

63.6

4%

68.1

8%

77.2

7%

9.09

%

68.1

8%

81.8

2%

54.5

5%

13.6

4%

27.2

7%

W

est

75.0

0%

40.0

0%

25.0

0%

59.5

2%

7.14

%

71.4

3%

61.9

0%

14.2

9%

2.78

%

9.52

%

4.88

%

7.14

%

2.38

%

11.3

6%

Al

l57

.91%

46

.82%

49

.65%

50

.71%

11

.06%

72

.71%

63

.29%

22

.12%

10

.98%

6.

60%

7.

78%

6.

20%

5.

46%

6.

95%

BSP

Avad

h

8.91

%

18.8

1%

22.7

7%

46.5

3%

9.09

%

Bu

ndel

khan

d

14.2

9%

23.8

1%

42.8

6%

33.3

3%

47.6

2%

66.6

7%

36.8

4%

Do

ab

0.

00%

1.

11%

13

.48%

21

.11%

22

.83%

51

.09%

22

.47%

East

8.33

%

10.0

0%

53.3

3%

13.3

3%

30.0

0%

63.3

3%

13.3

3%

No

rth-

East

13.5

1%

16.2

2%

16.2

2%

32.4

3%

20.0

0%

Ro

hilk

hand

3.70

%

0.

74%

3.

70%

6.

67%

18

.52%

8.

15%

Utta

rakh

and

9.09

%

Wes

t

4.

88%

26

.19%

54

.76%

38

.64%

All

3.

06%

2.

86%

16

.04%

15

.80%

24

.32%

51

.12%

19

.85%

Source

:Ada

pted

from

ECI

Rep

orts

.

117

Tabl

e2.

14Regionalseatshareperformanceofm

ainpartiesinUttarPradeshassemblyelections,1962-2012(continued)

Region

1962

1967

1969

1974

1977

1980

1985

1989

1991

1993

1996

2002

2007

2012

SP*

Avad

h0.

95%

6.

93%

11

.88%

10

.89%

8.91

%

12.8

7%

35.6

4%

8.91

%

41.5

8%

34.6

5%

41.5

8%

30.6

9%

74.7

5%

Bu

ndel

khan

d4.

76%

0.

00%

4.

76%

4.

76%

9.52

%

0.00

%

23.8

1%

0.00

%

4.76

%

14.2

9%

19.0

5%

19.0

5%

26.3

2%

Do

ab

13.0

4%

20.8

8%2

7.47

%

29.2

1%

18

.89%

28

.89%

66

.67%

13

.33%

31

.46%

26

.67%

33

.70%

20

.65%

56

.18%

East

16

.39%

13

.33%

15.

00%

45

.00%

6.67

%

26.6

7%

50.0

0%

10.0

0%

18.3

3%

31.6

7%

43.3

3%

25.0

0%

63.3

3%

No

rth-

East

17

.07%

12

.50%

20.

00%

24

.32%

21.6

2%

8.11

%

54.0

5%

2.70

%

18.9

2%

13.5

1%

32.4

3%

35.1

4%

45.0

0%

Ro

hilk

hand

5.

00%

0.

00%

4.

55%

9.

09%

4.55

%

9.09

%

22.7

3%

4.

55%

13

.64%

42

.00%

26

.00%

55

.77%

Utta

rakh

and

5.00

%

0.00

%

4.55

%

9.09

%

4.

55%

9.

09%

22

.73%

4.55

%

13.6

4%

Wes

t5.

00%

10

.00%

52.

50%

30

.95%

23.8

1%

38.1

0%

73.8

1%

5.56

%

4.76

%

14.6

3%

16.6

7%

4.76

%

22.7

3%

Al

l8.

84%

10

.35%

23.

06%

25

.00%

13.8

8%

19.7

6%

48.9

4%

8.11

%

25.7

1%

25.9

4%

35.4

8%

24.0

7%

55.5

8%

1977

JNP

Avad

h78

.22%

Bu

ndel

khan

d76

.19%

Do

ab

95.5

6%

East

75

.00%

No

rth-

East

89

.19%

Ro

hilk

hand

71

.15%

Ut

tara

khan

d81

.82%

W

est

90.4

8%

All

82.8

2%

Source

:Ada

pted

from

ECI

Rep

orts

.

*Inc

lude

sthe

mai

nbr

anch

oft

heso

cial

istt

radi

tion:

SSP

,PSP

,BKD

,BLD

,JNP

(SC)

,LKD

,JD,

JP,S

P.

118

Chapter3.ElectoralpoliticsinUttarPradesh:Therulesofthegame

Beforedwellingintothequestionofthechangingsociologyofelectedrepresentativesin

UttarPradesh,weneedtoconsiderthecontextwithinwhichtheyoperate,notablytheset

ofconstraintsthattheelectoralcompetitionimposesonpoliticalactorsandwhatimpact

thoseconstraintshaveonwhogetstocontestsuccessfully.Oneofsuchconstraintsisthe

competitivenessofelectionsthatcandidateshavetodealwith.

In this chapter, I examine what I would refer to as the rules of the electoral game.

Electionscanbeseenasaformofcompetitivegame,regulatedbyasetofrules,formaland

informal(Bailey1969).Someoftheserulesapplytopartiesandsometoindividuals.The

formalrulesareinducedbythepoliticalsystemitself,suchasthedisproportionalityofthe

electoral system, or the model code of conduct enacted by the Election Commission of

India,or thequalificationsrequired forcontestingelectionssetby theConstitution.The

informalrulesarenotablyimposedbypoliticalparties,whochoosewhogetstorunand

whocanre-runafterafirstelection.Someoftheserulesalsocomefromtheconfiguration

of thepolitical competition itself.Howmanypartiesare in fray?Howmanycandidates?

Howmanyeffectivecandidates?Votersalsocontributetoshapethepoliticalcompetition

by their choices, by nurturing expectations, by rewarding or punishing those they

previouslyelectedinpower.

The argument I would like to offer here is that all these rules put together create a

universe of constraints to which aspirant politicians must conform in order to stand a

chanceatwinningaseat.Theseconstraintstendinturntofilteroutcandidateswhoare

not in a position to meet parties and voters’ expectations in terms of fundraising or

redistributionofresources, forexample.Theyfinallyalsoimpacthowalargepartofthe

politicalpersonnelbehaves,inthepursuitandexerciseofpowerandinfluence.Theymust

therefore be understood not only in their diverse aspects and origins but also

longitudinally,inordertograspthechangingconditionsofthepoliticalprofession.

The risk of such an argument is to come up with a tautological explanation such as

“nothing succeeds like success”. I amnot claiming that the rules of the gameuniformly

119

apply to all with the same force, or that outlier candidates do not stand a chance at

winningelections.Butthedefinitionoftheserulescomesfromanexaminationofpolitical

data.TheyarebasedonwhatthenumberssayaboutthestateofcompetitivenessofU.P.

stateelections.Haveelectionsbeenmorecompetitivesincethe1990sthanbefore?Does

the stabilization of the party system in the 2000s mean that elections have become

somewhat less competitive? The second set of questions deals with the constraints

imposedbypartiesonindividualMLAs’careers?Whogetstocontest?Whogetstore-run?

HowmanyMLAslastinpoliticsandwhy?

3.1.HaveUttarPradeshelectionsbecomemorecompetitive?

Competitive elections are usually seen as a precondition to be a democracy, along side

their free and fair character (Dahl1961,Przeworski2000, Sartori1987).The literature

links competitiveness to better representation (Powell 2000), enhanced accountability

(Jones 2012) and reduced corruption (Coppedge 1993). In the context of recent

democracies, competitiveness is found to reduce the risk of democratic failure and the

incidenceofcivilconflict(Wright2008).

InthecontextofUttarPradesh,competitionisnotnecessarilyseenasapositivefactor.For

one, political competition in Uttar Pradesh is associated with instability, violence and

voters’intimidation.Thefragmentationofthepartysysteminthe1990swasaccompanied

by a sharp increase of electoral malpractices and violence (Seshan 1995). The

criminalization of politics that took place at the same time – that is the induction into

politicsofindividualsbelongingtoorganizedcrime–broughtunderworldviolencetothe

forefront of electoral battles. Despite the crackdown on electoral malpractices and the

introduction of electronic voting machines in the 2000s153, which have greatly reduced

electoralfraud,politicallifeinUttarPradeshremainsmarkedbyviolenceandintimidation,

andbythecriminalizationofeverydayeconomictransactions154.

153EVMs preclude booth capturing and have reduced vote invalidation. In fact, since theirintroduction,theElectionCommissionnolongerreportsinvalidvotes.154Evenvictories canbe celebratedwithviolence. InMarch2012, in thedays that followed theSamajwadi Party’s massive victory in state elections, violence erupted in various towns, in theformofpost-electoral vendettas andacts of intimidations.On thedayof the results, a groupofjournalists in JhansiwerecorneredandattackedbyamobofSamajwadipartysupporters,who

120

Themultiplicationofcandidatesinrecentyearsandthedispersionofvotesamongthem

hasalsomeantthatthevotesharerequiredtowinaseatisfairlylow(34.4%ofthevalid

votesonaverage),whichaffectsMLAs’representativenessand,intheeyesofsome,their

legitimacy.Sinceinafirst-past-the-postelectoralsystem,winningrequiressimplytohave

morevotes than theothers, the temptation to targetnarrowsegmentsof theelectorate

whileattemptingtodividetherestofthepoliticalspace, throughviolence, intimidation,

or bogus candidates, can be quite high. Thus, competition is often associated with

casteism,violence,instabilityandpoorgovernance.Manycommentatorssuggestthatthe

size of the competition be curbeddown, by imposing electoral threshold to reduce the

numberofpartiesor increasingfurthertheamountof thesecuritydeposit, togetridof

ineffectiveindependentcandidates(Jalan2005).

Competitiveness also means that political careers tend to be short, voters being often

inclinedtorejectthosetheypreviouslyvotedinpower.Thisaffectsthebehaviourofmany

MLAs, who are aware of the fact that they have only a few years to retrieve their

investmentinpolitics.

Beforewelookatcandidates’dataandatthenumberofeffectivepartiesinU.P.elections,

weneedtotakeameasurefirstofthesizeoftheelectorateandthereforeofthesizeofthe

population among which the competition takes place. Vote share of parties need to be

calibrated within the voting population and not simply within the registered voters

population. That is a more accurate measurement of candidates and parties’ actual

popularity.

protestedafterthedefeatoftheirleader.AkhileshYadav,notyetsworn-inChiefMinister,hadtointervene.Thenextday,SamajwadisupportersburntseveralDalithomesinSitapur,becausetheyhad voted for an independent candidate. There were also reports about women and childrenbeatenbyS.P.workersinBallia,becausetheirbastishadvotedforanothercandidate.Onthe8thofMarch,dayofthefestivalHoli,inavillagenearAgra,allegedS.P.workersbrutallymurderedaBSPPradhan with spears. In Ambedkarnagar, property of an ex-BSP Minister, Sanjay Rajbhar, wastorchedbyanangrymob(ThesameMinisterhadopenedfireonthesamedayinavillagethatdidnotvoteforhim).AnotherBSPworkerwaskilledinaclashinSandila,Hardoidistrict,onMarch10(NDTV.com,March10).

122

Contrarytomanyotherstates,UttarPradeshdidnothavealinearincreaseofturnout,as

there are important variations within decades. In the 2000s, for instance, the state

registered an all timehigh turnoutof 59.5%, in2012, against anearly all-time low five

yearsearlier,whenparticipationfellto46.4%.

UttarPradeshisalargediversifiedstateandweneedtoconsiderspatialvariationsaswell.

One way to do that is to project turnout on a map of the state’s constituencies, as

illustratedherebelow.

In 2007, the drop of turnout was quite uniform, sub-region wise, although more

pronouncedintheEast.TurnoutwasslightlyhigherinRohilkhandandWesternUP.There

areslightvariationsbetweenseatswonbydifferentparties.Theaverageturnoutinseats

won by the SP or the BSP was at 46.9 percent, while it was at 43.8 and 43.5 percent

respectivelyforseatwheretheBJPandtheCongresswon156.

156TheseatswonbytheBJPtendtobemoreurbanthanforotherparties,whichcanaccountforthevariationinturnout,participationbeinglowerincities.

123

The2012mapclearlyshowsanEast-Westdivide,withaturnoutabove60%registeredin

mostconstituenciesinRohilkhandandinWesternUP.AcrownofconstituenciesNorthof

Lucknow, inAvadh,also registeredahigh turnout,while theEasternandNorth-eastern

regionsofthestate,aswellaspartsofBundelkhand,areclearlybelowaverage.

Anotherwaytoconsiderspatialvariationsistoclusterconstituenciesintosocio-politically

salientsub-regions.Turnoutfiguresoftheeightsub-regionsrevealhowthesetrajectories

124

have followed a similar overall pattern and have rejoined in the recent years. In other

words, the gaps in turnout between sub-regions has reduced over time. At the

constituency-level,thegapbetweenthelowestandhighestregisteredturnoutusedtobe

of18%inthe1960s,15.5%inthe1970s,10%throughthe1980sand1990sandslightly

lower,at8.3%duringthe2000s.

In order to understand why these gaps have closed, we need to look at how different

categoriesofvotershaveparticipatedinelections.Therearefourcategoriesofvotersthat

usedtoparticipatelessinelectionsandwhohaverecentlyclosedornarroweddownthe

gap:votersinreservedseats,women,youngvoters,andurbanvoters.

In her dissertation on the political and developmental consequences of political

reservations, Francesca Jensenius shows that over time, participation in reserved seats

increases, a fact that she attributes more to increased mobilization of Dalit voters than

general category voters157. This is the case in Uttar Pradesh, where the gap between

reservedseatsandgeneralseatsnarrowsdownovertime,from11.26percentin1962to

4.6percentinthe1990sandthenaclosedownin2002(seetable3.2).

Table3.2DecadalaverageturnoutgapbetweenreservedandgeneralseatsinUttarPradeshstateelections

Decade 1960s 1970s 1980s 1990s 2000s

TurnoutGap 11.26 8.83 9.11 4.66 1.02

Source:AdaptedfromECIReports. *2012included.

Womenarethesecondcategoryofmarginalizedvotersthathaverecentlyclosedthegap.

Theturnoutgapbetweenmenandwomenhashistoricallybeenthelargest,asattestedin

figure3.6andtable3.3.

157Comparingturnoutinseatsbeforeandafterreservation,Jenseniusdetectsthattheturnoutgapexisted before the seats got reserved, indicating pre-existing causes for the observed gaps. InJensenius(2013).

126

What explains this convergence of marginalized categories of voters? The voters’

education programs set up by the Election Commission account for the recent jump in

participation. In 2010, the new Chief Election Commissioner, Dr. S.Y. Quraishi, made

voters’ education a priority, as a way to tackle the problem of participation deficit.

Targeted voters’ mobilization campaigns were held in direction of those categories of

voters that had lagged behind160. That being said, the convergence of marginalized

categoriesofvoters,notably forSCs,hadstartedearlier inU.P.aswellas inmanyother

states.TheECIeffortsacceleratedaprocessthathadalreadystarted.

Finally, why did turnout fall in 2007? As I noted earlier, participation fell by 8 percent

between2002and2007,andthenjumpedby13.3percentin2012.Oneexplanationcould

bethatthesupportersofpartieswhowerenotexpecttoperformstayedathome,while

thesupportersoftheBSP,whowaspoisedtowintheseelections,votedmassively.Thisis

notsustainedbythedata,whichdoesnotshowmuchvariations inaverageturnoutand

averagewinner’svoteshareinseatsbetweenseatswonbytheSPortheBSP(turnoutwas

significantlylowerinseatswonbyCongressorBJPcandidates,butthesecountanumber

ofurbanseats).

A significant feature of the 2007 elections was the strict enforcement by the Election

Commission of rules and restrictions on the display of campaign material. In order to

ensure a fair equality of treatmentbetween small and largeparties and candidates, the

Election Commission gradually introduced a series of restrictions on the displays of

visualsduringtheofficialcampaign.Thefirstnotificationswereissuedinthewakeofthe

2004Generalelectionsandwerestrictlyimplemented2007onwards.

Theserestrictionsdealwithagreatnumberofissuesandtheirrangehasexpandedwith

time.The2007Codeofconductandattachednotificationshaveprovisionsontheuseof

vehicles,onthenumberofhoardings,posters,banners,onthecampaignmaterialthatcan

bedistributed,onwhenandwhereitcanbedistributed,onwhenandwherepartiescan

160 The ECI used surveys, mass media, star campaigners and conducted a massive voters’registrationcampaign,to increasethesizeoftheelectoralroll.This ledtoanoverall increase inturnout in subsequent general and assembly elections in most states. For a comprehensiveaccountoftheECIvotereducationcampaigns,see(Quraishi2014)

127

erect temporary campaign structures, on the authorization and organization of rallies.

Therearestricterrulesinplaceonpollingday,preventingthedisplayofanypartyvisual

material,restrictingtheusageofbusestoavoidcandidatestoferryvoterstothebooths,

etc.

Thecodeofconductalsoregulatesparticipationinpubliceventsofpoliticiansduringthe

campaign. It decides whether a politician can attend functions for the anniversary of

JagjivanRamorAmbedkar,forexample,whichbothfallinApril,whethertheycanattend

mushairasornot,etc.Eachmove, interventionofpublicappearanceofseniorpoliticians

hastobesubmittedandauthorizedbytheECI,includingformembersoftherulingparties,

whohavespecificsetofrulesapplyingtotheminordertoensurethedistinctionbetween

partyworkandgovernmentwork.

Theserestrictionswereappliedstrictlyduringthe2007elections,whichwasdubbedby

parties, candidatesand themediaas the “colorless campaign”.Walking inLucknowand

otherpartsof thestate in thosedays, theonlyhoardingsonewouldseewere thevoter

awareness campaignhoardings of theECI, giving the impression that theECI itselfwas

contestingtheelections.TheECIwasthenlargelycriticizedforhamperingthecapacityof

candidatestoreachouttovoters.CandidatesIspoketoothencomplainedthattheycould

notcampaigneffectively,orcampaignthewayvoterswanttoseemthemcampaign.

“Peoplewanttoseestrongcandidates,howcanIcampaignwithonlytwovehicles?”

“Campaignsarelikefestivals,peoplewanttoenjoy”

SPcandidate

Candidates felt hindered in what they perceive as natural practices of campaigning. In

particular, the habit of filing their nomination heading a procession of supporters, as a

show of strength. They deeply resented to be imposed to file their nomination with a

maximumfourpersons(ECInotificationofFebruary9,2007).

Smallcandidatesinparticular,whodonothavemeanstoreachouttovotersotherthan

through regular campaigning instruments, complained that the restrictions were

particularlyunfairtothem.

128

To conclude this section, there have been massive transformations in turnout in Uttar

Pradesh recent elections, even though the long-term trends variations are not that

significant.Whatisworthretainingisthatmarginalizedcategoriesofvotershaveclosed

theirturnoutgap,andthatthesechangesaretobeattributestoprocessesanddynamics

extraneoustopartypolitics.

InowturntoananalysisofsomeoffeaturesandpatternsofpoliticalcompetitioninUttar

Pradesh, regarding the number of contesting candidates, and parties, the effect of the

electoralsystemonelectoraloutcomes,andananalysisof incumbencypatterns inUttar

Pradeshassemblyelections.These featuresandpatternwillprovide informationon the

kind of rules and constraints the political competiveness of U.P. election impose on

candidatesandparties,andreflectonsomeoftheirconsequencesonwhogetstowinand

lastinpolitics,aswellasonsomeaspectsofMLAs’politicalbehaviour.

3.1.2.Effectivenumberofpartiesandcandidates

ManypeopleandmanypartiescontestassemblyelectionsinUttarPradesh.Between1962

and 2012, there is an average of about 12 candidates per constituency, with a marked

increasedinthe1990sandthe2000s(seetable3.4).

Table3.4Decadalaveragenumberofcandidatesperconstituency

1960s 1970s 1980s 1990s 2000s6.65 8.31 13.13 17.27 15.27

Source:CalculatedfromECIdata.

Thesefigureshidevariationsacrossconstituencies,asshowninfigure3.7.Whileafew

MLAs used to be elected unopposed in the early days, as many as 48 candidates

contestedthe1993electioninFarrukhabad.

132

Whatwealsoseeisthatthereisnotmuchvariationacrossconstituenciesthroughtime,

which means that most seats are actually competitive. This means, in passing, that

elections during the Congress domination era were already competitive. The state has

alwaysbenmulti-partisanand the fact that thebalanceofpowerhasshifted in favor to

State-based parties or that the party system has fragmented does not mean that for

individualcandidates,electionshavebecomesubstantiallymorecompetitive.

Table3.5DecadalaverageENOP

1960s 1970s 1980s 1990s 2000s

AverageENOP 3.69 3.26 3.58 3.57 4.09Source:CalculatedfromECIdata.

Thefactthatnearlyallseatsarecompetitivemeansthattherearefew‘safeseats’forthe

mainparties,barringanumberofexceptions,notablysomeurbanseatssuchasLucknow

orAllahabadfortheBJP.

3.1.3.ThedisproportionalityeffectoftheelectoralsystemThe fact that the majority electoral system in the Indian context produces

disproportionality is well-known (Jaffrelot and Verniers 2011). In a fragmented party

system and electorate under a FPTP regime, it is the distribution of votes among the

candidatesaswellasthegeographicdistributionofvotesthatdeterminetheconversionof

votes into seats, more than aggregate vote shares. Thus, majorities of seats can be

obtained with as low as 29 percent of total vote share. Similarly, parties can be wiped

fromtheassemblydespiteobtaininga lotofvotes. Inthe1984elections,CharanSingh’s

party,theDMKP,gotonlythreeseatsisthestateassemblydespitehavingreceived22%of

thevotes.

The instabilityof the1990salsoresulted fromthewidedistributionofvotesamongthe

mainparties,whichreducedthedisproportionalityeffectoftheelectoralsystem.Inrecent

years, regionalpartieshavebenefited themost from thedisproportionalityeffectof the

electoralsystem.Thisisnotjustbecausetheygetmorevotesthantheircompetitors,but

alsobecausetheirvotebasesaremorespreadoutacrosstheterritorythanthenational

parties,whotendnottohaveaweakerpresenceinruralconstituencies.

136

Tosumup,whatwehaveseensofaristhattherehasbeenarecentriseofturnoutinUttar

Pradesh election, with various groups of marginalized voters reducing their gap in

participation,forreasonsthatarelargelyextraneoustopartypolitics.

Second, the inflation in the number of contesting individuals and parties does not

fundamentallyaffecttheoverallcompetitivenessofassemblyelections,whichhavebeen

competitivesincetheearly1960s.However,at3.4effectivecandidatesperseatovertime,

UttarPradeshelectionsarequitecompetitive.

Third,thelowaveragewinningthreshold–34.4%-encouragespartiestotargetnarrow

segments of the electorate rather than a majority of voters. This creates incentives for

narrowidentitypoliticsandpolarizationstrategies.Thislowthresholdisinducedbythe

distributionofvotesamongpartiesandcandidates,underthedisproportionalityeffectof

theelectoralsystem.

Andfinally,marginsofvictoriesarenarrowing,whichindicatethatthenumberofclosely

fought elections increases over time. Among parties, the BSP struggles more to win its

seatsthanothers.Theseobservationsconveytheimageofelectionsthatarecompetitive.

ThenextstepconsistsinexaminingaseriesofpatternsofindividualcareersofMLAs,as

another indicatorof the levelof competitivenessofUttarPradeshassemblyelectionsas

well as indicators of some of the constraints that the political system place on their

shoulders.Inparticular,Ilookatthedurationorlengthofindividualpoliticalcareers,the

turnoverofMLAsineachassembly.Ialsomeasurethephenomenonofturncoatsandwhat

Icallthesizeofthestablepoliticalclass.

3.2.Politicalconstraints

Inthissection,Ialsostartlookingatthepoliticaltensionsandobstaclesthatimpactand

shapepoliticians’careers,beyondthesystemiceffectsof theelectoralsystemandof the

voters’ choices. The combination of these two types of constraints – induced by the

137

politicalsystemandinducedbypoliticalactors–constitutethesystemicrulesofpolitical

engagement.

3.2.1.TheroleofIncumbencyAny observer of Indian elections knows that “anti-incumbency”, or the propensity of

voterstorejectwhomtheyvotedinpowerinthepreviouselection,isanimportantfactor

in Indian elections. One of the recurrent questions journalists ask before an election is

whethertheanti-incumbencywillplayagainsttheincumbentgovernment.

PartyincumbencyIncumbency is usually conceived in aggregate terms, that is to say the ability of a

governmenttostayinpower,orofapartytowintwoconsecutiveelections.SanjayKumar,

ShreyaSardesaiandPranavGuptashowthatinrecenttimes,governmentincumbencyhas

increased(Kumar,Sardesai,andGupta2013).

Earlier,thepercentageofelectionswonbytheincumbentdeclinedfrom85%inthe1950s

to51%inthe1970s,thento27%twentyyearslater,togoupto54%intheperiod2004-

2012.

In thecaseofUttarPradesh,governmentanti-incumbencyhasratherbeenandremains

the norm, as shown in table 3.9. The last party that was re-voted into power was the

Congress,in1985,whenN.D.TiwarisucceededSripatiMishra166.Infact,since1952,only

theCongressPartyeversucceededtowintwoconsecutiveelectionsinUttarPradesh.

This instability is further increased by the fact that many governments have fallen

between elections without necessarily precipitating early polls. Either President’s rule

wasdeclareduntilanewmajorityemergedorthepartyinpowerappointedanewChief

Minister (See Annexure 1). Despite the stabilization of the political scene in the 2000s,

governmentincumbencyremainshigh.

166WearenotcountingsuccessiveChiefMinisterswithinalegislature,ashappenedwiththeBJP,whohadthreeChiefMinistersbetweenthe1996and2002elections.

138

Table3.9Patternofproandanti-incumbencyinUttarPradeshstateelections167

Numberofelections

Numberofelectionswonbyincumbents(Pro-

Incumbency)

Numberofelectionswonbyincumbents(Anti-Incumbency)

%Pro-incumbent

%Anti-incumbent

1950s 2 2 0 100% 0%

1960s 3 1 2 33.33% 66.67%

1970s 2 1 1 50% 50%

1980s 3 1 1 33.33% 33.33%

1990s 3 0 3 0% 100%

2000s 3 0 3 0% 100%

Total 16 5 10 31.25% 62.50%Source:CalculatedbyauthorbasedonECIdata.

Governmentanti-incumbencybeingthenorm,itismoreusefultolookatthevolatilitythat

takesplace ingivenelections.Oneway todo that is to lookat thenumberof seats that

changehandsbetweenelections.Thiscanbedoneatthepartylevel,orattheindividual

level168.

Atthepartylevel,volatilityhastobemeasuredwithindelimitationperiods,fortheseats

comparedmustbesimilar169.Variations inparties’namesover timemustalsobe taken

intoaccount,toavoidcodingerrors.Forexample,thefollowingsequenceofparties: JNP

àJDàJPàSPwouldbecodedasthesameparty.Dissidentfactionsofaparticularparty,

such as INC(U) for example – have been coded as separate parties. Similarly, residual

factions,suchastheJPafter1993,havealsobeencodedasseparateparties.Fortheyear

1980,thesequenceJNPàBJPhasbeencodedasthesameparty,sincetheBJPwaspartof

theJanatacoalitionin1977.

167FormatinspiredfromKumarandSardesai(2013).168Adnan Farooqui and E. Sridharan show in their contribution on individual incumbency thatparty structures and strategies matter more than party ideology, in the decision or lettingincumbentMPsorMLAsre-run(FarooquiandSridharan2014).169About30%oftheseatschangetheirnameafterre-delimitation.Buteventheseatsthatkeeptheirnameseetheirboundariesmodified.Foradetailedaccountoftheprocedureandoutcomeofthe2008 re-delimitationexercice,onewill refer to theElectionCommissionof India’swebsite:http://eci.nic.in/delim/andto(India2008).

139

Theresultsarequitestriking,astheyrevealthat,intheperiod1980-2007,65.2%ofthe

seatsinUPAssemblyelectionshavechangedhands,onaverage170.Itisalsointeresting

tonotethatthepercentageofseatretentionisslightlyhigherduringthe1990sandthe

2000s.

Table3.10PercentageofseatschanginghandsinUttarPradeshassemblyelections(1980-2007)

1980 1985 1989 1991 1993 1996 2002 2007Changed 72.94% 52.71% 76.47% 60.63% 68.04% 64.54% 62.69% 63.68%Unchanged 27.06% 47.29% 23.53% 39.37% 31.96% 35.46% 37.31% 36.32%Total 100% 100% 100% 100% 100% 100% 100% 100%N 425 425 425 414 413 423 402 402Source:Calculatedbyauthor,onthebasisofECIdata.

Aggregatepartyperformancesoftenhideahighlevelofvolatilityatthecandidate’slevel.

In other words, a party’s performance is the difference between the numbers of

incumbentseatsitretainsandthenumberofnewseatsitwinsoverotherpartiesminus

thenumberofincumbentseatslost.Onewaytoplotthesevariationsistolookatparties’

seat retentioncapacity,or the shareof retainedseatsand thedistributionof the seats

lostacrossrivalparties171.Inthefollowingtables,thediagonalvalueindicatestheratio

ofseatsretainedbyapartybetweentwoelections.Thisparticularpresentationofdatais

useful since it revealswhichparties tend to registermore stableperformances,which

mayindicatethepresenceofparties’strongholds.

It also reveals who were the main two contenders in each constituency, therefore

helpingtounderstandthecompositionofconfrontationsatconstituency-level(inother

words,whoactuallycontestedwhomineveryseat).

170Wilkinsonestimatesthataveragetobeof40%atthenationallevelfrom1980tothelate1990(Wilkinson,2007:115).171Ramashray Roy suggested me this method of plotting party volatility, in a conversation inAllahabad,April2007.

140

Table3.11Seatretentioncapacityofpartiesbetweenthe1996and2002assembly

electionsinUttarPradesh

BJP BSP INC SP Others Seatsin2002

BJP51

(32.28%)38

(24.05%)11

(6.96%)46

(29.11%)12

(7.59%)158

(100%)

BSP8

(12.12%)30

(45.45%) 021

(31.82%)7

(10.61%)66

(100%)

INC8

(25.00%)6

(18.75%)5

(15.63%)7

(21.88%)6

(18.75%)32

(100%)

SP15

(14.02%)21

(19.63%)6

(5.61%)59

(55.14%)6

(5.61%)107

(100%)

Others5

(12.82%)3

(7.69%)3

(7.69%)18

(46.15%)10

(25.64%)39

(100%)

Seatsin1996 87 98 25 151 41Source:calculatedbytheauthor.

Inthiscase,theBJPsucceededinwrestling42seatstotheBSPin2007,despitethestrong

performanceofthelatter.BSPcandidatesdidmuchbetterwhentheywereopposedtoan

SP candidate (as their main opponent), which makes sense since the SP in 2007 faced

stronganti-incumbency.

Table3.12Seatretentioncapacityofpartiesbetweenthe2002and2007assembly

electionsinUttarPradesh BJP BSP INC SP Others Seatsin

2002BJP 23

(26.14%)42

(47.73%)3

(3.41%)18

(20.45%)2

(2.27%)88

(100%)BSP 5

(5.10%)64

(65.31%)5

(5.10%)19

(19.39%)5

(5.10%)98

(100%)INC 2

(8.00%)5

(20.00%)7

(28.00%)8

(32.00%)3

(12.00%)25

(100%)SP 16

(11.19%)71

(49.65%)6

(4.20%)44

(30.77%)6

(4.20%)143

(100%)Others 4

(12.12%)18

(54.55%)1

(3.03%)5

(30.77%)5

(15.15%)33

(100%)IND 1

(6.25%)6

(54.55%)0 3

(15.15%)6

(37.50%)16

(100%)Seatsin2007

51 206 22 97 27

Source:calculatedbytheauthor.

These two tables also reveal that the Congress has great difficulties to retain its seats.

EvenintheirstrongholdsofRaeBareliandAmethi,twoLokSabhaconstituenciesheldby

SoniaGandhiandbyRahulGandhi, respectivelycurrentPresidentandVice-Presidentof

141

theCongressParty,theCongresscannotwinamajorityoftheirseats’assemblysegments,

whichregularlychangehands172.

Table3.13SeatvolatilityinAmethiandRaeBareliLokSabhaconstituencies,2002-2012

PC ACsegments 2002 2007 2012Amethi Amethi BJP INC SP

Gauriganj INC SP SP

Jagdishpur INC INC INC

Salon SP INC SP

Tiloi BJP SP INC

RaeBareli Bachhrawan SP INC SP

RaeBareli INC IND PECP

Sataon BSP INC SP

Sareni SP INC SP

Dalmau BSP INC SP

Source:AdaptedfromECIreports.

TheSPand theBSP tend to retain ahigher shareof seats in recent elections.This is of

course linked to the fact that these twoparties dominate the state’s political scene and

thattheyareinapositionofwinninginagreaternumberofseatsthantheiropponents.

Butitrevealsnonethelessthateventhesetwopartiesinevitablyloseasubstantialpartof

theirincumbentseats,includingwhentheyregisteranoverallstrongperformance.Thisis

afirstindicationthatparties’performanceshidegreatervolatilitythanmeettheeye.Data

ontheindividualtrajectoriesofMLAsandcandidateswillprovideamoreprecisepicture

ofthatphenomenon.

IndividualincumbencyIn order to measure individual incumbency, I have matched individual names in an

originalconstituency-leveldatasetthatcontainsthename,constituency,yearofelection,

positionandperformanceofeverycontestantofallassemblyelectionsfrom1952to2012

(73,450 entries)173. I have coded individual incumbency into fivedistinct variables: the

statusofthecandidate(Incumbent,Ex-MLA,firsttimecontestant), thenumberofterms

172With theexceptionof theseatof Jagdishpur,which is firmlyheldbyRamSevak,anine-timeelectedCongressMLA,fromtheDhobicaste(SC).173The ‘rawmaterial’ comes fromECI reports that havebeenparsed and cleanedbyFrancescaJensenius. I have proceeded to further cleaning and verifications of the data before codingindividualincumbency.

142

served, number of time contested, whether contested in the same party (with specific

coding for turncoat candidates) and whether contested in the same constituency. The

coding was restricted to the candidates who won at least once, covering all party

affiliations.Thisenablesmetoknowforanyindividualwhofallswithinthisgroupinghow

manytimesheorshehascontested,withwhatresults,andwhether thatcandidatehas

switchedpartiesorseatbetweentwoelections.

OfficialECIreportscontainmanyerrors,rangingfromspellingmistakesandvariations174

tomissing entries orduplicate entries. There aremistakes in total and accumulationof

candidates.Further,namematchingisdifficultsincetherearemanycasesofsinglenames

candidates, homonyms175and candidates contesting in different constituencies through

time176.Delimitations,inwhichroughlyathirdoftheconstituenciesgetrenamed,further

addtothecomplexityoftracingindividuals’careertrajectory177.

Entryerrorshavebeencorrectedandspellingvariationsharmonized,withthehelpofa

fuzzy-namematching script and throughmanual checks and entries, based on personal

knowledge, press reports and interviews conducted over several years. In the case of

homonyms,weoftenmadecallsonthebasisofpartyaffiliation,consistencyofelectoral

scores through time.This is not a foolproofmethodbutwebelieve that it provides the

bestpossibleresults,giventhecomplexityofthetask.

174The ECI changes its spelling policy between elections. In 1967, first names in 1967 wherereducedtotheirinitial.175Partiesat timespaybogusor ‘ghost’candidatesbearing thesamenameas theiropponent inorder to cut into their vote base. In the 2014General elections, severalHemaMalini contestedagainstthe‘official’HemaMalini,BJPcandidate.176A few candidates also contest in several constituencies in the same elections. It is not morethana fewcaseandusually involvespartypresidents(MulayamSinghYadav,Mayawati,KalyanSingharecasesinpoint),whoseektosecuretheirre-election.177Thelasttwodelimitationexercisestookplacein2008and1974.Priorto1974,constituencieswere regularly re-delimited. In the dataset, candidates constesting in a different seat post-delimitationhavebeenmarkedas‘delimitation’.

143

Table3.14Sampleofindividualincumbencycoding

Year AC_name Position Cand Party status Mandates ContestedSameparty Turncoat

Sameconstituency

1969 BAHADURPUR 1ROOPNATHSINGHYADAVSSP

Firstelection 1 1

1974 JHUSI 2

ROOPNATHSINGHYADAVBKD Incumbent 1 2 ex-SSP No Delimitation

1980 JHUSI 7ROOPNATHSINGHYADAV JNP(JP)Ex-MLA 1 3 ex-BKD Yes Yes

1989 JHUSI 2ROOPNATHSINGHYADAV INC Ex-MLA 1 4

ex-JNP(JP) Yes Yes

1993 JHUSI 5ROOPNATHSINGHYADAV INC Ex-MLA 1 5 Yes No Yes

Source:Author,IndividualIncumbencyinIndianStateElectionsdataset(1962-Present).

Other methods are available to quantify individual incumbency. Yogesh Uppal, an

economist, has used a regression discontinuity design on close contests to study

incumbency effect in Indian elections (Uppal 2009). His identification methodology

consistsincodingincumbencyforcandidateswithavoteshareabove5%,withinsimilar

constituencies(thatisafterandbeforethe1974and2008re-delimitations).Hisanalysis

isthenlimitedtoconstituencieswherethevictorymarginiswithin70%ofthevotes.

Forourpurpose,weconsiderallelectionsbetween1952and2012.Sincewefollowallthe

main parties individual candidates, it is necessary to include individuals contesting in

differentconstituenciesovertimeandmainpartiescandidateswhomayfallbelowthebar

of5%votesharethatUppaluses,asarelevancythreshold.Thisdataset,asshownthrough

thesampleintable3.14,enablesseveralmeasurements.Firstandforemost,itenablesto

quantifytheturnoverofMLAsintheAssembly,thatistheratiooffirsttimeMLAsversus

thenumberofincumbentsorpreviouslyelectedMLAs.Inotherwords,itisthemeasureof

renewalofthepoliticalclass.

Ahighturnover

InthecaseofUttarPradesh,theturnoverofMLAs,fromitssecondelectionto2012,ison

average58.6%.Inotherwords,itmeansthatonaverage,nearly60%ofthemembersof

theLegislativeAssemblyare first timeMLAs, ineveryassembly.This isveryhigh ifyou

compare Uttar Pradesh with most of the cases covered by the literature (USA, United

144

KingdomandotherEuropeandemocracies),where incumbency is thenormrather than

theexception178.Thisstateofaffairscanbeexplainedbyavarietyofreasons.

Thefirstandmainreasoninthiscase is that lessthanhalfofsittingMLAsre-run inthe

firstplace.SomemaybefieldedbytheirpartiestoLokSabhaelectionsandthereforeleave

theirassemblyseatinthefollowingstateelection,ifsuccessful.SomeMLAsdieinofficeor

choosetotransmittheirseattoakin179.Butthesearenotcommoncircumstances.Inmost

cases,itisthepartythatdeniessomeofitssittingMLAsthechanceofgettingre-elected,

byfieldingdifferentcandidates.Theymaydosotopreventanti-incumbency,thinkingthat

changingheadswillpreventvoters to turnagainst them. In the2012elections, theBSP

fielded only 75 MLAs out of its 206 incumbents. Some had migrated to other parties.

Otherswereremovedunderpressureandprotestfromlocalpartycadrewhocomplained

againsttheirMLAs.

178ThereisnocomprehensivedatayetfornationalelectionsinIndia.Acomparativeoverviewofthe literature on incumbency can be found with (Erikson 1971, Collie 1981, Campbell 1983,Hayama 1992, Cox and Morgenstern 1993, Krashinsky and Milne 1993, Cox and Morgenstern1995, Cox and Katz 1996, Ansolabehere and Gerber 1997, Carey, Niemi, and Powell 2000,Weisberg2002,Ansolabehereetal.2007,Mayhew2008,Smith2013).179 Since the 2012 assembly elections, 10 MLAs have departed(http://uplegisassembly.gov.in/ENGLISH/memberListDead.htm).

145

T

able

3.1

5:In

divi

dual

incu

mbe

ncy

inU

ttarP

rade

shst

ate

elec

tions

(195

1-20

12)

2012

228

(5

6.58

%)

76

(18.

86%

)

46

(11.

41%

)

21

(5.2

1%)

16

(3.9

7%)

8

(1.9

9%)

5

(1.2

4%)

3

(0.7

4%)

40

3

(100

%)

175

(4

3.42

%)

99

(24.

57%

)10

4

(25.

81%

)74

(1

8.36

%)

225

46.2

2%

206

35.9

2%

Source

:Cal

cula

ted

byA

utho

r,In

divi

dual

Incu

mbe

ncy

inIn

dian

Sta

teE

lect

ions

dat

aset

(195

2-Pr

esen

t).

2007

220

(5

4.59

%)

79

(19.

60%

)

42

(10.

42%

)

27

(6.7

0%)

16

(3.9

7%)

10

(2.4

8%)

5

(1.2

4%)

1

(0.2

5%)

3

(0.7

4%)

40

3

(100

%)

183

(4

5.41

%)

104

(2

5.81

%)

145

(3

5.98

%)

46

(11.

41%

)

302

48.0

1%

233

19.7

4%

2002

189

(4

6.90

%)

95

(23.

57%

)

58

(14.

39%

)

26

(6.4

5%)

18

(4.4

7%)

10

(2.4

8%)

4

(0.9

9%)

1

(0.2

5%)

2

(0

.50%

)

403

(1

00%

)

214

(5

3.10

%)

119

(2

9.53

%)

140

(3

4.74

%)

74

(18.

36%

)

277

50.5

4%

246

30.0

8%

1996

191

(4

5.05

%)

109

(2

5.71

%)

53

(12.

50%

)

42

(9.9

1%)

18

(4.2

5%)

7

(1.6

5%)

1

(0.2

4%)

1

(0.2

4%)

2

(0.4

7%)

42

4

(100

%)

233

(5

4.95

%)

124

(2

9.25

%)

150

(3

5.38

%)

85

(20.

05%

)

297

50.5

1%

216

39.3

5%

1993

198

(4

6.70

%)

119

(2

8.07

%)

58

(13.

68%

)

26

(6.1

3%)

12

(2.8

3%)

6

(1.4

2%)

3

(0.7

1%)

2

(0.4

7%)

42

4

(100

%)

226

(5

3.30

%)

107

(2

5.24

%)

164

(3

8.68

%)

64

(15.

09%

)

324

50.6

2%

267

23.9

7%

1991

244

(5

8.23

%)

82

(28.

07%

)

42

(13.

68%

)

25

(6.1

3%)

19

(4.5

3%)

2

(0.4

8%)

3

(0.7

2%)

2

(0.4

8%)

41

9

(100

%)

175

(4

1.77

%)

93

(22.

20%

)14

0

(33.

41%

)37

(8

.83%

)

312

44.8

7%

177

20.9

0%

1989

207

(4

8.71

%)

113

(2

6.59

%)

60

(14.

12%

)

31

(7.2

9%)

6

(1.4

1%)

6

(1.4

1%)

2

(0.4

7%)

42

5

(100

%)

218

(5

1.29

%)

105

(2

4.71

%)

129

(3

0.35

%)

90

(21.

18%

)

271

47.6

0%

200

45.0

0%

1985

238

(5

6.00

%)

101

(2

3.76

%)

52

(12.

24%

)

22

(5.1

8%)

8

(1.8

8%)

4

(0.9

4%)

42

5

(100

%)

187

(4

4.00

%)

86

(20.

24%

)13

2

(31.

06%

)55

(1

2.94

%)

214

61.6

8%

172

31.9

8%

1980

269

(6

3.29

%)

81

(19.

06%

)

47

(11.

06%

)

19

(4.4

7%)

6

(1.4

1%)

3

(0.7

1%)

42

5

(100

%)

156

(3

6.71

%)

75

(17.

65%

)62

(1

4.59

%)

94

(22.

12%

)

224

27.6

8%

186

50.5

4%

1977

244

(5

7.41

%)

117

(2

7.53

%)

44

(10.

35%

)

10

(2.3

5%)

7

(1.6

5%)

3

(0.7

1%)

42

5

(100

%)

181

(4

2.59

%)

64

(15.

06%

)14

2

(33.

41%

)39

(9

.18%

)

271

52.4

0%

101

38.6

1%

1974

263

(6

2.03

%)

104

(2

4.53

%)

32

(7.5

5%)

17

(4.0

1%)

8

(1.8

9%)

42

4

(100

%)

161

(3

7.97

%)

57

(13.

44%

)10

8

(25.

47%

)56

(1

3.21

%)

230

46.9

6%

134

41.7

9%

1969

277

(6

5.18

%)

87

(20.

47%

)

38

(8.9

4%)

18

(4.2

4%)

5

(1.1

8%)

42

5

(100

%)

148

(3

4.82

%)

61

(14.

35%

)11

0

(25.

88%

)40

(9

.41%

)

282

39.0

1%

93

43.0

1%

1967

311

(7

3.18

%)

66

(15.

53%

)

37

(8.7

1%)

11

(2.5

9%)

42

5

(100

%)

114

(2

6.82

%)

48

(11.

29%

)92

(2

1.65

%)

27

(6.3

5%)

132

69.7

0%

33

81.8

2%

1962

318

(7

4.47

%)

77

(18.

03%

)

32

(7.4

9%)

42

7

(100

%)

109

(2

5.53

%)

32

(7.4

9%)

97

(22.

72%

)17

(3

.98%

)

200

48.5

0%

24

70.8

3%

1957

312

(7

2.56

%)

118

(2

7.44

%)

43

0

(100

%)

118

(2

7.44

%)

11

8

11

8

100.

00%

1951

430

(1

00%

)

430

(100

%)

Tabl

e3.

15

Individu

al

incumbe

ncyin

Uttar

Pradesh

state

elections

(1951-

2012)

1 2 3 4 5 6 7 8 9 10

Total

Total>1

Total>2

Total

Incumbents

TotalEx-

MLAs

Re-running

incumbents

Strikerate

Incumbents

Re-running

Ex-MLAs

Strikerate

Ex-MLAs

Source

:Ca

lcul

ated

by

Aut

hor,

Indi

vidu

al

Incu

mbe

ncy

in

Indi

an

Stat

eEl

ectio

ns

data

set

(195

2-Pr

esen

t).

146

Whatisstrikingisthatindividualincumbencywasactuallyhigherinthe1950sand1960s,

when the Congress dominated. The notion of Congress domination often conceals the

internalinstabilityandvolatilitywithinthatparty.TheCongresswasridbyfactionalism.

WiththeexceptionofGovindBallabhPant,ChiefMinisterfrom1946to1955180,noChief

Minister could impose his faction upon the others and constantly had to deal with

insubordinationandinternalcompetitionwithintheorganization,theassembly,andeven

theircabinets(Srivastava1976,330).

The induction of a large number of upper caste candidates extraneous to the party

organization in the2007electionshad createda rift betweenmanyMLAsand the local

cadre of the party, who complained that they were completely ignored after getting

elected.Thesituationwassosevere thatmany localbranch leadersof theBSPdeserted

thepartyandsoughtfortunewithotherparties,notablytheSPandtheCongressParty.

Table3.16SamajwadiPartyre-runningincumbentJD/JP/SP 1989 1991 1993 1996 2002 2007 2012Seatsinpreviouselection 84 208 126 109 110 143 97TotalNofcandidates 356 773 256 281 390 393 402Re-runningincumbents 19 63 29*** 61** 59 102 60Re-runningotherincumbents 8 6 6 25 11 40 8Ratioofre-runningincumbents 22.62% 30.29% 23.02% 55.96% 53.64% 71.33% 61.86%Source:CalculatedbyAuthor,IndividualIncumbencyinIndianStateElectionsdataset(1952-2012)*AddingJanataDalandJanataParty**IncludingMulayamSinghYadavcontestingintwoseats.***IncludingMulayamSinghYadavcontestinginthreeseats.

ThereislittlesignthatfieldingmanysittingMLAsprovidesanincumbencyadvantage.In

2007, the SP re-fielded 71.3% of its sitting MLAs, in addition to 40 other incumbents

candidateswhohadmigratedtowardstheSP(seetable3.16).Lessthanathirdofthem

won(56,including37incumbentSPcandidates).

180G.B. Pant had earlier beenPrimeMinister of theUnited Provinces, between1937 and 1939.Withtheabolitionofthezamindarisystem,thepassingoftheHinduCodeBill,theimpositionofmonogamyforHindus,heprobablyremainsU.P.’mostconsequentialChiefMinister

147

Tabl

e3.

17Mainpartiesre-runningincumbents

2012

22

355

11

4

50.00%

2012

51

398

27

1

52.94%

*89

JNP

cand

idat

esin

197

7ha

dpr

evio

usly

cont

este

don

aJa

nSa

ngh

ticke

t.

**In

clud

ing

Kaly

anS

ingh

cont

estin

gin

two

seat

s.

*I

nclu

ding

May

awat

icon

test

ing

inth

ree

seat

s.

Source

:Cal

cula

ted

byA

utho

r,In

divi

dual

Incu

mbe

ncy

inIn

dian

Sta

teE

lect

ions

dat

aset

(195

2-20

12)

2007

25

393

12

8

48.00%

2007

88

350

46

6

52.27%

2002

33

402

12

11

36.36%

2002

174

320

89

12

51.15%

1996

28

126

15

2

53.57%

1996

177

414

119*

*

21

67.23%

1993

46

421

33

3

71.74%

1993

221

422

165*

*

4

74.66%

1991

94

413

64

9

68.09%

1991

57

415

44

6

77.19%

1989

229

410

111

35

48.47%

1989

16

275 6 9

37.50%

1985

309

425

104

20

33.66%

1985

11

347 5 5

45.45%

2012

206

405

75

6

36.41%

1980

47

424

22

7

46.81%

1980

72*

400

49

3

68.06%

2007

98

403

54

6

36.41%

1977

215

395

115

14

53.49%

1977

61

89*

37

na

60.66%

2002

67

401

30* 3

44.78%

1974

211

415

92

33

43.60%

1974

49

401

24

33

48.98%

1996

67

296

22

7

36.41%

1969

199

424

119

14

59.80%

1969

98

397

58

2

59.18%

1993

12

165 9 1

36.41%

1967

249

425

80

7

32.13%

1967

49

401

10

1

20.41%

1991

13

386 6 1

36.41%

1962

286

429

129

na

45.10%

1962

17

286

16

0

94.12%

1989

na

372

na

1 na

Congress

Seat

sin

prev

ious

el

ectio

n

Tota

lNo

fcan

dida

tes

Re-r

unni

ng

incu

mbe

nts

Re-r

unni

ngo

ther

in

cum

bent

sRa

tioo

fre-

runn

ing

incu

mbe

nt

BJP

Seat

sin

prev

ious

el

ectio

n

Tota

lNo

fcan

dida

tes

Re-r

unni

ng

incu

mbe

nts

Re-r

unni

ngo

ther

in

cum

bent

sRa

tioo

fre-

runn

ing

incu

mbe

nt

BSP

Seat

sin

prev

ious

el

ectio

n

Tota

lNo

fcan

dida

tes

Re-r

unni

ng

incu

mbe

nts

Re-r

unni

ngo

ther

in

cum

bent

sRa

tioo

fre-

runn

ing

incu

mbe

nt

148

Shiftingalliances

The second reason for excluding incumbent MLAs is that parties shift their electoral

strategiesandlocalalliances,whichmayleadthemtoreconsidertheirticketdistribution

on thebasisof casteorothervariable.Again in2007, theBSPopted forastrategy that

targeteduppercastevoters,Brahminsinparticular.Thatshiftrequireddistributingmore

tickets touppercastecandidates,whichmeant thatsomenon-uppercastesittingMLAs

hadtobesacrificed.

Similarly,partiesmayengageinpre-electoralallianceswithotherparties,whichimplies

seat-sharingagreements.Aswesawearlier,pre-electoralallianceshavebeenratherrare

inUP,barringintheJanataphase.Theyusuallytakeplaceatthemargins,betweenmajor

partyandaminorplayer,suchastheRLDwiththeSPin2002,ortheApnaDalandthe

BJPin2014.Themainpartywouldconcedeafewseatstotheirpartners,usuallyintheir

limitedstrongholds,whichwouldnotdisruptmuchtheirownticketdistribution.

Partyfunding

The third reason for a party to field a different candidate comes from the need to

constantly search stronger candidates and candidates who can contribute more to the

party,includingfinancially.ElectionsinIndiahavebeendescribedasaformofauctions,

in which basic services and patronage are commoditized and traded against votes

(Chandra2004a).Similarly,partiesoftenengage insimilarbargainsand trade-offswith

potentialcandidatesandsittingMLAs.

TheBSPagainisanextremeexample,as it isknowntoauctiontickets. Intherun-upto

the2007elections,IinterviewedaBSPcandidateintheVaranasiarea,whohadjustgot

bailfromjailaftersecuringaBSPticket,whichhehadpurchasedfor1.25Croresofrupees.

When askedwhat hewoulddo in case of victory (hewon), the responsewas to “geta

returnonmyinvestment”181.Thisprincipleofauctionisalsovalidfor localelectionsand

somecandidates spendhugeamountsofmoney inorder to get a ticket. In some cases,

local elections tickets cost even more than MLA tickets. In the recent zilla panchayat

181InterviewinVaranasi,April2007.

149

elections,thatwereheldinJanuary2016,someBSPticketswerereportedtohavebeen

soldfor12Crores182.

Asageneralrule,candidatesareexpectedtofundtheirowncampaignandtocontribute

to party funding between elections, through legal and less legal means. This puts

tremendous pressure to candidates and elected representatives, who must constantly

seeknewsourcesof funding.Thefundraisingcapacityofcandidates isoftenayardstick

for their selection183, and sitting MLAs can be outspent by rival candidates in their

constituencies.

Factionalismandintra-partycompetition

AfourthreasonforsittingMLAstobeevictedisthatpartiesarethemselvesconstitutedas

ahighlycompetitivespace.Intra-partyfactionalismandrivalrycanleadtotherejectionof

sittingMLAs.Partiesmayalsowanttochangetheircandidatesoneelectionafteranother,

asameansofremainingcompetitivebutalsoasameantomaintaintheauthorityofthe

partyhighcommandovercandidates,whocouldusetheirpoliticalrootstogrowwithin

the party and eventually challenge the leadership, or constitute factions which could

threaten the party’s cohesion. Parties in India tend to be hyper-centralized. One

expressionofthisisthatthepartyleadershipoftenattemptstopreventtheemergenceof

alternatecentersofpowerwithinitsranksbyorganizingaturnoverofcandidates.Thisis

alsoawaytokeeptheotherMLAs‘incheck’andmaintainpartydiscipline,sincegetting

the ticket to re-run depends from the will of the leadership, who seek to maintain a

relationofdependencyvis-à-visitsMLAs.

There are however important variations between parties in that regard. The logic of

rotationprevailsintheBSP,whoisthemostinternallycompetitivepartyinUttarPradesh.

Ifwe compareparties’ ratioof re-running incumbents (see table3.17),we see that the

CongressandtheBJPtendtore-fieldalargershareoftheirsittingMLAsfromoneelection

to another. Since 1989, the BSP discards on average 62% of its sitting MLAs in every

1821.6MillionEuros,asper23April2016rate.ThatnumberwasreportedtomebymystudentRajkamalSingh,fromhisownfieldwork.183Exceptionsaremadeforimportantfigureswithintheparties,relativesoftheleaderorregionalbosses.

150

election,against41%and46%fortheBJPandtheCongress.TheSPissituatedinbetween,

witha54.4%ofrejectionratio.Thesenumbersalsovaryintime.InthecaseoftheSP,the

rejectionratiointhe1990swasof67%inthe2000s,from37.7%inthe1990s.

ThecaseofCongressandBJPisinterestingwhenweconsidertheirre-runningcandidates

patterns (including those who don’t win). Both parties tend to keep their candidates

longer, or to field the same candidates in the same constituencies over long period of

timesmorethantheBSPandSP,evenwhentheyarenotperforming.

The reason, as we shall see in chapter 3, is that their sociological recruitment pool is

comparativelymore limitedthanforthetwostate-basedparties,bothCongressandthe

BJPremainingskewedinfavoroftheuppercasteswhileboththeSPandtheBSPrecruit

fromabroadersociologicalspectrumofcandidates.

The Congress as an organization often retains old associational bonds. Past allegiances

and familyhistoriesmattersignificantly in theparty’schoiceofcandidates,oftenat the

costoftheirwinnabilityandthereforeperformance.Havingspenttimeinthe‘warroom’

of thepartyduringthe2007,2009and2012elections, Iwasstruckbythefact thatthe

same faces would evolve around the inner circle of the party’s state leadership,

comprisingoldnotabilities,politicalfiguresfromthepast,whowerestickingaroundand

continuedtopullstrings intheparty.Theroleofelders(or losers,assomebitterparty

workers refer to them) in the screening of candidates remains important within the

Congress, despite attempts at rationalizing and professionalizing the recruitment of

candidates. Inaddition, theparty’scampaignstrategistsareoften ‘imported’ fromother

states and lack information to foray into new recruitment pools of candidates. It is

literally as if thepartypreferred tomaintainold associationswith losing candidates to

inductingfreshbloodintheparty.Asitis,thepartyisreluctantto‘discardthedeadwood’,

asapartyworkerconfidedtomeduringthe2009campaign184.

TheBJPisthepartywhereMLAshavethelongestlinearcareersandwherecandidatesre-

runthemost.PartoftheexplanationliesinthefactthattheBJPcountsmorefromstable

strongholds than other parties, notably in urban seats, and is therefore under less

competitive strains than others. Another reason is that strong ideological ties tend to

184InterviewinLucknow,Congressheadquarters,May2009.

151

boundcandidatestotheparty,whichrecruitsmanyofthemandofitscadreswithinthe

RSS.BJPMLAsshowmoreallegiancetotheirpartythanothers,andtendcomparatively

migratelesstootherpartiesthantheirCongressandstate-basedpartiescounterparts.

Voters’choice

It isonlyafterhaving facedandovercomeall thesehurdles thansittingMLAs face their

last judge,thevoters. Inhisstudyonindividual incumbencyinIndianelections,Yogesh

Uppalfindsevidenceofasignificantincumbencydisadvantageinstatelegislativeelections.

He concludes, “Incumbents are not only less likely to win than their challengers, but the

adverseeffectofincumbencyhasincreasedafter1991”185.InUttarPradesh,Ifindarather

stablepatternofincumbent’sre-election.

Imeasuretheperformanceofincumbentcandidatesbycalculatingtheirstrikerate,orthe

successratiooftheincumbentsandex-MLAswhore-ran186.Weseethatthestrikeratefor

incumbent candidates is fairly stable, around 49% and that the ratio for ex-MLAs

decreasessharplyinthe1990sand2000s.Thisillustratesthefactthatvotersinthe1990s

andthe2000sgrewreluctanttogiveanotherchancestoMLAswhohadbeenoustedoutof

powerbefore.Asa consequence,partiesbecamealso reluctant togive tickets to former

losingcandidates.

Table3.18Strikerateofincumbentcandidatesandex-MLAsinUPassemblyelections 1960s 1970s 1980s 1990s 2000sStrikerateIncumbents

52.40% 49.68% 45.65% 48.66% 48.26%

Strikerateex-MLAs 65.22% 40.20% 42.50% 28.08% 28.58%Source:CalculatedbyAuthor,IndividualIncumbencyinIndianStateElectionsdataset(1952-2012).

This data confirms that parties and not voters constitute the main obstacle to the

longevityofpoliticians’careers.

185Uppal,ibid.,p.24.186Thestrikerateisatermborrowedtocricketthat indicatestheratioofseatswonagainstthenumberofseatscontested.Thismeasureenablestocompareparties’performancewhentheydonotcontestsimilarnumberofseats.Onthistopic,see(SircarandVerniers2016).

152

To sum up, voters’ anti-incumbency, in addition to the four motives of party rejection

mentioned earlier means that the probability for a incumbent to last after their first

election is quite low (roughly one chance out of four). This, as we shall later, has far-

reachingconsequencesonwhocontestsandonMLAs’politicalbehaviour.

3.2.2.TheroleofturncoatsAnother feature of electoral politics in Uttar Pradesh is the weakness of affiliation ties

betweenparties and candidates.Thephenomenonof turncoats –or candidates shifting

partyallegiancebeforeanelection–isanotherpopulartopicofconversationforpolitical

observersandjournalists.AboutathirdofUttarPradesh’sMLAshavechangedtheirparty

affiliationatleastonce(seetable3.19).Theymayhavedonesoforavarietyofreasons.

Table3.19PercentageofMLAswhochangedpartyaffiliationduringtheircareer,1952-2012

Changedonce

Changedtwice

ChangedmorethanTwice

Didnotchange

TotalMLAs

23.08% 7.00% 2.89% 67.03% N=4116Source:Calculatedfromtheauthor’sUttarPradeshincumbencydataset.

The first reasonmaybe thatcandidatesseekanewaffiliationafterbeingrejected from

theirparty,orleavebecausetheyanticipatethattheirpartywilldenythematicket.Their

status and strength will determine whether they get a strong competitive ticket or

whether they will have to satisfy themselves with some morganatic marriage with a

smallerparty.MLAswhoshiftfromastrongtickettoaweakticket–orasindependents–

losemostofthetime.

MLAsalsoleavetheirpartywhentheyfeelthattheydonotgettheplaceandstatusthey

deserve.Failurefromthepartytoaccommodateticketstokin,relativesandassociatesis

alsoafrequentmotiveforMLAstoseeknewaffiliations.Inthe2009election,thesonofa

prominent figure of the Congress party, former Minister and Speaker of the Assembly,

memberoftheparty’sStateCommittee,contestedonaBSPticketinaconstituencynear

Lucknow. In Western Uttar Pradesh, a tussle between the SP and a prominent Muslim

politicalfamilyledvariousofitsmemberscontestundervariousparties’affiliations.

153

Tabl

e3.

20–

Tur

ncoa

tsfi

elde

dby

mai

npa

rtie

sin

Utta

rPra

desh

ass

embl

yel

ectio

nsa

ndth

eirp

erfo

rman

ce(1

980-

2012

)

2012

11

399

2.76

%

1

9.09

%

27

356

7.58

%

1

3.70

%

11

403

2.72

%

1

9.09

%

30

402

7.46

%

16

53.3

3%

79

1562

5.06

%

19

24.0

5%

*In

clud

esth

eJN

P(SC

),LK

D,JP

and

JDfo

rthe

yea

rs1

980-

1991

.Source

:Cal

cula

ted

from

the

auth

or’s

Utta

rPra

desh

incu

mbe

ncy

data

set.

2007

26

350

7.43

%

3

11.5

4%

41

393

10.4

3%

6

14.6

3%

38

403

9.43

%

27

71.0

5%

55

393

13.9

9%

20

36.3

6%

160

1539

10.4

0%

56

35.0

0%

2002

21

320

6.56

%

6

28.5

7%

25

401

6.23

%

3

12.0

0%

21

401

5.24

%

6

28.5

7%

30

390

7.69

%

15

50.0

0%

97

1512

6.42

%

30

30.9

3%

1996

30

414

7.25

%

17

56.6

7%

11

126

8.73

%

5

45.4

5%

37

296

12.5

0%

19

51.3

5%

35

281

12.4

6%

22

62.8

6%

113

1117

10.1

2%

63

55.7

5%

1993

7 422

1.66

%

4

57.1

4%

13

421

3.09

%

1

7.69

%

2 165

1.21

%

1

50.0

0%

10

256

3.91

%

6

60.0

0%

32

1264

2.53

%

12

37.5

0%

1991

11

415

2.65

%

8

72.7

3%

9 413

2.18

%

0

0.00

%

4 386

1.04

%

0

0.00

%

7 399

1.75

%

1

35.0

0%

44

1613

2.73

%

15

34.0

9%

1989

11

275

4.00

%

4

36.3

6%

44

410

10.7

3%

19

43.1

8%

8 372

2.15

%

2

25.0

0%

72

356

20.2

2%

55

76.3

9%

135

1413

9.55

%

80

59.2

6%

1985

6 347

1.73

%

1

16.6

7%

30

425

7.06

%

26

86.6

7%

42

385

10.9

1%

23

-54.

76%

78

1157

6.74

%

50

-64.

10%

1980

10

400

2.50

%

2

20.0

0%

19

424

4.48

%

15

78.9

5%

10

399

2.51

%

4

40.0

0%

39

1223

3.19

%

21

53.8

5%

Turn

coat

s

Cand

idat

es

Ratio

Succ

esfu

ll

Turn

coat

s

Cand

idat

es

Ratio

Succ

esfu

ll

Turn

coat

s

Cand

idat

es

Ratio

Succ

esfu

ll

Turn

coat

s

Cand

idat

es

Ratio

Succ

esfu

ll

Turn

coat

s

Cand

idat

es

Ratio

Succ

esfu

ll

BJS/BJP

INC

BSP

SP*

Total

154

Andinanotherconfiguration,stillinWesternUttarPradesh,afigureoftheRashtriyaLok

Dal,six-timeelectedandsonofaformerdeputyChiefMinister,migratedhisentirefamily

totheSamajwadiPartybeforethe2012electionsafterhispartyPresidentdeniedaticket

tohisbrother.TheSamajwadiPartynominatedhimintheLegislativeCouncil.

Candidatescanalsoswitchpartiesontheirownbecausetheyanticipatethattheirparty

might not do well in the next election. In 1989, 57 Congress candidates, including 43

sittingMLAsand10ex-MLAs,leftfortheJanataDal.

In some cases, candidates are locally stronger thanparties andwin regardless of party

affiliation.Itisthecaseforsomecriminalpoliticianswhokeepchangingparties,attimes

contestingunderpartiesoftheirownorasindependents,andkeepwinningnonetheless.

Forinstance,AkbarHusain,asix-timeMLAfromKundarki,contestedtentimesundersix

differentpartyaffiliations187.

There is also evidence that parties attempt to poachMLAs fromvarious parties, either

ahead of an election, or after, as was current in the 1990s, when parties engineered

defections to break down the majority in place188 . The 1980s were also years of

recompositionforpartiesoftheJanataParivar.Chopraestimatesthat70%oftheJanata

DalandJanataPartyMLAshadchangedtheirpartyaffiliationateastoncethroughtheir

careers189.

Everyparty isaffectedby turncoats,although thephenomenon isnotaswidespreador

important as it is usuallymade tobe.Between1989 and2012, 6.3%of the candidates

fieldedbythefourmainpartiesand9.4%oftheirwinnershadcontestedunderadifferent

party affiliation in the previous election. Without surprise, turncoats tend to do better

when they join a strong party, or a party strong in the concerned election. Thus, 71

187AkbarHusainstartedhiscareerwiththeJanataPartyin1977.HethenlefttheLokDalfortheBSPin1989,joinedtheJanataDalin1991,whichheleftfortheBJP.Subsequently,hefoughtonaJanataDalticketagainin1996,thenonaBSPticket2002.HefinallyshiftedtoINLDin2007.188The1985Anti-DefectionLawwasmeanttopreventindividualMLAstodefecttootherpartiesafter theirelection,bystatingthatonlygroupsofMLAs,constitutingat leasta thirdofaparty’sstrength in the assembly could defect. As a results, parties organized mass defections, usingmoneyandattimesviolence.189Chopra,op.Cit.,p.100.

155

percent of the turncoats who had joined the BSP in 2007 were elected, while only 6

turncoatsoutof41wononaCongressticket.

Turncoatsplayedan importantpart in elections in the1990s,whenparties engineered

defectionsinordertotoppledowngovernments.Today,theymayhelppartiesprevailina

fewlocalcontestsbuttheirimpactonelectoraloutcomesotherwiseismarginal.

Turncoatsretainattentionbecausetheyserveasindicatorsofhowpoliticianspredictthe

outcomeofthenextelection,andbecausetheyarealsoasymptomoftheroleofmoney

andhorse-tradinginelectoralpolitics.Inanycase,theydorevealtheweaknessofparty-

candidatesaffiliationbonds.

Thetablesandfigures inthissectionhaverevealedthat thedefiningcharacterofMLAs’

careeristheirshortduration.MLAsfacemultipleobstaclesbeforeandaftertheirelection,

obstaclesthatcauseroughly60%ofthemtolosetheirpositionafterhavingservedonly

one term. Party politics accounts for the major part of the assembly’s turnover, while

votersrejectonaveragearoundhalfofthere-runningincumbents.ManyoftheMLAsthat

Iinterviewedwereacutelyawarehowdifficultitistolastinpoliticsandmanyconfessed

thattheyknewthattheyhadonly‘oneshot’tomakethemostoftheirpoliticalcareer.As

we shall see in the next section, these systemic constraints on political careers act as

structuralincentivesforpredatorybehaviour.

Another consequence is that political power tends to be concentrated within relatively

fewhands.Partiesarecentralizedorganizationsbutevencentralizedorganizationsneed

cadresand local figures tosustain their localpresenceandconnect theparty tovarious

sourcesoffunding.Onewaytofigurehowconcentratedpowerisistoestimatethesizeof

whatcouldbecalledthestablepoliticalclassofUttarPradesh.

3.2.3.Competitionfortheticket:ThelongroadtotheelectionAspiring politicians take sometimes years to prepare their candidacy, and must face a

competitionattimeslongerandharder,andinthelongruncostlierthattheelectionitself.

Theyofcourseneedtogathersufficientresourcestofundtheirownpoliticalcampaigns

156

andcareers.Theyareoftenexpectedtocontributetopartycoffers,notablythroughthe

purchase of their ticket. They must cultivate a caste network in order to develop a

following among the group they belong to. In order to develop their base of followers,

theymustbuildlocalpatronagenetworks.Sincethiskindofenterpriseoftencostsbeyond

anindividual’smeans,aspiringpoliticiansmustalsodevelopnetworkswithlocalspecial

interests, individuals or groups who will ‘invest’ in the political future of the said

individual andhelphimdeploy furtherhis orher fundraising anddistribution capacity

(Alm2010).

These localelitenetworkscanbecastebasedorcutacrosscaste,according to the local

demographyandthelocalconfigurationofpoliticalleadership.Throughthesepainstaking

efforts, aspiring politicians will slowly build for themselves a status of a local leader,

susceptibletoattracttheattentionofaparty.Someofthemattempttotakeshortcuts,by

consortingwithlocalcriminalelements,orbyresortingthemselvestoillegalactivities,in

ordertoacceleratetheprocessandbuildupanimageof‘effectiveleadership’.

Itisimportanttokeepinmindthelongroadtowardsthenomination,asitgreatlyimpacts

thecostofentryintopolitics.Aspiringcandidatesknowthecostoflosingafirstelection

andareconsciousoftheefforts–andresources–requiredtosucceed.

Datasuggeststhat76.3percentoftheMLAswontheirfirstseatontheirfirstattempt,16

percentintheirsecondattempt,and7.6percentbeyondtwoattempts,from1962to2012

(seetable3.21).Thatfigurehasrecentlyincreasedandstoodat87percentin2012.This

indicatesthatitisverydifficulttowinafteraninitialloss.Partiestendnottorecruittheir

candidatesamongpastlosers.Someofmyintervieweesconfirmedthatcandidatesfindit

hard to get a ticket a second time after having lost an election. Most of the aspiring

politicians Ispoketoowereveryspecificaboutwhentheywouldmaketheirattempt to

getaticket.Someofthemlookedasfarastwoorthreeelectionsahead,alreadybeingbusy

buildingtheirprofileattimestentofifteenyearsinadvance.

Another reasonwhyaspiringpoliticiansprepare their candidacywell inadvance is that

theyknowthatinallprobability,theywillhaveonechancetobecomeanMLA,andthatan

initiallosswillimpedetheirchancetogetaticketagain,andtogetvotesincasetheydo.

157

Tabl

e3.

21N

umbe

rofM

LAs’

cont

ests

bef

ore

elec

tion

inU

ttarP

rade

shst

ate

asse

mbl

yel

ectio

ns(1

952-

2012

)

2012

196

(87.11

%)

18

(8.00%

)

6

(2.67%

)

4

(1.78%

)

1

(0.44%

)

225

(100

%)

Source

:Cal

cula

ted

byA

utho

r,In

divi

dual

Incu

mbe

ncy

inIn

dian

Sta

teE

lect

ions

dat

aset

(195

2-Pr

esen

t)

2007

154

(72.99

%)

44

(20.85

%)

8

(3.79%

)

3

(1.42%

)

1

(0.47%

)

1

(0.47%

)

211

(100

%)

2002

142

(74.74

%)

28

(14.74

%)

12

(6.32%

)

3

(1.58%

)

2

(1.05%

)

3

(1.58%

)

190

(100

%)

1996

116

(61.05

%)

42

(22.11

%)

22

(11.58

%)

5

(2.63%

)

5

(2.63%

)

190

(100

%)

1993

116

(58.59

%)

50

(25.25

%)

21

(10.61

%)

7

(3.54%

)

3

(1.52%

)

1

(0.51%

)

198

(100

%)

1991

160

(65.04

%)

56

(22.76

%)

21

(8.54%

)

6

(2.44%

)

1

(0.41%

)

2

(0.81%

)

246

(100

%)

1989

144

(68.90

%)

42

(20.10

%)

19

(9.09%

)

2

(0.96%

)

2

(0.96%

)

209

(100

%)

1985

200

(82.99

%)

30

(12.45

%)

7

(2.90%

)

3

(1.24%

)

1

(0.41%

)

241

(100

%)

1980

205

(75.09

%)

50

(18.32

%)

14

(5.13%

)

3

(1.10%

)

1

(0.37%

)

273

(100

%)

1977

181

(73.28

%)

47

(19.03

%)

14

(5.67%

)

4

(1.62%

)

1

(0.40%

)

247

(100

%)

1974

225

(84.27

%)

31

(11.61

%)

8

(3.00%

)

2

(0.37%

)

1

(0.37%

)

267

(100

%)

1969

235

(83.04

%)

37

(13.07

%)

9

(3.18%

)

1

(0.35%

)

1

(0.35%

)

283

(100

%)

1967

258

(82.96

%)

49

(15.76

%)

4

(1.29%

)

311

(100

%)

1962

311

(98.73

%)

2

(0.63%

)

1 1

(0.32%

)

315

(100

%)

1957

309

(98.41

%)

5

(1.59%

)

314

(100

%)

1952

430

(100

.0%)

430

(100

%)

1 2 3 4 5 6 Total

158

Theymakecalculationsaboutwhichpartyorcandidatewillprobablywininthenext

election, when a particular seat will become vacant, which sitting MLAs stands a

chanceatcontestingaLokSabhaseat,etc.

ImetS.Thakur190,aCongresspartyworker inhisearly forties,at theCongressState

OfficeinLucknow,inJuly2011.HebelongstoasmalltowninAwadhandownsasmall

businessinLucknow.HespendsmostofhisfreetimeattheCongressoffice,wherehe

worksasacommunicationofficerduringcampaigns.

“I have been preparing my candidacy since long. See, I’m a Rajput. In my

constituency,thereare20.000RajputssoIhavetodeveloptieswithabout500

familieswho in turnwill touch other voters. Once I have their support, I can

bring it to the party who in turn, will ensure me the support of their own

supporters”.

Theprocessthathedescribesishighlycompetitive,sincemanyindividualsmayaspire

toastatusoflocalleaderwithinacasteandwithinalocality.Politicalcompetitionoften

starts by competing with one’s own caste member, for a position and status of local

leadership.Once theyhaveestablished thatposition, theymust seek supportbeyond

theirgroup,ifthatgroupisnotlargeenoughtoensuretheelection.

In many ways, aspiring politicians must act as elected representatives way ahead of

their election, to progressively build a status of a local leader. They actually often

behave,act,anddressuplikepoliticians,incarnatingtheroleaheadofbeingsanctioned

bythevoters.Inshort,theyfulfillthedutiesofapoliticianwaybeforetheygetachance

atrunning191.TheseindividualsgobymanynamesacrossIndia.DadasorComradesin

West Bengal (Banerjee 2010), dayals (‘intermediaries’) in Northern India, the

Pyraveekar(‘thefixer’)inSouthernIndia(ReddyandHaragopal1985).

190Namechanged.191Foravividdescriptionofthestylesadoptedbyaspiringpoliticians,see(Michelutti2010)

159

Thiskindoftrajectoryisessentiallyvalidforthoseaspiringcandidateswhohopetobe

selectedbyparties,thosewhomakepoliticstheirprofessionbeforeactuallybecoming

politicians. There are many other ways to obtain a ticket, some of them involving a

quickpurchase,orlongcareersinlocalpoliticsorlocalpublicorganizations.

IhaverarelyencounteredMLAswhohadbeenpreviouslyelected in localPanchayati

Rajinstitutions192.Butmanyofthemprofesstohavebeeninvolvedinstudentpolitics,

another common way to get into the political career. Leading agitations helps to

develop an individualized following and to attract parties’ attention. Local student

politicsishighlypoliticizedandstudentunionsgenerallyofficiateascampusbranches

of political parties. Many of them get into student politics with the prime motive of

developingindividualtieswithpoliticians,whorelyonthemformobilization,helping

with party events or create ruckus at rival parties’ events and rallies193. It is not

surprising that in large public universities, student elections emulate ‘real’ elections,

with rough campaign styles, hyper-personalization of the competition, bike rallies,

violentdemonstrationsof strength.Candidatesoftendressupaspoliticians, cultivate

brashmasculinesartorialstyleandattitude.

It is also important to contest for the first time on a strong ticket. Fewmain parties

MLAshavebeenpreviouslyelectedonaminorticket–alocalparty–or,morerarely,as

independentcandidates(barringin1957,anoutlierelection).Wesawearlierthatthere

hasbeenasurgeovertimeofthenumberofindependentcandidates.Thishasn’tledto

anincreaseofelectedindependentsintheVidhanSabha,asillustratedinFigure3.14.

192Thedataavailableonthisquestionislargelyinsufficient.Itessentiallycomesfromthebio-profileofMLAs,alsoknownas‘Who’sWhos’,containonlyself-declaredinformationbyelectedrepresentatives.193Foradetailedsociologyofstudentpolitical leaders,acrosscastegroupsandcommunities,see(Jeffrey2010a)

161

3.2.4.ThesizeofthestablepoliticalclassThe individual incumbencydataenables tomakeanothermeasurement,which is the

measurementofthesizeofthestablepoliticalclass.WhatIhavedonesofaristolook

atthetransientpoliticalpopulation,MLAswhocomeandgo.Theotherwaytolookat

thisdataistoclusterthosewholastinpolitics.These“stable”politiciansconstitutethe

class of professional politicians; those who tend to make of politics their main

occupation. It comprises individualMLAswhomatterwithinpartyorganizations, kin

and relatives of party leaders or party figures, and local political dynasties. These

lasting MLAs often usually head sub-regional patronage networks among other

politicians. This stable class of politicians also includes powerful individuals, who

succeed in lasting in politics on their own strength, at times regardless of party

affiliation.Theirnumber isameasureofhowconcentratedpoliticalpoweractually is

within the state and within parties, despite the alternance of power, the turnover of

MLAsanddespitetheoverallcompetitivenessoftheelectoralarena.

Mymeasure ofwho is part of this stable political is, quite simply, anyMLAwhohas

succeeded in being elected more than twice. One could contests the validity of this

measure, calling it arbitrary. The logic is the following. Many individual or political

entrepreneuraspire toget intopolitics,succeed ingettinga ticketandeventuallyare

elected.First-timeMLAswillbeinclinedtore-runiftheycan(orareallowedto)anda

certain number of them will succeed (28.8% of incumbent candidates on average

succeedingettingre-elected,inthecaseofUttarPradesh).

Tobere-electedasecond timehowever implies inmostcases that thecandidatehas

becomeaprofessionalpolitician,hassucceededtoovercomeallthehurdlesofpolitical

life more than once, matters to his or her own party and intends to remain in this

professioninthelongrun.Thismeasureisofindicativevaluemorethananythingelse.

MLAswhofall intothat ‘stable’categoryrepresentonaverageabout20%ofthetotal

population of MLAs. Over time, that number increases above 25% (see table 3.22),

whichmakesenseinaperiodofquadripartitionofthepartysystem.

162

Table3.22Decadalestimatesizeofthestablepoliticalclass

1960s 1970s 1980s 1990s 2000s

Stablepoliticalclass

11.05% 14.25% 20.86% 25.56% 26.63%

Source:CalculatedfromECIdata.

Ifwetakethisasameasureofinfluence,itcanbethussaidthattherearetodayroughly

aquarterofallregionalpoliticianswhoactuallymatterwithintheirparties, that isto

sayalittleaboveonehundredindividuals.Thisisn’tmany,forapoliticalsocietyoftwo

hundredmillionpeople.

Ifwebreakdown this databyparty,we see that twoparties – theBJP and the SP –

standout since1989.This isexpectedsince these twopartieshave risenduring that

period. But the BSP also rose in the 1990s and the 2000s and its share of ‘stable

politicians’remainsmuchlower,includingin2007,whenithadamajorityofseats.This

meansthattheBSPreliesmoreon“short-term’politicians,whichisconsistentwiththe

turnoverfigureswesawintheprevioussection.ThismeansthattheBSPdoesnothave

theimageofapartywhereonemakesalongcareer.

Table3.23Party-wisebreak-upofthestablepoliticalclass,1980-2012

Total>2 1980 1985 1989 1991 1993 1996 2002 2007 2012

BJP 4.65% 7.62% 25.81% 40.19% 45.97% 36.97% 16.35% 15.15%

INC 66.67% 58.14% 38.10% 26.88% 11.21% 8.87% 8.40% 7.69% 5.05%

BSP 0.95% 1.87% 10.48% 8.40% 26.92% 10.10%

SP 13.33% 25.58% 46.67% 44.09% 31.78% 30.65% 36.13% 39.42% 62.63%

Others 20.00% 11.63% 6.67% 3.23% 14.95% 4.03% 10.08% 9.62% 7.07%

Total 100% 100% 100% 100% 100% 100% 100% 100% 100%Source:CalculatedfromECIdata.

ThesenumbersalsoshowthattheSPisthepartythatcountsthelargestnumber–and

thereforeratio -of longstandingpoliticians.Mostof the longestpoliticalcareershave

been served within the Samajwadi Party and before 1993 within the Janata Parivar

parties. Mulayam Singh Yadav has been elected nine times, Mohammad Azam Khan,

Shivpal Singh Yadav eight times. Mulayam Singh Yadav’s family members figure

163

prominently among those longstanding politicians195but also other figures, such as

Mohammad Azam Khan, from Rampur, and Vijay Singh, from Milak, both eight-time

MLAs196.

ThelowernumbersoftheBJPinthelasttwoelectionsisnotonlythereflectionofits

declining performance, but also from the fact the party is losing many of its

longstandingpoliticians,whoeitherleavethepartyorarereplacedbynewfaces.Ajay

Kumar Poiea, a three-time BJP MLA from Govardhan, contested (and lost) on a BSP

ticket in 2007. Ajay Pratap Singh (alias Lalla Bhaiya), a four-time BJP MLA from

Colonelganj,contested(andwon)onaCongressticket in2007.Amarjeet JanSevak,a

three-time MLA from Bindki, contested (and lost) on a SP ticket that same year. In

recentyear, theBJPhas tended tocentralize its ticketdistribution fromDelhi,where

thenewofficebearerprivilegenewfacestoold-timers.

Thenumberofveteranfigureswithinpartiesisalsoindicativeofhowcentralizedtheir

organizationsare.Ofallthemainfourparties,theSamajwadiPartyisthepartythathas

thewidestbaseofstablepoliticians,despite thecontrolexertedby theruling family.

ThiswouldalsoexplainwhytheSamajwadiPartyisthelargestrecipientofturncoats.

Itissimplyamoreattractivepartyforthosewhoaspiretolongerpoliticalcareers.

3.2.5.PoliticalstrongholdsFinally, another marker of stability within volatility is the presence of stronghold

constituencies,whichIdefineasanyconstituencyheldatleastthreeelectionsinarow

by a party. There is a number of seats that parties succeed in retaining over time,

indicating either the presence of a strong MLA or a favorable constituency’s

demography,suchascitiesfortheBJPorahighproportionofMuslimvotersforsome

MuslimMLAs.

195Not necessarily in the MLA dataset since some of them, such as Ram Gopal Yadav, haveservedintheLokSabha.196SomeofthesecareershavebeeninterruptedformandatesservedeitherintheLokSabhaorintheLegislativeCouncil.Verylongcareersarerathertheexceptionwithotherparties.Someof these careers spanned across parties, like Kalyan Singh, ten times elected, who contestedundervariouspartybanners.

164

ThefirstmeasureofstrongholdsinUttarPradeshassemblyelectionsrevealsthatthere

are not that common, except in the 1990s, where more than one seat out of three

remainedinthehandsof thesameparty forthreeconsecutiveelections.TheBJPand

thesocialistpartiesaccountfor86percentofthestrongholdsinthe1990sandfor70

percentofthestrongholdconstituenciesinthe2000s.

ThehighnumberofBJPstrongholds in the1990s(82) indicatehowstrong theparty

was.Between1985and1993, theBJPcounts37strongholds, including7 inreserved

seatsand14inurbanseats.HalfofthesestrongholdsarelocatedinAwadhandDoab,in

and around Lucknow, Sitapur, Agra, Mathura and Aligarh. Outside these two central

regions,theBJP’sstrongholdsarelimitedtothemaincities,suchasMeerutintheWest,

BareillyCityinRohilkhand,VaranasiintheEast,andGorakhpurintheNorth-East.

Table3.24Decadalpartydistributionandratioof‘stronghold’seats 1970s 1980s 1990s 2000sBJP 7 82 34BSP 12 22INC 34 59 8 4Socialists 33 60 38JNP 37 5 CPI/CPM 1 1 2 1RLD 3IND 1 Totalstrongholds 73 105 164 102Notstronghold 351 319 321 301Strongholdratio 17.22% 24.76% 38.68% 25.31%Notstrongholdratio 82.78% 75.24% 75.71% 74.69%N 424 424 424 403

Source:CalculatedfromtheUttarPradeshIncumbencydataset.

In the1980s, theCongressstillhad59seats it couldcall strongholds,halfof them in

Avadh.Of these60strongholds,one threesubsisted in the1990s:Hardoi,RaeBareli,

Rampur Khas. It then lost the first two in the 2000s, maintained its hold of Rampur

Khas, and succeeded in preserving only two other strongholds, in Rae Bareli and

Padrauna, a city the North-East. These seats are held by prominent – or formerly

prominentfiguresoftheparty.

165

PramodKumarTiwari,aformerStateMinister,wontheRampurKhasseateighttimes

inarow,usuallywithlargemargins.Fornearlytwentyyear,Tiwariwastheheadofthe

CongressLegislatorparty,intheVidhanSabha.Hewasremovedfromthatpositionin

2012andreplacedby the three-termMLA fromMathura,PradeepMathur.Theparty

grewresentfulofTiwari’scross-partiesconnection–whichlednotablytheSamajwadi

Party to fieldweakcandidatesagainsthim–andofhisproximitywithSubrotoRoy’s

SaharaGroup,oneofthemaincorporatesponsoroftheSamajwadiPartyinthe1990s

andearly2000s197.TheCongressgothimelectedintheRajyaSabhain2013.

In Rae Bareli, the seat is held by Akhilesh Kumar Singh is a five-time MLA, who

contestedasanIndependentin2002andasaPeacePartycandidatein2012(hewould

then become the Peace Party’s leader in the Vidhan Sabha, after the removal of its

founder,Dr.Ayub).SinghstartedhiscareerintheCongressin1993.Hecurrentlyhas

45 criminal chargeson twelve cases againsthim, including seven charges for rioting,

sevenchargesforcriminalintimidation,tworelatedtomurder,threerelatedtoattempt

to murder, two charges or extortion one of dacoity and one charge related to

kidnappingforransom.

In Padrauna, the seat is held by Ratanjit Pratap Narayan Singh, Raja of Jagdishpur,

attachedtothePadraunaStateRulingfamily.HeisaformerUnionMinisterofStatein

the secondUPAgovernment (Road,TransportandHighwaysand thenHome), sonof

Congress MLA and MP Chandra Pratap Narayan Singh, educated in Doon School (of

whichhepresidestheoldschoolsociety)andSt.Stephen’s.HewaselectedintheLok

Sabha from Kushinagar (Padrauna Sagar) in 2009 and left his seat to Rajesh Kumar

Jaiswal,wholostagainsttheBSPcandidate,SwamiPrasadMaurya.

Inthe1970s,thesocialistpartieshad38strongholds(includingsixreservedseats)198,

scattered across the state, except in Uttarakhand and Bundelkhand, where they had

197TimesofIndia,June7,2012.19824ofthesewreheldbeforetheEmergency.

166

none. They retained fourteen of them through the 1980s, rural seats for the most

part199.

The Samajwadi Party’s strongholds in the 1990s were for the most part located in

centralUttarPradesh,inDoab(22),Awadh(10)andintheEast(11).Theyincludelong

term strongholds held by leading figures of the party, like Jaswant Nagar, held by

MulayamSinghYadavseventimesbetween1967and1993,thenbyhisbrotherShivpal

Yadav, from 1996 to 2012200or Rampur, held with one interruption since 1980 by

Mohammed Azam Khan, a prominent Muslim figure of the party and its longtime

GeneralSecretary.

SomeoftheseseatsareYadavstrongholds,suchasGunnaur,where12outof18MLAs

havebeenYadavs(includingMulayamSinghYadav,in2007),Jaswantnagar,where14

outof15MLAshavebeenYadavs,orAliganj,whereelevenoutofthelastthirteenMLAs

have been Yadavs. The strongman inAliganj is Rameshwar Yadav, a three-timeMLA

with110criminalchargesonhishead(for27differentcriminalcases).Someofthese

strongholds are located in reserved seats, like Tundla, Etmadpur or Kishni, where

RameshwarDayalBalmikiwaselectedwithoutinterruptionfrom1989to1996.

Finally, some are historical seats, like Chaprauli, where Chaudhary Charan Singh

contestedandwonbetween1967and1974. It isnowanRLDstrongholdsince2002,

thoughheldbythreedifferentMLAs201.

TheBahujanSamajParty,finally,hadfewstrongholdsinitially.Between1989and1996,

itonlyhadfour,allingeneralseats,andallheldbyOBCs.

RamLakhanVerma,aKurmipoliticians,heldtheJalalpurseat(EasternU.P.)between

1989and1993.HeservedasaMinisterforForestsinMayawati’sfirstCabinetandwas

199Rampur,Fatehpur,Kauriram,Maharajganj,Hata,Chilkahar, Jhunsi, Soraon,Bilhaur,Gokul,Hathras,Barnawa,Chaprauli,andKandhla.200ShivpalSinghYadavisalsoathree-timeMLAfromUnnao,nearKanpur,in1974,1980,and1991.201TwoJats,AjayKumarandVirpalRahi, in2002and2012,andaRajput,Dr.AjayTomar, in2007.

167

partoftheBahujanSamajDalbreakawayfaction,agroupdefectionengineeredbythe

SamajwadiPartyin1995.HequicklyjoinedtheSP.Hiscareerwasputtoabrutalend

whenhewaskilledbyLalBahadur,hisownpetelephant,in2003.TheBSPlostJalalpur

toRakeshPandey,fromtheSP,in1996.AfterVerma’sdefection,theseatwenttoSher

BahadurSingh,aRajputfromtheBJP(formerlyCongress),whodefectedtotheBSPin

2007,andthentotheSPin2012.

RamSevakSinghPatelisathree-timeMLAfromBara(Doab),aconstituencyonceheld

byH.N.Bahuguna,formerCongressChiefMinister.Helosthisseatin2002toUdaiBhan

Karvaria,aBrahminfromtheBJP.Hecontestedagain(andlost)onaSamajwadiParty

ticketin2007andthenonaBSPticket,inBadaun,in2012,wherehelostagain.

VishambarPrasadNishadwasathree-timeMLAfromTindwari(Bundelkhand).He’sa

leading figure among his Nishad (a caste of fishermen) who served as Minister in

Mayawati’s Cabinet on three occasions (Fisheries, Animal husbandry, Revenue,

Ambedkar Gram Sabha Development, External Aid and Mining). He climbed on to

becometheLokSabhaM.P.fromFatehpur(Doab)in1996.In2002,hedefectedtothe

SamajwadiParty,whoofferedhimthepostofNationalGeneralSecretary.Hecontested

(and lost) the2014generalelections fromHamirpur-Mahoba,andwaselected to the

RajyaSabhain2014202.

SriramPal isa three-timeMLAfromKalpi (Bundelkhand)andoneof the fourBaghel

communityfigureinUttarPradeshintheearly1990s203.After1996,theBSPwentonto

gain21strongholds(including6inreservedseats),scatteredacrossthestate.

It is difficult to state whether strongholds are held by popular parties or by strong

individuals.MostoftheMLAsdescribedinthesesectionsareeitherprominentpolitical

202HistwoclaimsoffamewerehisinterventionfortheliberationoftheBandit-QueenPhoolanDeviandtheattempttoinductseventencastesoffishermenintotheSClist.Thatattemptfailedwhen a 2005 ruling of the Allahabad High Court quashed that decision. He now serves asPresidentoftheSamajwadiKashyapNishadBindTurahaEktaMahasabha,acasteassociationworkingforthecauseoffishermen’scastes.203TheotherthreewereBhagwatPal(Majhwa),MathuraPrasadPal(Sarwankhera)andInderPalSinghPal(Auriya).

168

figures–stateMinistersorhighofficeholders in theirparties–orprominent figures

withintheircastes.Castetiesdomatter,asweseethatstrongholdsoccurmoreinseats

dominated by specific castes. In Western Uttar Pradesh, for example, several seats –

Saharanpur,MuzaffarnagarandShamli–aredominatedbyGujjars,whoarecourtedby

various parties. These are known as “Gujjar seats” even though there may be an

alternanceofparties204.Familyties–andfamilyfiefdoms–areanotherconfigurationof

strongholdconstituencies.Beyondtheprominentfamiliesrulingoverparties,thereisa

numberofpoliticalfamilieswhocontrolorholdtheirconstituenciesoveralongperiod

of time.KazimAliKhan (aliasNavaidMian) four-timeMLA inSuarTanda, isanother

example205.ThispoliticianbelongstotheBarechaRohilladynasty,whichusedtorule

the former Princely State of Rampur. He is the 15th ruler of this hereditary dynasty,

whichwasfoundedin1719.AShiaMuslim,he is thesonofNawabZulfikarAliKhan,

aliasMickeyMian,whowaselectedRampurMLAin1963,andfrom1980to1992206,

and of Noor Bano, the daughter of the Nawab of Loharu, also a prominent political

figure207.

KazimAliKhanjoinedtheCongressinthemid-1990s.Hewaselectedforthefirsttime

in Bilaspur, in 1996, then subsequently in Suar Tanda. In 2003, unhappy with the

party’s attitude towards minorities, he was among the eight members of a splinter

group within the Uttar Pradesh Congress: the Akhil Bhartiya Congress Party. The

formationthenmergedwith theBSPandKazimobtainedaportfolio in theMayawati

governmentasMinisterofMinorityWelfareandHaj.HedefectedagaininAugust2003

for theSPwhenMulayamSinghYadavwasable to formagovernment.A tusslewith

Azam Khan, another prominent Muslim political figure from Rampur prevented him

from becoming a Minister208. He contested on an SP ticket in 2007 and won a third

mandatebutrejoinedtheCongressforhisfourthterm,in2012.Notallpoliticalfamilies

belongtoformerrulers’families.Thelattercategoryoftenconsidertheirconstituencies

204RajkamalSinghattractedmyattentiononthispoint.205IbenefitedfrominputsfromJulietteGalonnierregardingthisMLA’sbiography.206Seehttp://164.100.47.132/LssNew/biodata_1_12/1818.htm207AkeymemberoftheAllIndianCongressCommittee,shewaselectedMPfromRampurconstituencyin1996andin19991999-2004.Seehttp://164.100.47.132/LssNew/members/former_Biography.aspx?mpsno=50208HewouldbegrantedthechairmashipoftheUttarPradeshTourismCorporationinstead.

169

as personal fiefdoms and use electoral politics as ameans tomaintain their political

status.

Looking at the caste composition of these strongholds, it would appear that they

containaslightlyhigherconcentrationofuppercastesanddominantOBCMLAsamong

the ‘stronghold MLAs’ than in the general MLA population. Around 13 percent of all

strongholds areheldbyMuslimMLAs (two thirdwith socialist parties and theother

thirdwithCongress,inthe1970-80s).Twentypercentofallstrongholdsarelocatedin

reservedseats.

3.3.ConclusionThis rather longdata enumeration reveals that thepath to apolitical career inUttar

Pradeshisscatteredwithobstaclesandpitfallsandthatthesecareerstendtobeshort.

Voters of course are the ultimate arbiters but parties act as powerful filters for

candidates, by determiningwho gets the ticket and by determiningwhomay have a

chancetore-runafteranelection.Thus,parties’assessmentonthe‘winnability’oftheir

candidates, and their assessment of what constitutes this winnability is crucial in

determiningthesociologicalcompositionoftheStateAssembly.

To bemore complete about the rules of political engagement, one should also speak

aboutthehighcostofentryintopolitics–thatistheindividualfinancialcommitment

required from the candidates.The fact that campaign costs spiral after eachelection,

thatlivingthelifeofapoliticianisitselfacostlyaffairalsohasagreatimpactonwho

getstobeanelectedrepresentative.

Andwhilethereisusuallyalotofattentionpaidtowhichcastegetstoberepresented

withintheAssemblyorwithinparties,weseldompayattentiontocriteriadetermining

whichindividualgetstobecometherepresentativeofhiscasteorcommunity.Besides

thehighcostofentry,candidatesmustalsoconfrontalongarduouspathofintra-party

competition–factionalism–andalsooftenintra-castecompetition.Whoemergesasa

localcaste leader isalso theproductof intensecompetition.Thesequestionsshallbe

addressedinchapterfourandchapterfive.Whatisimportanttorememberisthatstate

170

electionsinU.P.havealwaysbeencompetitive.Withanincreasedparticipation,parties

mustmobilizemorevoters thanbefore.Also, thecompetition thatused to takeplace

within the Congress party now takes place between a larger number of political

formations,eachconstitutingasiteofcompetitionontheirown.Moregroupsarealso

includedinthiscompetitionandwewillseeinthenextchapterthatifforatimecaste-

based competition seemed to take place through dedicated parties, caste-party

alignments have in recent times been blurred by parties distributing tickets across

castegroups.

171

Part.IIThechangingprofileofUttarPradeshStateLegislators

Chapter4.TheheterogeneisationofUttarPradesh’spoliticalclass

ThepoliticsofUttarPradeshismostfrequentlyanalyzedthroughthelenseofcaste.This

analysisismadefromtwodifferentangles,partiesandvoters.

From the parties’ point of view, caste is a central variable to the definition of their

electoral strategies. Parties “read” the electorate on the basis of these social divisions,

divide the electoral map and distribute tickets largely through their own assesment of

local caste demography and through the desire to maintain particular representation

balancesbetweenselectgroups.

Beyond this electoral arithmetic, caste is also presented as a fundamental political

variable for voters, who, as the saying goes, vote their caste while casting their vote.

Available survey data on voting behaviour confirms that there is a certain level of

congruenceoralignmentbetweencertain castesandcertainparties.But the samedata

also contains as we shall see the elements for a necessary nuance of that depiction,

indicatingamongotherthingsthatveryfewcastesorcommunities–ifany–seldomvote

enblocforanyparties.

It is largely understood and admitted that caste has been the main vehicle of political

mobilization in Uttar Pradesh for decades, the social unit around which demands for

social justice, equity, and dignity are articulated and crystallize. Descriptive

representation is seen as a measure and token of justice and dignity, an aim in itself

meanttoobtaintherecognitionofone’sgroupsocialandpoliticalimportance.Fromthe

pointofviewofdominantgroups,politicalrepresentationisalsooftenameanstoretain,

develop or expand a group’s social status, privilege, and hold over territories. Being

represented in theAssemblyalsomeans tohaveaccess topowerwielders,bothwithin

parties and within the bureaucracy, both organizations and institutions that play a

determinantrole in thedistributionofresources.Thetwofunctionsofrepresentation–

providingdignity andproviding resources – are certainlynotneatlydistributed among

socialgroupsoraccording toastricthierarchy.Thepursuitofpowerof lowercastes is

172

not devoid of materialistic considerations, and the elements of status and prestige

attachedtothepositionoflegislatorareapowerfulforceattractingmanyuppercastesor

membersoflocallydominantgroupsintotheelectoralfray.

The evolution of the sociological composition of the state assembly provides the main

empirical base for the measurement of the political empowerment of castes and

communities.Italsoprovidesthebaseforthebuildingofthenarrativethatdominatesthe

descriptionofU.P.politics,thatistosayastoryofdeclineoftheuppercastesandofthe

riseoftheOBCs,astoryofreversalofthesocialorder.

Thepurposeofthischapteristoexaminethisnarrativeby“unpacking”thedataoncaste

and community representation in the State Assembly. I do this by breaking down the

‘castedata’ in threemainways:by jati,bypartyandbysub-region.Thedivisionby jati

willreveal–asitisalreadylargelyknown–thatonlyahandfulofcastesdetainsamajor

share of the seats of the assembly, both within the upper castes and the OBCs. The

division by sub-regions will reveal, more importantly, that there are strong spatial

variationsinthetrajectoryofcastes,strongenoughtoquestionthedominantnarrativeof

declineoftheuppercastes.Recentdatafurthermorerevealshowtheuppercasteshave

regained a part of their past prominence, by beingwell representedwithin the parties

that initially rose against them. I also examine in this chapter the question of

inclusivenessofpoliticalparties,bylookingatcasterepresentationwithinthem,aswell

asinthestate’scabinet.

Thedata

The creation of an original database of caste representation among Members of the

LegislativeAssembly(MLAs)ofUttarPradeshtookconsiderabletime.Tobeginwith,U.P.

hasthelargestassemblyinIndia,with404members209.

Then,theoriginaldatasetcollectedbyJasmineZerininiforhercontributioninthe“Riseof

thePlebeians?”waslargelylostandalargepartofherdatahadtobecollectedagain210.In

209403 are directly elected and there remains one nominated members, from the Anglo-Indiancommunity.

173

ordertodoso,IconductednumerousfieldtripstoLucknowandacrossthestatebetween

2007and2012.DataonMLAsandcandidates’jatiwascollectedthroughinterviewswith

threedistinctsetsofsourcesandactors.Interviewswithpoliticiansthemselvesandparty

cadres were particularly useful to gather the information regarding contemporary and

past representatives.Correspondentsand local journalists inLucknowwere the second

source used to collect information. Long sessions of collective examination of the

candidates’ lists were held in newsroom or at the U.P. Press Club in Hazratganj. Local

correspondents from the vernacular press were contacted by phone through the Uttar

Pradesh Journalists’ Directory. The third source of information was composed of

colleagues, political observers and local research assistants211. I finally compiled the

secondary data that could be found online or in print about particular individuals or

constituencies. The data collected from these three sources has been juxtaposed and

validated when matching. In case of divergence, further investigation was done until

reachingaconvincingoutput.Insomecases,politicianswerecontacteddirectlybyphone

forconfirmations,inthemidstofabroaderconversation.

The original dataset was thus re-created and expanded, by adding data on the main

parties’candidatesforthe2007and2012elections.ThedataregardingtheMLAslyingin

theStateAssemblyarchiveandLibrarywerecollectedanddigitized212.Furthermore,the

Election Commission’ statistical reports on state elections were crawled, cleansed and

merged with this data, so that the profile of candidates may be linked to data on

performance,forfurtherresearch.

Choiceshadtobemadewithregardtothecodificationofjatis.Whilethereisanoverall

consensusonwhobelongstotheuppercastes,thejuridicalcategoriesofOBCs,Scheduled

CastesandScheduledTribesfluctuateintime,asitisthestate’sprerogativetodetermine

which caste belongs to which category. State governments have often changed the

210As a consequence, my data varies slightly from Zerinini’s, though not too substantially. Themain variation is that she found more OBCs in the assembly in 1993 than upper castes (32.39against26.98percent),while Iobtainedsignificantlydifferent figures (31.84 to34.43percent,respectively).Ialsohavelessunidentifiedcases(1against28),whichcanaccountforapartofthisvariation.211Inmorerecenttimes,RajkamalSingh,ResearchFellowattheTrivediCentreforPoliticalData,providedinvaluableassistancetofillthegapsinthedataandscoutforcodingerrors.212VidhanSabhaSecretariat,Who'sWhoinU.P.VidhanSabha(1952-2012),Lucknow.

174

denominationof certain castes, essentially for electoralist purpose, shifting castes from

theSClisttotheOBClistandvice-versa.

TheJatsareacaseinpoint.Thispeasantcaste,presentanddominantinWesternU.P.,was

included on the Central OBC list in March 2014, barely a few weeks before a general

election,tothedismayofothergroupsalreadyincluded213andtotheirritationoftheirJat

neighbors in Haryana, who do not benefit from the same treatment (the decision was

overruledbytheSupremeCourtayearlater)214.

Thedatasetdoesnotaccountforthesevariations.Thecodingofjatisintocastegroupsis

stable through time. This dataset should be seen more as a heuristic tools that help

accountingforsomeofthelargepoliticaltransformationsthathaveoccurredinthestate,

rather than an effort to classify identities whose definitions are necessarily plural and

shifting through time. With that purpose in mind, Jats for instance have been coded

separately,asIntermediatecastes,despitetheirrecentinclusionintheOBClist.

4.1.TheEvolutionofcasterepresentation

Accordingtothe1931Census,theuppercastesconstitute20.5percentofthepopulation,

the Brahmins and Rajputs taken together amounting for 16.4 per cent. The Other

Backward Classes (OBCs) represent 41.7 per cent of the population, the bulk of that

number being composed of numerous small-size and geographically dispersed castes,

usuallyreferred toaMostbackwardClasses(MBCs).The threemaingroupsamongthe

OBCs,namelytheYadavs,theKurmisandtheLodhis,accountforrespectively8.7,3.5and

2.2 per cent of the population.Other locally dominant groups, such as the Jats and the

Gujjars,representrespectively2and0.7percentofthepopulation.

These numbers have only an indicative value since we do not know how they have

evolvedintime,intheabsenceofcastecensus.Further,thecarvingin2000ofUttaranchal

213TheywerenotincludedintheoriginalOBClistdrawnbytheMandalCommission.214Interestingly,theJatpoliticalleadershiphadinitiallyopposedtheideaoftheinductionofJatsamongtheOBClist.See(Jaffrelot2010b).

175

(now Uttarakhand) would have changed the overall demographic balance in U.P., the

uppercastesbeingdemographicallydominantintheHillareas.

Table4.1CastesandCommunitiesinUttarPradesh,1931(in%)CastesandCommunities

UpperCastes

Brahmin

Rajput

Bhumihar

Bania/Jain

Kayastha

Khatri

Tyagi

Intermediarycastes

Jats

OtherBackwardClasses(OBC)

Yadav

Kurmi

Lodhi

Teli

Koeri/Kacchi

Kewat/Murao

Gujjar

Others

ScheduledCastes(SC)

Chamar

Pasi

Dhobi

Bhangi

Others

ScheduledTribes(ST)

Muslims

Sikhs

Anglo-Indians

20.5

9.2

7.2

0.4

2.5

1.0

0.1

0.1

2.0

2.0

41.7

8.7

3.5

2.2

2.0

3.1

2.4

0.7

19.1

21.0

12.7

2.9

1.6

1.0

2.8

<1.0

15.0

<0.5

<0.1

Total 100.0

Source:Census1931,UnitedProvincesofAgraandAwadh,Part2,ProvincialandImperialTables,1933,reproducedinHasan1989.QuotedfromZerenini(2009).

176

WedohavemorerecentdataontheMuslimandScheduledCastepopulations, fromthe

Censusof2011(seechapter2).Whenwecomparethetwocensuses,thetotalshareofSC

population decreases from 21 to 20.7 per cent. Among them, the Chamar / Jatav

populationdecreases from12.7 to11.3per cent, thePasis increase from2.9 to3.3per

cent215.TheshareofMuslimsincreasesfrom15percentto19.26percentin2011216.

4.1.1.Thefourphasesofcastegrouprepresentation

In her contribution to the Rise of Plebeians?, Jasmine Zerinini laid down in 2009 the

empirical and contextual groundwork for the study of the evolution of caste-based

representationfromtheearly1960suntil2002(Zerinini2009).Shedividesthisevolution

in three periods, or phases217. The first one, from 1952 to 1967, is marked by the

ascendencyoftheuppercastes,withinandalongsidetheriseoftheCongressParty.The

uppercastesthengraduallydeclinedinasecondperiod,markedbytheriseofCongress

opposition,between1967and1989.During thatphase, a greaternumberof castes are

represented in the assembly, particularly among the OBCs and among the Scheduled

Castes.Therewereonaverage26and32castesrepresentedintheAssemblyinthe1960s

and1970s.Thenumbersincreaseto40inthe1980s,46inthe1990sand43post-2000218.

Zerininiseesin1989aturningpoint,markingthebeginningofathirdperiodinwhichthe

representationofOBCsintheAssemblyrisessharply.Weseeinthefollowingfigurethat

theriseofOBCsisinfactanteriorto1989.ThevictoryofIndiraGandhiin1980meanta

surgeofrepresentationofuppercastes,at47.8%,andcausedareductionofthepresence

ofOBCs in theAssembly,downfrom18.6percent in1977to14.3percent threeyears

later.Fromthere,theOBCswouldbeonacontinuousriseuntil1993,wheretheypeaked

at31.8percentoftheseats,nearlyatparwiththeuppercastes,at34.4percent.Zerinini

rightly points at the SP-BSP alliance of 1993 as the reason for the surge of OBC

215TheseoverallvariationsamongtheSCscanbeexplainedbydemographicsdynamicsandalsobythefactthatthecompositionoftheSClistchangesintime.Further,thecensusattimesclusterscertainsub-castestogetherwhichmayalsoaccountforvariations.216Onthequestionofreligious-baseddemographictrends,see(JefferyandJeffery2006).217Ibid.,p.33.218The variations come from the induction of small castes among theOBCs and the SCs. ThesefiguresdonottakeintoaccountthediversityofgroupsamongMuslims.

178

nottovoteforDalitcandidateswhentheyhavealternativesandduetothefactthatinany

case,partiesfieldveryfewDalitcandidatesingeneralseats220.Wecannoticethatthede-

reservationofmanyconstituenciesafter30yearshasnotproducedanyeffectintermsof

electabilityofDalitcandidatesingeneralseats.Therewereinfactaverysmallnumberof

Dalitcandidatesinthede-reservedseats.

Overtime,theuppercastesremaintheonlycastegroupover-representedintheassembly

(above30percent).TheOBCdeclinedfrom31.9percentin1993to26.9percentin1996.

Theyhavesincethenremainedstableat26percentoftheseats.TheSCsremainstable

due to their quota, which ensures their descriptive representation. But the most

remarkablechangeoverthepasttwentyyearsistheriseofMuslimsintheAssembly,due

totheperformanceofthemaintworegionalparties–theSPandtheBSP–butalso,aswe

shallsee,duetoasignificantevolutionoftheirvotingbehaviourinrecentelections.

Obviously, these castegroupsarebroadcategories thatmustbedifferentiated.Noneof

themvoteenblocnoreevenconstituteacohesivesocialentityastheyaredividedintoa

largenumberofjatis,sub-castesorotherformsofclanickinship221.

4.1.2.Uneventrajectoriesamongtheuppercastes

Among the upper castes, the Rajputs and the Brahmins clearly dominate. These two

groupsused tobe sideby side in shareof seatsuntil1985, afterwhich theydriftedon

divergenttrajectories.TheBrahminsdeclinedearlierwhiletheRajputsmaintainedtheir

overallrepresentationuntiltheearly2000s.Oneofthereasonsforthisdivergenceisthat

the Brahmins were more associated with the Congress and the BJP, which declined at

various stages, while the Rajputs divided their votes across other parties as well, and

werewellrepresentednotablywithintheSP.

Collectively,thesetwocastesoccupynearly35percentofalltheseatsthroughthe1970s,

37percent throughthe1980s,30.5percentinthe1990sand26.5percentafter2000.

220Thishasbeenthecasesincethefirstelections,asnotedbyPaulBrass(1984),p.23.221Foranoverviewoftheearlydebateoverthedefinitionofthebackwardclassescategory,see(Galanter1984).

181

roughtlythevotesharecurveoftheBJP.TheBJP’ssuccessinthe2000wasbasedonits

ability toattract thesupportofmanyMBCvotersbydistributing ticketsamongvarious

non-dominantOBCgroups.

Table4.2CasteandcommunitycompositionoftheU.P.VidhanSabha,1969-2012(%)

1969 1974 1977 1980 1985 1989 1991 1993 1996 2002 2007 2012

UpperCastes 46.71 41.65 39.49 47.42 41.08 39.53 41.19 34.35 36.71 35.80 34.07 30.86

Bania/Jain 3.29 3.76 3.74 1.17 0.70 2.35 2.38 2.59 2.35 3.21 3.46 2.72Bhumihar 0.47 2.35 1.40 1.64 0.70 1.18 0.95 0.71 1.18 1.98 0.99 0.74Brahmin 20.6616.71 16.82 20.42 19.01 14.82 14.76 9.65 10.59 10.12 13.83 11.60Kayastha 1.88 2.59 1.87 0.94 1.17 1.18 0.71 0.71 1.18 0.74 0.99 0.99Khatri 0.47 0.47 0.47 1.41 0.70 0.47 0.71 0.94 0.94 1.23 1.231.23Rajput 19.48 15.29 14.72 20.66 17.84 18.35 19.5218.35 18.5917.28 13.33 13.33Vaishya 0.23 0.47 0.47 0.23 0.70 0.47 1.19 0.71 0.94 0.99 - 0.25Others 0.23 - - 0.94 0.23 0.71 0.95 0.71 0.94 0.25 0.25 -Intermediarycastes 2.11 2.82 1.64 2.58 2.58 3.53 3.10 3.06 2.59 3.46 2.96 2.47

Jat 2.11 2.82 1.64 2.58 2.58 3.53 3.10 3.06 2.59 3.46 2.96 2.47OtherBackwardClasses 12.21 17.65 16.59 13.85 20.19 23.53 26.67 31.29 26.59 26.17 25.93 25.93

Baghel 0.23 - - - - - - - 0.24 0.25 - -Bishnoi - - - - - - - - - 0.25 0.25 0.25Chauhan - - - 0.47 0.47 0.47 0.48 0.24 0.47 0.49 0.49 -Chaurasia - - - - - - - - 0.00 0.25 0.25 0.25Gadariya - - - - - - - - 0.24 0.49 0.49 0.74Garedia - - - - - - 0.24 - 0.47 0.25 - -Goswami 0.23 - 0.23 - - - - - - - 0.25 -Gujjar 0.70 1.65 1.87 1.64 2.35 2.82 1.90 2.12 1.88 1.48 1.98 1.73Jaiswal 0.47 - 0.23 0.23 0.23 - - 0.24 0.24 0.25 0.49 0.74Kapur - - - - - 0.24 0.24 0.24 0.24 - - -Kashyap - - - - - 0.24 0.24 0.24 - 0.49 0.25 -Katiyar - - - - - - - - - - 0.25 -Kevat 0.23 - - - - - - - - - - -Koeri - - 0.23 - - - - 0.24 - 0.25 - -Kumhar - - - - - - 0.24 - 0.24 0.25 - 0.25Kurmi 1.41 3.29 3.04 2.58 4.23 4.94 5.48 6.12 4.24 4.94 5.68 3.46Kushwaha 0.23 0.71 0.70 0.70 0.47 0.47 0.24 0.47 0.47 0.74 1.23 1.48Lodhi 1.64 1.65 0.93 0.47 0.94 1.41 1.90 3.29 2.59 1.73 2.22 3.70Mallah - - - 0.23 0.47 0.47 - 0.24 0.24 - 0.25 0.25Maurya - - 0.23 - - - 0.24 - 0.47 0.49 0.74 0.74Muraon 0.23 - - - 0.23 - - - 0.47 - - -Nishad - 0.24 - 0.23 0.70 - 0.71 0.24 1.18 0.99 0.74 0.99Rajbhar - - - 0.23 - - 0.48 0.47 0.71 0.99 0.74 -Saini - 0.24 - - - 0.24 0.24 0.24 0.47 0.49 0.25 0.25Shakya - 0.24 0.70 0.23 - 0.24 0.48 0.24 0.71 0.74 0.99 1.23

182

Soni - - - - - - 0.48 0.24 - - - -Teli - - - - - - - - - 0.25 - -Yadav 6.34 8.24 7.24 4.69 7.51 10.12 9.52 11.76 9.41 10.12 8.40 9.88Unidentified 0.47 1.41 1.17 2.11 2.58 1.88 3.57 4.71 1.65 - - -ScheduledCastes 21.6 22.82 21.96 21.60 22.07 21.65 21.90 21.65 22.12 22.22 21.98 21.23

Arya - 0.71 0.47 0.23 0.23 0.71 - 0.47 0.24 0.25 - 0.25Bairagi - - - - - - 0.24 - - - 0.25 0.25Barwar - - - - 0.23 0.24 - 0.24 0.24 0.25 0.25 -Baudh - - - - - - - - 0.24 - - -BelldaraChauhan - - - - - - - - - - 0.25 -Beria - - - - - 0.24 0.24 0.24 - 0.25 - 0.25Chamar/Jatav 2.11 1.88 2.57 3.76 6.10 5.65 4.76 5.18 4.94 11.11 12.10 7.41Dhanuk - - - - - - - - - 0.49 0.49 -Dhobi 0.47 0.71 0.47 0.47 0.47 1.18 1.67 0.47 1.18 1.98 1.48 0.74Dhrikar - - - - - - 0.24 - - - - -Dhusia 0.23 0.24 - - - 0.24 - - - - - -Dusadh - - - - 0.23 - - - - - - -Gond - - - - - 0.24 0.24 0.24 - 0.25 0.25 0.25Goud - - - - - - 0.24 0.24 - 0.25 0.25 -Katheria - - - - - - - - - - - 0.25Katik - - - - - - 0.24 - - - - 0.49Katoriya - - - - - - - - 0.24 - - -Kewat - - - - - - - - - 0.25 - -Khatik 0.23 0.24 0.23 0.47 1.17 0.71 0.95 0.94 1.88 1.23 0.49 0.74Kol - - - - - - - - 0.47 0.25 - 0.25Kori 0.47 - 0.23 0.47 0.94 - 0.48 0.71 1.41 0.25 - 1.73Kureel 0.23 0.47 0.70 0.23 0.47 0.24 0.48 0.24 - 0.25 - -Lonia - - - - - 0.24 0.24 - - - - 0.25Pasi 1.64 1.18 1.64 3.05 3.99 3.06 2.86 1.88 2.59 4.20 4.69 5.93Rajak - 0.24 0.23 0.23 - - - - - - - -Ravidasya - 0.24 - - 0.23 - - - - - - -Shilpkar - - - 0.23 0.47 0.24 0.24 - 0.24 - - -Valmiki 0.23 - 0.23 0.23 0.70 0.47 0.71 0.47 0.24 0.25 0.25 1.73Visharad 0.23 0.71 - 0.47 - - 0.00 - - - - -Visvar - - - - - 0.24 0.24 0.24 - - - -Unidentified 15.73 16.24 15.19 11.74 6.81 8.00 7.86 10.12 8.24 0.74 1.23 0.74ST - 0.24 0.23 0.23 0.23 0.24 0.24 0.47 0.24 0.25 0.25 -

Muslims 8.22 9.65 11.21 11.74 12.21 9.65 5.48 7.76 9.18 11.36 13.83 16.79

Ansari 0.47 0.94 0.93 1.41 1.41 0.47 0.48 0.71 0.94 0.74 0.74 0.99Bhangi - - - - - - - 0.24 0.24 0.25 0.25 -Gaddi - - - - - 0.47 0.24 0.24 0.24 0.74 0.25 0.49Khan/Khansab 2.11 2.12 1.87 1.64 2.82 2.59 0.95 1.65 1.88 1.73 0.74 0.49Lohar - - - - - - - - 0.24 - - -Malik - - - - - - - 0.24 - 0.25 0.25 0.25Momin - - - - - - 0.24 - - - - -

183

MuslimGujjar - - - - - - 0.2% 0.24 - - - -Qazi 0.23 - - - - - - - - - - -Qureshi - - 0.23 - 0.23 - - 0.24 - 0.74 0.25 0.25RaeenKunjda - - - 0.23 0.23 - - - - - - -Sayyed 0.23 - - 0.23 0.23 0.24 0.24 0.24 0.94 0.99 0.49 0.99Sheikh 0.23 - - - 0.23 0.24 0.48 0.47 0.47 1.23 0.99 1.48Shia 0.23 0.94 0.70 1.41 0.94 1.18 - 0.47 0.47 0.74 0.49 0.49Turk - - 0.23 0.23 0.23 - 0.24 0.24 0.24 0.49 0.49 -TyagiMuslim 0.23 - 0.23 0.23 - - - - - 0.25 0.25 0.25Unidentified 4.46 5.65 7.01 6.34 5.87 4.47 2.38 2.82 3.53 3.21 8.64 11.11Otherminorities 0.23 0.47 0.70 0.94 0.70 0.24 0.24 0.47 0.47 0.49 0.25 0.25

Christian - - - - - - - -

0.25 - -

Sikh 0.23 0.47 0.23 0.94 0.70 0.24 0.24 0.47 0.47 - - -

Nominated 0.23 0.24 0.23 0.23 0.23 0.24 0.24 0.24 0.24 0.25 0.25 0.25

Unidentified 8.69 4.47 7.94 1.41 0.70 1.41 0.95 0.71 1.88 0.00 0.49 2.22

Total 100 100 100 100 100 100 100 100 100 100 100 100Source:Author’sfieldwork.

Other dominant groups, such as the Jats (clubbed with the OBCs here for practical

purpose)or theGujjars areonlypolitically significant in relatively small pocketsof the

territory,intheWest.SomesmallOBCcastescandrawadvantagefromtheirdemographic

weightwhentheyaregeographicallyconcentrated.SuchisthecaseoftheLodhis,present

acrossthestatebutconcentratedinafewpocketsinLowerDoab.ButmostOBCcastes–

theMBCsinparticular–aretoosmall,scatteredandpoortomatteraspoliticalgroups224.

Wecanalreadyseeatthisstagethatcastepoliticsisagameoffewratherthanmany.By

my estimate, about forty per cent of all repertoried castes in U.P. have never been

representedintheAssembly225.ThesearemostlylowerOBCgroupsandsmallSCcastes,

whoselownumbers,povertyandgeographicaldispersionkeepsthemawayfromparties’

attentionorstrategies.

224Partiesoften“adopt”lowerOBCfigureheads–usuallyreferredtoas“posterboys”or“mascots”–whoseroleconsistingivingtheircasteatokenrepresentation.IndividualssuchasBabySinghKushwahaorSwamyPrasadMaurya,whohaveshiftedpartyallegianceseveral timealong theircareers, are twoexamples.There isnoevidence that such “castehints”actuallyhavean impactbeyondthelocalcircleofinscriptionofthesecandidates.225I estimate that figure by running the list of castes represented in the Assembly against theCentralandStatelistsforOBCsandSCsandSTs.Onthatquestion,seeEssaDoron’schapterontheMallahcommunityinUttarPradesh(Doron2014).

184

4.1.4.ThesteadyriseofMuslims’representation

TheothermajorphenomenonthathasmarkedU.P.politicsoverthepasttwentyyearsis

thesteadyriseofMuslims’representation.Accordingtothe2011Census,Muslimsmake

19.26percentofthepopulationofUttarPradesh.InthecurrentAssembly,theyoccupy

17percentoftheseats,anearproportionalrepresentation.In1991,therewereonly23

Muslims in the Assembly (5.5 per cent). Since then, each election has seen their share

increasing.

This trajectory is significant sinceMuslimshavehistorically beenunder-represented in

electedofficeatboththenationalandthestatelevelinIndia(Ansari2006,Jaffrelotand

Kumar2009, Jensenius2013).For instance,Ansari shows thatbetween1952and2004

Muslims held about 4% of the seats in the Indian Parliament after the 1952 election,

whichincreasedtoabout9percentinthe1980electionsandthendeclinedsomewhatto

5to7percentintheelectionsbetweenthemid-1980sand2004226.Inthe2014elections,

thenumberofMuslimsintheLowerHouseofParliamentdecreasedfurthertomerely4

per cent (Jaffrelot and Verniers 2014a). Ansari concludes that Muslims have been

consistently under-represented and that parties' unwillingness to nominate Muslim

candidatesisoneofthemainreasonsforthisunder-representation227.

This trend is also significant since Muslims are also among the most economically

disadvantaged groups in India and generally do not benefit from affirmative action. A

recent government report, set up to investigate the social, economic and educational

status of Muslims in India identified “deficits and deprivation” in practically all

dimensions of development (Sachar 2006, 237), including literacy, access to education,

employmentinthegovernmentsector,andaccesstocreditandloans.Thesituationwas

found to be ‘particularly grave’ in States with large Muslim populations, such as Uttar

Pradesh. Moreover, the report found that in addition to the 'development deficit', the

perception among Muslims that they are discriminated against and excluded is

widespread, which exacerbates the problem. Uttar Pradesh has a long history of

226Ibid.,p.64.227Seealso(Jensenius2013).

185

communal violence between Hindus and Muslims. According to the Sachar Report228,

Muslims still fear for their safety and security and there is an underlying feeling of

injustice towards the compensation to riot victims, with a perception of government

discriminationagainsttheirclaims.

InUttarPradesh, through the1990s, thepost-Mandalpoliticsand theriseof theHindu

right did not leave much political space for the Muslims. Caste-based politics, quota

politicsandtheFPTPelectoralsystemensuredthattheyremainedconfinedtotheroleof

secondbaseforthesecularpartiesfightingthecommunalforces.Buttheneedtoexpand

theirsocialbases ledbothSPandBSPtodistribute largernumbersofticketstoMuslim

candidates229. The Congress usually distributes few tickets to Muslims and the BJP

practicallynone230.

Table4.3Muslimcandidatesnominatedbymainpartiesinthe2012StateElection

BSP SP Congress BJPMuslimcandidates 88(21.9) 83(20.8) 20(6.4) 1(0.3)Muslimselected 15(18.8) 43(19.4) 3(12.0) 0Muslimsrunner-up 38(42.2) 13(16.7) 3(11.5) 0Source:UttarPradeshStateAssemblyLegislators'dataset.Quotedfrom(Heath,Verniers,andKumar2015,13).

The BJP rose in Uttar Pradesh largely by antagonizing the state’s largest minority. The

Congress,havinglostalargepartofthesupportitusedtogetfromMuslims,seesnopoint

indistributingthemmanytickets.TheCongressStatePresident,Mrs.RitaBahuguna,told

meonce that “AllpartiesgiveticketstoMuslimsandtheirwinnabilitythusreduces.Then,

someHinducontestsandwins”231.

DatashowninTableXXrevealshoweverthatmainparties–barringtheBJP–havebeen

consistently distributing tickets to Muslims over time. Even if these numbers have

228Ibid,p.13.229With respectively 21% and 18% of potential vote share, and with an average winning voteshare of 36%, SCs and Muslims offer to the BSP a powerful combination. However, recentethnographicworkonthesubjecthasshownthatthetransferabilityoftheDalitvotebankinfavorofMuslimcandidatescannotbetakenforgranted(Guha2008).230TherewasonlyoneMuslimBJPcandidatefieldedin2012,inSahaswan.Helosthisdeposit.231InterviewheldattheCongressheadquarter,inLucknowonJuly27,2011.

186

increased in recent years (they fielded 251 Muslim candidates in 2012), they do not

account for thevariation in time232. In fact, several factors–demographic, institutional,

socialandpolitical–havebeenlimitingMuslims’representationinthepast.

ThefirstlimitingfactoristheunevengeographicdistributionofMuslimpopulationacross

the state, which confines their political strength to specific sub-regions (Western U.P.,

Rohilkhand, Awadh and a few districts north of Poorvanchal). Without surprise, the

geographical distribution of elected Muslims matches roughly their demographic

distribution233.

ThepresenceofreservedseatsinconstituenciescountingalargeshareofMuslimvoters

constitutesasecondlimitingfactorfortheirrepresentation234. It isaregularcontention

that constituency delimitation, added to the presence of 90 reserved constituencies in

Uttar Pradesh, plays against Muslim representation, as acknowledged in the Sachar

Report235.

232These numbers are below what they used to be throughout the 1990s, when the level ofrepresentationofMuslimswasaroundathirdorlessthanwhatitistoday.233Inafragmentedpolity, themajoritarian,Firstpastthepost,electoralsystemhastheeffectofdispersingtheminoritieswhentheydonotconstituteasolidvoteblock(Verniers,2011).234 In 2007, Muslims represented more than 25% of the total electorate in four reservedconstituencies (Hapur, Koil, Khalilabad and Jansath) and more than 40% in two of them(Najibabad and Nagina). These are however specific cases. In her study on the effects ofdelimitation, Jensenius finds that Muslims are not over-represented in reserved seats andthereforenotdiscriminatedagainstinthatregard(Jensenius2013).235Ibid.,p.25.

187

Tabl

e4.

4RepresentationofMuslim

sintheUttarPradeshLegislativeAssembly,1962-2012

Muslim

s

elected

(%)

6.98

%

6.28

%

8.24

%

9.67

%

11.2

9%

11.7

6%

12.2

4%

9.65

%

5.49

%

7.78

%

9.20

%

11.4

1%

13.9

0%

16.8

7%

Source

:Com

pile

dby

aut

hor.

*P

opul

atio

nda

taco

mes

from

the

Cens

us2

011

for2

012,

from

the

Sach

arC

omm

ittee

Rep

ortf

or2

007

and

adap

ted

from

(Ans

ari2

006)

forp

re-2

007

year

s.T

hese

figu

ress

houl

dbe

trea

ted

asin

dica

tive

and

nota

sact

ualr

epre

sent

atio

nof

the

Mus

limsp

opul

atio

nin

thos

eye

ars.

Indepen

dent

Muslim

s

elected

0 0 6 1 4 3 4 8 2 1 1 2 2 0

Muslim

s

Elected

30

27

35

41

48

50

52

41

23

33

39

46

56

68

Muslim

candidates

(%)

8.24

%

9.19

%

9.82

%

10.2

3%

12.2

8%

10.9

1%

10.4

2%

13.2

1%

11.4

0%

10.9

0%

9.05

%

10.6

8%

11.5

9%

13.2

2%

Total

Candidates

2620

3015

2871

4039

3012

4619

6019

6102

7845

9602

4429

5524

6085

6859

Total

Muslim

candidates

216

277

282

413

370

504

627

806

894

1047

401

590

705

907

Indepen-

dent

Candidates

75

146

108

150

254

255

416

544

604

768

156

184

240

177

Nom

inated

byother

parties

28

29

56

120

21

52

68

37

41

62

117

196

236

479

Nom

inated

bymain

parties

113

102

118

143

95

197

143

225

249

217

128

210

229

251

Muslim

population

(%)*

14.6

3

14.6

3

15.4

8

15.4

8

15.4

8

15.9

3

15.9

3

17.3

3

17.3

3

17.3

3

17.3

3

17.3

3

18.

02*

19.2

6**

Seats

430

430

425

424

425

425

425

425

419

424

424

403

403

403

Year

1962

1967

1969

1974

1977

1980

1985

1989

1991

1993

1996

2002

2007

2012

188

A third factor is related to the split voting in areas where Muslims have a strong

presence. Voters split their votes between several Muslim candidates, opening

thereforetolosingtheseattoaHinducandidate.

Table4.5SeatspossiblylostduetosplitMuslimvote(1996)Constituency %

MuslimsNumberof

relevantMuslimcandidates*

Beneficiary Winner’scaste

Seohara 39.43 2 BJP ThakurAfzalghar 41.60 2 BJP ThakurKanth 37.37 3 BJP OBCAmroha 55.80 2 BJP OBCSuartanda 56.89 2 BJP KayasthBaheri 44.09 3 BJP KurmiPilibhit 31.68 2 BJP SikhJalalpur 13.43 2 BJP ThakurKaimganj 15.95 2 BJP ShakyaGarhmukhteshwar 30.94 2 BJP Jat

*Candidatesgatheringmorethan2%ofvoteshare.Source:ElectionCommissionofIndiaandadaptedfromAnsari(2006)

Ten seats were lost in the 1996 State elections, to the sole benefit of the BJP, the only

mainstreampartythatdoesnotfieldanyMuslimcandidate.In2002,theBJPwonthirteen

seats in the same way, while six other seats were also lost due to the dispersion of the

Muslimvote,inconstituencieswheresometimesMuslimsareinnearmajority.

Table4.6SeatspossiblylostduetosplitMuslimvote(2002)Constituency %

MuslimsNumberof

relevantMuslimcandidates


Afzalghar 41.6 2 BJP ThakurMoradabad 50.88 2 BJP BrahminThakurdwara 45.97 2 BJP ThakurUsehat 31.91 2 SP YadavLaharpur 28.63 2 SP KurmiShahabad 19.32 2 BJP ThakurTiloi 50.78 2 BJP ThakurSultanpur 13.87 2 BJP BrahminTanda 27.85 2 BSP KurmiMasauli 24.82 2 SP KurmiKaiserganj 30.92 2 BJP KurmiNanpara 35.4 2 BJP ThakurSadullanagar 26.36 3 BJP ThakurItwa 37.2 2 SP BrahminShyamDeurwa 14.59 2 BJP ThakurGeneralganj n.a. 2 BJP VaishyaPatiali 17.9 2 BSP ChauhanBulandshahr 23.97 2 BJP YadavMeerut 44.72 3 BJP Brahmin

Source:AdaptedfromElectionCommissionofIndiaandadaptedfromAnsari(2006)

189

Thisphenomenonwasmuchmorecircumscribedin2007,whereonlyfiveseatswerelost

duetothesplittingoftheMuslimvote.Therewerepracticallynoinstancesofsplitvotingin

2012,when68MuslimswereelectedMLAs.

Table4.7SeatspossiblylostduetosplitMuslimvote(2007)

Constituency %Muslims

NumberofrelevantMuslimcandidates


Bijnor 46.99 2 BSP RajputTulsipur 23.01 2 BJP BrahminUtraula n.a. 3 BJP KurmiItwa 37.2 2 SP BrahminSarsawa n.a. 2 BSP Soni

Source:AdaptedfromElectionCommissionofIndiaandadaptedfromAnsari(2006)

This observation points out to the main factor explaining the recent rise of Muslims’

representation,whichisthatMuslimsvotefarmorecohesivelyattheconstituencylevel.

We should not deduct that Muslims have suddently started to vote en bloc. In fact, it is

quitethecontrary.Muslimsdodispersetheirvotesacrossparties,asshowninTable4.8.

EveniftheMuslims’supportfortheSPremainsimportant,ithaserodedinrecentelections,

asshownherebelowintheLokniti/CSDSdata.TheBSP,whohasalsobeenfieldingmore

andmoreMuslimcandidatesrecently,received30.4percentoftheMuslims’votesin2010.

Table4.8Muslimvoters'partypreferencesinfourstateelections

1996 2002 2007 2012

Congress 12.20% 10.00% 14.10% 18.00%SP 48.00% 53.00% 47.70% 39.40%

BSP 12.30% 9.70% 17.60% 30.40%BJP 1.90% 1.70% 2.40% 6.60%Source:CSDS/LoknitiNESData.

Further, research conducted soon after the 2012 state elections showed that a large

numberofMuslimcandidateswereelectedwithvotesharebeyondthedemographicshare

of their co-religious electors (Heath, Verniers, and Kumar 2015). This contradicted the

190

widespreadnotionthatfieldingMuslimcandidatesleadstoaHindubacklash,resultingin

lostseats236.

Finally, to further discard the notion that Muslims in Uttar Pradesh are a homogeneous

grouporactpoliticallyassuch,wemustconsidertheirinternaldivisionsandstratifications.

There are sectarian differences among Muslims, notably between Sunnis and Shias, the

latter beingparticularly under-represented in politics (Gayer and Jaffrelot 2012).Within

theSunnimajority,thedivisionbetweentheBarelviandDeobandischoolsofthoughtisalso

important237.

ButthecrucialdivisionamongMuslimsinU.P.,asintheIndianpolityingeneral,remains

caste. With the conversions to Islam of Hindus from different backgrounds, the Indian

Muslimcommunityhasgenerateditsowncastesystem238.Arapidexaminationatthecaste

compositionofMuslimMLAsshowshowtherepresentationofMuslims isclearlybiaised

towardstheuppercastes.

236This research was conducted on the basis of survey data collected before the 2014 Generalelections, in which the BJP swept the state of Uttar Pradesh, and before the August-September2013Muzaffarnagarriots.ThereligiouspolarizationthathastakenplaceinUttarPradesharoundthesetwoeventsislikelytoalterourfindings.237The former, numerically dominant, defends the popular and traditional practices of IndianIslam (including thevisitingofmazars anddargahs) andhashistoricallyopposed the reformiststance of the later. Bothmovements emerged inUttar Pradesh,withAhmadRazaKhan (1856-1921), thefounderoftheBarelviSchool,hailingfromBareillyandtheDeobandi thoughtcomingfromthefamousDarulUloomseminaryinDeoband.238At the top of the hierarchy stand the Ashrâfs. They are the alleged descendants of Muslimmigrants:amongthem,theSayyidscomefromthelineageofProphetMuhammad;theSheikhsarethedescendantsofotherArabgroups;theMughalshaveTurkishancestorsandthePathansclaimanAfghanlineofdescent.Hinduconvertsfromhigh-castegroups(suchastheKshatriyas)alsofallin thisuppercategory.Thencome theAjlâfs,Hinduconverts frombackwardcastes, suchas theSaifis(blacksmiths),theQureshis(butchers)ortheMalis(cultivators).Atthebottomoftheladderare theAzrâls, issued fromDalitfamilieswho converted to Islam.Theyareoften referred to asDalitMuslimsorPasmandaMuslims.

192

and that caste biases have in fact perdured among the national parties, who claim to

represent all sections241. This exercise helps to discard some common misconceptions

aboutpartyandcastealignments.

Firstandforemost,andcontrarytopopularperception,therepresentationofOBCsisnot

theonlypreserveoftheSPandtheBJP.TheyhaveactuallyformedthebulkoftheBSP’s

MLAsformorethan20years,beingoutweighedbytheuppercastesonlyin2007.Sixty-six

yearsaftertheimplementationofpoliticalquotasinfavorofScheduledCastes,itremains

difficult (but not impossible) for a Dalit candidate to be elected in a general (non-

reserved) seat242 , creating the need for the BSP to distribute tickets to candidates

belongingtolocallystrongcastes,hencetoOBCcandidatesinmanyinstances.

Second,thedistributionofSCMLAshasshiftedintime.In2012,theSPwon58reserved

seats in 2012 (including 11 Jatavs). The share of BSP Dalit MLAs in Uttar Pradesh is

usuallyquitelow,asseeninTable4.9243.TheBSPgetsusuallyathirdofitsseatsamong

thereservedconstituencies,exceptin1991,whereitwonnone,and2007,whereitwon

sixty-oneseatsoutofeighty-six.

Table4.9.CasterepresentationamongamongBSPDalitMLAs(1989-2012)

1989 1991 1993 1996 2002 2007 2012

GEN 8 12 45 47 74 145 65

SC 5 23 20 24 61 15

Total 13 12 68 67 98 206 80Source:AdaptedfromECIdata

Third,both theCongressand theBJPremainbiased towards theuppercastes.Over the

last fiveelections,amajorityof thecandidateswhowononCongress ticketswere from

241Partsofthissectionhaveappearedin(JaffrelotandVerniers2012).242ThereasonbeingthatSCcandidatesstillfacediscriminations,firstandforemostfrompoliticalpartiesthemselves,whoarereluctanttogivethemticketsoutsidereservedseats.Thatbeingsaid,81 SC candidateshavebeen elected in general seats since1962: 17onCongress tickets (in the1960sand1980s),12onBJP tickets (in the1990s) andaboutoneor twoperyear for theBSPsince1993(includingMayawatiherself,whousuallycontestsintwoseatsatatime).243The election of a Dalit MLA depends in most cases from the support provided by non-Dalitvoters, which can play in disfavor of an overtly pro-Dalit party. This observation validatesAmbedkar’s critiqueof the systemof political reservations. In a text titled «WhatCongressandGandhi have done to the Untouchables», Ambedkar argued that the outcome of the election inreservedseatswouldremaindeterminedbythevoteofthecasteHindus.

193

the upper castes. The induction of several prominent OBC figures, such as former SP

MinisterBeniPrasadVerma,didnothelpittomakeadentamongOBCvoters.Similarly,

theBJPisalsobenttowardstheuppercastes,includingunderthestewardshipofitsOBC

figurehead,KalyanSingh.Hereagain,therecoursetocastemascots,suchasKalyanSingh,

UmaBharti(bothLodhis)andBabuSinghKushwahafailed.In2012,theBJPfaredbadly

amongtheOBCs,includinginareasdominatedbythesetwocastegroups.

TheSP,withafairlybalancedshareofrepresentationofthedifferentcastegroups,isthe

new catch-all party.Upper castes have a significant presence among the SPMLAs since

2002.DespiteadecreaseinsupportfromtheYadavs,themajorityofitsOBCMLAsarestill

drawnfromthisgroup(34outof58MLAsin2012).

Casterepresentationamongcandidates

The“rainbowcoalitions”ofthe2000shadgeneratedhopesthatsocialdivisionswouldbe

assuaged and that parties would be incited to devise inclusive strategies and design

policies that benefit people on the basis of needs rather than on the base of their

ascriptiveidentity.

Anexaminationofthecasteprofileofparties’candidatestellsadifferentstory.Ifwelook

at the distribution of tickets across caste groups, the data seems to validate the

impression that parties have indeed become inclusive, even if some preferential

distributionremains.In2012,boththeCongressandtheBJPdistributedahighershareof

ticketstouppercastescandidates(34.4percentand47percentrespectively),whilethe

BSPandtheSPgavethemnearlyathirdoftheirtickets(28.8percentand29.3percent

respectively).

194

Tabl

e4.

10Representationofmajorcastegroupswithinmainparties(1989-2012)

2012

59.5

7%

8.51

%

23.4

0%

6.38

%

2.13

%

100.

00%

N=47

2012

53.5

7%

3.57

%

10.7

1%

14.2

9%

14.2

9%

3.57

%

100.

00%

N=28

Source:

Aut

hor’s

fiel

dwor

k.

2007

58.8

2%

3.92

%

23.5

3%

13.7

3%

0.00

%

100.

00%

N=51

2007

63.6

4%

4.55

%

9.09

%

22.7

3%

0.00

%

0.00

%

100.

00%

N=22

2002

62.5

0%

1.14

%

17.0

5%

19.3

2%

0.00

%

100.

00%

N=88

2002

60.0

0%

4.00

%

12.0

0%

8.00

%

12.0

0%

4.00

%

100.

00%

N=25

1996

50.0

0%

2.30

%

22.9

9%

21.8

4%

2.87

%

100.

00%

N=17

4

1996

63.6

4%

6.06

%

12.1

2%

9.09

%

9.09

%

0.00

%

100.

00%

N=33

1993

53.6

7%

3.39

%

20.9

0%

20.3

4%

0.56

%

1.13

%

100.

00%

N=17

7

1993

53.5

7%

3.57

%

14.2

9%

21.4

3%

3.57

%

3.57

%

100.

00%

N=28

1991

52.0

4%

1.81

%

19.4

6%

25.3

4%

1.36

%

100.

00%

N=22

1

1991

52.1

7%

0.00

%

17.3

9%

17.3

9%

8.70

%

4.35

%

100.

00%

N=46

1989

56.1

4%

0.00

%

19.3

0%

22.8

1%

1.75

%

100.

00%

N=57

1989

44.6

8%

1.06

%

18.0

9%

20.2

1%

11.7

0%

4.26

%

100.

00%

N=94

BJP

UC

IC

OBC

SC

Mus

lims

Othe

rs

Tota

l

INC

UC

IC

OBC

SC

Mus

lims

Othe

rs

Tota

l

2012

21.2

5%

1.25

%

36.2

5%

20.0

0%

18.7

5%

2.50

%

100.

00%

N=80

2012

26.7

9%

0.00

%

27.2

3%

25.8

9%

19.2

0%

0.89

%

100.

00%

N=47

2007

27.6

7%

2.43

%

24.2

7%

30.5

8%

14.0

8%

0.97

%

100.

00%

N=20

6

2007

28.8

7%

1.03

%

36.0

8%

12.3

7%

21.6

5%

0.00

%

100.

00%

N=51

2002

24.4

9%

1.02

%

33.6

7%

26.5

3%

14.2

9%

100.

00%

N=98

2002

25.8

7%

0.70

%

32.1

7%

25.1

7%

16.0

8%

0.00

%

100.

00%

N=88

1996

14.9

3%

1.49

%

37.3

1%

29.8

5%

16.4

2%

100.

00%

N=67

1996

22.7

3%

0.91

%

36.3

6%

17.2

7%

20.0

0%

2.73

%

100.

00%

N=17

4

1993

4.41

%

42.6

5%

35.2

9%

17.6

5%

100.

00%

N=68

1993

17.4

3%

0.00

%

49.5

4%

21.1

0%

11.0

1%

0.92

%

100.

00%

N=17

7

1991

8.33

%

83.3

3%

8.33

%

100.

00%

N=12

1991

22.8

3%

9.78

%

31.5

2%

21.7

4%

13.0

4%

1.09

%

100.

00%

N=92

1989

23.0

8%

38.4

6%

38.4

6%

100.

00%

N=13

1989

35.5

8%

6.73

%

26.9

2%

24.0

4%

6.25

%

0.48

%

100.

00%

N=20

8

BSP

UC

IC

OBC

SC

Mus

lims

Othe

rs

Tota

l

SP

UC

IC

OBC

SC

Mus

lims

Othe

rs

195

AllpartiesdistributedaboutaquarteroftheirtickettoOBCcandidatesanddivided

therestbetweenSCandMuslimcandidates(barringtheBJPforthelattercategory).

Table4.10showsthereturnafterthepollsandconfirmsthattheCongressandthe

BJPhavegreaterdifficultiestogettheirnon-uppercastecandidateselected.Onthe

overall, the caste profile of BSP and SP candidates is converging, each caste

categoryorcommunityreceivingbetween20to30percentoftheticketsaswellas

representation.

But does this inclusiveness mean that parties no longer rely on caste in the

elaboration of their strategy? Is the party appeal or party leaders’ appeal strong

enough that the caste identity of the candidates no longer matters? If we break

down these caste categories, we see that some alignments between certain jatis

andpartiespersist.TherearemoreYadavs,Rajputs,PasisandUppercasteMuslims

candidates on the SP side, while there are more Brahmins, Jatavs, Kurmis and

Lodhs on the BSP side. The BJP favors non-Yadav candidates amongst its OBC

candidates(KurmisandLodhisinparticular),whiletheSPdistributesnearlyhalfof

itsOBCticketstoYadavcandidates(53outof108).

These caste-party alignments indicate that caste remains important to parties’

strategies.Thevariations in time indicate that thesealignmentsaren’t stableand

thatpartiesadjusttheirdistributionofticketsaccordingtocircumstances.

Whatisimportanttoretainhereisthatthedistributionofticketsisprimarilymade

according to local circumstances, and not from any pre-defined caste

representation balance. Parties seek to maximize their chances to win seats by

givingticketstocandidateswhocanbringenoughvotesfromtheirowncommunity

(andbeyond),inadditiontothesupportoftheircoresupportbase,whentheyhave

one.Thelocalizationofelectoralstrategiesenablespartiestodevelopageneralist

discoursewhile letting thecandidatesdo thecasteappeal locally.Partiesneedto

displaysuchadiscourse inordertoattract thesupport fromfloatingvoters,who

tend not to respond to caste appeal. Parties who give preferential treatment to

specific castes or who seek to attract the vote from other groups through caste

mascotstendtoloseelections.

196

Table4.11Castesandcommunitiesamongmainpartiescandidatesinthe2012U.P.Stateelections

INC BJP BSP SP

Candidates Elected Candidates Elected Candidates Elected Candidates Elected

Uppercastes 34.37% 51.72% 46.98% 59.57% 28.78% 21.25% 29.18% 26.79%Bania/Jain 2.25% - 4.52% 10.64% 1.99% 1.25% 2.74% 2.23%Bhumihar 1.13% 3.45% 1.51% 2.13% 1.24% - 1.25% 0.45%Brahmin 13.24% 13.79% 19.60% 21.28% 17.37% 12.50% 11.47% 9.82%Kayasth 1.13% 6.90% 0.75% 2.13% 0.50% - 0.25% 0.45%Khatri 0.85% 6.90% 1.01% 4.26% 0.25% - 0.50% 0.45%Rajput 15.49% 20.69% 19.10% 17.02% 7.44% 7.50% 12.22% 12.95%Sindhi - - - - - - 0.25% -Tyagi 0.28% - 0.25% - - - 0.25% -Vaishya - - - - - - 0.25% 0.45%Unidentified - 0.25% 2.13% - - - -Intermediarycastes 1.41% 3.45% 4.02% 2.13% 1.74% 5.00% 1.25% 0.45%Jat 1.41% 3.45% 4.02% 2.13% 1.74% 5.00% 1.25% 0.45%OtherBackwardClasses 22.25% 10.34% 25.38% 23.40% 23.33% 36.25% 26.93% 27.23%Gujjar 0.85% 3.45% 1.51% 4.26% 1.24% 3.75% 1.25% -Kurmi 2.82% - 6.03% 2.13% 4.71% 7.50% 3.99% 3.13%Kushwaha 0.85% - 1.26% 2.13% 1.99% 5.00% 0.75% 0.45%Lodhi 2.54% 3.45% 5.03% 4.26% 4.22% 5.00% 2.74% 3.13%Maurya 0.85% - 1.01% 2.13% 1.99% 2.50% - -Nishad 1.41% - 0.50% - 1.74% 2.50% 1.00% 0.89%Jaiswal 1.41% 3.45% 1.26% 2.13% 0.50% - 0.25% 0.45%Shakya 0.56% - 0.75% - 1.49% 3.75% 0.50% 0.89%Yadav 7.89% - 4.02% 4.26% 1.99% 1.25% 13.72% 16.52%Baghel - - - - 0.25% - 0.50% -Banjara 0.28% - - - - - - -Bishnoi - - 0.25% 2.13% - - - -Chauhan - - - - 0.25% - - -Chaurasia - - - -

- 0.25% 0.45%

Gadariya - - - - 0.50% 2.50% 0.25% 0.45%Garedia 0.56% - 1.01% - - - - -Gosain 0.28% - - - - - - -Goswami - - 0.25% - - - - -Kahar - - 0.25% - - - - -Kashyap - - 0.25% - 0.25% - 0.25% -Khadagvanshis 0.28% - 0.25% - 0.25% - - -Kumhar 0.28% - - - - - 0.50% -Mali 0.56% - 0.50% - 0.25% - - -Mallah - - - - - - 0.25% 0.45%Rajbhar 0.28% - 0.50% - 1.24% - - -Saini 0.28% - 0.75% - 0.25% 1.25% 0.50% -Teli - - - - - - 0.25% -Unidentified 0.28% - - - 0.25% 1.25% - -ScheduledCastes 24.23% 13.79% 21.36% 6.38% 21.84% 20.00% 20.70% 25.89%Jatav 10.99% 6.90% 7.29% 2.13% 14.39% 17.50% 4.74% 5.36%Pasi 5.63%

4.27% 2.13% 2.98% 2.50% 5.74% 8.93%

Dhobi 1.41% 3.45% 1.01% - 0.25% - 1.50% 0.89%Khatik 0.85% - 1.51% - - - 1.25% 1.34%Kori 2.25% - 1.76% - 0.50% - 1.75% 3.13%Valmiki 1.13% 3.45% 2.01% 2.13% 0.50% - 1.25% 1.79%Others 1.13% - 1.76% - 0.99% - 2.49% 3.13%Unidentified 0.85% - 1.76% - 2.23% - 1.75% 0.89%Muslim 16.34% 13.79% 0.25% - 21.34% 18.75% 20.95% 19.20%Christian 0.28% - - - - - - -Unidentified 1.13% 3.45% 2.01% 2.13% 2.98% 2.50% 1.00% 0.89%Total 100% 100% 100% 100% 100% 100% 100% 100%

N=355 N=29 N=398 N=47 N=403 N=80 N=401 N=224

Source:adaptedfrom(Zerinini2009)andauthor’sfieldwork.

197

Eventhemuch-advertisedDalit-Brahminallianceof2007wasinawaymisleading,since

theBSPdistributeda largernumberof ticketstoOBCcandidates.The logic followedby

the BSP was to distribute tickets according to local caste configurations, without pre-

conceivednotionsonwhoshouldgethowmanytickets.The inclusionofBrahmins into

theparty’spublicdiscoursewasawaytoillustratethatitworksforall,inaformulation

reminiscentoftheCongress’‘coalitionofextremes’.Theinclusivecharacterofpartiesisa

by-productofthelocalizationoftheirelectoralstrategies,morethantheconsequenceof

anideologicalshifttowardsinclusiveness.

Thisevolutionbearstwoimportantlessons.Thefirstoneisthatstatewidecasteappeals

are less likely to reap political dividends today than twenty years ago. The tropes of

backward caste mobilization of the 1990s – quota politics – have drained out their

efficacy.Votersarenolongermobilizedbythethemeofreservations.Twenty-fiveyears

after the implementation of the Mandal report, the benefits of reservations have not

trickleddownvery lowamong thebackwards, farmorenumerous than thenumber of

public jobs anduniversity seatsmade available (therehasbeen also a good amount of

elitecaptureamongthebackwardsofthosebenefits).

Thesecondlessonisthatthisevolutiondoesnotmeanthatcasteappealhasdisappeared

altogether. It remains in factat theheartofparties’strategiesandremainsanessential

vehicle for political mobilization but locally. First because local demographic still

determine to a large part a candidate’s chance at winning, and second, more

fundamentally,becauselocalnetworksofpowerandinfluenceremainlargelyorganized

aroundlocalsocialstructurelargelybasedonascriptiveidentities.Wewillseeinthelast

two chapters of this dissertation that caste is deeply enmeshed with local economic

networksandthatlocalpowerandinfluenceremainslargelyexertedthroughcaste.

4.1.6.Casterepresentationincabinets:persistenceofbiases

Thenextquestion is to seewhether the inclusive characterofpartieshas led to actual

power sharing between groups. After all, parties could very well provide token

representation to various caste groups and yet retain thepositions of influence among

theircoresupportgroup.Onewaytolookatthisquestionistoexaminethecomposition

198

oftheCabinetovertime,toseewhetherchangesintheVidhanSabhaarereflectedinthe

Cabinet’scomposition,andwhetherweobservepartywisevariations.

Traditionally, theupper casteshavehad the lion shareofCabinetportfolios, consonant

with their domination of the parties in power and of their over representation in the

Assembly (Jaffrelot 2003b). In the first Cabinets of G.B. Pant and Dr. Sampurnanand,

Brahmins hold above 30 per cent of the portfolios. Some numerically marginal upper

castessuchasKayasthasorBaniaswerealsowell-represented.Zerininipointsatthefact

thateachChiefMinisteroftheearlydaystendedtopromoteitsowncaste.Theshareof

Bania ministers rising from 12.5 per cent to 29.4 per cent in 1960, when C.B. Gupta –

himselfaBania–ledthegovernment244.

Overtheperiod,Muslimsaregivenamorethanproportionalrepresentation,whileabout

12percentof theportfoliosareallotted toSCMinisters.BackwardMinistersarequasi

absent from these Congress governments. Besides Charan Singh, the first backward

Ministers inducted asDeputyMinisterswereRamSwaroopVerma and Jai RamVerma

(bothKurmis), in the1957Sampurnanandcabinet. In the followingC.B.GuptaCabinet,

onlyRamSwaroopVermaremainedaMinister.TwootherBackwardMinistersjoinedthe

1963 C.B. Gupta Government but were not maintained in the Kripalani Cabinet. There

were no backward Ministers in the 1967 C.B. Gupta Government (Mathur 2004).

BackwardrepresentationstartedwiththefirstCharanSinghGovernment,althougheven

thefirstnon-Congressgovernmentsremainedbiaisedtowardstheuppercastes.

The return to power Congress in the 1980s also meant a return to old practices. The

averageshareofportfoliosallottedtobackwardMinistersduringthatdecadewasoften

per cent, while the upper castes trusted nearly sixty per cent of the berths. The

representationofSCsandMuslimsremainproportionaltotheirdemography.Thisbiasis

reflective of the fact that party apparels were dominated by upper castes, including

withinthesocialistparties.

Therupturewouldcomeintheearly1990s,withtheinstallationofthefirstSPandBSP

governments.Themajority of portfolioswereheldbyupper castes in the1991Kalyan

Singh’sgovernment.Theirsharedroppedto6.7percentand6.25percentinthenexttwo

244Op.Cit.,p.55.

199

cabinets, formed by the SP-BSP alliance. Nearly three quarter of the portfolios were

dividedamongtheSCsandtheOBCs,Muslimsgettingnorepresentationatall245.

The secondhalf of the1990swas aphaseof ascensionof theBJP. It became theBSP’s

coalitionpartnerin1997andruledthestateonitsownfrom1999to2002.Consequently,

theshareofuppercasteministersroseagain,tonearlyhalfoftheCabinet.Post-2002,the

strategiesoftheSPandoftheBSPcreatedaspaceforuppercasterepresentation,which

remainedover-representedintheCabinet,comparedtotheirshareofseatsintheVidhan

Sabha.

One needs however to make the distinction between the types of portfolios allotted.

TherearebasicallythreecategoriesofMinisters:CabinetMinisters(CM),MinisterofState

andMinisterofState(MoS)withIndependentCharge(MoS-IC). Inthe2002SPCabinet,

Upper castes make for a third of the portfolios, but sixty per cent of the Cabinet

Ministerships. They are also clearly over-represented among the MoS (IC), positions,

consideredmoreprestigiousthanMoSsincetheyarenotplacedundertheauthorityofa

CabinetMinister.However,OBCsanduppercasteshare thebetterportfolios.TheOBCs

and SCs get about a quarter of the portfolios, while the Muslims get a proportional

representation,includingintheCMcategory.

245op.cit.,p.59.

200

Tabl

e4.

12CastesandcommunitiesinU.P.Cabinets

2012(SP)

Total

33.30%

8.90

%

22.2

0%

2.20

%

- -

31.10%

15.6

0%

4.40

%

4.40

%

2.20

%

-

13.30%

4.40

%

2.20

%

2.20

%

4.40

%

-

22.20%

17.8

0%

4.40

%

-

100.

00%

N=45

Sour

ce:A

utho

r'sfi

eldw

ork.

MoS

40.00%

-

40.0

0%

- - -

20.00%

-

20.0

0%

- - -

20.00%

- -

20.0

0%

- -

20.00%

-

20.0

0%

-

100.

00%

N=5

MoS(IC)

21.70%

4.30

%

13.0

0%

4.30

%

- -

34.80%

8.70

%

4.30

%

8.70

%

4.30

%

4.30

%

17.40%

8.70

%

4.30

%

-

4.30

%

-

26.10%

21.7

0%

4.30

%

-

100.

00%

N=2

3

CM

47.10%

17.6

0%

29.4

0%

- - -

29.40%

29.4

0%

- - - -

5.90%

- - -

5.90

%

-

17.60%

17.6

0%

- -

100.

00%

N=1

7

2007(BSP)

Total

34.40%

7.50

%

10.8

0%

3.20

%

2.20

%

2.20

%

29.00%

12.9

0%

6.50

%

-

3.20

%

6.50

%

20.40%

7.50

%

2.20

%

1.10

%

5.40

%

4.30

%

14.00%

9.70

%

4.30

%

2.20

%

100.

00%

N=9

3

MoS

20.00%

2.00

%

8.00

%

6.00

%

2.00

%

2.00

%

34.00%

18.0

0%

4.00

%

-

4.00

%

8.00

%

30.00%

14.0

0%

4.00

%

-

6.00

%

6.00

%

12.00%

8.00

%

4.00

%

4.00

%

100.

00%

N=5

0

MoS(IC)

25.00%

8.30

%

16.7

0%

- - -

8.30%

8.30

%

- - - -

16.70%

- -

8.30

%

-

8.30

%

50.00%

33.3

0%

16.7

0%

-

100.

00%

N=1

2

CM

51.90%

33.3

0%

11.1

0%

-

3.70

%

3.70

%

37.00%

7.40

%

14.8

0%

-

3.70

%

11.1

0%

7.40%

- - -

7.40

%

-

3.70%

3.70

%

- -

100.

00%

N=2

7

2002(SP)

Total

33.30%

7.80

%

11.1

0%

2.20

%

2.20

%

2.20

%

26.70%

13.3

0%

- -

5.60

%

7.80

%

23.30%

7.80

%

2.20

%

1.10

%

5.60

%

6.70

%

16.70%

6.70

%

5.60

%

-

100.

00%

N=9

0

MoS

18.90%

3.80

%

7.50

%

3.80

%

1.90

%

1.90

%

32.10%

18.9

0%

- -

5.70

%

9.40

%

32.10%

13.2

0%

3.80

%

-

5.70

%

9.40

%

17.00%

5.70

%

7.50

%

-

100.

00%

N=5

3

MoS(IC)

60.00%

20.0

0%

40.0

0%

- - -

10.00%

10.0

0%

- - - -

20.00%

- -

10.0

0%

-

10.0

0%

10.00%

-

10.0

0%

-

100.

00%

N=1

0

CM

51.90%

18.5

0%

22.2

0%

-

7.40

%

3.70

%

22.20%

7.40

%

- -

7.40

%

7.40

%

7.40%

- - -

7.40

%

-

18.50%

18.5

0%

- -

100.

00%

N=2

7

Uppercastes

Brah

min

Rajp

ut

Bani

a

Bhum

ihar

Othe

rUC

OBC

Yada

v

Kurm

i

Lodh

Nis

had

Othe

rOBC

SC

Jata

v

Katik

Kori

Pasi

Othe

rSC

Muslim

s

UCM

uslim

s

BCM

uslim

s

Unid

entif

ied

Total

201

The2007BSPGovernment follows the samepattern.Upper castesareover-represented,

particularlyamongtheCMpositions.Interestingly,theshareofportfoliosallottedtoOBCs

increasesslightly,to29percent,andthepercentageofportfoliosallottedtoSCministers

decreases,from23.3to20.4percent.ThemaindifferencewithregardtoSCrepresentation

inSPandBSPcabinetsisthefactthatSCsgetbetterportfoliosundertheBSPrule,butnota

largeroverallnumberofberths.

TheSCrepresentationdropsto13.3percentinthe2012SPgovernment,despitethestrong

performanceof theparty in reserved seats.Here again, theupper casteshave theupper

hand,evenifOBCministersgetcomparativelystrongerportfolios.

Looking at jatis, we observe the same alignment that we saw in the Vidhan Sabha and

amongthecandidates.TherearemoreRajputandYadavMinistersinSPCabinetandmore

Brahminandnon-YadavOBCMinisters in theBSP candidates.Among theDalits, there is

not much differentiation between the two parties. Both distribute portfolios across the

main SC castes and both of them exclude Jatavs from major posts (there are no Jatav

CabinetMinisters,withthenotableexceptionofMayawati).

AlookatthedenominationoftheportfoliosalsorevealsacommontrendinIndia,whichis

that the upper castes and dominant OBCs trust the major remunerative positions

(industrial development, natural resources, transport, sugar and cane industry, public

works,etc.)whilethe“subalternministers”areconfinedto“subalternportfolios”,suchas

SC/ST welfare, or minority welfare. This inequality in the distribution of portfolios is

significantsinceitdoesnotonlyreflectadifferentiationofstatus.Remunerativeportfolios

are those who provide access to vast resources – collection, subsidies, contracts and

tenders – which can be used or rather misused to nurture patronage and clientelistic

networks,andbuildelectoralsupport.

ConcentrationofpowerwithinCabinets

Finally,ifthecastecompositionofcabinetsisareflectionoftherepresentationequilibrium

soughtbythepartyinpower,itshouldnotbenecessarilyinterpretedasaindicatorofcaste

202

empowerment.The realityofpower sharingwithin cabinets is thatpower is in factvery

muchconcentratedinthehandsoftheChiefMinisterandahandfulofimportantministers.

Thus, Chief Ministers retain most of the major portfolios (and a large number of less

significantones).Inthecurrentadministration,AkhileshSinghYadavretainsnearly38per

centoftheportfolios,includingmajorcompetencessuchasHome,Finance,Energy,Excise,

Education,SugarandCaneDevelopment,and IndustrialDevelopment (TheHindu,2012).

Inthepreviousgovernment,Mayawatiheld27percentoftheportfolios,includingHome,

Finance,Justice,Appointments,StateRevenue,IndustrialDevelopment,andsoon.

Alookatpastdatashowsthatthisisnonewphenomenon.Infact,CongressChiefMinisters

inthe1980susedtoconcentratepowermorethantheirsuccessorshave.Inthe1980s,N.D.

TiwariandVirBahadurSinghusedtoretainupto60percentoftheportfolios.

Table4.13PortfolioconcentrationinU.P.Cabinets,1980-2012*

Year ChiefMinister PartyNoofPortfolioheldbyCM TotalPortfolios CabinetSize

%heldbyCM

1980-1982 V.P.Singh Congress 56 93 15 60.2%1982-1984 SripatiMishra Congress 29 98 26 29.6%1984-1985 NarayanDuttTiwari Congress 74 226 48 32.7%1985-1985 NarayanDuttTiwari Congress 44 106 36 41.5%1985-1988 VirBahadurSingh Congress 56 92 17 60.9%1988-1989 NarayanDuttTiwari Congress 54 89 13 60.7%1989-1991 MulayamSinghYadav JD 44 89 13 49.4%1991-1993 KalyanSingh BJP 50 144 47 34.7%1993-1995 MulayamSinghYadav SP 65 121 27 53.7%1995-1995 Mayawati BSP 63 146 33 43.2%1996-1997 Mayawati BSP 69 222 46 31.1%1997-1999 KalyanSingh BJP 19 113 43 16.8%1999-2000 RamPrakashGupta BJP 20 107 50 18.7%2000-2002 RajnathSingh BJP 19 169 86 11.2%2002-2003 Mayawati BSP 45 117 24 38.5%2003-2007 MulayamSinghYadav SP 33 200 97 16.5%2007-2012 Mayawati BSP 40 147 57 27.2%2012- AkhileshSinghYadav SP 50 133 46 37.6%

Source:Compiledbytheauthor,UttarPradeshGazette.*Measuretakenatthefirstgovernmentformation.

TheincreaseofCabinetsizeandofthetotalnumberofportfoliosinlateryearsaccountfor

the lower per centages. If we add the fact, not represented here, that the government’s

203

numbertwoalsotendtoconcentrateanumberofmajorportfolios246,wecanascertainthat

Ministerstendnottomattermorethanthedescriptiverepresentationtheyprovideandthe

numberofsupportersthattheybringtothepartyinpower.

WealsoseethattheCabinetsizecanvarygreatly,from13membersintheN.D.Tiwariand

Mulayam Singh Yadav governments between 1988 and 1991 to 97 in the second SP

government.Theinflationofportfolioscancomefromtheneedtoaccommodatecoalition

partners, maintain regional and factional balance, induct new faces in the party, give

prominencetosomecasteleadersaheadofelectionsorsimplylureadversariestodefect.

The inflation of cabinet size also meant an increase of meaningless competences, such

Minister of State for Awards and Trophies. These are only meant to grant the status of

Minister, with the perks and advantages, material and symbolic, that come with it247.

Similarobservationscanbemadeaboutthenumberofdepartments,whichvariesfrom89

to 226, signaling the pressures put to the bureaucracy who must adapt rapidly to those

changesoforganizationandnomenclature.

What we see here however is the persistence of biases that indicate that upper caste

politicians have resisted rather well to the rise of so-called backward parties. An

examinationofsub-regionalvariationsinthecastecompositionoftheStateAssemblywill

providefurtherconfirmationofthatfact.

4.2.Sub-regionalvariations

WhatIhavedonesofaristolookatvariationsintime,onthebasisofcaste,jatiandparty

lines.ButtheStateofUttarPradeshisavastterritorycontainingapopulationgreaterthan

Brazil’s. We must therefore pay heed to spatial variations and see how these variations

affectthenarrativesbuiltonthebasisofaggregatedata.

246In 2007, Nasimuddin Siddiqui held eight “remunerative” portfolios, including Public Works,Irrigation, Sugar andCaneDevelopment, Excise, andUrbanLand. In thepreceding government,thebrotherofMulayamSinghYadav,alsoheldeightportfolios.247In2004,aSupremeCourtorderedfixedthelimittoCabinetsizeto15percentoftheAssemblyseats.MulayamSinghYadav’s jumbocabinetwasmadetoresigninitsquasientiretyinordertocomplytothenewrule.

204

What emerges from that exercise is that caste groups have been set on very diverse

trajectoriesinvariouspartsofthestate,theuppercastesnotablyresistingdeclineinlarge

populated sub-regions.And in the sub-regionswhere theupper casteshave indeedbeen

challenged,theyhaven’tnecessarilybeensupplantedbyOBCs.

Theevolutionofparties’electoralstrategiesafter1996hasalteredpre-existingcaste-party

alignments,pavingthewaynotablytoaresurgenceofuppercastesrepresentationinthe

Assembly.

Sub-regional trajectories can be sorted into three categories. The first one includes sub-

regions where the upper castes and the OBCs have indeed fallen and risen. The second

trajectory includes sub-regions where the upper caste domination has remained

unchallenged over time. The third trajectory includes a particular case – Rohilkhand -

wheretheuppercasteshavebeensupplantedintwoperiodsbyMuslimMLAs.

4.2.1.WheretheSavarnashavefallen

Thefirsttrajectoryincludesfoursub-regionswheretheuppercastes’dominationhasbeen

successfullychallengedbytheOBCS:Doab,theEast,BundelkhandandRohilkhand.

InDoab,theshareofuppercastesMLAswasforalongperiodstable,around40percent,

until 1985when theygraduallydeclined.Theyhit their lowest representation, at25per

cent, in 1996. The OBCs are on a reverse mirror trajectory over the same period, and

peakedat42.3percentoftheseatsin1993.Theyhavegraduallydeclinedsince,duetothe

riseofMuslims’representation,buthaveremainedaheadoftheuppercastes.

DoabcontainsanumberofconstituenciesthathavebeenSPandBSPstrongholds.Etawah,

JaswantNagar,KannaujorAligarhareimportantYadavstrongholds,controlledbytheSP.

Overthelastfiveelections,theBSPhashadastableholdover13seatsinthatarea,though

notnecessarilyheldbythesameindividualsovertime.

212

It isworthnoting thatRalphC.Meyer,who conducted the first study of the sociological

profileofU.P.MLAs inthe1960s,notedthatwhathecalledthe“NorthwesternersMLAs”

wereamongthemostdistinctgroupinthestate(Meyer1969).Hedescribesthemasbeing

“highly non-agricultural in occupation as well as in heritage (…) half of them (…) in

professions252”. He adds, “no other plain regions had as high a proportion of lawyers,

educatorsanddoctors”253. TheseMLAswere alsomorewealthy than others – due to the

relative prosperity of the region, and counted more Muslims and few upper castes. The

mostrepresentedHinducastesweretheBanias,Khatris,Jats,TyagisandChamars254.

With such variations, it does not make much sense to speak any more about overall or

aggregate trends. The various groups in competition rely on the grip they have over

particularconstituencies,duetotheirdemographicstrengthandtothesocial,politicaland

economiccontroltheyexertovertheseterritories.

4.2.4.Explainingvariations

What accounts for these variations? I have already hinted at two possible types of

explanation:demographic,andpolitical.

Thesimplestexplanationwouldconsistinsayingthatcastesnotbeingequallydistributed

across the territory, their representation in various sub-regions is determined by their

demographicconcentration.Thus,thehighnumberofMuslimMLAsinRohilkhandorthe

highnumberofuppercasteMLAsinUttarakhandderivesfromtheirrespectivenumerical

strength. The SP also tends to be stronger in constituencies where there is a high

proportion of Yadavs, such as Mainpuri, Azamgarh, Ghazipur or Kheri255. But a high

252Ibid.,p.287.253Ibid.,p.293.254Ibid.,p.293.255The 1931Census provides caste data at theDivision and the district level. It provides someindications. Yadavs have a stronger presence in an area that covers four contiguous divisions:Faizabad (11.50per cent),Allahabad (11.71per cent),Benares (14.36per cent)andGorakhpur(13.85per cent). They represent nearly 10per cent of the population in the Lucknowdivision.TherearefewYadavs(lessthan3percent)inRohilkhandandMeerut.Thehighestconcentrationsof Yadavs (above 20%) are found in the districts Mainpuri, Azamgarh,, Ghazipur and Kheri.

213

demography does not necessarily translate into political representation and, more

importantly,doesnotaccountforvariationsintime.

Thesecondexplanationissimplypolitical.Bytheticketdistributionchoicesthattheymake,

parties determine to a large extent who is to be represented where. They may be

determinedbydemography in theirchoices,but the localcompetitivesetupalsoplaysa

partintheirchoiceofcandidates.

Thisissignificantsinceitisoftenassumedthatcasterepresentationemergesfrombelow,

fromthepoliticizationandmobilizationofcastesonthebasisofidentityandtheirdemands

for representation. I am not denying that these dynamics take place, but I surmise that

these types of mobilization dynamics only concern a handful of caste, relevant and

powerfulenoughtoconstitutethecorebaseofaspecificpoliticalparty.Formostcastes–

and therefore for most voters –, the political choice they make is largely determined by

whatpartieshavetooffer,arepresentationheavilyskewedinfavorofdominantgroups.

There is further ground to explore to account forwhy certain castes are in apositionof

dominanceinsomeareasandnotinothers.Itisnocoincidencethattheareathathasseen

themostpoliticalchurning–WesternU.P.–isalsotheareathathasdevelopedthemost.It

is also the area where the land tenure system (based on the baichara system) favored

smallerlandholdingpeasantcommunitiesevenbeforetheGreenrevolution.Landreforms

havebeenlesssuccessfulintheEastandthevariousclassesoftraditionallandlordsmore

successfulatpreservingtheirassetsandprivilegesthanintheWest256.

Therearealsocommonalitiesacrossthesesub-regionaltrajectoriesthatareimportantto

ourmainargumentonthelocalizationofelectoralpolitics.

Thefirstcommonalityisthatthephaseofdeclineoftheuppercastesandriseofbackwards

wasinfactanepisodelimitedintime–tentofifteenyearsatthemost,roughlythroughthe

Incidentally,theSPPresidentMulayamSinghYadavcontestedbothinMainpuriinAzamgarhseatsin2012.256As Rohini Guha demonstrates in her doctoral dissertation, access to land remains the mainissueforDalitsinEasternU.P.,whileeducationistheprimaryconcernofDalitsintheWest(Guha2008,117).

214

1980s and the early1990s. In fact, thedeclineof theupper castesbefore all follows the

declineoftheCongressParty.TheuppercasteMLAshadafirstresurgencewhentheBJP

wontwoelectionsinthemid-1990s,andasecondresurgencewhenthestate-basedparties

startedtoopentheirdoorstothem.

The second commonality is that even in sub-regions where the OBCs have risen, it has

usually not been uniformly behind a single party. Some crucial aggregation remain –

YadavswithSPforinstance–butotherwise,theOBCrepresentationisdispersedbetween

manygroupswhodonotformacohesiveensemble–notsociallynorpolitically.

This trend is also observable at the sub-regional level. Parties distribute tickets across

castesalsowithinsub-regions,whichisafurtherindicationthatparties’strategiesfollow

localcasteconfigurationsandnotpre-plannedbroadcastealliances.

There were some caste-party alignments in the 1980s and early 1990s, but post 1993,

thesealignmentswereblurredbythetransformationofparties’electoralstrategies.Inthat

context, caste representation has become a by-product of localized party strategies. In

terms of mobilization and political discourse, state-level caste oriented narratives have

ceased to operate to the benefit of local arrangements, negotiations, and transactions

between groups and individuals embedded in specific socio-economic contexts. Party

appealsmatters,state level issuesmatters,but it isat the local level that thearticulation

betweencasteandpoliticstakeseffectivelyplace.Whatmakesaparticularcastepolitically

relevanttopartieslocallybecomesthenthenextquestiontoaddress.

4.3.Beyondcaste

The Vidhan Sabha’sWho’sWho’s mention the education and occupation background of

MLAs.Wewillseeinthissectionthatthecontentofthesedocumentsisoflimitedhelp,but

dorevealhoweversomeinformationaboutthesociologicalchangesthathavetakenplace

intheAssembly.

215

4.3.1.Education

Thefirstvariableiseducation.Inhercontributiononthesubject,Zerinininotedthatsince

the1960s,theproportionofMLAswithauniversitydegreewasquitehigh,aphenomenon

sheattributes to theelite characterof the twomainparties’organizations,Congressand

socialists (Zerinini 2009, 48-49). Among the degrees pursued by the legislators, the LLB

diplomawasthemostcurrent.

The proportion of MLAs with a university degree increases over time. in a state where

college enrolment in 2012 was four points below the national average (16.8 per cent

against20.4percentnationally).

Thereisn’tmuchdifferenciationofeducationalbackground,party-wise.CongressandBJP

MLAs tend tobemorehighly educated than theothers, but theyare alsomuch fewer in

numberandmoreurban,asfarastheBJPisconcerned.

Table4.14EducationofU.P.MLAs

1980 1985 1989 1991 1993 1996 2002 2007 2012

UptoClassVIII 6.57 3.76 4.68 3.57 2.81 8.96 3.23 6.45 6.20ClassIXtoIntermediate 25.59 26.76 23.89 23.24 21.12 7.31 9.43 25.56 32.75Graduation 15.72 20.19 23.42 25.23 35.7 48.35 52.85 41.44 38.21Postgraduation 36.39 39.67 36.06 32.38 25.35 19.58 26.05 22.08 19.11PhD 0.23 0.23 0.23 0.24 1.88 2.12 2.73 1.74 1.99Non-conventional 0.23 0 0.23 0.24 0 0.24 - - -Unidentified 15.25 9.39 11.47 15 13.14 13.44 5.71 2.73 1.74

Total 99.9% 100% 99.9% 99.9% 100% 100% 100% 100% 100%

N=426 N=426 N=427 N=420 N=426 N=424 N=403 N=403 N=403

Source:quotedfromZerinini(2009,52)until1993.Post-1993,adaptedfromVidhanSabhaWho'sWho(Affidavitdatapost2002).

Thevariationsthatpre-existedbetweencasteshavealsofadedintime.UppercasteMLAs

stilltendtobemorehighlyeducatedthanothersbutthegapisclosing.Overthepastfour

elections,JatshavehadthehighestshareofMLAswithuniversitydegrees(75.5percent),

followedcloselybyuppercasteMLAs(73percent).67percentof theOBCMLAshave

attended college, while that number lowers to 62 and 60 for SCs and Muslims,

respectively.ZerinininotedtheBJPSCMLAstendedtobemoreeducatedthantheBSPSC

MLAs.Thatisnolongerthecasesincethemid-1990s.

216

Wedohaveinformationalsoontheplaceofeducation.Thedataisonlypartialbutdoes

confirm that a large number of MLAs got their degree from a limited number of

universities, among them the universities of Agra, Lucknow and Allahabad. A large

number of Muslim MLAs have received their higher education at the Aligarh Muslim

University.Thedeclineoftheseinstitutionsthroughtimehasnotdiminishedtheirroleof

producerofpoliticalelites,thoughtheseelitesemergemorethestudentunionsthanfrom

theclassroomperse.

Overall,theU.P.Assemblydoeshaveanelitistcharacter,whichconfirmsthateducationis

alsoapoliticalresource.However,oneshouldbecautioushandlingthisdata,sincemany

MLAs simply do not declare their education level, nor is there a standardized

nomenclatureforthetypeofthenomenclature257.

4.3.2.Theproblemofoccupation

Finally,assessingtheprofessionalbackgroundofMLAsisacomplicatedtaskforaseriesof

reasonsthathavetodowiththequalityandreliabilityofthedataavailableonthatsubject.

ThetraditionalwayconsistinginlookingatVidhanSabha’sWho’swhosisunsatisfactoryfor

atleastsixreasons.

The first reason is that the socio-professional categories used are too broad and

undifferentiated. In particular, the categories of agriculturists, business or political or

social workers are virtually meaningless, if one does not cross them with other

information, on landholding sizes, thenumberofpeople employed, and soon,which is

difficulttoobtain258.

257Fewactuallydonotdeclarethetypeofdegree,whichmakesitdifficulttoassesstheactualratiooftypeofeducation.258Onecouldtechnicallydothatbycrossingtheoccupationvariablewiththerevenueinformationwe get from affidavit. But this data being also self-declared and generally recognized assystematically under-valued, it is in fact of little help. Candidate’s income is individualwhile inmost cases, the relevantunitwouldbe thehousehold. In recentyears, theaffidavit includes thespouse’s occupation and revenue declaration. However, it remains however easy for anycandidatetoconcealtheiractualwealth,especiallysincenoverificationsareconductedbytheECIorotherstateagency.

217

Theseconddifficultycomesfromthefactthatmanycandidatessimplydon’tdeclaretheir

occupation. Jayant Chaudhury, son of Ajit Singh, for example, does not declare any

particularprofession.

Thethirdissueisthatitisfrequentthataprofessionconcealsothersourcesofrevenue.In

myyearsof fieldwork, I have encounteredanumberof lawyerspoliticianswhoowned

factories,ranprivateschools, tookparticipation intheirpartnersor friends’businesses.

Formostofthem,legalfeeswereaminorpartoftheirincome.

Until the 2000s, few MLAs would declare themselves as ‘businessmen’, for an

‘agriculturist’tagfitsbetterwiththeimagetheywishtoprojecttotheirconstituents259.A

numberofMLAsstartdeclaringpoliticsorsocialworkastheirprofessionaftertheirfirst

election260.Forexample,BhagwanSharma,anMLAfromDebaiconstituency,ranin2007

onaBSPticketasanindustrialistandin2012onaSPticketasapoliticalworker.

AfourthproblemcomesfromthefactthatmanyMLAsdeclaremorethanoneprofession,

such as lawyer and industrialist, or socialworker andpetrol pumpowner,whichblurs

socio-professionalcategoriesandposescodingissues.

Afifthproblemcomesfromthepresenceofpoliticiansinvolvedinillegalactivitiesorin

theblackeconomymoregenerallyspeaking.Inmanyinstances,localelectedmafiadons

areengagedintoallsortsofundeclaredinterestsinarangeofeconomicactivities–liquor,

construction,illegalmining,trafficking,etc..Itwouldbeunrealistictoassumethatpolitics

does not reflect in part the fact that a large share of the economy belongs to the

shadows261.

259AndthefactalsothatagriculturalincomeisnottaxedinIndia.Thereisthusagreatincentivetodeclareoneselfas“agriculturist”sincemoneyearnedfromotheractivitiescaneasilybelaunderedthroughthefarm.IthankPhilipK.Oldenburgforremindingmeofthatfact.260In quite a few cases, the spouse’s occupation declaration provides a hint. Number of socialworkersaremarriedtowomenwhoownsbrickkilnsorpetrolpumps.261There isdisagreementonthesizeof theblackor“shadow”economyin India.Recentreportsplaceitsomewhere50to75percentoftheGDP.ForanattempttomeasurethesizeoftheblackeconomyinIndia,see(Chaudhuri,Schneider,andChattopadhyay2006).

218

Finally, the occupation data suffers from the general unreliability of self-declared

informationbypoliticalactors262.Thatinformationcannotbetrustedoreasilyverified.In

myexperience,thatappliestoallparties263.

Ifweassume that theaffidavitdatacontainsmore reliable information, the2012Vidhan

SabhagivesusastrikinglydifferentpicturefromthedatacontainedintheWho’swho for

thepreviousyears.

AlongstandingtrendisthatalargepartoftheMLAsdeclaresagricultureasoccupationand

sourceorincome.Theywereslightlybelow40percentinthe1950sand1960s,aratiothat

increased to a near majority of the House in the 1980s, at the height of Kisan politics.

Zerinini notes a decrease of the ratio of farmers in the Assembly from the early 1990s

onwards.In2012,only28.4percentoftheMLAsdeclaredthemselvesasfarmer264.

Asecondlongstandingtrendisthestablepresenceofself-declaredlawyersthroughtime,

around 18 per cent according to theWho’swhos. In 2012, they were only 3 per cent265.

Generally speaking, liberal professions, or white-collar professions, represent only 6 per

centoftheMLAs(includingthelawyers)inthe2012Assembly,anotherdeparturefromthe

past.

Inthe1980s,7to8percentoftheMLAsusedtodeclaredbusinessastheiroccupation,a

proportion that doubled in the followingdecade. In 2012, therewere33.4 per cent self-

declared business in the Assembly. And if we club to that category those who declare

themselvesasindustrialists(3.5percent),builders,contractorsandpropertydealers(8.7

262Withthenotableexceptionofself-declareddataoncriminalcharges,whichistheonlyformofdatathatcanactuallybereliablycross-checked.263 The situation has improved somewhat since the Supreme Court made it available forcandidatestodeclaretheirrevenueandcriminalchargesituationinanaffidavit,beforefilingtheirnomination forms.The issueof self-declarationremainsbutwedohaveaccess tomoredata. Inthe case of Uttar Pradesh, the affidavit data did not contain information on occupation before2012.What Ididwastoverify that informationthrough interviews.Thismethodenabledmetoget a comparatively more reliable picture of the socio-economic profile of the members of thecurrentassembly.264The legislators declaring as occupation agriculture and business have been coded in the“business”category.265Lawyerswhodeclaredtobealsobusinessmenhavebeencodedasbusinessmen.

219

percent)andtraders(1.75percent),thatratioincreasesto47.4percentoftheMLAs.7.7

percentoftheMLAsdeclaredsocialserviceorpoliticsastheirprofession.

Table4.15ProfessionofMLAsinthe2012U.P.Assembly,perparty

INC BJP BSP SP Others

Agriculture 32.14 17.02 15.00 36.16 20.83Allbusiness 35.71 46.81 66.25 41.96 54.17Business 28.57 29.79% 35.00 33.04 45.83Construction,Contractor,Builder,Realestate - 12.77 22.50 4.02 8.33Industry 3.57 2.13 7.50 3.13 -Trade 3.57 2.13 1.25 1.79 -LiberalProfession 7.14 8.51 3.75 4.02 4.17Service - - 1.25 2.23 -Education 7.14 4.26 1.25 2.23 -Politicsandsocialwork 7.14 14.89 10.00 5.36 4.17FormerCivilServant - 2.13 1.25 0.89 4.17Others* - 4.26 1.25 3.57 8.33Undeclared/unidentified 10.71% 2.13 - 3.57 4.17Total 100% 100% 100% 100% 100%

N=28 N=69 N=133 N=318 N=37

Source:Candidate'saffidavits,procuredbytheAssociationforDemocraticReforms.*includeshousewives,religiousmissionaries,pensioners,employeesandZamindars

Someof thesevariationsmayhave todowithchanges insemantics.Manytraderswould

declarethemselvesasbusinessmen.Otherswouldcreatetheirowncategory,suchas this

MLAfromLoniwho,in2012,registeredhisprofession“financierandtraditionalelite”.

There are interesting variations between parties. The BSP is the party with the highest

shareofbusinessmenalongitsMLAs,notablyintheconstructionbusiness.Italsohasthe

smallest share of farmers (15 per cent). The BJP, too, has few farmers, which is not

surprisinggiventhefactthatmostofitsMLAsareelectedinurbanorsemi-urbansegments.

It alsohas thehighest ratioof self-declaredpoliticiansor socialworkers (14.9per cent).

TheCongress,whichalsohasthesmallestnumberofMLAs,countsnobuilderswithin its

ranks.BoththeBSPandtheSPhavefewerliberalprofessionsamongtheirrepresentatives

thanthenationalparties.

In terms of occupation distribution among castes, we see that businessmen are most

represented among the Jats, the OBCs and the Muslims. There are slightly fewer

220

businessmenamongtheuppercastes(46.8percent)andleastamongtheSCMLAs(36per

cent)266.

Table4.16ProfessionofMLAsinthe2012U.P.Assembly,percastegroup

IC Muslims OBC SC UC Unidentified

Agriculture 20.00 23.53 27.36 34.88 29.37 14.29Allbusiness 60.00 51.47 53.77 36.05 46.83 57.14Business 50.00 38.24 33.02 27.91 32.54 57.14Construction,Contractor,Builder,Realestate 10.00 7.35 13.21 5.81 7.94 -Industry - 1.47 5.66 1.16 5.56 -Trade - 4.41 1.89 1.16 0.79 -LiberalProfession - 5.88 3.77 5.81 4.76 -Service - 2.94 1.89 1.16 0.79 -Education - 1.47 1.89 3.49 3.17 -Politicsandsocialwork - 4.41 7.55 5.81 10.32 14.29FormerCivilServant 10.00 2.94 1.89 - - -Others - 1.47 0.94 9.30 1.59 14.29Undeclared/unidentified 10.00 5.88 0.94 3.49 3.17 -Total 100% 100% 100% 100% 100% 100%

N=10 N=68 N=106 N=86 N=126 N=7


We also see some caste, occupation and party aggregations. 60 per cent of the Dalit

businessmenarewith the SPand29per centwith theBSP.62.9per centof theMuslim

businessmenarewiththeSP,against31.4withtheBSP.HalfofthebusinessmenOBCare

withtheSP,against38.8withtheBSP.TherearepracticallynoOBCbusinessmenwiththe

BJPandbarelytwowithCongress.

And finally, there are also patterns emerging if we break down the data by sub-regions.

Withoutsurprise,theshareofbusinessmenMLAsisgreaterinthemoredevelopedpartsof

the state – Doab and the West. Their proportion is the highest (63.16 per cent) in

Bundelkhand267.

26660percentof theDalitbusinessmenarewith theSPand29percentwith theBSP.62.9percentoftheMuslimbusinessmenarewiththeSP,against31.4withtheSP.HalfofthebusinessmenOBCarewiththeSP,against38.8withtheBSP.TherearepracticallynoOBCbusinessmenwiththeBJPandbarelytwowithCongress.267 Most of these businessmen-MLAs from Bundelkhand are either contractors or in theconstructionbusiness,whichwouldindicatethattheyarenotsmallbusinessowners.

221

TheratioofbusinessmenMLAs is the lowest in theNortheast,wherewesawthatupper

castedominate.Inthisregion,mostMLAstendtobelandlordsanddeclarethemselvesas

agriculturists.

Table4.17ProfessionofMLAsinthe2012U.P.Assembly,persub-region

Avadh Bundelkhand Doab East Northeast Rohilkhand West

Agriculture 30.39 15.79 20.45 24.59 45.95 38.46 25.00Allbusiness 43.14 63.16 56.82 45.90 32.43 38.46 59.09Business 33.33 42.11 38.64 32.79 27.03 26.92 34.09Construction,contractor,builder,realestate 3.92 21.05 10.23 9.84 - 7.69 18.18Industry 4.90 - 4.55 1.64 2.70 1.92 6.82Trade 0.98 - 3.41 1.64 2.70 1.92 -LiberalProfessions 6.86 5.26 2.27 1.64 5.41 9.62 2.27Service 2.94 - 2.27 1.64 - - -Education 2.94 - 1.14 6.56 - - 4.55Politicsandsocialwork 7.84 5.26 9.09 13.11 5.41 3.85 2.27FormerCivilServant - - 1.14 - 2.70 1.92 4.55Others* 1.96 5.26 1.14 4.92 2.70 7.69 2.27Undeclared/unidentified 3.92 5.26 5.68 1.64 5.41 - -Total 100% 100% 100% 100% 100% 100% 100%

N=102 N=19 N=88 N=61 N=37 N=52 N=44

N=19 N=88 N=61 N=37 N=52 N=44


Thedata for2007remainstooscarce(occupationdidnot figure inthenominationform)

butifwecompareincumbentMLAs,wecanmakeaseriesofobservation.First,nearlyhalf

ofthemdeclareadifferentprofessionwhentheyre-run.

For example, Deepak Yadav, from Garoutha, ran in 2007 as farmer and in 2012 as

businessman. Dharam Singh Saini, a BSP MLA from Nakur, Western U.P., ran in 2007 as

medicaldoctorandin2012asabusinessman.GendaLalChaudhary,anotherBSPMLA,ran

in 2007 as an ‘educationist’ and in 2012 as a real estate dealer and contractor. Iqbal

Mahmood,anSPMLAfromSambhal,ranasafarmerin2007andasabusinessmanin2012.

SameforJagdishSonkar(SP)inMacchlisharhorKailashNathChaurasia(SP)inMirzapur.

Between2007and2012,amongthe95re-electedMLAs,45declaredthesameprofession,

41shiftedfromanon-businessrelatedprofession(agriculture,teaching,medicaldoctor)to

222

abusinessrelatedoccupation(business,realestate,contractororindustrialist).Intotal,36

MLAswhoregisteredin2007asagriculturistsdeclaredanotheroccupationfiveyearslater

(sevenof themdeclaredpoliticsas theirprofession).Whatwehavehere isan indication

that in 2012, nearly half of the incumbent MLAs geared towards some business activity

aftertheirfirstelection.

Therearefinallysomespatialvariations.MoreMLAsdeclarefarmingastheiractivityinthe

East than in the West. There are more MLAs who declare a profession linked to

construction or real estate inWesternU.P., a reflection of the economic transformations

thathavetakenplaceinthatsub-region.

4.4.Conclusion

WhatlessonscanwedrawfromthechangingportraitoftheStateAssembly?Thefirstone

isthatcastepoliticsisagameoffewandnotofmany.Onlyahandfulofcastesconstitutea

core support base to a party they can claim their own. The others are too small or too

geographicallydispersedtoconstituteacoresupportbasetoanypartyorcandidate,even

locally.Thesesmallanddispersedgroupsconstituteafloatingelectoratethatisgenerally

insensitivetocasteappeals.

Second, caste-partyalignmentshavebecome loose,at thebest.Lokniti/CSDSsurveydata

already informed us that it was the case for party-voters alignments. Only a few groups

votecohesivelyforspecificparties.Thesearethegroups–JatavsandYadavs,essentially–

that have both numerical strength and a party of their own.Other groups, including the

uppercastes,havebeensplittingtheirvotesbetweenpartiesandlocalcandidateselection

afterelection.

Third,nurturingacoresupportbaseonthebasisofcastemaybenecessarytowinelections

but surely cannot be enough. The politics of the 1990s has shown that campaigning by

wooingspecificcastesattheexclusionofothersdoesnothelpwinningelections.Infact,in

recent times, majorities have been built precisely on the capacity of parties to mobilize

votersbeyondtheircoresupportbase,thatistosayacrosscastes.

223

Fourth,parties–bychoosingtheircandidates–playaprominentroleinthedetermination

of thesociologicalcompositionof theAssembly.Casterepresentationoutcomesareoften

characterizedasthetranslationofpoliticalandsocialmovementordynamicsamongvoters

– i.e. the mobilization of caste groups and the political alignment of voters according to

ascriptive identities.This is forgetting thatparties act aspowerful filtersbydetermining

whogetsrepresentationornot,byselectingcandidatesonthebasisofcaste.Inotherterms,

theexplanationforthevariationinthesocialcompositionoftheassemblyispoliticaland

notonlyormerelysociological.

Fifth, sub-regional variations tell us that narratives based on aggregate data must be

discardedorifnotthencriticized.InthecaseofU.P., thenarrativeofageneraldeclineof

upper castes and rise of OBCs does not hold, concomitant with the decline of national

partiesandriseofregionalparties,doesnothold.

And finally, these transformations can be interpreted as a reconfiguration of the

relationship between caste and politics. Caste mobilizations were used in the pre-

liberalisationperiod topoliticize social groupsandconstituteblocksof votersacross the

state, bound by their ascriptive identity. In the post-liberalization period, these identity-

based mobilization tropes – quota politics, for instance – lost their efficacy. The need to

expandtheirsocialbasesrequiredpartiestoadoptamoreinclusivegeneralistdiscourse.

Thisdoesnotmean thatcasteasapoliticalvehicleofmobilizationdidnotdisappear, far

from it. It in factbecameconfinedat the level atwhich it effectivelyoperates, that it the

local level, where caste social, economic and political interests clash or coalesce. One

elementofproofthatcasteremainsimportantisthatitstilldeterminesthedistributionof

tickets,whichinturnshapestherepresentationcastegroupshaveintheassembly.Ticket

distributionisnotdonefollowingpan-statecastecombinationstrategiesbutaccordingto

local,constituency-level,circumstances.

Inthenextchapter,Iwillpaymoreattentiontothefunctioningofthisrelationshipbetween

casteandpolitics,byexamininglocalcontextsinwhichthisrelationshipisinscribed.Iwill

finallycomparepartiestrajectoriesandorganizationinafinalchapter.

224

Chapter5.Whowieldspower?LocalperspectivesonUttarPradeshelectoralpolitics

Thenextquestion I seek to address is thequestionof inscription anddiffusionof local

power.OneofthethemesthatIkeptatthebackofmymindeverytimeIwasonthefield

was:whowields powerhere?Who are the individuals possessingpolitical influence or

exertcontroloverlocalinstitutions?Whatistheirsocialstatusandwhatpositiondothey

occupy in the socio-economic environment of the locality in which they operate? And,

finally,howhasitevolvedintime?

5.1.Thesourcesofpoliticalpower

Thedatausedinchapter4pointstowardstwotypesofmajorchangesinthesociological

profileofMLAs.The firstone is thegrowingheterogeneisationof casteandcommunity

representation.Overtime,moregroupsaregettingrepresentedintheassemblyaswellas

withinthepartiesthathavebeenwinningelections.Therearesub-regionalvariationsand

variations between parties but overall, the U.P. State Assembly today is more

representativethanitusedtobeinthepast.

The second transformation suggestedby thedata is that important changeshave taken

placeintermsoftheoccupationalbackgroundofMLAs,notablyariseofMLAsdeclaring

business-related activities, including in regions that are poorly developed such as

Bundelkhand. This data however does not capture the detail of this information,

occupational categories being vague and ambiguous. This data also certainly does not

informusaboutthemechanismsthroughwhichbusinessfigureshaveriseninpolitics,or

donot tellusanythingabout themotives thatmighthaveattracted these individuals to

contestinthefirstplace.

Wesawinchapter3thattherulesoftheelectoralcompetitiontendtofavorcandidates

with certain attributes, notably thosewhohave the resources to compete.Wealso saw

that the shortness of political careers filters in aspiringpoliticianswho expect tomake

short-termgainsfromtheirelection.Butwecannotderivecandidates’motivationsfrom

thesesystemicrules.

225

In order to examine this question, one needs to adopt a more qualitative approach,

examiningthecareertrajectoryofindividualpoliticiansandtheirinscriptionintheirlocal

social,politicalandeconomicalcontext.Inordertoavoidalocalitybias,Icomparesome

of these trajectories in Western U.P. and in the East, where I conducted most of my

fieldwork.

This comparisonwill enableme to elaborate someof the aspects touched in chapter3,

regardingtherulesoftheelectoralgame.Ihavelookedsofarattheconstraintsinduced

by the formal rules of electoral politics and by the overall competitiveness of electoral

politics. I have also looked at the constraints that parties impose on candidates, by

decidingwhogetstobeinpoliticsandveryoftendecidingwhogetstostayinpoliticsor

not.

Two other factors contribute to the competitiveness of a candidate and need to be

accounted for: the centralityofmoneypower, or the resources required to contest and

win an election; and the compliance with the expectations trusted upon elected

representatives by voters. Consorting with illegality comes as a third component of a

candidate’scompetitivenessbutforreasonsthatwillbeelicitedfurther,Idonotmakea

generalruleoutofit.

Theargumentdoesnotconsisthere insayingthat theseattributes–moneyandmuscle

power and compliance to voters’ expectations – are sufficient conditions to get elected.

Surely, individual qualities such as individual charisma and eloquence have their own

importance. So do party and state-level considerations. But in many instances, these

attributesareanecessaryconditiontohaveastandingchanceofwinningaseat.Infact,I

find thatmany, ifnotmost, longstandingMLAscumulate theseattributes,which iswhy

partiespickedthemupascandidatesinthefirstplace.

Inthenexttwosections, Icomparethetrajectoriesof locallyprominentpolitical figures

and will try to situate these trajectories into a broad context of political and economic

transformation.

226

5.1.1.ThecaseofWesternU.P.268

WesternU.P.hasundergonedeep transformations since theearly1980s, largelydue to

theeconomicdevelopmentof thissub-region,adjacenttotheNationalCapitalRegionof

Delhi.Thosechangeshavebeendrivenmostlybyurbanization,byindustrialdevelopment,

and by the marketization and diversification of the rural economy (Jeffery, Jeffrey, and

Lerche2014,Kumar2014,SharmaandPoleman1994,SinghandMehrotra2014).

WesternU.P. is therichest,mosturbanizedsub-regionofUttarPradesh. It concentrates

muchofitsindustry(seechapter2)andofitsprivatecapital,concentratedintheurban

sprawlsofNoida,GreaterNoida,FaridabadandGhaziabad,whichsurroundthenational

capital.

It also contains pockets of rural poverty, signal that the economic transformations that

tookplace in this regionhavealso contributed to social andeconomic inequalities.The

conversionof largeamountof landintocommercialor industrialusehasthrownonthe

marketa largenumberofyoungeducatedpeoplewhostruggle to find satisfactory jobs

(Jeffrey 2009, 2010b, 2013). Land ceiling, which prohibit the indefinite expansion of

familylandholdings,alsoinduceafragmentationoflandholdingsbetweensiblingsatthe

timeofinheritanceortransferofpropertytitles.

Politically speaking, the region is highly volatile with a competitive five-party system.

Castesalignmentskeepvaryingbetweenparties,whodistributeticketsacrosscastesand

communitylines.Aswesawinchapter4,mostcastegroupsobtainrepresentationfrom

differentparties,whichprobablymakeofWesternU.P. themostcompetitivesub-region

inUttarPradesh.

Butifthesocialalignmentsofcastesandpartiesarerelativelyblurredatthelevelofstate

politics,theyarenotatthelocallevel.AbriefincursionintolocalPanchayatpoliticswill

268This section is based on interviews conducted over five years with political actors, andfieldwork conductedduring themonths ofNovember andDecember2012 in 11 villages acrossMeerutandBaghpatdistricts,inwhichPaulBrass,hispartnerSueandIinterviewedmostlylocalpoliticians(contestingPanchayatelections)andlocalcasteleaders.

227

showhowandwhycasteremainsrelevantasapoliticalvariable.Itwillalsorevealhow

localpoliticsandthepoliticaleconomyoflocalitiesaredeeplyintertwined.

Aminagar,urfBhurbarai,MeerutDistrict

The village of Aminagar, also known as Bhurbarai, sits on the outskirts of the city of

Meerut,onthehighway35thatconnectsthedistrictheadquartertoDelhi.Accordingto

the2011Census, thevillagecomprises1044households, fora totalpopulationof6141,

17percentofwhichareDalits.Accordingtolocalestimates,Muslimsrepresentabouta

thirdofthepopulation.Brahminsrepresentaquarter.TherestisdividedamongGujjars

and Jats, who used to own most of the land. There is also a small number of Bania

households,whorunshopsandsmall-scalebusinesses.TheDalitsaredividedintothree

maingroups: Jatavs,whodominatenumerically,ValmikisandKoris,whobarelycounta

fewhouseholds.

DuetoitsproximitywithMeerutandthehighway,mostoftheagriculturallandhasbeen

soldforindustrialdevelopments.Theseparcelsweresoldthroughpropertydealerswho

werenot fromthevillage.Theplotsdirectlyadjacenttothehighwaycountanumberof

large structures, mostly education institutions, undergraduate colleges and technical

schools. Most of these structures have remained empty, either waiting for more

construction or for a re-allocation of the land-use, which will enable their owner to

convertthosebuildingsintocommercialorindustrialventures.

As a result, few villagers are engaged into agriculture (about 20%, according to the

Pradhan).ThelocalGujjarshavesoldtheirlandorleasedittoindustrialgroups.TheJats

havealmostentirelyleftthevillage(onlythreehouseholdsremain),tomigratetoMeerut

or beyond. Most people are employed in service jobs – private or government. Others

havesmall-scalebusinessinthevillage.AccordingtoaBrahminlandlord,themajorityof

thevillageinhabitantsarelandless.Brahminshaveretainedtheirpropertyandnowown

mostoftheremainingland.BesidestheJats,fewhavemigrated.

Duetourbanization,thepopulationofAminagarhasincreased.AccordingtothePradhan,

thevillagedidnotcountmanyasmanyMuslimstwentyyearsago.Theyhavecomemostly

from other towns and villages in Western U.P.. They live in a distinct neighborhood,

228

separatedbythehighway.Theyaredividedamongfourgroups:theHalvis,theSaifis,the

Sakhe(tailors)andtheAbbasi(whosetraditionaloccupationconsistsinbringingwaterto

homes).TherearealsoafewFaqirs,loweramongMuslimsandconsideredasbeggarsby

our Halvis interlocutors. Most Halvis are engaged in local businesses. The others work

mostlyasconstructionlabourerss.

None of the land thatwent to themarketwas acquired by local landless dwellers. The

agriculture labourerss – mostly Dalits – who used to work for the local landlords now

commutetoMeerutoritssurroundingsfordailywageconstructionorfactoryjobs.Some

work as local masons or as security guards in Meerut. With an increased access to

education,manyDalitshave left thesedailywage jobs to joincivil service.According to

oneofourJatavinterlocutor:“Around50boysareingovernmentjobs:incometax,salestax,

police, LIC. One has become an IPS officer. He topped is rank. Four are working in bank

branches”.HeaddedthatwhenMayawatiisinpower,thereisnoharassmentorbribesfor

appointments.

The fact that Dalits are now unwilling to work on local farms has led the larger

landowners to import seasonalagriculture labourerss fromBihar.Thosewhohavesold

theirlandorpartoftheirlandhaveeitherinvestedinbusinessventuresoracquiredland

inareaswherethepriceshaven’tsoaredasmuchastheyhaveintheirvillage.Manyhave

invested in real estate–mostly residentialpropertiesor shops– that theyhaveputon

rent.

Politically speaking, the village is dominated by Gujjars. Between 1995 and 2010,

AminagarhashadthreedifferentGujjarPradhans.Before1995,thevillagehadthesame

Muslim Pradhan, Mohammed Yamin, for twenty-two years. Mohammed Yamin’s family

usedtobeonethelargestlandholderinthevillage.WhenYamin’sfatherpassed,heand

hisfivebrothershad200bighas(about40acres)tosplitamongthemselves.

The main rivalry used to take place between Muslims and Gujjars. Mohammed Yamin

prevailedbecausetheGujjarvotewassplitbetweenvariousfactions,andbecauseother

groups,includingtheBrahminsandtheDalits,didnotwantaGujjartobecomePradhan.

In2010,theseatbecamereservedandaKori(SC)namedLalluwaselected.EighteenDalit

candidatescontestedthatyear,whichledtoascatteringofthevotes.Lallu’srunnerup–a

229

Jatav – came 44 votes short. Some of my interlocutors described Lallu as an outsider

(baharkaaadmihai),asaGujjarstooge,andasatantricwhohealspeoplewithsweets.

Inter-grouprelationshavebecometenseinrecentyears,particularlyattimeofelections.

Thereisalong-standingrivalrybetweenGujjarsandMuslims,whocompetetogarnerthe

support from other groups, notably the Dalits. Traditionally, Dalits do not vote for the

Gujjars,theirformeremployers.ThefactthatnoDalitsinAminagarworkontheirfarms

anymorehasgiventhemacertainpoliticalautonomy,whichtheyusetotradefavorswith

othergroups.S.K.Jatav(namechanged)contestedtheelectionin2010.Aself-professed

Ambedkarite,heownsasmallbusiness. Inourconversation,hedecried thereservation

system,whichletsstoogesbeingelectedonbehalfofthelocalstrongmen.

“Lallu’selectionisharmfultoDalits.HewasmadePradhanbytheJats,theBanias

andtheGujjars.Allthedevelopmentworkisdoneintheaffluentareas(…)Heis

notaccessibletosigndocuments.He’snotdoinganything”.

Aquick surveyof recentPanchayatprojects confirmed thatmostof the resources– for

drainage and road construction – have been concentrated in the Gujjar neighborhood.

They even obtained that a new road be built, connecting their basti directly to the

highway,bypassingthevillage.

Inordertocountertheconsolidationofvotesbehindaboguscandidate,theJatavssought

toforgeanalliancewiththeMuslims.

“We have good relations with the Muslims. We made a pact with them. They

support us in this election and we will support them in the next Panchayat

elections”.

Despitethisarrangement,thehighnumberofcandidatesledtoadispersionofthevotes.

TheGujjarsvotedsolidlybehindtheirchosenDalitcandidate.

Local Gujjars easily admitted that Lallu was a ‘proxy candidate’. They supported Lallu

because they did not want a Jatav to becomePradhan. A young Gujjar, member of the

youthwingoftheSamajwadiParty,confirmedittomeinthoseterms:

230

“Lalluisanamesake,we(theGujjars)haveinfluence.Wehavethemusclepower

soothergroups follow(…)Musclepower is inourblood.Wehavemoneypower

too”.

OurinterlocutorbelongstoalocalGujjarfamilythatsupportstheBJPinstateandnational

elections.HeenlistedwiththeSP,againsthisfamily’swish,becauseheseesitastheonly

contender that can defeat theBSP.His uncles and cousins are still engaged in farming,

mostly sugarcane. He used the money he inherited after his father’s passing to buy

propertiesinthearea.Henowlivesofftheirrentandshareshistimebetweenproperty

dealsnegotiationsand“doingpolitics”.

ThebreakdownoftheeconomictiesbetweenlocallandlordsandtheDalitshasn’thelped

toimprovetheirrelation.StillaccordingtoS.K.Jatav:

“Muslims teach their children that their enemies are Hindus. We teach our

childrenthatourenemiesareGujjars”.

Subsequent fieldwork in other villages revealed similar stories of how political

competition remains entrenched in caste antagonisms – particularly among Dalits and

landed OBCs – and how economic transformation – mostly dynamics around land

transfers and acquisition – affected thebalance of powerbetween groups.What varied

literallyfromavillagetoanotherwasthecastecompositionandtheidentityofthelocal

dominantgroup(s).

Anotherrecurrentfeatureofchangeinthesevillageswasthefactthatmostofthosewho

had sold their land and had migrated were from the upper castes. They often lost the

powerandinfluencetheywieldedinthepasttootherbackwardgroups,whousedtheir

numericaladvantageandresourcestogaincontroloflocaldemocraticinstitutions.

Amroli,urfBaraGaon,MeerutDistrict

In the village of Amroli (Urf Baragaon), for instance, Muslim Mirs dominate the local

politicalscene.Theyownmostoftheland,orchardsandfishponds,aprivilegetheyuseto

231

sharewith theRastogis (Banias). In recent years,manyRastogi families sold their land

and migrated to cities. There used to be a number of Brahmin families but they have

migratedgradually.ThelastBrahminfamilytoleavethevillagewasthefamilyofthefirs

lastBrahminPradhan,Kaushik.

The Mirs acquired the land of these upper caste migrants and consolidated their

properties.Theyalsoincreasedtheirrevenuebycultivatinghighervalueproduces,such

as fruitsand turmeric,andbygainingcontrolofmostof the localbrickkilns.Theyalso

mademoneyby selling landbordering the village toGujjars and Jats from surrounding

localities.Therichestlandlordamongthemisreputedtoownmorethan100acresofland,

wayabovetheauthorizedceiling,throughbenamiproperties(propertiesownedthrough

a strawman). The Mirs finally use their dominant position to bag most of the public

contractsandlocalpublictenders.

AstheMirsgrewmoreandmoredominant,theysplitintotwofactions,oneledbyTanzim

AkhtarsonofaformerPradhan,theotherbyoneofhisrelatives,namedJafar.Thesplit

started as a family dispute in which the head of the family and village pradhan Abdul

Hamidwasmurdered.Violenceensuedandadozenmorepeoplewerekilled ina short

spanoftime.Sincethen,thetwofactionshavebeenfightingeachotherthroughelections,

leadingtoamorepacifiedrivalry(oneofJafar’sgunmenwassittingonaplasticchairin

thecorneroftheroomwhenweinterviewedTanzimAkhtar).

Their divisions however cost them politically. Tanzim complained to me that they (the

Mirs)couldnotcommandothercommunitiestovoteforthemliketheyusedtointhepast.

ThevillageofAmroli,locatedintheMewanaMandal,Meerutdistrict,countedin2011786

households, for a total population of nearly 4600. By their own admission, Muslims

representoverfortypercentofthepopulation.TheyaremostlyMirs(uppercastes)but

therearealsoanumberofsmallbackwardMuslimcastes,suchastheNais(barbers).The

Censussaysthattwenty-fivepercentofthepopulationisDalit.Therestofthepopulation

isdividedamongSainis(OBC),Gujjars(OBC),Banias(Rastogis)andJogi-Upadhyay(OBC).

MostoftheSCsareJatavs.ThereareadozenofValmikihouseholds.

232

In termsofemployment, theDalitsare splitbetweenagriculture laborandmenial jobs.

Mostof themownsmallplotsof land.Only twoDalitsownsufficient landtogeneratea

surplus. NREGA jobs come as a revenue complement but many complain of the

dysfunctionalpaymentsystem,whichleadstoconsiderabledelaysinthepaymentofdues.

Sainis are engaged into business activities. They also own one brick kiln and the cold

storagefacilities forthearea.Theyarerelativelywell-off,comparedtootherOBCs,who

mostlyworkasgovernmentclerks,securityguards,masonsorconstructionworkers.

ThefirstPradhanrememberedbyourinterlocutorswasaBrahminnamedKaushik.The

first Muslim Pradhan, Abdul Hamid, was elected before 1990. He was succeeded by

TanzimAkhtar,aliasTannu,oneofhisrelative.Akhtar’swife,Fakhra,thensucceededhim

whentheseatbecamereservedforwomen.In2000,thewifeoftheotherfactionleader,

Jafar, became Pradhan. The seat became reserved for SC women in 2005. The votes of

dominantgroupsweresplitthatyearandawomannamedRajesh,aJatav,won.In2010,

the seat got de-reserved and was won for the first time by a Saini, Harbir Singh. He

defeatedaMuslimnamedYasinSalmani(aNai),withthesupportoftheTanzimfaction.

Before the election, Harbir had made a pact with Tanzim, offering his support for his

affairsatthedistrictlevelagainstthevoteofhisfaction.Thecombinationoftheirsupport

to the Saini vote led him to a comfortable victory. Asked about state elections voting

patternsinthevillage,HarbirSinghexplains:

“SainisvotefortheBJP,MuslimsandGujjarsfortheSP,theSCsfortheBSP.Butif

the BSP candidate is a Saini, Sainis will vote for him. If there are several Saini

candidates,theygetconfusedandtheylose”269.

Thisexampleillustrateshowcastevotingremainscohesiveatthelocalitylevel,butcanbe

distributedacrosspartiesacrossconstituencies.Aggregatesurveydatamayshowthatthe

vote of a particular group – here the Sainis – may be split across parties, but it also

obfuscatesthefactthatmembersofaparticulargrouptendtolargelyvotecohesivelyat

thelocallevel,wherecastetiesremainstrongandrelevanttoindividuals’lives.Thisisnot

specific to particular groups and local caste groupsmaydecide to vote for one of their

269InterviewwithHarbirSinghSaini,PradhanofAmroli,athisresidence,13November2012.

233

ownnomatterwhattherestofthecompetitionlookslikeordecidetovotestrategically

forthecandidateofastrongpartywhomaynotsharetheircasteidentity.Itremainsthat

voterstendtocoordinatetheirchoicewithintheirascriptivegroup.

Dhanauria,Baghpatdistrict

AsweprogressedwestwardtowardsBaghpatdistrict,weentered into Jat territory. Jats

representseventeenpercentofthepopulationinWesternU.P.buttheyareparticularly

concentrated ina fewconstituenciesscattered in themost ruralpartsofWesternU.P. -

Agota,Barnawa,Khatauli,SiwalKhasandChaprauli– longstretchesofsugarcanefields

dottedwithbrickkilns.

Inthevillageswevisited,thecontrolofJatsoverlocalinstitutionswasalmostcomplete.

Whentheseatsarereserved,theydetermine,barringafewexceptions,whichDalitsgets

tobecomePradhan.

TheseJatdominatedvillagesalsotendtobefarawayfromthemoredevelopedandmore

urbanized parts of the region. Take a right angle on a highway and travel a fewdozen

kilometersinroadsandthesignsofurbanlifequicklydisappear.ThevillageofDhanauria

forinstance,locatedabout80kmNorthofDelhi,nearthetownChaprauli,hasremained

unaffectedbythekindofeconomictransformationthathasmarkedtheregion.Sugarcane

isthemaincropandindustryandthosewhodon’tcultivatetheirownlandtillthelandof

others.

Asa result, theold inequalitypatternshavepersisted.Dalitsownno land inDhanauria

(exceptonehousehold)andtheynearlyallworkaslabourerssintheJats’canefields.Shoe

making and NREGA jobs come as revenue complements. The Valmikis, who tend to be

unemployed,getmostoftheNREGAjobs,whichessentiallyconsistinlandfillingandother

odd jobs. Inrecentyears,only threeDalitshave left thevillage towork in thepolice, in

Delhi,asconstables.OnehasjoinedtheAirForce270.

270InterviewwithDhaniRam(namechanged),aDalitshoemaker,athisresidence,Dhanauria,17December2012.

234

By their own admission, over sixty per cent of the population is Jat. Other groups

compriseKashyaps(OBC),Muslims,Pandits (Brahmins)andBanias (a fewhouseholds).

Amongthelowergroups,thereareNais,Kumhars,JatavsandValmikis.Mostofthenon-

uppercastenon-Jatcommunitiesaresmallanddonotpossessanyland.Theuppercastes

dobuttheystartedmigratingtocitiesfifteenyearsago,householdperhousehold,leaving

thepoliticalfieldopentotheJatswhosawtheirdominanceincreasetothemeasurethat

uppercasteswereleaving.

In local elections, non-Jat voters tend to vote against the Jats’ preferred candidate (the

seathasbeenreservedformanyyears,generatingfrustrationandresentmentamongthe

Jats).

LocalGujjarshavesucceededtogainemploymentincitiesandhaveusedtheremittances

tobuildsmalllocalbusinesses.Theyownmostofthegeneralstoresaswellasafewbrick

kilns(mostbrickkilnsareownedby Jatsanda fewbyBaniasandGujjars).Theyraised

theirpositionbysellingpartsoftheirlandandinvestingthecapital.

TherelativeprosperityoftheGujjarscreatesjalousies.Oneofourinterlocutors,M.Rana,a

relative of a former Jat Pradhan told us that “Gujjars are richer because they haveNCR

relatives.We[theJats]don’thaverelativesinDelhi”271.

TheseareasareandhavebeenRashtriyaLokDalstrongholdsformanyyears.Chaudhary

CharanSinghcontestedfromChapraulifrom1967to1974,whenhewasattheheightof

hispopularity.WhilemostofourJatinterlocutorswerecriticalofhisson,Ajit,theyhave

largelyremainedfaithfultotheRashtriyaLokDal.

Casterelations–particularlybetweenJatsandDalits–areharshandmarkedbyviolence.

SomeofourJatinterlocutorstoldusthatthebeatingofJatsisstilldoneinChaprauli(but

not inBaghpat,where theyretaliateandwhere thepoliceworks for them).Theyspoke

disparaginglyof their shoemaking, chastising them for fabricating “outof fashion” jutis

(shoes). They blamedMayawati for the necessity of beatingDalits,whom they still call

Harijans.Oneofinterlocutortoldus:

271InterviewwithM.Rana,athisresidence,Dhanauria,17December2012.

235

“Before[Mayawati],therewasnoneedtobeatthem.Theywerenaturallyterrified”272.

EveniftheygettopicktheirDalitcandidateinPanchayatelections,theystillresentbeing

representedbyanSC.“Howcanwemeetourrelativesinothervillages,whenourPradhan

isanSC?”,confidedthesameinterlocutor273.

Caste violence and violence related to elections isn’t new in Dhanauria. In the 1988

Panchayatelections,RameshChandBhairagi,anOBC,defeatedVirender,aJatcandidate.

LocalJatsgotangryandattemptedtomurderhim.Theyblockedaccesstotheirfieldsto

hiscastefellows,preventingRamesh’ssupportersfromreachingtheirworkplace274.The

army (more likely the paramilitary) had to be called to pacify the village. In the next

election,theJatsdistributedmoneyandthreatenedpeople,inordertodividethenon-Jat

vote.TheyevenbribedacousinofRameshChandtorunagainsthim.

When I askedhimwhether thiswas allworth it, he smiled first, and then toldme that

beingPradhanprovidesaccesstoalotofresources,muchmoresoafter1998,whenthe

decentralizationreformsstartingproducingtheireffects.BeingPradhanhasoverallbeen

a profitable experience. He invested in buffaloes and sold land to send his children to

Delhi,togeteducated.HiseldersonisnowapropertydealerinDelhi(andhasbeencited

inamurdercase).Whenhesoldhisland,hemadesurethathesoldittooutsiders,andnot

tolocalJats.

Localelectionsmatterbecausewhatisatstakegoesbeyondthematerialbenefitsthatan

individual and his community can gain from winning. From the dominant groups’

perspective,controllinglocalinstitutionsisconsideredtobeanentitlementowedtotheir

status andposition.The languageused todescribe this entitlement constantly refers to

theircontrolofinstitutionsasanaturalstateofaffairs.Fortheindividualcontesting,the

electionisaboutpersonalprestigeandstatuswithinthecommunity,beforematerialgains.

272InterviewinDhanauria,17December2012.273InterviewinDhanauria,17December2012.274InterviewwithRameshChandBhairagi,athisresidence,Dhanuria,17December2012.

236

From the perspective of the dominated, the election is the opportunity to challenge

oppression and contest the dominants’ claim over local institutions. I find a lot of

resonance with Mukulika Banerjee’s argument that elections matter first and foremost

because theyrepresenta rareandvaluableopportunity toexperiencepoliticalequality,

regardlessofwhoiscontesting(Banerjee2007).

However,seenfromthevantagepointofcandidatesandtheirsupporters,theelectionis

firstandforemostagamewhosestakeislocalterritorialcontrol.Thosewhosucceedto

‘dethrone’ a dominant group candidate tend to replicate the same proclivity for elite

capture. This tends to show that if voters are drawn to the polling booth by their

attachment to political equality, the outcome of the act of voting rarely offers the

possibilityofanenactmentoftangiblepoliticalequality.

Localcontestsarealsomorecomplicatedthanasimpleconfrontationbetweendominant

and dominated.Dominant groups are often divided into factions,who tradewith other

groupsinordertosurpasstheirrivals.

Also,materialgainsfromofficearenottobediscardedatall.Forone,thedecentralization

reformsintroducedbythe1992’sseventy-thirdAmendmenttotheConstitutioninitiated

aprocessofdevolutionofpowerandresourcestoPanchayatiRajInstitutions(PRI).The

implementation of various schemes – and the resources that come with them – also

passes through local institutions. But more importantly, control over local political

institutions isamean fordominantgroup toenterabroaderpolitical arena, to set foot

intodistrict-levelpolitics,wheremoreresourcescanbetappedforthebroadobjectiveof

development.Itisalsoameanstoengagewithotherpublicinstitutionsandactors–the

police,theSub-DivisionalOffice–fromapoliticalandlegitimatestandpoint.Landowning

groups or those engage into various businesses or contracting must frequently engage

withpublicauthorities.Havingthestatusofvillagerepresentativehelpsinthosedealings.

When Iaskedvariouspradhans and local caste leaderswhether theywerecoordinating

with theircastebrethren fromothervillages todeterminehowtheywouldvote inZilla

Panchayat or state elections, they toldme two things. The first is that there are formal

consultation mechanisms taking place ahead of elections between chosen caste

representatives. Discussions take place at the occasion of caste meetings (or Khap

237

meetings, in thecaseof the Jats).Localpoliticiansor theirassociatesoftenattendthose

meetings,tokeeptrackofpeople’svotingintentions.Thesecondaspectisthatincaseof

local alliances between groups, a particular caste may trade support to another caste’s

candidateforthePanchayatelectionagainstthatcandidate’scastesupportfortheirown

candidateinanotherelection,aswesawinthecaseofAmroli.

Politicsasaninstrumentofterritorialcontrol

Theselocalobservationsmayseemdisconnectedfromstatepoliticsorconstituency-level

political considerations. They are however important to keep in mind for two main

reasons.First,itisatthatlevelofobservationthantheinterweavingofcaste,community,

politicsandtheeconomyisvisible,tangible.Andsecondly,constituency-levelpoliticsisin

finetheaggregationoftheselocalcontexts,theproductoflocalquidproquoandalliances

betweengroups,whooftentradetheirsupportinoneelectionforsomeoneelse’ssupport

inanother.Itisthereforeimportanttograspthedynamicsatworkatthevillagelevel.

What I learned from these various cases is that political competition remains deeply

shapedbyagonisticcasterelations,andthatsocialandpoliticalalliancesareformedalong

the fracture lines of these caste-based antagonisms. Caste distinctions remain deeply

entrenched and other forms of social stratification remain largely organized around

castes.Thisdoesnotmeanhowever that the trajectoryof families, their riseand fall in

termsoftheirland-wealth,andtheirotherresources,arenecessarilytiedtothecastethey

belongto,orcastedirected275.

I also learned that local politics is as much about preserving one’s social status and

positionorcontestingsomeoneelse’ssocialstatusandpositionthanfightingforaccessto

resources.BothinfactgohandinhandandconstitutethebasicelementsofwhatJeffrey

Witsoe, inhisworkonBihar,astermedas“territorialdemocracy”(Witsoe2009,2012),

which he defines as “theways inwhichelectoraloutcomesare influencedby relationsof

dominanceandsubordinationwithinspecificterritorialspacesand,conversely,thewaysin

whichelectoralpracticereinforces,andcanevenproduce,territorialdominance”.

275IamthankfultoPhilipOldenburgforpointingthisouttome.

238

AccordingtoWitsoe,theinterplaybetweenelections,electoralpractices,andrelationsof

dominationandsubordinationaccountsfortheviolencethatmarkslocaldailypubliclife

andpoliticalprocesses,suchaselections,inparticular.

“Elections, therefore, tend to not be about competing policy platforms or

ideologies;fromtheperspectiveofthedominant,electionsareaboutmaintaining

standing in thevillage, controlling labor,andensuringcontinuedaccess to state

patronage; for the subaltern, elections are about challenging the dominance of

oppressive landlords (to whom one’s parents and grandparents may have also

been bonded in subservience), demanding minimum wages (that are already

legallymandatedbutnotenforced),preventing indiscriminateviolencebyupper

castes,andassertingone’scaste’slong-trampledhonor”276.

InthecaseofUttarPradeshtoo,localpoliticsisfundamentallyaboutterritorialcontrol,or

themaintenance,expansionorchallengetosomeoneelse’sholdoveraparticularterritory

–avillage,atown,atehsil,anAssemblyconstituency,adistrict.Themorelocal,themore

clear-cutorentrenchedcasteantagonismsare.Thehigherwego,themorecomplexand

diffusethingsbecome,duetothemultiplicityofactorsandduetothecompetitivenessof

theelectoralprocess.

Territorial control passes through control over resources (such as land), as well as

through social and political dominance. These three aspects of influence are self-

reinforcing. The resentment against reservations is high because reserved seats in

Panchayat elections areoften seenby thedominant as an affront and as a challenge to

theirauthority.

As the region developed economically, the sources of political and economic influence

havechangedanddiversified.Thecontrolovereconomicassetssuchasbrickkilns,petrol

pumps, cold storages or transport companies is crucial as it enables local dominant

groupstoexertpressureand/ordeveloppatronagevis-à-visthepopulationwhodepends

fromthosesectorsofeconomicactivity.

276Ibid.,p.66.

239

Atthesametime,thediversificationoftheeconomyandurbanizationinparticularcreate

new opportunities of employment to subaltern groups, who can develop their own

political agency from the autonomy they gained vis-à-vis the local dominant groups.

Access toeducationhasenabledmany individuals to join thecivil services. In this case,

economic autonomy precluded political agency, something that individual from other

groupsreadilyadmit.“MayawatihastransformedtheJatavsintoJats”wasarefrainthatI

wouldoftenhearinmyinterviews277.

Ialsolearnedthatpoliticizationwasnotenoughtogainactualpoliticalpower.Inthecase

ofAminagar,socialandeconomicinequalitieshavepersistedandpoliticaldominationhas

faded. However, local dominant group still capture local democratic institutions by

supporting proxy candidates and running these institutions in their stead and to their

benefit. An illustration of that came from the fact that our interview with the Pradhan

quicklyturnedtoafarce,asBrahminandGujjarmenquicklysurroundedus.Theyinitially

whisperedanswerstothePradhan’sears,keptcorrectinghisanswers,andafterapoint,

didnotevenbothertolethimspeakandansweredtoourquestiondirectly.

Finally, thiswholeprocessofpolitical competition isdeeplymarkedbyviolence. In the

elevenvillagesthatweinvestigated,ninehadarecentstoryofviolencetotell.Mostcases

involved murders, for motives of land, caste or family disputes (and sometimes a

combination of these two or three). In an extreme case, a young twenty-four year old

Pradhanadmittedthathehadmurderedhisowngrandfather(whobyhisadmissionwas

very old anyway) in order to frame the leader of a rival faction. As we saw in the

introductionof thisdissertation,daily life inUttarPradeshremainsmarkedbyconstant

forms of social, political and religious tensions that can turn banal incidents into full-

fledgedconfrontations.

Theriseofbusinessmen-politicians

Politics at the assembly constituency level does not fundamentally differ from local

politics. I find that themotives for political engagement of contestants froma business

277Interview with the brother of former Pradhan of Bhurbaral village, near Meerut, at hisresidence,on13November2012.

240

background similarly consist in using politics as a tool to expand individual or group-

basedsocialstatusandeconomicinterests.

We saw in chapter 4 that in the 2012 state election, over sixty per cent of the MLAs

declaredtobeingengagedinsomeformofbusinessactivity.Wealsoknowthatathirdof

those are engaged in real estate or construction. For the rest, the business category

remainsvague.

A cursory look at the economic profile of some of the main parties’ contestants across

severalconstituenciesrevealsthattheeconomicprofileofMLAstendstobequitespecific.

Theyusuallybelongtothesectorsofeconomicactivitythathavebeengrowingthemost

inrecentyears,andthatarekeytothedevelopmentofothersectorsofeconomicactivity.

InthecityofLoni,threeofthefivemaincandidatesinthe2012electionswererealtors.

The RLD candidate, Madan Gopal, alias Madan Bhaiya, a Gujjar, is a moneylender who

specializesinpropertydisputes.Heisreputedtohaveamassedvastpropertiesinthearea

andisknownasalocaldon.TheBSPandtheSPcandidates,ZakirAliandAuladAli,are

also both property dealers. Aulad is Zakir’s uncle. Zakir, 34 year old, has strong Delhi

connectionsandisacquaintedwiththeBSPMuslimfigureNaseemuddinSiddiqui.Aself-

declaredagriculturist,Alilivesfrompropertydealingsanddeclared16Croresofassetsin

his2012nominationaffidavit.

Loni isacityofhalfamillion inhabitants, locatedat thenorthernborderof thestateof

Delhi. It is the gateway to the capital for those travelling on the Highway 57, from

Saharanpur,Shamli,BarautandthenBaghpat.Asabordertown,itisamajortransitpoint

forgoodsandtrade.Majortransport,storageandlogisticscompaniesoperatefromthere.

Loniisalsotheterminalpointoftheworld’slargestgaspipeline,connectingthegastown

ofJamnagar(Gujarat)toDelhi.Majorgascompanieshavesettheirdistributionterminals

thereandmostofthecookinggassuppliesforNorthIndiaoriginatesfromLoni.Duetoits

strategic location, thecityhasseenaboomof therealestatesectorandof its industrial

activity.

Thecity,alongsideGhaziabad, isalsohost toanumberofcriminalorganizationswhose

membershavemigrated from the ruralhinterlandsover thepast twodecades.Thecity

offersthemanonymityandopportunitiestoapplytheirmethodstolucrativebusinesses.

241

Thus,businesscompetitionhasbecomequitecriminalizedandviolent.MadanBhaiyahas

beentheobjectofseveralattacksandhasbeenaccusedofviolentretaliations.In2001,his

bodyguardwascriticallyinjuredafteraattemptonhislifeinbroaddaylightinGhaziabad.

Thefourassailantswerelaterfounddead,allegedlykilledbyangryvillagers.In2012,he

contested with four criminal cases on his head, including murder, rioting and criminal

intimidationcharges.In2007,hecontestedwith59casesagainsthim(Hindu2007).

MadanBhaiyaisafigureamongthelocalGujjars(heisnicknamedthe“GujjarTiger”278)of

the localunderworld inLoni.Based inGhaziabad,hehails fromanearbyvillagenamed

Jawli.Hestartedhiscareerintheearly1980sasasecondknifeinalocalgang,headedby

Sunil Tyagi and Mahinder Fauji. He gradually rose within the organization as Tyagi’s

closestlieutenantsdisappeared,eitherkilledorarrested.Hemadeanameforhimselfas

theleaderofanabductionracket,inwhichlocalbusinessmenwerekidnappedforransom.

MadanBhaiyacontestedandwonforthefirsttimein1991,intheseatofKhekra(Khekra

becameLoniafterthe2008re-delimitation).Thatyear,heshowedupattheDM’sofficeto

filehisnominationpapers,accompaniedby2,000supporters.Gunshotswereexchanged

as the police tried to keep them outside the office. In protest, Madan’s men went on a

rampageinthestreetsofMeerut,thethendistrictheadquarters.

Madancontestedeverysubsequentelectionandwonthreeothertermsin1993,2002and

2007. After his 2002 victory (he contested as an Independent), he went undercover,

wantedbythepoliceinaseriesofmurdercases,notablyforthemurderofarival,SriRam,

alsoknownasSiriyaPahalwan,neartheAshokHotelinDelhi,barelytwohundredmeters

awayfromthePrimeMinister’sresidence(hewouldalsobechargedunderTADA).Rumor

had it that he developed ties with the BSP in order to trade his support against police

protection.HewasarrestedinFebruary2003butwasimmediatelyreleasedonbailbya

DistrictJudgeinMeerut.Hecontestedagain,andwon,onanRLDticketin2007.In2012,

helosttoZakirAli.

Fifty kilometers North sits the town of Baghpat, headquarter of the recently created

eponym district (carved out from Meerut district in 1997). It is surrounded by vast

278Oneofhiscampaignsloganis“DekhoDekhoKonAayaGurjarBiradriKaSherAaya”(“Look,lookwhoiscoming.It’sthetigeroftheGujjarcommunity”).

242

stretchesof agriculture land, essentially sugarcane,wheat and rice.Themain industrial

activitycomes fromtheBaghpatCooperativeSugarMills, locatedat theoutskirtsof the

town,ontheroadleadingtoMeerut.

From themid-1980s to2012, the seatofBaghpatwasheldby the localnawab,Kaukab

Hameed Khan, son of former Congress MLA Shaukat Hameed Khan. He contested eight

times, starting in 1980 and won five mandates under four party affiliations (Lok Dal,

Congress, BKKGP andRLD).Hismain opponent through the 1990swasVedPrakash, a

farmercontestingonaBJPticket.In2002and2007,hismainopponentwasSahabSingh,

aJatcandidatewhomadehismoneyonlanddealsandpropertydisputes.Singhcontested

in2002onaBSP ticket and in2007ona SP ticket.TheBSP candidate that yearwasa

BrahminnamedMukeshPandit,alsoknownasGudduPandit.Panditstartedhiscareeras

a puncture mechanic in Baghpat and went on to work as driver for a famous criminal

politician,AmarmaniTripathi.Heusedthewealthgatheredduringhisyearsofserviceto

invest inpropertiesandnowlivesasapropertydealer inBaghpat.Overtheyears,both

Congress and BJP have nominated outsider candidates (candidates who hailed from

Baghpatbutwerenotresidingthem).

In 2012, Kaukab Hameed Khan was defeated by Hemlata Chaudhary, wife of Prashant

Chaudhary,alocalfigureoftheGujjarcommunity,BSPpolitician,andaformerMemberof

theLegislativeCouncil.TheChaudhary’srunasuccessfulcontractingbusinessthatdeals

with sand, mud and road digging. They also are property dealers. Khan described her

opponent tomeasanewcomer inpolitics,andattributedhisdefeat to the fact thatshe

and her husband had outspend him many times during the campaign (he mentioned a

figureinseveralCroresofRupees).HelamentedthattheflowofmoneytheChaudhary’s

injectedintothecampaignmadehimpowerless279.

The seat of Muradnagar, northeast of Ghaziabad, is a Tyagi stronghold. Eight of the 10

winnersandrunnerupofthelastfiveelectionshavebeenTyagis.Theseatwasheldfor

eighteen years by Rajpal Tyagi, a property dealer who invested in schools and petrol

pumpsafterhisfirstelectionin1989.In2002,aGaddiMuslim(backward)namedWahab

Chaudhary,challengedRajpal,contestingonaBSPticket.Wahablostbyamarginof3,000

279Interviewwiththeauthor,KhurshidManzil,Baghpat,15November2012.

243

votes. After his defeat, the BSP appointed him Chairman of the Muradnagar Municipal

Corporation.Fromthisposition,hedefeatedRajpalin2012.

Five years earlier, Wahab was a newcomer to politics. He had declared then to be an

agriculturistwith30,000Rupeesasassets. In2012,hedeclaredowning3.3hectaresof

agriculture land, a commercial centerwith forty-two shops, and residential property in

Delhi for a total worth of 9.2 Crores of Rupees. The resources he gathered during his

tenure as Chairman helped him consolidate his position among his community, who

backedhimstrongly.AddedtothesupportofalargechunkoftheDalitvotershissocial

baseandhisresourcesmadehimanMLA.

OnecouldmultiplyexamplesbutwhatweseehereisthattheelectoralgameinWestern

U.P.seemstofavorcandidatescomingfromfairlyspecificeconomicbackgrounds.Mostof

the businessmen politicians in Western Uttar Pradesh tend to come from a limited

number of sectors of economic activity: property dealing and construction, contracting,

transportcompanies,brickkilns,liquorproductionanddistribution,ortheownershipof

assetssuchaspetrolpumpsorcookinggasdistributioncompanies.Besides,manyofthem

havebuiltorowneducation institutions.Therearealsomanufacturersofvariouskinds

buttheytendtobelesserinnumbers.

This similarity of business background is not coincidental. In fact, these sectors share

three important characteristics, which also help us understand why these

businesspersonsshouldwantto‘invest’inpoliticsinthefirstplace.

The firstcharacteristic is that thesesectorsand theiractivitiesareheavily regulatedby

the state, essentially in the form of licensing and tenders. According to the scale and

investment amounts involved, district or state-level authorities preside over the

allocationofvastamountsofpublicresourcesintheformoftendersandlicenses.

Despiteliberalization,statecontrolovereconomicactivitiesremainsstrong.Duetoalack

ofprivateinvestment,theorganizedprivatesectorissmallandalotofprivateactivities

depend fromstate regulations and interventions.Thismeans thatbusinessowners and

local industrialists must engage with the state – and its actors – on a frequent if not

244

constantbasis.Gettingintopoliticsorgettinginacquaintancewithpartiesandpoliticians

isameanstonavigatethatinterface.

Thesecondcharacteristicisthatthesesectorsofeconomicactivityputtogetherconstitute

whatcouldbetermsas thebackboneof the localeconomy. Inother terms, theyarethe

sectorsofactivityonwhichothersectorsaredependent.Overthepasttwodecades,many

householdsinWesternU.P.havebeensellinglandtoinvestinmoreproductivenon-farm

activities. Be it building schools or small workshops, starting shops or dispensaries, all

thesesmall investorsneedthesamecommodities:bricks,sand,mortarandpebbles,etc.

So do public and private infrastructure projects. Farmers depend from transport

companies tocarry theirproduction to themillor to thedistributionmarket.Theyalso

depend from cold storage owners to store their produce before they get distributed.

Liquordistributorsusuallyenjoylocalmonopolies,orcontrollocalsegmentsofthelocal

distributionmarket.

Thosewhoownorcontrolthesekeysectorscaniftheywishderivepoliticalcapitalfrom

these, for they are not only lucrative, the also generate employment and create

dependency by providing indispensable services to a large number of households and

economic agents. They constitute an ideal position to start building patronage or

clientelisticnetworks,byredistributionorbyassociation.

The third characteristic is that despite the amount of state regulation, these sectors of

economicactivitiesaredeeplycriminalized.Likemostoftheinformalandblackeconomy,

theyrestessentiallyoncash-basedtransactionsandgeneratevastamountofblackmoney.

Thecompetitionforpublicandprivateresources isharshandoftenviolent.Thecontrol

overtheseactivitiesisoftenenmeshedwiththecontrolexertedbylocaldominantgroups

–andtheircompetitors–overlocalterritories.Theinfluenceofagroupderivesnotonly

from demographics or land ownership but also from the control they exert over local

political and economic institutions. Competition over such control through engagement

intothedemocraticprocessisameanstomaintainorchallengethepositionandstatusof

localdominantgroups,dependingonone’svantagepoint.

Fewpoliticiansexemplifyhowapoliticalcareercanbeusedto furtherprivate interests

more thanHaji Iqbal, abackwardMuslimbusiness figureborn in the townofMirzapur

245

(Behat Tehsil). Iqbal rose within the BSP and developed a local business empire that

stretchesfromminingtoconstruction,includingthesetupofalargeprivateuniversityin

Saharanpur.

Haji Iqbal started his career as a small trader, selling utensils from village to village in

Saharanpur district. His father helped him set a permanent store, which he and his

brothersAarifandMahmoodusedasabasetoexpandanddiversifytheiractivitytowards

the smuggling of timber, notably of poplar trees, a lucrative illegal activity that has

proliferatedinthisarea,setinthevicinityoftheforest-coveredHimalayanfoothills280.

Iqbalcontestedthe1996electionsinMuzaffarabad,Saharanpurdistrict,onaBSPticket.

He lost to the incumbentSPMLAJagdishSinghRana,aRajput.His inductionwithin the

BSP however enabled him however to expand his business interests, and develop

connections that would help him develop his activities further, notably through the

protectionofhisillegalactivitieswhentheBSPwasinpower.

Inthelate1990s-early2000,HajiIqbalexpandedhisbusinessactivitiestosandandstone

mining (pebbles), both illegal activities in this part of the state. He and his brother

invested the money they made into real estate and companies, allegedly created to

launder their massive wealth. They appointed friends and relatives of other local and

state-level politicians in the boards of these companies, and further developed their

politicalnetworks.

Within the BSP, Iqbal became close to one of Mayawati’s confidante and Minister for

FamilyWelfare,BabuSinghKushwaha.Kushwahawasexpelled fromtheparty in2012,

duetohisinvolvementintoamassivescamaroundtheNationalRuralHealthMission281.

According to media reports, some of Iqbal and Kushwaha’s relatives were business

associatesinaDelhi-basedcompany.

280GlocalUniversity(http://www.glocaluniversity.edu.in).281This massive scam involves massive embezzlement from NHRM funds (allegedly near 1.5billion USD), destined to health care delivery in rural areas, by a nexus of politicians andbureaucrats associatedwith theBSP,underMayawati rule.TwoMinisters, includingKushwaha,were indictedaswellasanumberofbureaucrats.BetweenOctober2010andFebruary2011,6civil servants, including four Chief Medical Officers, were murdered or died in unexplainedcircumstances.Foradetailedaccount,see(Bhalla2012).

246

In 2009, Haji Iqbal was nominated by the BSP as a Member of the Legislative Council,

Uttar Pradesh’UpperHouse.His brother,MahmoodAli, also became anMLC,while his

wifeisanelectedmemberoftheSaharanpurZilaParishad282.

In2011,hesetupamassivecampusona300acresplotintheShivalikFoothills.Healso

runsintermediatecollegesinSaharanpur,Deoria,andBadshahiBagh,ontheborderofthe

Rajaji National Park. When asked about his credentials, Haji Iqbal usually introduces

himself as an educationist. His personal description on his Facebook page reads: “Haji

MohdIqbal isawellknowSocialworkerandeducationreformerofDistt.Saharanpur,His

contributiontoeducationandsocialworkwillberememberedforgenerations”283. Similar

hagiographicdescriptionscanbefoundonvariouswebpagesdedicatedtohim.

Haji Iqbal’s term in theLegislativeCouncil ended in January2016.Perhapsas a signof

howfastonecanlosepoliticalprotection,IqbalbecamesubjectedtoaCBIinquiry,under

theallegationthathehadamassedcloseto10,000CroresofRupeesof illegalmoney284,

and created 111 shell companies to launder it (Mahapatra 2012). In his 2009 affidavit,

Iqbaldeclaredassetsworth6.5CroresofRupeesanddidnothaveanycriminalcharges

againsthim.

Hiscasemaybeparticular,notablyowingtothescaleofhisoperationsbutitiscertainly

notuniqueandheexemplifieswellthenexusthatcanexistbetweenapartyliketheBSP

and localbusiness figures.Bothdevelopamutuallybeneficial relationshipbasedon the

trade of position, influence and protection, against resources for party and campaign

funding. Haji Iqbal acquired the protection of the BSP by funding it, and developed his

own patronage networks within the party by supporting financially a number of BSP

MLAs in the Behat area. He also exemplifies how local political figures expand their

political grasp by having their close relatives elected in positions at various levels of

representation,includingthedistrictadministration,throughtheZilaParishadCouncil.

282IthankRajkamalSinghforprovidingthisinformation.283https://www.facebook.com/HajiMohdIqbal/info?tab=page_info(accessedonJuly13,2016).284Approximately1.34billionEuros.

247

ItiseasytounderstandwhyapartysuchastheBSPmaybeinterestedinnominatingan

individualsuchasHajiIqbal.Putitsimply,hehastheresourcestofundhisowncampaign

andcontributetopartyfunding.Healsohasthecapacitytobuildafollowingthroughhis

business and patronage networks. And lastly, he also likely to respond to the various

incentivesthatapoliticalcareer–howevershort–canoffertoindividualsofhisset.

These incentives work in different ways, or offer various types of benefits. Developing

political ties enables businessmen to acquire protection, from the police and from

competitors.Italsoprovidesaccesstonewresourcesthatcanbeusedtofurtherdevelop

businessandpatronagenetworks.Politicsalsoservesasastatusenhancer,amarkerof

powerandinfluencethatgoesbeyondtherestrictedrealmofpolitics.

The riseofbusinessmen inpolitics thusmeans that aprocessof integrationofpolitical

and economic elites is taking place. While political representation becomes more

heterogeneousonthebasisofcaste, italsobecomesmorehomogeneousonthebasisof

theclassbackgroundofthecandidates(aswewillseeinChapter6,thisisessentiallytrue

fortheSPandtheBSP,andlesssofortheCongressandtheBJP).

Inthiscontext,politicsbecomesaneffectivemeantofurtherprivateeconomicinterests.

Accordingtolocalcontextsandindividuals,theseinterestscanbeindividualorcollective.

Theycanalsobecaste-basedorcross-castebased.

Onecouldarguethatithasalwaysbeenthecaseandthatthetraditionaluppercasteelite

individuals contestingonaCongressor JanaSangh ticket in the1950sand1960swere

also‘inpolitics’asameanstofurtherprivatecasteorclassinterests.

WhathaschangedinWesternU.P.,andindeedinotherpartsofUttarPradesh,isthatthe

composition of these elites has changed. Local economic elites have become a more

diverseorheterogeneousensemble.Also, someof the fastest-growingeconomicsectors

havebecomemoreandmoreintertwinedwiththestate,thestatecontrollingamajorpart

of available resources for business development. It is therefore no surprise that a new

class of politicians as emerged from these sectors.Getting intopolitics is away to stay

competitiveinaharshagonisticsocialenvironment.

248

One of the results of economic transformation in Western U.P. is that inter-group

competitionhasincreased,assegmentsfromgroupsthatwerehistoricallylaggingbehind

have become upwardly mobile. This enhanced competition has also affected relation

betweengroupsthathadahistoryofcollaboration,suchastheJatsandtheMuslims.

WhileMuslimsoverall lagbehindothergroupsintermsofsocio-economicdevelopment

(Sachar 2006), segments among them have gained from the economic transformations

thathave takenplace inWesternU.P.Particularly among thebackwardMuslims, a tiny

elitehasbenefitedfromthegrowthofsectorssuchastrade,manufacturing(metalwork,

brassandglassindustry)orthemeatindustry285.

BackwardMuslimsaspiringpoliticiansfoundintheBSPapartythatwouldprovidethem

withspaceandopportunitiestocontestelectionsonstrongtickets.Theconsolidationof

backward Muslim support behind BSP candidates partly accounts for the good

performanceoftheBSPinthisregion(Heath,Verniers,andKumar2015)286.

Theyhavealsousedtheirdemographicadvantageincities,wheretheyfrequentlymake

upmorethanthirtypercentofthepopulation,toconquermunicipalities(Verma2012a)

aswellasZillaParishads.

TheeconomicandpoliticalriseofMuslimsinparticularhasgeneratedresentmentfrom

rivalpartiesandsocialgroups,whichtranslatedintoariseofcommunaltensionsandtoa

full-fledged riot in Muzaffarnagar in August and September 2013 (Berenschot 2014,

Chishti2013).

SixmonthsaheadtheMuzaffarnagarriots,aBJPMLAfromtheareahadtoldmehowthe

issue of (cow) meat and Muslim assertion were seen as a sensitive provocative issue,

generatingangeramongstHindus:

285Indiahasbecome theworld’s secondexporterofbeef (buffalomeat)anda largepartof thattradeislocatedinWesternU.PIfsomeofthemaintradersareHindus,theprofessionofbutcherremainsassociatedwithMuslimcommunities.286Inthe2012stateelections,theBSPremainedaheadoftheSPonlyinWesternUttarPradesh.

249

“ThebiggestbusinessinWesternU.P.,aftermining,iscowslaughter.Youcanbuy

acowfor4,000Rupeesandsellitcutfor20,000.Cowslaughterisbannedbutthese

meat-processingplantsarehugelyprofitable.Theyarerunbystrongmen(…)This

createsadeepdivideamongthepeople”287.

My Muslim interlocutors in Aminagar had explained how their political situation had

changednowthatmostpartieswerecourtingtheMuslimvote:

“Muslims are not scared. They have their own will. Muslim leaders have joined

differentparties.Theyareprotectedbecausetheyareavotebankforeveryparty,

excepttheBJP”288.

Itisgenerallyexpectedthateconomicgrowthwouldleadtotheattenuationofsocialand

communal tensions. In reality, the assertionofdeprivedgroups tends to increase these

tensions. Moreover, the fact that economic change, crime, and the antagonistic social

interplay of castes and communities all converge and get intertwined in the field of

electoralpoliticsbreedthosetensionsratherthandefusethem.Partypoliticsalsoplaysa

contributiverole,throughtheactivationofsocialandpoliticaldivisionsamonggroups.

5.1.2.ThecaseofEasternU.P.:morecontinuitiesthanchange

EasternU.P.contrastsvastlywithWesternU.P.assomeofthemajorprocessesofchange

described earlier have not taken place in that region. Eastern Uttar Pradesh, and the

North-East in particular, are poor, under-developed, poorly connected and is, with

Bundelkhand, one of the least urbanized part of the state. Poverty reduction has been

slower than in the rest of the state, particularly among the Dalits and Muslims who

represent a higherproportionof theBPL (below thepoverty line)population than any

othersub-region(Akarsh2015).

In2012,byCSDSestimates,EastandNorth-easternU.P.togethercounted86ruralseats,

11 semi-urban seats and only 4 urban seats (three inVaranasi and one inGorakhpur).

287Interview with P. Singh (name changed), BJP MLA from Rohilkhand, in Delhi, 10 February2013.288Interview with a former Muslim Pradhan’s relative, Bhurbaral urf Aminagar village, nearMeerut,13November2012.

250

Thatis85,11and4percentrespectively,against59,25and16percentrespectivelyfor

WesternU.P.Asperthe2011Census,EasternUttarPradesh is themostpopulatedsub-

regionofthestate(nearly40percent).

The economy of Eastern U.P. remains largely based on agriculture. The few industrial

sectorsthatflourishedinthisregion–sugarandtextile–havedramaticallydeclinedover

thepasttwentyyears,owingtoa lackof investmentandunderthepressureofnational

andinternationalcompetition.Theregionthatusedtoleadthestateinsugarproduction

has seen many of its sugar mills closing, generating unemployment. As a result,

outmigration ishigh, creatingadditionalpressuresonwomen,who tend to staybehind

(Thelmaetal.2005).

SudhaPaidocumentedinthemid-1980showthelargelandlordssucceededinmitigating

the impactof successive landreformsandmaintained therefore theirpolitical influence

(Pai 1986). As a result, Eastern U.P. has not seen the kind of political empowerment

inducedbyshiftsinlandownershipasWesternU.P.experienced.

Lieten and Srivastava describe how agriculture productivity in the East, before

Independence,usedtosurpassthatoftheWesternregion,duetothefertilityoflandand

abundance ofwater resources. They also describe how the combine effect of the green

revolutionandlandtenancyreformsintheWesthaveledtoareversalofthatsituation.

Agricultureandlandreformsalsoledtotheempowermentofthemiddlepeasantryinthe

Westandtoareshufflingofthecardsofpowerbetweenerstwhiledominantgroupsand

newly ascendinggroups (LietenandSrivastava1999,85-90).Thisprocessdidnot take

placeintheEast,leavingthelandowninggroupsinapositionofpreeminence.

Land in the East has remained unequally distributed between big landlords – mostly

uppercastes–medium-largelandowners–mostlyOBCs–andamultitudeofoccupancy

tenants,who could gradually acquire small parcels (between1 and3 acres, orbelow1

acre)289. As a result also, agriculture has remained concentrated around food crops,

mostly forsubsistenceagriculture, rather thandiversified towardsmoreprofitablecash

crops.

289Ibid.,p.86.

251

Theconsequenceofthisstagnationisthattheoldeliteshaven’tbeenchallengedtheway

theyhavebeen in theWesternparts of the state, leading to the resilienceof theupper

castes.

ThisisparticularlytruefortheNorth-East,wherefifteenoutoftheeighteencaste-based

strongholds are held by upper castes, mostly Rajputs (11). In the East, there are six

Muslim strongholds, mostly located within the Azamgarh-Varanasi-Mau triangle. Seven

seatshavealsoconsistentlyreturnedYadavcandidates,scatteredinAzamgarhdistrict,in

Mughal Sarai, and in Phulpur, near Allahabad. Beyalsi, Kolasla and Varanasi South are

Bhumiharstrongholds.

The observation of caste-based strongholds has an indicative value about patterns of

changeattheconstituency-level,butdoesnotinformusmuchaboutanythingelse.Inthis

nextsection, Iexaminemoreclosely thesocio-demographicprofileof someof the long-

standingnortheasternpoliticians,toseewhetherotherpatternsthancastearise.

Theresilienceoftraditionalelites

Aswe saw in chapter4, if theoverall shareof upper casteMLAs in theNorth-Easthas

beendecreasing since1996, it remains that oneMLAoutof two is still anupper caste.

Over thepast fiveelections, thirty-fivepercentof therunner-upshavealsobeenupper

castescandidates(theirsharehasactuallyincreasedoverthepastthreeelections,aftera

dropintheearly1990s).Infact,ifoneonlyconsidersthegeneralseats,theaverageratio

ofuppercastewinnersandrunner-upsoverthepastfiveelectionsincreasesto67and44

percent.

Since1991,37MLAsoutof138havebeenelectedthreetimesormore.Outofthese37,23

are upper caste, 6 Dalits, 5 OBCs and 3 Muslims. The Janata Dal – SP concentrates the

largestnumberofuppercastelongstandingpoliticians(eight,halfBrahmin,halfRajputs),

followedbytheBJP(six,mostlyRajputs).FiveoutofsixCongresslongstandingpoliticians

areuppercastes.TheBSPcountsonlyfoursuchMLAs,twoOBCsandtwouppercastes.

252

BarringthethreelongstandingMuslimpoliticians,whoareallwiththeSP,allothersare

infactquitedispersedbetweenparties.UppercastesarewellrepresentedintheNorth-

East because they are well-represented within all parties, and not because they have

alignedwithanydominantparty.Ifwelookfurther,wecanobserveothercommonalities

betweenthem.

The first one is that none of these longstandingpoliticians is elected in urbanor semi-

urbanseats,whichisnotsurprisingsince90percentoftheseatsinthisregionarerural.

Second, all but five declare agriculture as their profession. Janardhan Prasad Ojha, in

Shyam Deurwa and Abdul Kalam, in Mehndawal, are self-declared traders. Shiv Pratap

Shukla, inGorakhpur,and JagdambikaPal,areboth lawyer.HarivanshSahai, inBhatpar

Rani,presentshimselfasaneducationist.

A cursory look at their assets declaration reveals that they all declare possession of

agriculture land, often beyond three hectares, and that they all draw revenue from the

possessionofcommercialbuildings.Noneofthemdeclareanyindustrialactivity,although

somearereputedtoownfactoriesinotherpartsofthestate,suchasNoidaorAzamgarh.

Intermsofeducationlevel,twenty-twoaregraduateormore.Mosthavestudiedinlocal

universities,inGorakhpur(7),BastiandVaranasi(BHU).AfewhavestudiedinAgraand

Bombayuniversity.

Twenty-onebelongtopoliticalfamilies(14uppercastes,4SCs,2OBCsandoneMuslim),

mostly through horizontal ties (cousins, brothers and sisters-in-law). There are few

verticalpoliticalfamiliesandthesetendtobeeitheroftheroyalvariety(RatanjitPratap

Narain Singh, from the royal family of Padrauna, son of former Union Cabinet Minister

C.P.N. Singh)orof the criminalone (AmarmaniTripathi’s son,orHari ShankarTiwari’s

father).AdozenoftheseMLAsareorhavebeenMinisters,amajorityofthemwiththeBJP

(a largenumberofBJP’sMinisters,bothat thestate levelor intheUnionCabinet,come

fromEasternU.P.andfromtheNorth-East).

Infact,prominentpoliticalfamiliestendtocumulatearoyallineageandministerialberths,

such as R.P.N. Singh from Padrauna and Kushinagar, Jai Pratap Singh from Paniara,

253

ShivendraSinghfromSiswa,orVirBahadurSingh,formerChiefMinister.Severalofthese

“royalMLAs”havebecomeLokSabhaMPsandattimesMinisters,likeR.P.N.Singhandhis

fatherbeforehim.

Mostoftheseroyalpoliticalfamiliesdonothavecriminalrecordsandwhentheydo,itis

usuallyforfamilyfeuds.MostofthemaligneitherwiththeCongressortheBJP.Ministers

alignedwith theSPor theBSPtendtobemoreof thecriminalkind,suchasAmarmani

Tripathi,HariShankarTripathi,orRaghurajPratapSingh,alsoaliasRajaBhaiya,inKunda.

Onealso findssimilarprofileamongpoliticianswhohadshortercareersorwho indeed

lostelectionssothisisnottosaythatpossessingthesefeatures–land,uppercastestatus

andtitlesoracriminalrecord–istheonlythingthatgetstheselongstandingpoliticians

elected.Individualqualitiessuchascharismaandeloquence,thedemographicandsocio-

economicconfigurationoftheconstituencyandthefeaturesofthecompetitorsalsoplaya

role,tosaynothingofstatelevelpoliticalconsiderations.Buttheydoappearasnecessary

conditionstoholdlongpoliticalcareers.

JaiPratapSinghisasix-timeMLAfromBansi,Siddharthnagardistrict.ARajputbycaste,

he is theheirof theBansiestate.Agraduate fromK.C.College inBombay,Singhdidhis

secondaryeducationattheMayoCollegeinAjmer,aboardingschoolwellattendedbythe

aristocracy.Heisaself-declaredagriculturistanddeclaredin2012possessing52acresof

land.Healsoowns largecommercialspacesaroundBasti,aswellasamarriagehall.He

alsoownsabungalowintheposhneighborhoodofGomtiNagar,inLucknow.

JaiPratapSinghstartedhispoliticalcareerasanindependent(1989and1991),andthen

contestedsubsequentlyasaBJPcandidate.Hewasbrieflyexpelledfromthepartybefore

the2012stateelections,whenhiswife,VasundharaKumaricontestedthe2012election

in Domariaganj on a Congress ticket (the former Congress MLA from Domariaganj,

JagdambikaPal,wontheseatonaBJPticket).

His longstanding local opponent is an SP politician named Lal Ji Yadav, a medium

landowningfarmerwithfourcriminalcasesagainsthim, for intimidation,electionfraud

(personation),andrioting.SinghlosthisseattoYadavin2007butregaineditfiveyears

254

later against a BSP candidate, Vinay Shankar Tiwari, son of the renowned criminal

politicianHariShankarTiwari.

AnotherlongstandingroyalpoliticianisShivendraSingh,aliasShivBabu,wonfiveterms

inSiswa,1985,1991,1996,2002and2012.Hebelongstothelocalroyalfamily.

Shiv Babu is a ‘party-hopper’. He contested eight times under four consecutive party

affiliations:Congressfrom1985to1996,thentwiceonaBSPticket,asaBJPcandidatein

2002andthenasanSPcandidatein2007and2012.Heownslargetracksoflandsaround

thetownofSiswaandanumberofcommercialbuildings.

ShivBabuhashadtocompeteagainstseveralstrongopponents.Inrecentyears,hismain

rival was a Brahmin named Avnindra Nath Dwivedi, alias Mahant Dubey, a local

strongmanchargedformurderanddacoity,whocontestedundervariouspartybanners

(andwontheseatin2007).TheothereffectivecandidatewasRakeshKumar,amedium

landowningfarmer,whocontestedfirstonaBSPticketandthenonaPeacePartyticket.

In 2012, his cousin, RaghavendraPratap, aliasAnkit Singh, contested against him, on a

Rashtriya Lok Manch ticket (Amar Singh’s short-lived party). Ankit is the son of

Shivendra’suncleandformerstateMinisterDevendraSingh,whowasallegedlymurdered

byShivendra’selderbrother,Manvendra.Ankitfinishedfifth.

TheCongressalsohasitsshareofroyalcandidates.ThemostnotoriousCongressroyalin

Northeastern U.P. is Ratanjit Pratap Narayan (R.P.N.) Singh, a three-time MLA from

Padrauna,whowentontobecomeamemberofthe15thLokSabha,aswellasaMinister

ofState(firstRoad,TransportandHighway,andthenHomeAffairs).Singhbelongstoa

SainthwarRajputfamily,custodianoftheestateofPadrauna.Hisfather,C.P.N.Singh,was

aMemberofParliamentandaformerMinisterofState(Defense)inIndiraGandhi’s last

Cabinet.

R.P.N.studiedatDoonSchool(heisthePresidentoftheDoonSchoolOldBoySociety)and

at St. Stephen’s College, two of India’s premier elite institutions. He ambitioned a

cinematiccareerbuttookthefamilypoliticalmantleafterthemurderofhisfather,byone

255

ofhiscousins.R.P.N.’smother,MohiniDevi,initiallycontestedtheseatbutlostbadly(she

finishedfourth).

WhileanMLA,R.P.N.Singhattemptedtwicetoregainhisfather’sseatofKushinagar.He

wononthethirdattempt,defeatingSwamiPrasadMaurya,oneoftheleadingMBCfigures

of the BSP. He lost his seat in 2014 against the BJP candidate, Rajesh Pandey, son of a

CongressmemberRajmangalPandey.

R.P.N.haslongbeenoneoftheCongress’mostprominentfaceinNorth-EastU.P.Heisa

formerYouthCongressPresident(1997-1999)andAICCSecretary(2003-2007).

Therehasbeenarecentacademicand journalisticattentionpaidto thephenomenonof

political dynasties in India (Aron 2016, Chandra 2016a, Malhotra 2004). Due to the

difficultyofgatheringdataonthesubject,theyhavefocusedonnationalpoliticsandbig

familybiographies,inthecaseofthejournalists290.

In Kanchan Chandra’s book, Jensenius notes that the constituencies that send dynastic

MPstotheLokSabhadonotdifferparticularlyfromthosewhodon’t,butthat“royalMPs”

tendtobelongtopoorerandmoreruralareas(Jensenius2016b,101).Itiscertainlythe

caseinEasternU.P.aswellasinotherpartsofthestate.Butthen,thewholeregiontends

tobepoorandrural.

Itisthereforenotthelevelofbackwardnessthataloneexplainswhyvotersmaywantto

sendtheseroyalpoliticianstotheStateAssembly,butthefactthattheypossessqualities

andattributesthatattractvoters’support:anameandareputation,ahighsocialstatus,

party linkages, resources, a history of closeness with the local and district-level

administration.Inshort,theytendtopossesstheresourcesthathelpwinningelections.

But it is not a guarantee that they position is secure through time. Many ‘dynastic

politicians’loseelections,orfailtogettheirfamilymemberselected.Vasundhara,wifeof

JaiPratapSingh,finishedfifthinhercontestinDomariaganj,withonlyninepercentofthe

290Romain Carlevan, at City University of Hong Kong, is currently undertaking an extensiveresearchofpoliticalfamiliesinMadhyaPradesh.

256

votes). Similarly,manyex-zamindari familieswhouse tohavesomepolitical clouthave

sunkintoanonymity.

Otherlongstandingpoliticianstendtoconformtotheportraitsketchedearlier.Theytend

tobelongtotheuppercaste,declarefarmingastheiroccupation.Theydrawmostoftheir

resources from agriculture and rent of commercial buildings. They belong to various

partiesandnearlyathirdofthemhaveahistoryofsuccessfullyswitchingpartyallegiance,

whichmay indicatethat it is theirattributesratherthanpartyaffiliationalonethatgets

themelected.Thepoliticianswhoconformmore to theportrayoutlined in theWestern

U.P.sectionofthischaptertendtobecriminals,whoareengagedinillegaltradeorexerta

criminal control over economic activities such as brick kilns, transport of raw material

suchassandandstones,miningandquarrying.

The profile of politicians from Western and North-Eastern U.P. do not fundamentally

differfromeachother.Onbothsides,thosewhopossesstheresourcestofightelections

standabetterchanceatgettingelectedthanthosewhodon’t.Cumulatingvarioussocial

and economic attributes – such as a high caste status, an aristocratic lineage, land and

property – helps to become even more competitive. In both sub-regions, local political

elites tend to be drawn from groups and communities who exert control over local

politicalandeconomicalterritorialentities.

Whatdiffershowever is thenatureof the resources fromwhichpolitical capital canbe

drawn, and the distribution of these assets between social groups. In the West, a fast-

developing urban economy and a transforming rural economy had led to the

empowermentofalargenumberofsegmentsofthepopulation,ortotheriseofeconomic

andpolitical eliteswithinmanysocial groups.Evenunderprivilegedgroups suchas the

Muslimsdohaveaneconomicelitewhosesomemembershaveriseninpolitics,including

fromthenon-traditionaluppercasteelitecommunities.

Economic change has contributed to the emergence of a very volatile and competitive

politicalscene,inwhichcaste,votersandpartyalignmentshavebecomeblurred.

In the North-East, the lack of economic transformation has meant that the traditional

landowningeliteshaveretainedmuchoftheirpastinfluence.EveniftheOBCshaverisen

257

intheEasterndistricts,thefactthatallmajorpartiesco-optalargenumberofuppercaste

candidateswithintheirranksensurestheirresilience.

Similarprocessescontributetoexplainalsowhyandhowtheuppercasteshaveretained

muchof theirascendency inAwadh,a formerPrincelyState inwhicha largenumberof

ex-ZamindariandaristocraticfamilieswentintopoliticsafterIndependence.Thispattern

however is not uniform. There are cases of politicians who have risen through the

mobilizationofbackwardvoters.OneexampleisPankajChaudhary,aKurmileaderfive-

timeMPfromMaharajganj.

Chaudhary emerged through municipal elections in Gorakhpur. A traditional Rajput

stronghold,thecityhasbeenformanyyearsundertheinfluenceoftheGorakhnathMath,

areligiousinstitutionandmonasticorderthatrunsseverallargetemplesinGorakhpuras

well as in Nepal, in the district of Gorkha. The head of the Math (Mahant) has been

involved inpolitics since the1920s.MahantAvaidyanathhas been a localMLA for five

termsinthe1960sand1970s,onaHinduMahasabhaticket,thenathree-timeMPfrom

Gorakhpur,firstasaHinduMahasabha,thenasaBJPcandidate.ThecurrentMahant,Yogi

Adityanath,succeededhimbothattheheadoftheMathandasaMemberofParliament.

HeisoneoftheBJP’sstrongestandmostcontroversialfiguresinU.P.(Jaffrelot2014).

Pankaj Chaudhary challenged the Rajput’s hold over the municipality by mobilizing its

largeKurmipopulation,andbygatheringthesupportfromsmallergroupswhoopposed

orresentedthepoliticaldominationoftheRajputs.Attheageof25,Chaudharybecamea

memberof theMunicipalCorporation in1989,andthentheDeputyMayor, in1990.He

joinedtheBJPthesameyearandwasnominatedfortheseatofMaharajganj,atraditional

uppercastestrongholds.

HeappliedthereasimilarstrategythaninGorakhpurandsucceededinoustingtheupper

castesfrompower.TheuppercastesinMaharajganjaredividedamongfourmaingroups,

competingwith eachother:Brahmins,Banias,Rajputs andKayasths.Eachgroupwould

try towinover theotherby forming local allianceswithothernon-upper castegroups.

PankajChaudharyconsolidatedsupportamongthenon-uppercastevotersandwonfive

terms(helostin1999).

258

ChaudharystartedhiscareerasasonoffarmerinMaharajganj.HemigratedtoGorakhpur

andbecameabusinessman(henowownshairoilfactoriesinNoidaandAligarh).During

hispoliticalascension,hepushedotherfamilymemberstobeinvolvedinpolitics.Hisown

father,BhagwatiPrasadChaudhary,wasaZilaParishadmember.Hiselderbrotherwas

thefirstZillaPanchayatofthedistrictofMaharajganj,createdin1989.Hissister,Sadhna

Chaudhary, contested three times in Shohratgarh (Siddharthnagar district),

unsuccessfully291. She is also a former Zila Panchayat head. He even has cross-border

politicalconnection,sincehisbrother-in-lawisalegislatorinNepal292.Thepoliticalclout

of Pankaj Chaudhary over his constituency is thus completed by the control of local

democratic institutions,whichenablesthefamilytoaccumulateresourcesandputthem

inapositiontoredistributethemclientelistically.

Therearelimitsthereforetothevalueofacross-regioncomparison,asnoneconstitutesa

homogeneous social, economic and political landscape. This furthers the argument that

electoral political is before all local politics. Once again, local configurations do not

necessarily trump state level or national level considerations. Nor do they exhaust the

complexityof social, economic andpolitical antagonisms thatmark the lifeof localities.

Buttheselocalconfigurationofpowerandinfluencedoplayacrucialroleindetermining

thepoliticalsupplyvotersmustchoosefrom;whogetstocontest inthefirstplace,who

getstobemorecompetitivethanothers,andwhomaylastinpolitics.

In thenext section, I examine in furtherdetail thequestionofwhatvoters expect from

their candidates and elected representatives and how the capacity to meet those

expectations determine whether other attributes of political competitiveness – such as

crime–helpornotwinningelections.It isnotsufficienttoexplainwhybusinessmenor

criminalsmightbeinterestedtocontestelections.Onealsoneedstofactorinwhyvoters

wouldalsowanttosupportsuchcandidates.Asimpleanswertothatquestionconsistsin

saying that the candidates who succeed are not those who possess the attributes that

makes them competitive. But equally importantly, successful candidates are also those

whoconformtowhatvotersexpectsfromthem,bothretrospectivelyandinanticipation.

291In2012,shecontestedagainstherformerhusband,ChaudharyRavindraPratapaliasPappu292IamthankfultoRajkamalSinghforprovidingmewithinformationaboutPankajChaudhary.

259

5.2.Whatdoelectedrepresentativesdo?Ajobdescription

IgrewsorichthatIwassentByapocketboroughintoParliament.

Ialwaysvotedatmyparty'scall,AndIneverthoughtofthinkingformyselfatall.

Ithoughtsolittle,theyrewardedmeBymakingmetheRuleroftheQueen'sNavee!

SirJosephPorter293

Anyonewhohasfollowedanelectedrepresentativeonaconstituencyvisitoronaregular

dayofbusinesshasfelttheexhaustionhittinghardbytheendoftheday–andsometimes

earlier.

TheregularlifeofmostelectedrepresentativesIhavemetstartedearly–usuallyatdawn

–andendedlate–oftenpastmidnight.Betweenthesetwomoments,theymighthavemet

andtalkedto literallyhundredsofpeople(sometimesthousands,during thecampaign),

visiteddozensofplaces,homesandvillages,receivedanequallyhighnumberofrequests

from petitioners camping in front of their personal or official residences, and taken a

countlessnumberofcallsfromoneofthethreeorfourhandsetstheypossessandthatare

usuallyhandledbyaidesandassistants,whomustmanageat the same time the fluxof

peopleandofincomingcalls.

TheConstitutionlaysdowntherulesofqualificationstobecomeanMLA294butissilenton

theirduties.Soistherulebookofthestateassembly,whichdescribestheproceduresand

the general mode of functioning of the Assembly and its proceedings, but does not

mentionexplicitlywhatitsmembersareactuallysupposedtodo.TheMLA’soathofoffice

doesmentionthatanMLAwill ‘faithfullydischargethedutyuponwhich[heis]aboutto

enter'but,asanRTIactivistinMaharashtrafoundoutin2011,noonecouldtellexactly

whatthesedutiesformallywere295.

293From"SirJosephPorter'sSong"("WhenIwasaLadIservedaTerm")fromH.M.S.Pinafore,or,TheLassthatLovedaSailor(1878),W.S.Gilbert.294Art. 173 of the Constitution mandates that in order to be qualified to fill a seat in theLegislatureofaState,onemustbe(a)acitizenofIndia,(b)lessthantwenty-fiveyearsofageand(c)possessingsuchotherqualificationsasmaybeprescribed […]byorunderany lawmadebyParliament,suchasbeingaregisteredvoterinthatstate,andsoon.295‘RTIrevealsMLAshavenoduties’,TimesofIndia,29December2011.

260

ItisunderstoodthoughthattheresponsibilitiesofmembersoftheLegislativeassemblies

areoffiveorders:legislative(makinglawsonitemsfiguringintheStateandConcurrent

lists), financial (the Assembly approves the state’s budget and state’s funds allocation),

executive (the Executive is accountable or responsible in front of the Legislature),

electoral (MLAs are part of the electoral college for Presidential and Vice-Presidential

elections)andconstitutional (Somepartsof the IndianConstitutioncanbeamendedby

theParliamentwiththeapprovalofhalfthestateLegislatures).

However,asChoprafoundout inhissurveyof legislatorsacrossfivestates,notasingle

MLAsurveyedinUttarPradeshmentionedanyofthesedutieswhenaskedabouttheirjob

description296.

Whenaskedabouttheirroleandfunctions,mostMLAsandcandidatesIspoketothrough

myyearsoffieldworkhaveusuallyusedthetermseva(“service”)todescribetheirduties

ofelectedrepresentatives.Manyofthemexplainedthatonceelected,theirmaindutywas

toprovidereliefandassistancetotheirconstituents,andthattheirlifebasicallyconsisted

inreceivingpleasanddemandsfordirectassistance,orrequestsformediationindealings

with the localbureaucracy.Someof these requestsare individual.Othersarecollective,

carriedbyvillageorcasterepresentatives,whosometimesstormtheMLA’sofficeingreat

number,asashowofstrength.Villageheadsusuallycomewithpleasforpublicwork,or

tocomplainwhensanctionedprojectsarenotimplemented.AsPaulBrassnoted,elected

representatives in India are often expected to “care for the material interests of their

followers”(Brass1990,96).

It is therefore not surprising that many elected representatives actually declare “social

work” as their profession, often used interchangeably with the “political worker”

denomination.

Many legislators complain about the gruelling routine that they are subjected to when

theyvisit theirconstituencies. InterviewingMLAs in theirconstituenciesalone is in fact

296Chopra,op.cit.

261

virtually impossible, since they are constantly surrounded by aides, assistants, party

workers,visitors,petitioners,andassailedwithphonecalls.

This routine follows them outside their constituencies as well. People travel long

distancestomeettheirrepresentativeinthestatecapital,Lucknow,orinDelhi.Theposh

leafygreenavenuesofLutyensDelhioftenofferthesightofdhoti-cladfarmersknocking

at the door of their representative’s home, early in the morning, before the inrush of

vehiculartraffic.

Thefactthatelectedrepresentativesareexpectedfirstandforemosttosolveproblemsis

areflectionandaninheritanceofthedysfunctionsoflocaladministration,ofthepaucity

of local resources, and of the general difficulty faced by the public to access goods or

servicestheyareentitledto.

Sometimes, this role of facilitation of bureaucratic processes gets institutionalized by a

division of labor established between the local bureaucrats and the politicians. Elected

representativescanattestcertainqualitiesofindividualsinlieuofthebureaucracy.They

canrecommendindividualcasesforpensioncards,rationcards,MNREGAdutiesordues.

Theycanalsoattestthecasteofindividuals,ortheirstatusasawidower,thecertification

ofwhichisvitaltoaccessavarietyofbenefits.Bureaucratsoftensendpetitionerstotheir

electedrepresentatives,forwrittenattestationsorlettersofrecommendations.

InNovember2013,PaulBrass,hiscompanionSueandIweresittingintheantechamber

of a makeshift party office in Meerut, waiting to meet a prominent Muslim Cabinet

Ministerfromthearea.Weweresharingthecrampedroomwithabouttwentyindividuals,

allPradhans(Heads)ofPanchayats from thedistrict.Whenaskedabout thepurposeof

theirintendedmeetingwiththeMinister,theylaughedatthenaïvetéofthequestion.One

ofthemlookedatusstraightintheeyeandsaid“Electionsarelooming.We’veallcomefor

onething(…)pistollicense”(saidhe,afteradramaticpause,andpointingatmewithhis

handraisedmimickingagun).

Theothersideofthatroleoffacilitationistheimperativetoprovideaccesstoresources

toone’sconstituents.Notonlythosewhodependonthestateforlivelihood,butalsothose

whoengagewiththestateforeconomicpursuits.Electedrepresentativesareexpectedto

landprojectsfortheirconstituency,tofacilitatetheirsupportersandassociates’accessto

262

contracts, sub-contracts,public tendersandaswell credit.This is the foundationof the

transactionalrelationthattiesthemtotheirsupportbase–whatisusuallyreferredtoas

patronage(Chandra2004c). Inthisformofquidproquorelation,(electoral)supportis

tradedagainstaccesstoresources.

This role of intermediation of politicians has been well documented in the literature

(Jeffrey 2002, Manor 2000, Oldenburg 1987, Reddy and Haragopal 1985). Politicians

themselvesemployorrelyonlocalrelaysofinfluencebybuildingmutuallybeneficialties

with local dayals (“intermediaries”, or “brokers”), caste leaders, head of local political

institutions, Zila Panchayat Chairpersons, or any other individual susceptible to garner

themvoteinexchangeoffavorsandprotection.

Thesepatronagenetworks,indispensableforthebuildinganddevelopmentofapolitical

career,areusuallybuiltonbothsidesof legality.Theyalsoconstituteapre-conditionto

becomeapoliticianinthefirstplace.

Through my years of fieldwork, I have met scores of aspiring politicians who were

preparing their future candidacy by building their own local patronage networks, by

assistingasittingMLAorMP inbuildingormaintaining theirownpatronagenetworks,

andbydevoting time to “partywork” in thehope to climbwithin theorganization and

attracttheattentionoftheparty’sleadership.

Itishardtoassesswhethertherespectoftheseobligationsandtheactualeffectivenessof

politicianstoactasdayals(intermediaries)aredeterminanttoelectoraloutcomes.Aswas

mentionedearlier,thereareonlyacertainnumberofhoursinadayanddaysinaweek,

andpoliticianscannotpossiblyattendtheneedofevenasignificantportionoftheirvoters.

But I have alsomentioned earlier thatnot conforming to that role is a surewayof not

goinganywhereinpolitics.

5.2.1.Sevaasapoliticalandsocialobligation

Thistransactionalrelationthatbindsarepresentativetohissupportersorconstituentsis

notdefinedalonebythematerialityofthecommoditiesexchanged.Itisfirstandforemost

263

asymbolicobligation thatweighson thepoliticians’ shoulders,an imperative toappear

bothaccessibleoreffective.

Effectiveness in particular can be attested by the service rendered, but they are also

ritualizedintheformofimage,practices,languageandcodethatsignaltovotersthatthe

candidateswhocomestothemisdeterminedandabletosolvetheirissues.

For top party leaders, these encounters with voters/citizens are often staged and

ritualized,intheformofcollectiveaudiences,ofdurbar,reminiscentofapracticecommon

at the time of monarchy, in which the Prince would appear in front of “his people”

assembled. It is rare that actualworkgetsdone in these assemblies, unlessparty aides

collecttherequestsanddofollowthecasesup.Forpoliticians,thesedurbarsaremorean

opportunitytodisplaytheiravailabilitytoageneralaudience.

After his 2012 victory, the Chief Minister Akhilesh Yadav resumed the practice of the

weeklyJantadurbar(“People’sDurbar”).Tenthousandpeopleflockedtothegatesofhis

residence on the first day, transformingwhatwas advertised as a popular consultation

andanopportunity toaddressgrievances into the stagingand the spectacularizationof

theChiefMinister’saccessibilityandpopularity297.

The imperativeofappearingaccessibleandeffective isconstantandpermeates into the

dailylifeofelectedrepresentatives,whocanbedisturbedatanymomentofthedayand

are perpetually called to grace various sorts of social and political events with their

presence.

MLAsareboundbyaseriesofsocialobligation,vis-à-vistheirconstituents,familiesand

thelocalitythatsentthemtotheAssembly.Theymustmaintainavisibilityandpresence

in their constituency by attending social events such as weddings, funerals, religious

festivals, castesammelans (assemblies)andastringofofficialevents– inaugurationsof

projectschiefamongthem.

297Thereareconflictingreportsaboutthenumber.Kushnercitessourcesevoking25.000people(Kushner2015).

264

ThisiswhatpoliticiansinWesternU.P.referstoasthesukhdukh(literallythehappyand

thesad), that is therangeofsignificantsocialeventsmarkingthe lifeof individualsand

communitiesthatrequirethepresenceandthecontributionofa localpolitical figure.In

thiscase,thesukhreferstobirths,andthedukh tofunerals,tomarkthecomprehensive

rangeofeventspoliticiansareexpectedtoattend.Thus,theweddingseasons,whichtakes

placetwiceintheyear,inwinterandinsummer,areaparticularlygruellingtime.AnMLA

caneasilyattendseveraldozensofweddinginasingleday,blessthegroomandthebride,

attendtotheparentsandtheirrelatives,totheotherlocaldignitariespresent.

Forthehost,havingthepresenceofthelocalMLAisnotjustseenasamatterofprestige

oratributetotheirownstatus,butasamatterofobligationthattherepresentativehas

towards them.The cost of refusing an invitation canbehigh.OneMLA fromAllahabad

describedthisthefollowingway:

“Representativesarenotonlyexpectedtobeaccessible, theyshoulddemonstrate

theirforwardnessbycomingtopeople,ratherthanwaitingfortheirrequests’298.

Beyond attending those events, MLAs are also expected to contribute to these events,

eveninamodestmanner,byprovidingmonies,food,byhelpingwiththeorganizationof

theevents,byprovidingworkersortransport.AnMLAinWesternUPcomplainedthathis

daily expensesamounted toat least fifteen thousand rupeesaday,whichamounts toa

heftysumattheendofthemonth.

Enteringpubliclife,asaprivatebusinessfigureorasapublicpoliticalfigureimpliesthat

oneentersintoasystemofdeeplycodifiedobligations.Theremustbecounterpartstobe

paidforthebenefitofrisinginsociety.

Candidatesare firstand foremostevaluatedontheircapacity toredistribute.Theyhave

an obligation of redistribution. What politicians present as ‘service’ (seva) is in fact a

codifiedobligationtoredistribute,toprovideaccesstoresourcesinordertodevelopand

maintainahigh status.This is evocativeof theancientRomanevergetism,apractice in

whichtherichclasses legitimizedtheirdominantpositionbyfundingpublic institutions

or funding public work from their own resources. Wealthy citizens who sought high

298InterviewinAllahabad,April2007.

265

magistrate or elective positions had the moral obligation to distribute a share of their

wealthtothecommunity(Veyne1992).

Asaresult,beinganelectedrepresentativeisacostlyproposition.Complainingaboutthe

costofapolitician’slifeisanoldrefrain,whichF.G.BaileyrecordedinOrissa,inthe1950s,

withoneoftheAssembly’swealthiestmember:

“I’mnotinitforthemoney,thatiscertain.Iwasaministerinthelastgovernment,

and after paying one hundred rupees to the Assembly party ad fifty rupees to the

party, and bills for electricity and water and all that, I was getting a clear eight

hundredfiftyrupees.Whoonearthcanlivelikethat?Ihavetoentertain.Morarjiand

Mrs. Gandhi and other people come down and staywithme. (…) There are some

MLAs that rent out their quarters and themselves live in the servants’ roomor the

garage.Tobeinpoliticsyouhavetobeaveryrichperson,orverypoor”299.

Inmyyearsof fieldwork, I have rarely encounteredMLAswhowereverypoor, even if

some ledapparentlyasimple lifeand ifmanyothersclearly livedabovetheirmeansor

hadtostruggletomeettheirdailyexpenses.WhentheBSPwonitsfirstmajorityin2007,

some itsMLAs found it difficult to find adequate housing in Lucknow.The government

then set up a housing scheme for MLAs in an apartment blocks near Hazratganj, in

Lucknow.

5.2.2.Oncandidate’seffectiveness

Themessageofeffectivenessisconveyedthroughthedeedsoftherepresentationbutalso

through awhole symbolic grammar that interweaves language and eloquence, sartorial

choices, body language and attitudes. Politicians, and aspiring politicians in particular,

tendtooverstateoremphasizetheirprojectedeffectivenessbyadoptingcertain‘styles’of

politicalleadership,meanttostrikepeople’sattentionandimagination.

299Quotedin(Bailey1998,95).

266

Muscularpolitics is one such register.Resorting to violenceorprojecting visual signals

thatconnecttoauniverseofheroicmachoreferencesisonewaytoconveythemessage

thatthecandidateisreadytodowhatittakestomeethisvoters’expectations.

Political campaigns are marked by these signals of strength: bike rallies, jewellery and

gold-rimmedsunglasses,aswellasafieryrhetoric.Itisnotsurprisingthatpartiesliketo

bring movie stars, incarnating heroic figures on the screen, onto the campaign trail. In

2009,oneofthestarcampaignersoftheSamajwadiPartyinthegeneralelectionswasthe

actorSanjayDutt,alarge-heartedmuscularherofigureinIndiancinema,withatroubled

relationwiththelaw.

The muscular register can take sinister turns when violence gets unleashed during or

betweenthecampaigns.Thecriminalswhopopulatepartiesdonotshowrestraintinall

circumstances and the newspapers frequently relate tales of road rage, intimidation,

kidnappings or even murder involving politicians. Ahead of the 2012 elections, and in

ordertoamenditsimageofapartyharbouringcriminals,theSamajwadiPartyimposeda

strictdresscodetoitscandidates,meanttocodifyandregulatetheirappearance.

Thedresscodeincludedalonglistofmandatoryitems,suchasanicelyironedwhitekurta

pajama(ofgoodcutandfinematerial),ablacksabri(Nehrujacket,optional),well-tucked

shirts(“Nokeepingbuttonsopenwiththehaircomingout”),personalgrooming,trimmed

beardsandnounkemptorfloatinghair.Inaddition,itprohibitedpaan(betelandtobacco)

chewing,thevisibledisplayofgunsandrifles.Itspecificallyaskeditscandidatestoavoid

drivinginthestreetsflashingtheirgunsoutofwindows.Therewerealsorestrictionson

jewelry,darkandgold-rimglassesandrecommendedthosesportingsandalstowearthen

withwhitecottonsocks.ACabinetMinisterandpartystrategistexplained therationale

behindthedresscodeasfollows:

“WewantedtoprojecttheimageofanewSP,notassociatedwitholdbias,prejudices,

withvices.Wewantedourcandidatestoprojectaneatandcleanimage,inacampaign

267

led by a well-groomed, modern, foreign educated leader. The choice of candidates

reflectedthatchoiceofnewimage”300.

Accessibilityandeffectiveness is linked to thenecessity tobuildpatronagenetworks to

buildanddevelopapoliticalcareer.Thesenetworksarenotnecessarilybasedondirect

interactionbetweenpoliticiansandvoters,butratherbetweenpoliticiansandarangeof

intermediaries,taskedwiththebuildingofdirecttieswithvotersandlocalcommunities.

Thesenetworksoftenincludeindividualswhoaredrawnfromlocalelitefamiliesand/or

localdominantgroups.Theselocalelitenetworksusuallycutacrosscastes,followinglocal

demographyandlocalconfigurationofpower.

Through painstaking efforts and at great personal cost, aspiring politicians will slowly

build for themselves a status of a local leader, susceptible to attract the attention of a

party.Someofthemattempttotakeshortcuts,byconsortingwithlocalcriminalelements,

orbyresortingthemselvestoillegalactivities,inordertoacceleratetheprocessandbuild

upanimageof‘effectiveleadership’.

One meets many of these aspiring politicians in party offices, where they seek the

protectionandpatronageofaseniorpartymember.Manyoffer theirservice toexisting

candidatesandrepresentatives,helpingthemtobuildandmaintaintheirownpatronage

networks,with thehopeofusing themoneday for theirownbenefit.Theyoftenactas

gatekeepers, or intermediaries, between a senior political figure and their own

community.Establishedpoliticiansoftenseekto‘reachout’toothercastevialocalleaders

thattheypatronize,inexchangeforaccesstotheirownbaseofsupporters.Thus,aspiring

politicians cultivate their own networks and use these roles to ensure resources to

themselves(Harriss2011).

Thereareotherroutestothecandidacy,usuallythroughindividuals’inscriptionintolocal

networkof influence,organizedaround institutionsor socialorganizations.ManyMLAs

startedtheirpubliccareerinKisanunions,cooperativeorganizations,studentunionsor

teachers’ associations. There is no data to quantify howmanyMLAs have been elected

300InterviewwithAbhishekMishra,MinisterforProtocolintheAkhileshCabinet,athisresidence,Lucknow,November7,2013.

268

first at the Panchayat levels butmost politicians I interviewedmaintain close tieswith

localdemocraticbodiesintheirconstituency.

Theyactasspokespersonoftheircommunityandconveytolocalpoliticiansandthelocal

bureaucracy requests related to thewelfareof their communities.Manypublicworkor

public project have started with the mobilization campaign of caste or communities

leaders: building roads, digging tube-well, fixing power connections, repairing or

expandingirrigationcanals,providinggovernmentjobs,etc.

Induetime,thesenetworkscanbeconvertedintoresource,attheserviceofapartyorat

the service of the individual who contributed to build those networks and accessed a

positionof leadershipwithinthem.Sometimes, local leadersacquirethestrengthtowin

electionswithouteventhesupportofpoliticalparties,becauseoftheirpositionorsocial

statuswithintheirlocalityandcommunity301.

5.2.3.Thehighcostofentryintopolitics

Buildingandmaintainingapoliticalcareerinvolvesignificantinvestments.Therearethe

costsofcampaigning,ofcourse,thatarespiralingaftereachelection.Mostpartiesexpect

theircandidatestofundtheircampaignsaswellastocontributetopartycoffers.

Thecostofcampaigninginfactcanbesmallcomparedtothecostofbuildingapolitical

stature,orprofile.Buildinglocalsupportnetworks,throughmobilizationandpatronage,

consumes a lot of time and resources. Many candidates indeed start investing in their

politicalcareeryearsaheadofanelection,byacting,looking,speaking,andspendinglike

electedrepresentatives.

Someparties, such as theBSP, offer shortcuts for aspiring politicians, by selling tickets

(auctioning them, in fact).Months aheadof thepolls, rumors start spreading about the

cost of BSP tickets, fixed by the party high command. The closer from the date of

nomination,thehigherscale.

301 InterviewwithSibagtullahAnsari,MLAMohammadabad,inYusufpur,13thJune2007.

269

InMay2007,aBSPcandidateexplainedtomehowhehadboughthisBSPticketfor1.25

Crores of Rupees, contested from jail (where he was awaiting trial for murder), won

thankstothelocalDalitvoteandthevotesofhissupporters,andobtainedbailsoonafter

theelection.Whenaskedabouthisplanofactionfromthere,helaughedatmeandsaid

“Well,nowIhavefiveyearstoregainmyinvestment”302.

Once theyhave the ticket, the candidatesmust fund their campaigns andbemindful of

what theircompetitorsarespending.There isa lotofpressure to try tooutspendone’s

opponents. As F.G. Bailey put it, “to be successful as a leader is to gain access to more

resourcesthanone’sopponentandtouseitwithgreaterskill”(Bailey2001,35).

Andonceacandidateiselected,heorsheentersintoaworldofconstantexpenditures,for

events,people,staff,thehiringofvehicles,hostingofleaders,gifts,etc.

The combination of high cost of entry, cost of competition, uncertainty of winning and

further uncertainty of serving more than a term create many incentives for predatory

behaviour.Particularlywhenthesectorssusceptibleofgeneratingcashforelectionsare

themselvescriminalized.

Mostcandidatesdon’thavetheresourcestofundtheirpoliticalcareerontheirown,and

thereforedevelopbusiness activities or business tieswith individualswho can support

them in turn. Patronage serves the double purpose of building support among voters,

largely through intermediaries,and todevelopasupportnetworkamong localbusiness

elites who can contribute to party and campaign funding against the ‘facilitation’ of

businessdealingsandtransaction.

Candidates who cannot follow up or who cannot count on their parties to fund their

campaigntendtobefilteredoutofthecompetition.Generallyspeaking,failingtoconform

to evena fewof themultipleobligationsbestowed to elected representatives can cut a

candidacyorapoliticalcareershort,regardlessoftheotherqualitiesorattributesofthe

person in question. This partly explain why powerful individual, who possess both

302InterviewwithaBSPMLA,Varanasi,May2007.

270

financialandcriminalresources,canverywellloseelections,shouldtheyfailtoconform

tovoters’expectations.

Inthenextsection,IexaminethequestionofcriminalizationofpoliticsinUttarPradesh

andcomparethetrajectoriesofcriminalorganizationsinWesternandEasternU.P.

5.2.DonorDayavan?Divergenttrajectoriesinthecriminalizationofpolitics

TheGoondaRaj – or the reign of the brigands – is a central feature of politics inUttar

Pradesh.Thetermreferstothecriminalizationofpublic lifeandof institutions,through

theinductionofcriminalelementswithinpartiesandtotheuseofviolenceintheconduct

ofpower.Thetermisalsousedtorefertoageneralsenseof lawlessness,arbitraryand

violence in public life, as well as to the impunity that such a climate offers to criminal

individualsororganizations.

Dataonthe ‘criminalprofile’ofcontestants isavailable for thepast twoStateAssembly

andGeneralElections inUttarPradesh.While the treatmentof thatdata isproblematic

(JaffrelotandVerniers2014b),onecanstillpointat interestingvariations.According to

theaffidavitdata,27.3percentofthecandidatesfieldedbythefivemainpartiesin2007

hadpendingcriminalcharges.Thatnumberroseto37.6percentin2012.

TheBSPandtheSPtendtofieldmorecandidateswithcriminalchargesthantheBJPand

Congress,althoughtheratiofortheBSP,BJPandCongressin2012wereofsimilarorder

(between 32 and 36 per cent). In 2012, 49 per cent of the SP candidates had criminal

charges. These percentages tend to increase among the winners and the runner-ups,

which confirms Vaishnav’s observation that criminality improves electoral prospects

(Vaishnav2012,88).

That being said, the fact that the criminal ratio among runner-up candidates and other

losingcandidatesindicatesthattherelationshipbetweencrimeandvictoryshouldnotbe

seenasdeterministic.Manycandidateswithcriminalchargeslosetheirelection,including

against“clean”candidates(88and71respectively,for2007and2012).

271

Intermsofregionalvariations,therewasin2007ahigherratiooftaintedcandidatesand

winners in the East and the North-East than in the West and Rohilkhand. Those

differencesfadedin2012,theoverallrationbeinghigherandmoreequallydistributed.

Thisdatashouldbeseenasmerelyindicativeandcertainlynotisolatedfromotherfactors

contributingtothecompetitivenessofacandidate.

Withthesecaveatsinmind,Iwishinthissectiontomakethreeobservations.Thefirstone

isthatthereareavarietyofcontextsfromwhichthecriminalizationofpoliticsemerges

andoperate,andthatthesecontextsfollowthekindofeconomictransformationthathas

occurred in various sub-regions. In a nutshell, criminality and the criminalization of

politics in Western Uttar Pradesh is more intertwined with the urban context and the

capitalist economy that has grown over the past two decades, while criminal

organizationsintheEasthaveremainedmoreassociatedwithtraditionalformsorrural

criminality.

ThesecondpointIwishtostressonisthattheparticipationof‘criminalelements’inthe

electoralprocessdoesnotfundamentallydifferfromtheparticipationofbusinessfigures

inpolitics,fortheyrespondtosimilarincentive:accesstoresources,fameandprotection.

Just as the businessmen politicians, criminals use the resources they have at their

disposition–muscleandmoney–asacompetitiveadvantage.

Thelastpointisthattheadvantagesthattaintedcandidatescandrawfromtheircriminal

profilework in conjugationwithother factors contributing to their electability, such as

personal reputation, eloquence, accessibility and so on. Criminals might have a

competitive advantage in the electoral race but they still need to conform to voters’

expectationsinordertowinandinordertolastinpolitics.Thereareanumberofcasesof

dreadedcriminalslosingelectionsoncetheyfailtolivetotheirsupporters’expectations.

272

5.3.1.“Wehaveturnedcorporate”:thereconfigurationofcriminalorganizationsinWesternUttarPradesh

TheMLAs I interviewed throughmy fieldwork frequently admitted that it is extremely

difficulttostartordevelopapoliticalcareerwithoutdealingwiththeworldofcriminality.

The reason quite simply is that in order to build up support, raise funds, expand their

influenceanddeveloparedistributivecapacity thatwillmakethemattractive tovoters,

theyneed to cultivate tieswitha rangeof economicactors andgroupswhowield local

influenceandpower.AndinWesternU.P.,anumberofkeyeconomicactivitysectorsare

deeplycriminalized.

The size of the black and grey economy and the absence of effective policing of the

region’s fastestgrowingeconomicsectorsmean that individualsandgroupswho invest

capitalineconomicalorbusinessactivitieshaveanearfreehandatusingillegalmeansto

further their interests. Thehigh entry cost topolitics is anotherpull factor to resort to

illegalwaysofraisingfunds.

With the liberalization that took place post-1991 and with the explosion of the

development of Delhi NCR, opportunities for enrichment have greatly increased,

enhancingthecompetitionoverresourcesandinfluenceandincitingeconomicagentsto

resorttocriminalmeansinordertoexpandtheirbusinessactivities.Thus,thesourcesof

party funding for politicians have changed. Pre and post-Independence, local dominant

farmingcommunitiesdrewtheirresourcesfromland,whichgeneratedbothrevenueand

opportunities forpatronage, through labor relations. In the contextofurbanizationand

diversificationof theeconomy,politiciansnowdrawtheirresources fromthesectorsof

construction,realestate,transport,brokerage,liquorandtheprovisionofutilitiessuchas

water, electricity, cable television, and, famously, sand303. As we saw earlier, many

politicianswerealreadyactiveinthesebusinessesbeforegettingintopolitics.Andaswe

alsosaw,anumberofpoliticiansusedtheirelectivepositionto further their interestor

startventuresinthesesectorsaswell.

Thesenewsectorsandhubsofeconomicdevelopmentareequallyattractivetocriminal

elements and criminal organizations, who in Western U.P. have diversified their

303See(KapurandVaishnav2011)

273

traditional activities – smuggling, trafficking and the extraction of resources from

impoverished rural populations – by investingheavily into thesenewbooming sectors,

more profitable. Their criminal proclivity provides them with nearly unbeatable

competitive advantages, particularly since the development of these sectors of activity

remainlargelyunregulated.

Inshort,localpoliticallifeisalmostinextricablylinkedwithlocalcriminality.Notonlythe

pressures and constraints of electoral competition create incentives to resort to illegal

means, but the very sources that fuel political life with resources are themselves

criminalized.Unsurprisingly, inWesternUttarPradesh, those sources are equatedwith

thesectorsofactivitythathavegrownthemostinrecentyears.Theyarealsothesectors

fromwhichmostofthebusinessmenpoliticianscomefrom.

The criminalization of these sectors of activity was accelerated by the migration of

gangstersfromvillagetothecities.

WhileconductingfieldworkinvillagesacrossBaghpatandMeerutdistrict,Iwasstruckby

thedifficultyoffindingthedacoitstheregionwassoinfamousfor.Ask(almost)anyonein

thesevillageswhothelocaldreadedcriminalfiguresare,andtheywillprovideyouwitha

longlistofnames.Askwheretheseindividualscanbefoundandtheanswercomesinthe

formofaquestionmark.Tothequestiondakhukahahain?(“Wherearethebandits?”),the

commonresponsewasthattheythereusedtobedacoitsinthearea,butthattheyhadleft

recently.

Indeed,most of the leading figures of gangs, or criminal organizations,whoused to be

based in the countryside and operate in villages, have migrated to nearby cities, in

Ghaziabad,NoidaorLoni,orsometimesDelhiandbeyond.Having tracedsomeof these

individuals,twoexplanationswereprovidedforthisruralexodusofcriminals.

Thefirstandmainmotivefortheirmigrationisthefactthatcitiesofferbetterreturnsto

criminal activities than villages. Investing in booming business sectors and using their

criminalskillsasacompetitiveadvantageisfarmorerewardingthanextractingresources

fromimpoverishedlocalitiesandtheirlow-incomeinhabitants.

274

Criminalorganizationsusuallyliveofftheplunderingofstateresourcesbycontrollingthe

world of contracting and distribution of raw materials, such as stones, sand or timber.

They usually use criminal pressure to win contracts, bribe local police forces and

intimidatetheirlegitimatecompetitors,ordrivethemoutoflocalmarkets.

Theseareprofitableactivitiesbuttheydonotcomparewiththereturnsthatcanbemade

fromplunderingtheprivatesectorinfast-growingcities.Targetingtheprivatesectoralso

hastheadvantageofavoidingpublicscrutiny.

Cities also offer a relative anonymity, compared to villages where everyone knows

everyone’swhereabouts.

Thesecondincentiveforcriminalstomigratetocitiesisthefactthatruralcriminalityhas

becomeamorecompetitivespace.Theassertionofbackwardgroups,thefragmentation

ofthepoliticalspace,theriseofnewpartieswhodeveloptheirownlocalnetworkshave

alsoledtoanincreaseofcompetitionamongcriminalsandamongcriminalorganizations.

This competition is often regulated through violence, which, at a certain point, call for

stateintervention.

There isaparticulararea inWesternUttarPradesh,atrianglebetweenBaghpat,Baraut

andMeerut,wherecriminalgangshavebeenatwarwitheachother throughthe1990s

andearly2000s,overterritorialcontrol,competitionoverresourcesandvendettas(TNN

2006).UndertheMayawatiregime,thepolicecrackeddownonanumberofthesegangs.

Severalprominentcriminalswerekilled inencounterswithpoliceforces304.Therewere

alsoinstancesofconnivancebetweenpoliceofficersandgangs,leadingtofurtherviolence.

In that context of gangwar,many sought refuge to cities and their anonymity, and are

consideredabscondingfromthevillagestheyoriginatefrom.

304InApril 2016,47policemenwere sentenced life terms for fakeencounterskillings that tookplaceintheearly1990s(Rashid2016).

275

“Lifewasbecomingdifficult [in thevillage]as thepolicewasafterme. Idecided to

turn corporate, since there is moremoney to bemade here [in Ghaziabad] in the

constructionbusiness”305

Gangsters often speak euphemistically when describing their occupation. “Turning

corporate” is a term often used by criminals to describe their conversion to the new

economy,wheretheyusetheircriminalresourcestocarveforthemselvesortheirpatrons

apieceoftheurbancake.

Onlyafractionofthesegangstersbecomepoliticians.Mostofthemhoweverdevelopties

withpartiesandpoliticians, inordertosecurebothopportunitiesandprotection.There

are famous and spectacular figures that illustrate this process of integration of the

spheresofpolitics,businessandcrime.OnesuchfigureisthelateliquorbaronGurdeep

SinghChadha,alsoknownas“Ponty”Chadha306.

By the timeofhisdeath inNovember2012(hewaskilledbyhisownbrotherduringa

shootout at his South Delhi Chattarpur farm house), Chadha had acquired a quasi

monopoly on liquor distribution in Uttar Pradesh, occupied a dominant position in the

liquorretailingbusiness,hadstartedmakinginroadsintheliquormarketinneighboring

states, andwasrunningaWesternU.P.basedrealestateand industrialempire that cut

across construction, education institutions, malls, cinema halls, paper mills and film

distribution.

TheChadha’swerePartition refugees (fromRawalpindi)who settled inMoradabad, an

industrial settlement 180 kilometers East of Delhi, near Rampur. The Chadha family

invested in a local liquor store, and proceeded to expand their business, eventually

gaining control of the liquor market in Moradabad. During the Emergency, the family

moved into the sugar trade (the base product for liquor making), setting up several

factoriesinthearea.

305InterviewwithananonymoussourceinGhaziabad,January2013.306Thissectiondrawsheavilyfromadetailedportraitwrittenin2013byMehboobJeelaniinTheCaravan(Jeelani2013).

276

Their business started to really flourish and expand once Gurdeep’s father, Harbhajan,

developedtiesinthelate1980swithanemergingpoliticalfigure,MulayamSinghYadav.

The Chadha family contributed generously to the Janata Dal 1989 campaign. Once

MulayambecameChiefMinister,theChadhasstartedapplyingforpublictendersandgot

a number of public contracts for sand and pebble mining, two activities key to the

industrial development and construction boom that was taking place in Western U.P.

Theyused themoney theymade through theseactivities to expand further their liquor

business.

LiquortradeinU.P.isperhapsoneofthemostcriminalizedsectorsofeconomicactivity.

In order to cope with the harshness of competition, the Chadhas hired local criminal

figures tooverseeandprotect theiroperations.Themixofcriminalmeansandpolitical

protection enabled them to expand their business aggressively to a growingnumberof

districtsinUttarPradesh,notablythroughtheriggingoftendersandlicensingsystem.

Thepolitical instability of the1990s incited theChadhas todevelop ties acrossparties.

TheyconnectedwiththeBSP,whichwasgrowinginWesternU.P.,aswellaswiththeBJP

(they allegedly helped Kalyan Singh’s son Rajvir to develop his own liquor business in

Aligarh). The way to connect with politicians was to provide financial support to their

campaigns,butalsotodeveloplocalbusinesstieswiththeirrelativesorassociates,thus

ensuringpoliticalprotectionwherevertheyexpandedtheiractivities.

PontyChadha’sbusinesspeakedundersuccessiveMayawatigovernments.SincetheBSP

needed to consolidate its hold over several districts in Western U.P. and surrounding

areas,theyreliedheavilyonPontyChadha’sinfluenceandmoneypower,tradingsupport

against further business activities. Mehbood Jeelani sums up the trajectory of Chadha

undertheMayawatiregimeasfollows:

“(…)underMayawati’srule,Chadhawasawardedamonopolyoverdistributionforthe

state’sRs14,000-croreliquormarket.Inaddition,hewasgivencontrolof30percentof

the alcohol retailers across the state, and was allowed to purchase a number of

distressedbutviablestate-ownedsugarmillsatapricebelowtheirfair-marketvalue.

He also received a Rs 10,000-crore contract for distributing food under the state’s

middaymeals scheme for children and pregnant women—in violation of an earlier

277

SupremeCourtorder—andlandedvasttractsofprimerealestatejustoutsideDelhiat

a loss to thepublic, stateCongress leadersclaimed,ofRs40,000crore.BothChadha

andtheBSPgovernmentmadeenormoussums fromthebooze trade inparticular—

excisetaxonthe10millioncasesofliquorsoldeveryyeargeneratedroughlyRs10,000

crore annually for Mayawati’s government—and within the state administration

Chadhabecameknownas“Mayawati’sfinancier””(Jeelani2013).

Onceheacquiredhis liquordistributionmonopoly,Chadha increased the retailpriceof

liquor,agesturethatwasknownasthe“Pontytax”.

InNovember2013,afamilydisputeoverthedivisionoffamilyassetsandproperty,found

its conclusion in a shootout in which Ponty and his brother Hardeep died. Since then,

Ponty Chadda’s son, “Monty”, took the reins of the company and has started a cleanup

drive of the family businesses, aiming to convert his father’s empire into a company

followingmoreestablishedcorporatenorms.AfterChadha’sdeath,theU.P.government,

ledbyAkhileshYadav,orderedthependinginvestigationsagainstthefamilytobeputto

anend,andrenewedthecompany’sliquorlicense.

PontyChadhanevercontestedanelectioninhislife,nordidanyofhisrelatives.Buthis

ties with various parties through time and his engagement with local and state-level

political figuresmadehimaprominentpolitical figure.Chadha’sbusinessacquisitions–

notably the sugarmills –wereoftenmade through screen companiesor through cartel

organizations in which shares were generously distributed to local power holders,

blurringtheboundariesbetweentheworldofpolitics,businessandcrime.

ThestoryofPontyChadhaisofcoursequitespectacularandinvariouswaysunique.But

one finds a large number of “smaller Ponty Chadhas” across the region, individual or

networks of entrepreneurs who reap the benefits of a developing capital region, and

develop both political and business ties in order to fight their way through a highly

competitive,violentandthereforeriskybusinessenvironment.

These ties enables them to gain access to the vast resources that the state distributes

through its system of tenders, as was well illustrated by Kapur and Vaishnav in their

paperonthequidproquorelationshipbindingpoliticiansandbuildersinU.P.(Kapurand

278

Vaishnav2011).Thestateretainshugestakesandstillcontrolanumberof industries–

notably sugar mills and the range of Public Service Undertakings (PSUs). And while it

shouldnormallybethetaskofthebureaucracytoorganizeandsupervisethosetenders,

partiesandpoliticiansingovernmenteffectivelyoverseeordirectthoseprocesses.ABJP

MLAfromRohilkhandconfidedtomethat:

“The sand contracts are decided by the Minister of Mining but usually by the Chief

Ministerhimself,oroneofhisclosefamilymember”307.

Asaresult,politicalconnectionsarenearlymandatorytohopetowinthosetenders.

IinterviewedaBJPMLAfromtheneighboringregionofRohilkhand,whoreflectedonthe

profileoftheSPandBSPpoliticiansinthefollowingmanner:“ThecandidatesoftheSPand

the BSP are not business people. They are exploiters of state resources”. He went on to

describethemas“mafia,thatisthetermIwouldchoose”308.

Thisgoesontoillustratehowtheworkofpoliticiansinterfereswiththemissionsofthe

administration. If one goes by the institutional book, legislators make law and the

bureaucracy apply them. In reality, those distinctions can be quite blurred since

individualinterestsgetofteninterwovenwiththepursuitofthegeneralorpublicinterest,

throughregularinterferencesofpoliticalactorswithinpublicsystems.

This also calls on to reflect about the fact that despite 25 years of gradual opening or

liberalizationoftheeconomy,thestateremainsengagedinanumberofkeyeconomicand

industrialsectors.Itstillalsopresidesoveramazeofbureaucraticregulations,systemsof

tendersandlicensingthatorganizesthestate’seconomiclife.Aspoliticalanthropologist

AkhilGuptademonstratedinhisstudyoftheworkingofalocalbureaucracyinMandi,a

smalltowninthevicinityofMuzaffarnagar,theworkingoflocaladministrationisnotonly

politicizedbut also guidedby arbitrariness (Gupta2012),making “political guidance” a

necessaryhelptonavigatethiscomplexuniverse.

307InterviewwithP.Singh(namechanged)BJPMLAfromRohilkhand,inDelhi,10February2013.308Ibid.

279

5.3.2.TheresilienceofruralgangstersintheEast

ConsideringthestateofderelictionandcorruptionofpublicinstitutionsinUttarPradesh,

it isnotdifficult to imaginewhyvoterswouldvote for taintedcandidates.Theypresent

themselves–andoftenare–ashavingthewillandcapacityto“getthingsdone”.AnIAS

officerinLucknowoncetoldme,underthepromiseofanonymity,thatsomeofthebest-

performing areas in the state were those controlled by criminal politicians, who could

coercelocalbureaucratstodeliver309. IntheU.P.context,markedbypoverty,scarcityof

resources,lawlessnessandviolence,thereisasociallegitimacytoresorttogreymethods

–orplainillegalone’s–inordertofulfillpeople’slegitimateneeds.

Onthisquestion,IrecallaparticularconversationwiththeMuslimownerofamillinthe

outskirtsofVaranasi,staunchsupporteroftheSamajwadiParty.Iaskedhimwhatdidthe

SP ever do for Muslims in Uttar Pradesh, and if it bothered him that the party was

harboringknowngangsters.Hisresponsewasasfollows:

“See,whatmattersthemosttousisbijli(electricity).Withoutbijli,wecannotrun

themills.Attheendofthedays,they[thepoliticians]areallcorrupt.Butatleast,

withtheSP[referringtothelocalSamajwadiPartyoffice],youcancountonthem.

Whenthepowerisgone,youcangotothem.Theywillsendsomegoonstobeatup

thelocalengineeruntilthepowerisrestored”310.

Thus,criminal incursion intopolitics isnotaone-waystreet.Afterall, taintedcriminals

must confront themselves to the ballot test and win support from voters. There was a

periodinUttarPradeshpolitics,particularly inthe1980s,wherepartiesandcandidates

called on criminal gangs to influence local electoral outcomes. Criminal intimidation,

coercion, booth capturing, political assassinations were quite common. The rise of

criminalizationofpoliticsfollowedtheprocessofdeclineoftheCongress,whocalledon

thesecriminalelementsasawaytomitigatetheirflailingpopularity.

Thecriminalizationoftheelectoralprocesswaslargelycheckedinthemid-1990s,under

theimpulseofT.N.Seshan,thethenChiefElectionCommissioner.Seshancountermanded

309InterviewinLucknow,March2009.310InterviewinVaranasi,March2007.

280

elections in constituencies that had seen fraud, arrested preventively “trouble makers”

and“historysheeters”,anddeployedthearmytosecuretheballot.Inthe1996elections,

nearly one hundred fifty thousand people were preventively arrested in Uttar Pradesh

alone.AfterSeshan’sretirementin1996,theElectionCommissionofIndiawouldpursue

theeffortofsecuringandpolicingtheelectoralprocess,aswellasplacecandidatesunder

tight scrutiny (Verma 2005b). Today, voters are far less likely to be intimidated by

criminalpoliticians.Asapopularsayinggoes,sabkothodehimaarsakteho(“onecannot

possiblybeatupeveryone”).

Insteadofcoercion,thecapacitytosolveproblems,perceivedgenerosityandareputation

ofaccessibilityiscrucialforbuildingandmaintainingpopularsupport.Asalways,theway

thisprincipleworksinrealityismorecomplicatedandambiguousthatitinitiallyappears.

Ihadademonstrationoftheambiguityofthecriminal-voterconnectionduringthe2007

StateElectionscampaign.

InthemonthofMarch2007,IpaidavisittotheAnsaribrothers–AfzalandSibagtullah–

in Mohammadabad, a small dusty town bordering Ghazipur, hundred kilometers east

fromVaranasi,whereIwasbasedtocoverthestateelectioncampaign.FriendsfromDelhi

hadprovidedmewithacontactnumber,whichIusedtosecureanappointmentwiththe

familyofoneofUttarPradesh’smostprominentcriminalpoliticalfigure,MukhtarAnsari,

at that time jailed in the Jhansiprison for theallegedmurderofKrishnanandRai,aBJP

opponentandheadofrivallocalcriminalorganization.

The story of the family is well known. Born in 1960, Mukhtar Ansari hails from an

illustrious political family. His grandfather, Mukhtar Ahmad Ansari, served as Congress

Presidentin1927-28andwasoneofthefoundersoftheJamiaMilliaUniversity.Heisalso

arelativeofIndia’scurrentVicePresident,MohammadHamidAnsari.Hiselderbrother,

Afzal,isafive-timeMLAfromMohammadabad,whereheranfirstin1985asaCongress

candidate.Hethenservedthreetermsasacommunist(CPI),andwonafifthtermin2002,

onaSamajwadiPartyticket.In2004,hewontheseatofGhazipur,againonaSPticket,but

losthisseattwice,firstunderaBSPticketandthenunderthebannerofthefamily’sown

politicalparty,theQuamiEktaDal(QED).

281

A four-time MLA from Mau constituency, Ansari has been elected under various party

labels (includingunder the flagofhisownparty, theQuamiEktaDal).Hewas inducted

into state politics by Mayawati during the 1996 elections. The BSP leader saw in the

strongman of Mau the ideal candidate to win the seat of this volatile locality (before

MukhtarAnsari,noMLAsucceededinbeingre-electednotanypartymanagedtowinthe

seattwiceconsecutively).Ansari’sgangster’sreputationwasthenalreadywellearned.He

was accused of the murder of Nand Lal Rungta, a local VHP Treasurer, of extortion,

kidnapping,cheatingbyimpersonationandrioting.

Ansari isknown forhisdisposition todeal violentlywithhis competitors,notably from

thenotoriousBrijeshSingh’sgang.OnNovember29,2005,agroupofarmedmenkilled

KrishnanandRai,theMLAfromMohammadabad,onhiswaybackfromafamilywedding.

Seven people died in the attack. Rai, a Bhumihar, was a medical school dropout from

BenaresHinduUniversity.HesettledinVaranasiwherehegrewaflourishingbusinessin

infrastructurecontractingandrealestate.Hethenallegedlygrewtieswiththelocalmafia,

notablytheBrijeshSingh’sgang.Hewas inducted intopoliticsbyManojSinha, thenMP

fromGhazipurandfellowBhumihar311,andcontestedafirstinMohammadabadin1996.

He lost against SP candidate Afzal Ansari, Mukhtar’s brother and MLA from this

constituency since 1985 (first on a Congress ticket, then on three CPI tickets). Rai

defeatedAnsari fiveyears laterandwasassassinated threeyearsafter that.TheAnsari

brotherswerequicklyaccusedofthemurder,aswellassomeoftheirassociates.Thecase

hadtobetransferredoutsidetheregion,sincenoonedaredtopresschargeorinvestigate

thetwobrothers.Eventually,Mukhtarwasarrestedandjailed,firstinJhansi,theninAgra,

wherehestillawaitstrial312.

Throughtheirpoliticalcareers,theAnsaribrothershaveexpandedtheiractivitiesthrough

theMau,Ghazipur,VaranasiandJaunpurdistricts.Bythetimetheyrangeneralelections,

theyhadestablishedthemselvesaskeyplayersintheareaforthecoalminingbusiness,

railway contracts, scrap disposal, public works and liquor distribution business. Their

statusofpolitician,clearly,hadhelpedthemtofurthertheirbusinessandillegalinterests.

311AndcurrentlyMinisterofStateforRailwaysintheNarendraModigovernment.312This imprisonmentdoesnotapparentlypreventhim fromconductingbusiness, sincehewasbooked inMarch2010 for themurderof a local contractor, in theDakshintola areaof theMaudistrict.

282

Itnotablyhelpedthemandtheirassociatestobagpubliccontracts.Theircloutalsohelped

them to develop (or buy) strong rapport with regional party leaders across the board

(he’sreputedtobeclosetoMulayamSinghYadavandtosomeseniorcongressmen).Both

CongressandtheSPkeptfieldingweakordummycandidatesagainstMukhtar313.

Theirpositionof influencealsoenabled themtobuildstrongpatronagenetwork,which

theywoulduseforelectoralpurpose.I interviewedAfzalandSibagtullaAnsariinMarch

2007, a few weeks before the declaration of the elections’ results. Afzal described his

brother as a do-gooder, using his strength and influence to help the lives of common

citizens. He described the long queues that would form every morning in from of his

residenceinMau,wherepeoplewouldregistertheirpleadsandgrievancestoMukhtar’s

constituencycaretakers.

Criminal politicians such as the Ansari’s like to present themselves as Robin Hood

incarnations, insisting that they serve interests greater than themselves, and excusing

their recourse to criminalmeans to the obligations set upon themby “their” people or

supporters.

Popular culture and movies in particular have reinforced this stereotype. In 1988, the

movie Dayavan (in English, “the Compassionate”), a remake of a popular Tamil film,

featuredVinodKhannaasayoungboyturningtocrimeafterthemurderofhisfamilyand

thedestructionofhishomebythepolice.Hischaracter,Shakti,killshisfamilyandfriend’s

murderersandacquiresareputationofaDonwithagoodheart(“dayavan”),leadinghim

tobecomethekingofBombay’sunderworld.

Thisposturingofcoursefoolsfewpeopleinthearea,whoarewellawareofthefamily’s

exactwhereabouts.Whatmattersmore is that the familydoesnotdeviate fromwhat is

expectedfromthem,ascommunityleaders,politicians,andrepresentatives.Thequidpro

quo relationship between voters and representatives is marked by pragmatism and

opportunism, from both sides, as voters are also often in a position to bargain with

candidates.Thisdoesnotpreventpoliticianstousetheirowncapacityofredistributionto

313In2007,alltheCongresscadreoftheMaudistrictresignedcollectively,inprotestagainsttheparachuting of Gopal, a dummy candidate imposed by the PCC, who lost his deposit againstMukhtarAnsari.

283

bargainwithvoters,bydistributingresourcesselectively.Ihadanillustrationofthisform

ofopportunismandpragmatisminacampaigneventorganizedatSibagtulla’sresidence,

inMohammadabad.

The event consisted in handing 20,000 Rupees checks to a large number of women,

queuing in front of the gate of his residence. These checks were part of a last-minute

Samajwadi Party scheme that rewarded families whose girls attended schools. The

womeninlinehadbroughtdocumentsattestingoftheirdaughters’enrolment.Somehad

brought their daughters along in fear that the documents might not suffice or be

dismissed.

Inthecourseoftheproceedings,InoticedthatsomeofSibagtulla’sassociates,whowere

guarding thegate toverify thewomen’sdocuments,wererebukingsomeof themwhile

lettingtheother in thecompound.WhenIaskedonwhatbasis theydecidedwhocould

getinandwhocouldnot,IwascasuallyexplainedbyoneofSibagtulla’saidthatonlythe

womenwhohadcomefromthe ‘right’basti (‘neighborhood’)couldcomein–the ‘right’

meaning thosewhohadvoted forhisbrotherAfzal in thepreviousstateelection (Afzal

had contested and lost against Krishnanand Rai). The other women, who resided in

pockets thathadvoteddifferently,weretoldtogoawayandtovotethe ‘rightway’ this

timeiftheywantedtobenefitfromtheMLA’slargesseinthefuture.Whenaskedhowthey

knewforsurewhichbastivotedforthem,thesameassociateproducedabunchofprinted

document–pollingboothdatafromthepreviouselection–whichtheyusedtodetermine

wheretheyvotersreside.

Sibagtulla prevailed in that election, on a CPI ticket, over Alka Rai (BJP), the widow of

Krishnanand, by thirty-four hundred votes. He thus inherited, through the ballot, the

constituencythathadsendhisbrotherAfzalfivetimestotheStateassembly.Twoyears

after his 2002 defeat, Afzal contested the Lok Sabha elections, from Ghazipur. He won

against the BJP candidate Manoj Sinha, who had held the seat in 1996 and 1999, and

regaineditin2014,onaBJPticket(hewasappointedMinisterofStateforRailwaysthen

Telecominthe2014NarendraModicabinet).SinhaisanRSSpracharakandformerABVP

leaderatBenarasHinduUniversity(hepresidedtheuniversity’sstudentunionin1982),

aswellasanengineerbytraining.HeisoneoftheBJP’smajorfiguresinUttarPradesh.

284

Mukhtarcontestedgeneralelectionstwice,unsuccessfully,in1996(inGhosi)andin2009,

inVaranasi,both timeson theBSP ticket. In2009,he contested from jail, againstMurli

ManoharJoshi(BJP).HeattemptedtocontestagainstNarendraModiin2014butpulled

outoftheracebeforethenominationdeadline(healsocontestedfromGhosi,andfinished

third,behindtheBJPandtheBSPcandidates).

PartieshavekeptanambivalentattitudetowardstheAnsaribrothers.In2010,Mayawati

expelledbothMukhtarandAfzalfromtheBSP,inanefforttocleanuptheparty’simage314.

Ayearearlier,inanelectoralmeetingheldinVaranasi,MayawatidescribedMukhtaras“a

Messiahforthepoor”.Shealsoreferredtohisprofessionalactivitiesaspartofa“crusade

against affluent and powerful landowners to uphold the cause of the poor and the

downtrodden”315.

Bythen,MukhtarandAfzalhadbecomepoliticalpariahs.Asregionalpartiesattemptedto

clean their image of parties harboring criminals, they turned into the caricature of the

criminalization of politics. They formed with their elder brother their own party, the

QuamiEktaDal, in2010,which in2014 joined theEktaManchplatform, a groupingof

Eastern U.P. micro-parties, convened by Om Prakash Rajbhar, former BSP MLA from

KolaslaandleaderofatheSuheldevBharatiyaSamajParty(SBSP),atinyformationthat

commandssomesupportamongtheMBCs(mostlyRajbhars)andsomesegmentsof the

DalitsinvariousdistrictsinEasternU.P.316

One should of course be wary of the ‘Robin Hood’ type of discourse that criminal

politiciansliketooffertothosewholistentothem.Discussionswithrandomindividuals

across the constituency revealed that people were fully aware of the family’s criminal

whereabouts. But these revelations did not sound as indictments. It was actually

314The BSP press release indicated candidly “Ansari's involvement in criminal activities led ourpartypresidenttotakethedecisiontoexpelhim”andthathe(Mukhtar)hadfailedtoliveuptotheexpectationsoftheBSPwherehewasgivenentryonthepromisethathewouldmendhisways”In http://www.hindustantimes.com/Mayawati-expels-Mukhtar-Ansari--brother/Article1-532149.aspx315HindustanTimes,April16,2010.316TheSBSPcontested13seatsin2004and16seatsin2009.Theygetabove10,000votesinhalfoftheseatstheycontestin.Theirstrategyconsistsinallyingwithotherpartiesandprovidethemwiththeadditionalvotersthatmayhelpthemwintheirseats.Theyusuallytradetheirsupportformoneyorfavors.

285

enunciatedrathermatter-of-factly,orpresentedasanobviousaspectoflocalpoliticallife.

Severalrespondentsalsonotedthatmostoftheircrimeshadbeengearedtowardsother

criminals, and not against ‘common people’. But for all their known wickedness, the

Ansari brothers forged themselves a reputation of accessibility and helpfulness, which

surpassedthefearthattheirnot-soundergroundactivitiesmightinduce.Theyalsomade

surethatsomekeysegmentsandindividualsfromtheircoresupportbasebeincludedas

beneficiariesoftheirredistribution.MauisaconstituencywithalargeMuslimpopulation

(11outofits14MLAshavebeenMuslims)andMukhtarAnsarihasbeencarefultocater

tothatparticularsegmentoftheelectorate,aswellastoincludeinhisorganizationlocal

figures from other castes and communities. Thus, the “social engineering” that marks

electoral strategy also applies to the organization of patronage and daily business

activities,inthiscaseoftheillegaltype.

Eveniftheirpoliticalfortunaseemsatthemomentbehindthem–theyhavebeenpublicly

outcastbymostparties–theyretainalotofstrengthintheirarea,astheirreputationfor

generosity has not eroded among their core supporters. This is not the case for other

famouscriminalpoliticians,whosecareersankthemomentvotersspreadthewordthat

they had become ‘greedy’, a euphemistic term used to say that they weren’t sharing

anymore.

ThecaseofAtiqueAhmad,a formerSamajwadiPartyMP inAllahabad,providesagood

example. A local figure of the Allahabad crime scene, Ahmed rose into politics by

contestingandwinningtheAllahabadWestseatin1989,asanIndependentcandidate.He

was re-elected twice, with large margins, still as an Independent, until the Samajwadi

Partyco-optedhim,forafourthtermin1996.Ayearafterhiselection,Ahmaddefectedto

theApnaDal,alocalLodhpoliticalparty.Hethenproceededtowinhisfifthterm.Bythat

time, Ahmad was controlling much of the illegal activities in Allahabad and faced little

opposition.TheSamajwadiParty luredhimback into theparty foldbyofferinghimthe

PhulpurLokSabhaticketinthe2004election,whichhewon317.Bythen,histroubledpast

andpresentcaughtupandhewasjailedin2008onvariouschargesofmurder,attempt

317That seat was once held by Jawaharlal Nehru and V.P. Singh. Ram Manohar Lohia alsocontestedinPhulpur,in1962,andlostagainstNehru.

286

murder,kidnappingandabduction.RejectedbyboththeSPandtheBSP318,hecontested

fromjailinthe2009electionsonthePratapgarhseat,onanApnaDalticket.Hefinished

fourth.

Inthe2012Assemblyelections,hecontestedagainfromhisoldconstituency,Allahabad

West,againonanApnaDal ticket.He lostbyamarginof43,000votesagainstPujaPal,

WidowofBSPleaderRajuPal,whohadbeenkilledinbroaddaylightbyAhmadin2005,

onRepublicday.RajuPal,whowentbythenicknames«TigerofAllahabad»or«Garibo

kaMasiha»(‘Messiahofthepoor’)hadlostagainstAtiqueinthe2002election,butwon

overhisseatinaby-electionheldin2005,againstAtique’sbrother,MohammadAshraf.

Jailed once again, this time for the murder of Raju Pal (and a score of other murder

charges),Ahmadcontestedthe2014electionsinShrawasti,furtherEast,onanSPticket.

He lost by an 86,000 vote margin against Daddan Mishra, former Minister of State for

MedicalEducationinthe2007MayawatiCabinet,whohaddefectedtotheBJP.

Contesting in seats where his criminal influence did not reach cost him his job of

representatives. He could have contested in Allahabad on his own, as an independent

candidateorundersomeothersmallpartybanner,ashehaddoneinthepast.Butbythen,

his popular support in Allahabad had shrunk, under the reproach that he had become

greedyandinaccessible.Regularthreatstohislifehadledhimtocuthimselfawayfrom

publiccontacts.Besides,thelossofhispoliticalprotectorscuthisownaccesstoresources

andunderminedhiseffectiveness.Hisauthorityintheunderworldwasalsochallengedby

rival organisations. As a result, he lost both the legitimacy and the aptitude to build

electoralsupport.

Otherpolitical criminal figureshave shownextraordinary resilience,despiteall sortsof

odds.Theytendtobelongtomoretraditionalformsofcriminalorganizations,ruralbased,

livingofftheextractionofruralandpublicresources.Thesefigurestendtoalsobelongto

traditional elite groups that is the upper castes, contrary to Western UP where the

compositionofthecrimeworldismoresociologicallydiverse.

318AtiqueAhmadwasformallyexpelledfromtheSPforhavingbrokentheparty’swhip,byvotingagainsttheIndo-USnucleardeal,whichhadbeensavedinParliamentbytheSamajwadiParty.Onthissubject,see(SasikumarandVerniers2013).

287

These criminal figures combine several attributes that help them resist the pressures

frompartieswhoattimeswishtodistanciatethemselvesfromthesecontroversialfigures.

One example isAmarmani Tripathi, a four-timeMLA fromLakshmipurwho started his

politicalcareerin1989withtheCommunistParty,beforejoiningtheCongress.Heserved

a second term in1996before leaving theCongress to joina smallparty, theLoktantrik

Party,whichmergedwiththeBJPaheadofthe2002elections.Hecontestedandwonhis

thirdtermthatyear,butonaBSPticket.

Tripathi’s influence grew far beyond the limits of his constituency. He developed ties

across parties and became a master schemer, peddling defections on behalf of the

CongressParty,theBSPandtheSP.TiwariwaspartofKalyanSingh’sinfamousgangof19

criminal cabinetmembers, asMinisterof State for ScienceandTechnology. In2001,he

joinedRajnathSingh’scabinet in2001,asMinisterofStateforInstitutionalFinance,but

wasdismissedsoonafterbecauseofhisinvolvementinacaseofkidnappingforransom,

involving the son of a businessman from Basti319. He became a Minister in Mayawati’s

Cabinetin2002butwasrapidlydismissedonaccountoftheaccusationsmountedagainst

him,inthecaseofthemurderofMadhumitaShukla,a24year-oldpoetess,whohappened

to be Tripathi’s mistress (Tripathi 2003)320 . The Samajwadi Party welcomed him

immediatelyandhelpedhimwinafourthterminLakshmipurin2007.

Tripathi is a knownassociateof another criminalpolitician,Hari ShankarTiwari, a six-

timeMLAfromChillupar(North-east).Tiwaristartedasarailwaycontractor,andended

upheadingoneofthetwolargecriminalgangsofGorakhpur(hewasoneofthefirstMLA

tobeelectedfromjail,in1985).TiwariservedfourmandatesundertheCongressbanner

and then created his own outfit, the All-India Indira Congress (Tiwari), a short-lived

platformofdejectedCongressmen321andfellowuppercastes322.

319ThesonwasrecoveredfromTiwari’sbungalowinLucknow.320AmarmaniTripathiandhiswifeMadhumaniwereconvictedinOctober2007forthemurderofMadhumitaMishra,whowassevenmonthpregnant.321Tiwari transformed his party into the Akhil Bhartiya Loktantrik Congress (ABLTC) and ranthreetimes,unsuccessful,underthatbanner.32238 candidates ofAIIC(T) contested the 1996 elections, including seven incumbent or formerlegislators, essentially in the North-east and in Uttarakhand, a state with a large proportion of

288

Tripathi and Tiwari are also the head of political families who have contested under a

variety of party affiliations. Hari Shanker Tiwari’s son, Bhishma Shanker Tiwari, was

elected in the14th Lok Sabha in a by-election in2008, on aBSP ticket.His second son,

Vinay Shanker, contested and lost against Yogi Aditynath in Gorakhpur, in 2002. Vinay

Shankercontestedagaininthe2008Balliaby-election,whichhelostagainagainstNeeraj

Chandrashekhar, son of the late Prime Minister Chandra Shekhar. Tiwari’s nephew,

GaneshShankarPandey,waselectedMLAinDalmauin1993.Pandeywasre-electedthree

timesandwaselectedSpeakeroftheHouseinhisfourthterm.

In 2012,AmarmaniTripathi’s son,Amanmani, contested in his fathers’ seat (Nautanwa

waspartlycreated fromLakshmipur in the2008Delimitation)and lostagainstKaushal

Kishore, a first-time candidate from Congress, by a small margin. After his lifetime

conviction in 2007, in the Madhumita Shukla case, he appointed his brother Ajit Mani

Tripathi,ashissuccessorfortheMaharajganjLokSabhaSeat323.

Bothcarriedmultiplecriminalchargesontheirheads,onaccountofmurder,attemptto

murder,riotinganddacoity,amongothers.Formanyyears,theirproximitytopartyheads

protected fromthereachof the law.UnlikeTripathi,whogota lifetimeconviction,Hari

ShankarTiwariemergedcleansedfromallcharges324.

Therearemanyothersuchfigures,acrossparties.Inthe2012stateelections,therewere

only 92 seats out of 403 where none of the first three candidates did not have any

criminal charge. If we consider only the winner and the runner-ups, the number of

constituenciesincreasesto143.

The five-time MLA from Kunda, Raghuraj Pratap Singh, alias Raja Bhaiya, is another

example.RajabelongstothelocalroyalfamilyofKunda.Hebuilthimselfareputationof

strong man and won his first election as an Independent candidate. Like Amarmani

uppercastepopulation.Fourofthemwontheirseatthatyear,includingJagdambikaPal(Basti),aturncoat from Congress, Hari Prasad Tiwari himself, in Chillupar, K.C. Singh (alias Baba), inKashipur,andShyamSunderSharma,inLucknowEast.323«InUP,masterdefectorsbackinbusiness»TheAsianAge,29March2009.324AshishKhetan,thenreporterforTehelka,uncoveredthatTripathiwasregularlyleavingprisontoholdcourt,managehisbusinessaffairsandmaintainhispatronagetiesalive,undertheguiseofmedicalcheck-ups.See«JailhouseRocks»,Tehelka,June2012.

289

Tripathi, Raja Bhaiya also started his political career as a self-professed Marxist. He

quicklyturnedintoamajorcriminalfigure,acquiredvasttractsoflandandengagedinto

variousillegalactivitiessuchassandmining.HeservedfivetimesasaMinisterinvarious

BSPandSPgovernments.HistenureasMinisterforFoodandCivilSupplies,from2003to

2007andhislootingofthePublicDistributionSystem(PDS)wastheobjectofafrontpage

Tehelkareportage,whenRajaBhaiyaregainedthatportfolio, intheAkhileshCabinet, in

2012 (Khetan2012)325.Prior tohis enrolment in theSP,RajaBhaiyahad served in the

KalyanSingh’1996Cabinet,wherehewasclose toAmarmaniTripathi. In2002,hewas

bookedunderPOTA(PreventionofTerrorismAct)afteracacheofarmsandexplosives

wasdiscoveredinhisresidence.Hereceived44criminalchargesthatyear.Afterthe2002

elections, Mulayam Singh Yadav obtained him bail and inducted him in his Cabinet,

allegingthatthechargesagainstBhaiyahadbeenfabricatedbyMayawati.Hewasforced

toresigninMarch2013forhisinvolvementinthemurderofaDistrictSuperintendentof

Police326.

The reasons attracting these Eastern U.P. criminals into politics and the mechanisms

throughwhichtheybuildandmaintainsupportdoesnotfundamentallydifferfromtheir

WesternU.P.counterparts.Whatdiffersistheirsocialandeconomicinscriptionandtheir

sociologicalcomposition,whichreflectboththeeconomicchangesthathaveoccurredin

theirregionofinscription,andthesociologicalprofileofthepoliticalclassatlarge.

Thesethreespheresofactivity–politics,businessandcrimes–havenotjustbuiltties–

or a nexus – with each other. They have integrated each other, functionally and

sociologically, as an outcome of the broader transformations that have taken place in

UttarPradesh.

Theintegrationofthesethreefieldsdoesnotcovertheentiretyofthepoliticalfield,which

remainsmorediverse.Buttheydooccupyacentralplaceandplayacentralrole inU.P.

politics,asthestablepoliticalclass,asIhavedefineditinChapter3,isusuallyconnected

intothesethreespheresofactivity.

325Ironically,hewasalsoappointedMinisterforPrisons,thatyear.326«DSP’sKilling:RajaBhaiyaResignsFromUpCabinet»,inTehelka,March4,2013.

290

5.4.Conclusion

Three observations can be made at this stage. The first is that electoral politics is not

simplyaspacewherecontendingpoliticalandsocial formationsandidentitiesmeetand

competeforpower,statusandinfluence.Electoralpoliticsisalsodeeplyembeddedinthe

politicaleconomyofthelocalitiesandtheregionswherethatcompetitiontakesplace.

Inanumberofconstituenciesanddistrictsacrossthestate,therulesoftheelectoralgame

andchangingeconomiccontextshavecreatedincentivesforpeoplefromvarioussortsof

businessbackground to invest inpolitics.BecominganMLAenhancesaperson’s status,

providesaccesstoresourcesandnetworksofpower,andtoprotection.Itisintheparties’

interesttonominatecandidateswhocanwinelections,thosewhocombinetheresources

and thequalities that attract votes.Many voters seek to elect a representativewho can

effectively defend their interests. The conjunction of these three rationales favors the

integrationoflocalsocialandeconomicelitesinthedomainofpolitics.

Thisishardlyanewphenomenon.PoliticalpowerafterIndependencewasalreadybased

onasimilarsortofconjunction.Candidateswhocombinedahighcastestatuswith land

andaCongresspartyticketstoodgreaterchancesthanotherstogetelected.Traditional

elites and other landed groups used the resources they drew from land ownership and

theircastestatustodominatelocalinstitutionsandwinelections(Brass1983,1984c).

Theydidsountilthesuperiorityoftheircastestatuswaschallengedfrombelowanduntil

achangingeconomycreatedneweconomicresourcesavailabletoelectoralpolitics.Asa

result,castestatusandlandaspoliticalresourceshavebeensubstitutedbycastenumber

(ortheabilitytomobilizewithinandacrosscastes)andbytheinscriptionofpartiesand

candidatesintolocaleconomicstructures.

Thesecondobservationisthatthishasnotbeenauniformprocess.Theeconomicsources

ofpoliticalpowerhavediversifiedtothesameextentastheeconomydid.InWesternU.P.,

urbanizationandthetransformationof theruraleconomyhaveproducedanewclassof

politiciansembeddedinspecificsectorsofeconomicactivity.

291

Inotherpartswhereeconomicchangehasbeenslower,andwheretheeconomydidnot

diversify as much, the old patterns of social, economic and political domination have

resistedtothepressuresfrombelow.Thus,traditionaluppercasteelitescontinuetoexert

aninfluence.Theywerealsohelpedbytheverypartieswhoinitiallyroseagainstthemand

whohavenowpragmatically opened their gates to anyonewho canhelp themwinning

seats.Thepartieswhountil2012dominatedthepoliticalscenewerepreciselythosewho

succeededinfieldingcandidatesdrawnfromthelocalelites,inalltheirdiversity.

Somescholarshadalreadynoted that “thepersistenceoflocalpatternsofdominanceand

subordinationexplainswhy the riseofaparty like theBSPhasnot translated intoadeep

structural transformation of local caste hierarchies and into a redistribution of economic,

socialandpoliticalopportunities”(Jeffrey,Jeffery,andJeffery2008a,1365).

If social and economic inequalities have indeed persisted, I find however that the

transformationoflaborrelationsbetweenlandedgroupsanddaily-wageearnerclasses,or

the breaking of economic dependency between these two groups, has facilitated the

development of an autonomous political agency among subaltern group, leading to an

electoralcompetitionlessaffectedbytraditionalcastehierarchies.

ItmaybesothattheriseoftheBSPhasnotledtoamajorredistributionofeconomicand

socialopportunities,asindicatedbyJeffreyandtheJeffery’s.Butpolitically,theriseofthe

BSPhasledtoabreakdownoftraditionalbondsofsubjugationwhichhasledlocalDalit

communities to make their own local political choices. It may not be reflected in the

sociology of the candidates of the BSP, which does not leave much room to Dalits to

become representatives, but it is reflected in the support that the BSP gets from Dalit

voters,andthroughthepreferentialdistributionofpublicjobstoDalitsbysuccessiveBSP

governments.

The third and final observation is that these tectonic economic andpolitical shifts have

beenaccompaniedbyatremendousamountofviolenceofvariouskinds.Theprofessional

domains from which this class of business politicians has emerged are also the most

criminalized sectors of the economy. The violence that regulates business life and

transactions transpired to theirpoliticaldealings.Thepolitical assertionof lowercastes

292

hasalsobeenmetwithviolent reactions fromdominantgroups,whodonothesitate to

resorttoviolentmeanstomaintaintheirascendency.

Finally,thecontextofexacerbatedcompetitionhascreatedresentmentamongthelanded

dominantgroups,whofacecompetitionfromotherbackwardgroupsandtheminorities.

InEasternU.P.,criminalorganizationsdominatedbyuppercastegangsterpoliticianshas

alsocontributedtothepoliticalresilienceoftraditionalelites.Wecannotsayhoweverthat

thepolitical class isundergoing aprocessof classhomogenization.There is still a large

diversityofbackgroundamonglegislators,notablythosecomingfromreservedseats.But

thereisaprocessofintegrationoflocalpoliticalandeconomicelites,thatiswellaligned

withtheevolutionofparties’electoralstrategies.

Inthelastchapterofthisdissertation,Icompareparties’organizationsandthewaythese

organizationsrelatetolocalelites.

293

Chapter6-Interpartycomparison

In this final chapter, I examinehowparties relatewith local elitesandnotablywith the

newlocalelitesthathaveemergedsincetheearly1990s. Idividethisquestion intotwo

sub-sections.The first sub-sectiondealswith themethodsparties followtorecruit their

candidates, including the methods they follow to source information on potential

candidatesandtherolepartyorganizationsplayinthatprocess.

Inthesecondpart,Idiscussthemainconsequencesofpartyorganizationalvariationson

thecomparativestrengthofregionalpartieswithregardtonationalparties,onparty-elite

linkages,onthecriminalizationofpoliticsandonthequestionofdemocratizationof the

politicalsystem.

TheargumentIwishtoofferisthatparties’organizationalstructuresandstrengthimpact

thewaytheyrelatewithpoliticalelites,dependingonhowopenorclosedtheyare.Some

partiesforinstancehaveacandidaterecruitmentprocessthatismoreopentoindividual

outsidersandindividualpoliticalentrepreneursthanothers.Theseoutsidersgenerallydo

nothave thevocationofbecomingpartycadres,or formalmembersof theorganization

(asopposedtomembershiptotheparty).ApartyliketheBSPmustrelyonoutsidersto

win seats, which leads to a division of labor between the party’s organization, mostly

populatedby JatavDalits, and the candidates,whoremainoutsiders to theorganization

evenaftertheirelection.

This leads to a second argument,which is that the success of regional parties in recent

yearsislinkedtotheirabilitytoattractstrongcandidates,drawnfromthelocalelites,and

particularlyfromthenewelites.Putsimply,regionalpartieshavebeenmoresuccessfulat

winning seats than national parties because of their ability to attract strong candidates

whohaveincentivestocontestandtheresourcesthathelpwinningseats.Intheprocess,

the old Congress political elite – still inscribed in the old forms of political and social

dominance–hasbeensidelinedbymembersofthesenewsocialandeconomicelites.

The third argument is that there are drawbacks to this elitist recruitment by regional

parties, which is that it has contributed to the formation of a largely predatory rent-

294

seekingpoliticalclass,inclinedtomisconduct,corruptionandabusesofpower.Thismay

alsopartlyexplainwhytheturnoverofelectedrepresentativesissohigh.

Finally, the integration of political and economic elites within the regional parties is a

significant transformationas itputsunderquestionthedemocratizationpotentialof the

riseofbackwards.WhiletheAssemblybecomesmorerepresentative,italsoremainsvery

elitist,althoughindifferentwaysthanbefore.

Political parties in India are peculiar bodies. They tend to be highly centralized,

personalized, and weakly organized (Manor 2003, 2005, Wilkinson 2015, Wyatt 2013).

Thereisagooddealofdifferenceshoweverbetweentheirorganizationandtheirlevelof

institutionalization.Theabilityofparties todevelopastrong localpresence iscrucial to

their ability to connect with local elites, as they become part of the daily life of a

constituency,ratherthansimplybeamachinetocontestelections.Astronglocalpresence

isalsocrucialtothesourcingofinformationthatguidethenominationprocess.

In all these aspects, regional parties have had considerable comparative advantages

againstnationalparties,whichsufferfromtheircentralizedcharacterandtheirurbanand

uppercastebiases.

6.1.Candidatesselectionprocess

Inthissection,Icomparehowthemainfourpartieshavebeenselectingtheircandidates

in recent elections.As a rule, candidatesnomination tends tobe centralized andplaced

under the control of parties’ leadership (Farooqui and Sridharan 2014). Where parties

mostlydivergeisonthecriteriatheyfollowfornomination,thekindofinformationthey

rely on to identify and nominate candidates, and the processes used for sourcing that

information.Thosedifferencesreflectvariationsinorganizationalstructuresandstrength

betweenparties.

As far as nomination criteria are concerned, one can identify four main selection

principles327. The first is affiliation. Some parties insist that their candidates be long

327IamthankfultoNeelanjanSircarfordiscussionsonthistopic.

295

affiliatedmembersof thepartywhile othershavenoqualmsnominatingoutsiders.The

second principle is personal loyalty, or personal ties between the candidates and the

party’sleadership,orbetweenthecandidatesandtheleadersoffactionswithintheparty.

Nepotismandthedistributionofticketstokithandkinalsofallunderthatcategory.The

third principle is a broad pragmatic category of ‘winnability’. Parties who follow this

principle tend to select candidates on the basis of their expected ability to win a seat,

following a number of pre-set criteria, such as money, muscle, personal following and

reputation.Thelastprincipleisticketauction,aratherextremecaseinwhichticketsare

simplysoldtothehighestbidder.

Iamdeliberatelynotincludingideologyinthislist.Iamnotimplyingthatcandidatesare

not at times expected to adhere to some basic commonly instituted party values or

aspirations, but ideological considerations do not constitute a determining criterion for

candidate selection. It is commonly admitted that ideologydoesnot serve asdistinctive

factortodifferentiatepartiesinIndia,barringafewexceptions(Hasan2010).Ifanything,

ideological ties between candidates and parties can be a component of the category of

affiliation.

It is never the case that only one criterionprevails over all others. For instance, tickets

maybeauctionedamongcandidateswhoarealsodeemed ‘winnable’.But the fourmain

parties in Uttar Pradesh are positioned differently with regard to those criteria, some

predominating over others. These variations offer a mean to compare parties’

organizations.

Thesourcingof information is thesecondimportantaspect tocandidatenomination.All

parties claim to follow a systematic information sourcing process that starts from the

ground, inconsultationwith theircadreandrank-and-file.Theseprocessescanbequite

sophisticated and all parties now do extensive data gathering on potential candidates,

insisting on the scientific character of their methods. Parties differ on the treatment

processing theydoof that information.Sometend to followthe indications theyreceive

fromtheirgroundorganization.Othersapplyacentralizedtop-bottomcandidateselection

process,orrelyoninformationgatheredfromsourcesexternalfromtheparty.

296

TheCongressandtheBJPfallinthatlattercategory.TheBSPreliesonthestrongestlocal

informationsourcingmechanismamongallparties.TheSPfallsbetweenthesecategories.

Itsnominationprocess reliesonground informationbut isheavilybiasedby leadership

intervention.

6.1.1.TheCongressParty:stickingwiththedeadwood

Until the party’s split of 1969, the process of Congress candidates’ selectionwas highly

centralized, under the control of the party’s high command. Congress leaders in Delhi

werethenwarythatlocalMPsandMLAsdevelopedasenseofpersonalownershipoftheir

constituency and organized a rotation system under which a third of the party’s

representativeswasforcedtoretireafteraterm(Graham1986,211-12).Atthesametime,

thepartyleadershiphadtograpplewithacomplexsetofcriteria,includingcaste,personal

claims, regional claims, the need to accommodate demands from the party’s ‘frontal’

organizations (YouthCongress,Mahila Samithi,, theunions, etc.).Taking these local and

regionalfactorsintoconsiderationanddealingwiththeconsequencesofdecisionstaken

wasa complicatedbalancingact that, to theeyeof the leadership justifieda centralized

process(Roy1966,1967a,b).InhisstudyoftheCongressParty,Kochanekunderlinesthat

local caste and religious configurations were also key variables for the choice of

candidates(Kochanek1968).Otherscholarshavealsohighlighted theroleof factions in

thedistributionoftickets(Brass1964a,1965,Weiner1967).

After1968, state leaderswere “givenwidediscretiontonominateasmanyincumbentsas

theywished,tobuilduponexperienceandtoexcludedefectors”328.Thisencouragedfurther

aclientelisticdistributionoftickets.

The splits in the party and the rise of Indira Gandhi led to a new phase of centralized

controloverthenominationprocess,inwhichloyaltytothehighcommandandtheruling

family often prevailed over other considerations. Party loyalists were also entitled to

distribute tickets in their area of local influence, often bypassing the State Committee’s

rulesandprocedures.Thisamongotherfactorsledtoadeclineoftheparty’sorganization,

andofitscapacitytoreachvoterslocallythroughitscadre(Manor2003).Afterthedeath

328Graham,ibid.,p.212.

297

of Indira Gandhi, there have been a series of attempts to introduce some measure of

internaldemocracy,whichremainedlargelyinconclusive329.

Despite theestablishmentofa formal listofselectioncriteria, theCongress traditionally

recruiteditscandidatesamongthelocaltraditionalelitesandnotabilities,mostlyfromthe

upper castes (Meyer 1969). After Independence, the largest number of Princes who

enteredpoliticscontestedonCongresstickets(Richter1977).AswehaveseeninChapter

4,theCongressneverdepartedfromitsuppercastebias.Eventhoughitnowdistributes

tickets across caste groups, the remaining strongholds are usually held by upper castes

MLAs.

TheCongressnominationprocess inrecentyears(post2003election)has followedtwo

divergentandsomewhatcontradictoryroutes.The first route is thedeliberatechoice to

useelectionsasameantoreviveitsorganizationandconnectwithitslostsubalternvoter

bases,byselectingcandidates“fromthegrassroots”.Betweenelections,thepartyspendsa

considerable amount of time scouting for potential candidates from local communities,

individuals involved in the lower strataofdemocratic institutionsorpeople involved in

localmobilization.Theyseektoinductthemwithinthefrontalorganizationsoftheparty–

especiallytheYouthCongress–withtheavowedobjectiveto‘groom’futuregenerationsof

newpoliticians.

Thesecondfeature,moreatraitthanadeliberatestrategy,isthecultivationofoldties,or

thedistributionofticketsaccordingtothelongstandingaffiliationofsomeofitsmembers

and their own affiliates. Having experienced a severe decline in Uttar Pradesh, the

Congress party seeks to reward the loyalty of its longstanding associates, even though

some were associated to the decline of the party. As a result, the party nominates

candidatesregardlessoftheirabilitytowinseatsandoftencontradictsitsotherobjective

ofrejuvenationoftheorganization.

In the2012stateelections,aCongresspartyworkerexplained thatacertainnumberof

tickets (about ten) in Rohilkhand had to be distributed to followers of N.D. Tiwari, a

former Minister, Congress U.P. President and Governor, whose career had ended in

329Manor,op.cit.

298

disgrace threeyearsearlier330.Leadersof thepast thusretainsome influencebyhaving

“their people” contest election in their past areas of influence.Manypartyworkers and

aspiringcandidatescomplainedtomethattheentrytothenominationwas‘cloggedupby

deadwood’thatthepartywasunwilling,orunable,togetridof.

The continuing influence of past leaders is not the only way the party contradicts its

internaldemocracyagenda.Inthe2009and2012elections,theCongresspartyattempted

toorganize anambitiousground consultationeffortbyorganizingand institutionalizing

theconsultationofvariouscomponentsofitsorganization,settingupacomplexnetwork

ofparallel informationchannels. In the runup to the2012stateelections, theCongress

soughtinformationandsuggestionsonpotentialcandidatesfromfourdistinctsources.

The first level of consultation is the local level – block level committees and ward

committees, for the cities.These committees includeall formalCongresspartyworkers.

They report to anAssembly segment committee, itself divided into various subsidiaries

representing segmentsof theelectorate (ayouthwing, anOBCwing, aminoritywing, a

women’s wing, and so forth) 331 . These subsidiaries also report to their mother

organization,atthestatelevel,andthenatthenationallevel.Thus,interferencesbetween

these different layers of the Congress subsidiaries are frequent. Constituency-level

committeesareexpectedtosendalistofonetofivenames,asproposedcandidates.

Thesecondlevelofconsultationisagroupcomposedoftheparty’sofficeholders(Pradesh

Congress Committee), enlarged with the U.P. members of the All India Congress

Committee (AICC), as well as a selection of appointed former MPs and MLAs. This is

probably the most dysfunctional level since each member of that group, expected to

provide counsel to the party, is rid by factional divisions and individual conflicts of

interests.

330InterviewwithaCongresspartyworker,Lucknow,March2009.331AmemberoftheCongressPradeshCommitteesharedwithmethatthefactthatthepartyhascaste-basedwingsforallsegmentsof thepopulationbutnonefortheuppercasteswas itselfanacknowledgment that theparty remainedupper-castedominated. Interview inLucknow,March2009.

299

The third level is agroupof10observersappointed fromandby theAICC.The state is

dividedintotenterritorialunits,eachobserverresponsibleforone.Theyareexpectedto

traveltheirassignedterritoryforaperiodofsix-months,conductlocalmeetingswithlocal

partybranches, supervise theorganizationof campaignyatras, inwhich theyassess the

mobilization capabilities of potential candidates.At the endof the exercise, they send a

confidentialreporttotheparty’shighcommandastowhomshouldgetanomination.

TheseobserversdonotnecessarilyhailfromUttarPradeshandareappointedonthebasis

of theirperformance in their stateoforiginand loyaltywith theparty’shighcommand.

Theyusuallyconformtothreetypesofprofile:seniorAICCmembersputinchargeofthe

campaign,youngpoliticianswhohavedistinguished themselves in theirhomestateand

are ‘groomed’ by the party leadership for higher responsibilities, and young foreign-

educatedsonsofpoliticalfamiliesclosetothehighcommand.

Theirpresenceisoftendeeplyresentedbylocalcadres,whoconsiderthemasoutsiders,

illegitimateandincompetentonlocalpoliticalmatters.

Finally, a list of proposed candidates is submitted by the MPs, sitting MLAS and other

‘senior leaders’. Sitting MLAs are asked to suggest who should contest in the

constituenciessurroundingtheirown.

The information provided by these four sources is then compiled for a State Election

Committee,whichconvenes inDelhi.Thisstateelectioncommittee,chairedbytheparty

President,SoniaGandhi, counts21members,allnominatedby theChair.Theyclear the

names proposed into panels of candidates. These panels are then sent to a central

screeningcommitteethat“seeifthebalancesarecorrect”332,meaningthatitensuresthat

variouscastegroupsandfactionsareappropriatelyrepresented.

Thisratherelaboratestructureleavesplentyofroomfordiscreetinterventionsfromthe

partyleadership.Interferencefromthetopcreatesdeepresentmentintherank-and-file,

particularly when the candidates nominated do not even figure in the original lists of

names sent by the local branches. As a result, and despite all the efforts to deploy a

332InterviewwithUttarPradeshPCCPresident,RitaBahugunaJoshi,inLucknow,July27,2011.

300

participativenominationprocedure, the tickets’distributionprocess remainsmarredby

opacity and nepotism, which leads to further internal divisions and organizational

disarray. As Kanchan Chandra puts it: “Theprocessof ticketallocation in theCongress is

virtuallyunconstrainedbyformalrules”(Chandra2016b,227).

Aswith other parties, the nominationprocess gets oftendisturbed in the lastmoments

preceding the nomination filing deadlines. Once other parties publish their list of

candidates,theytrytopoachcandidates,nominateincumbentrejectedbyotherparties,or

tomakelastminutestrategicchanges,inviewofotherparties’nominationstrategy.

In 2012, a party functionary part of the Congress mobilization team for reserved seats

explainedtomeinthedetailhowtheirinitialstrategyconsistedinwooingnon-JatavDalits

–Pasisinparticular–whocouldbemoreeasily“detached”fromtheBSP.Butwhenlocal

BSPJatavleadersexpressedtheirinterestforcontestingundertheCongressbanner,they

quicklychanged their list,effectivelyundermining themobilizationefforts that theyhad

consentedweeksandattimesmonthsbeforetheelections.

Thecommitmenttointernaldemocracyandinclusionisfurthermorecontradictedbythe

socialcompositionoftheparty’sleadership,whichremainsprimarilydominatedbyupper

castes and Delhi loyalists. The party may distribute tickets across castes, but the

organizationremainsreluctanttoincludelowercasteleadersinapositionofleadership.

One finds party leaders from lower castes but they are few and often sidelined. One

example isP.L.Punia,a former JatavDalitChiefSecretaryunderMayawati,whohad left

thecivilservicetojoinCongress(hewonhisLokSabhaseatinBarabankiin2009butlost

in2014totheBJPcandidate). Insteadofprojectinghimastheparty’sDalit face,hewas

largelysidelinedwithintheorganization,tothepointthattheDalitmobilizationprogram

(labeled“Mission89”forthe89reservedseats)wasentrustedtoayoungprincelingfrom

Rajasthan, son of a former Congress Minister. A party worker at the Congress office

complainedtomethatthepartycouldnothavefoundalesslegitimatefiguretomobilize

DalitsinU.P.333

333InterviewwithaCongresspartystaffer,Lucknow,March2009.

301

Theover-representationofuppercastewithintheCongressorganization inthepasthas

beenwelldocumented(Jaffrelot2003b,Zerinini2009).Since1988,theCongresshashad

tenstatePresidents.Sixofthembelongedtouppercastes,onetoanaristocraticMuslim

family(SalmanKhursheed),twoOBCs(SreeprakashJaiswalandBalramSinghYadav)and

one Scheduled Caste (Mahavir Prasad, from Bansgaon). All of them are loyalists to the

Gandhifamily.Theycollectivelycontestedforty-fiveLokSabhaseatsandtheylosttwenty-

sixtimes.Oneofthem,JagdambikaPal,defectedtotheBJPin2014.

The ratio of upper castes members of the Pradesh Congress Committee is roughly

equivalent.Zerininiestimatesthatbetween1991and2000,57percentoftheparty’svice-

presidentsandgeneralsecretariesbelongtotheuppercastes334.

Thatratioisexactlythesamein2916,as22of38Vice-Presidentsareuppercastes.There

isonlyoneScheduledCasteVicePresidentandonlyfiveMuslims.AttheCityandDistrict

Presidents (CCC/DCC), 55 per cent of the office holders are upper castes, 25 per cent

MuslimsandbarelyafewScheduledCastes.

Besides,theAICCSecretariesandGeneralSecretaryinchargeforUttarPradeshusuallydo

notincludeanyonefromthestate.Thelogicistoavoidconflictofinterestsandfactional

feuds spreading to national bodies. It also has the effect of marginalizing the state

leadership.

In short, formal rulesandpledgesof inclusivenessare thwartedby thecentralizationof

decision processes, as well as by the upper caste and traditional elite biases that have

subsistedintheparty.

6.1.2.TheBJP:withinthefamily

TheBJPisusuallydescribedasacadre-basedpartywhosemembersarestronglyattached

tothepartythroughideologicalbonds.Inreality,theorganizationalstrengthoftheBJPis

usually overstated, as it as been observed in other states (Manor 2005). Also, deeply

334Op.cit.,p.58.

302

entrenched political rivalries and internal caste conflicts often trump the ideological

cohesivenessoftheparty.

TheBJPcompensates itsorganizationalweaknessbyrelyingontheRSSformobilization

and, in recent times, by relying on modern forms of communication and new forms of

campaigntechnologies.

Formallyspeaking,candidatesareselectedbytwobodiesinstitutedforthatpurpose:the

StateElection’sCommittee (SEC), and theCentralElectionsCommittee (CEC), located in

Delhi.Forstateelections,theSECmerelysuggestslistsofpotentialcandidatestotheCEC.

Thatlistisscreenedandevaluatedbyasmallgroupofappointedfunctionaries.Inreality,

the process is evenmore centralized,with a limitedmembers of senior partymembers

andtheparty’spresidentcallingtheshotsonwhogetstocontest.

BJPticketsaredistributedaccordingtothreemaincriteria.Thefirstoneiscaste,following

amethodthatdiffersfromotherparties.Ratherthanadaptingticketdistributiontolocal

caste configurations, theBJPpre-defines the caste or the caste combinations it seeks to

mobilize and then distribute tickets accordingly. Thus, in the 2000s, the BJP started

distributing many tickets to non-Yadav OBCs such as Kurmis, Lodh, Rajbhars and

Kushwahas; and to non-Jatav Dalits, such as Pasis, Sonkars and Rawats. The party also

appointedKalyanSingh,aLodh,asStatePresidentandthenChiefMinisterin1991-92,and

between1997and1999.

Asaresult,themanagementofcasteequationswithintheBJPisabalancingactbetween

groupsthatareoften inconflictwitheachother.UnliketheBSPwherethecoresupport

base dominates the organization, the BJP’s organization is frequently undermined by

internal caste conflicts, particularly between upper castes leaders – who can display a

strongsenseofownershipoftheparty–andOBCleaders,oftenconsideredbytheformer

second-ratepartyfunctionaries.

The second criterion is loyalty and compatibility with the RSS, the BJP’s parent

organization.MostoftheBJPbackwardandDalitcasteleadersareRSSmembers,orhave

beensocializedandeducatedthroughHindunationalistorganizations.Manyofthem,such

asKeshavPrasadMaurya,arepracharaks(“propagandist”)orVHPfunctionaries.

303

TheRSSvets theBJPcandidatesandprovide thepartywith theirownsuggestions. It is

entitledtodothatnotonlybecauseofthehistoricaltiesthatbindsthepartytoit,butalso

becausetheBJPremainsdependentfromtheRSStomobilizevoters,eitherbyproviding

campaignworkersduringelections,orthroughitsmultipleground-levelorganizationsin

between.Infact,theBJP’sforaysintothelowerstrataoftheelectorategoesthroughthe

deploymentofRSS-linkedorganizations,thatprovidesocialservicestothemoredeprived

sectionsofthepopulation(Thachil2014).

Like in Madhya Pradesh or other states, RSS pracharaks form the core of the party’s

organizationandmaintainitscohesivenessandinternaldiscipline(Jaffrelot1998).

The third criterion is a general definition of winnability, essentially based on an

assessment of past performances. In recent years, the BJP relies more and more on

technology and private technology firms and survey companies to generate the data

necessary to make those assessments. These companies also provide campaign design

services,surveydata,andcampaigncoordinationservicesthatareseenassubstitutefor

traditionalformsofcampaigning.

TheBJPalsowelcomesdefectors fromotherparties,usually tosendthesignal tovoters

aheadof theelectionsthattheseturncoatsswitchedtotheBJPbecausetheyexpect it to

wintheincomingelection335.

Inthatorganizationalsetup,stateorganizationandlocalbranchesareonlyoneofvarious

sourcesprovidinginformationtotheleadershiponwhoshouldgettickets.UnliketheSP

or theCongress, fewBJP leadersareentitled todistribute tickets to their followersona

discretionarybasis.Therearefewcasesofleaderswhohavethepowertoweighonticket

distributionintheirownarea,duetotheirparticularstatusanddominationoflocalparty

structures. Yogi Adityanath, in Gorakhpur, is one example. He keeps pushing for the

distributionofticketstofellowRajputsinNortheasternU.P.andisusuallyobligedbythe

party,whofearstocrossthisall-powerfulpoliticalandreligiousfigure.

335IamthankfultoShivamVij,whopointedthistome.

304

HukumSingh,inSaharanpurisanotherexample.InanareawheretheBJPdoesnothavea

strongpresence,hecallstheshotduetoitsprominenceamonglocalGujjars,thestrongest

groupinthearea.

Butasarule,thepartydoesnothavemuchtosay.Inrecentyears,theBJPhasdeveloped

thehabitofappointingweakstatePresidents,whodotheCentralCommand’sbidding.

TheBJPalsodoesnotaccommodatepoliticalfamilieseasily.Allegiancetothepartyprimes

and theparty leadershipdoesnotwant to see strong regional leaders institutionalizing

themselvesthroughdynasticpolitics.RajnathSingh,formerChiefMinister,BJPPresident

and Union Minister, has not been able yet to obtain a seat for his son, due to internal

resistanceandrivalries336.

Overall, the logic of affiliation binding the candidates to the party is strong. Candidates

withanRSSbackgroundtendtobeprivilegedtoothers.Aswesawinchapter3,careers

within the BJP tend to be longer. There are also less people leaving the BJP for other

partiesthantheotherwayaround(Jaffrelot1998,Manor2005).

The dependence of the BJP to the RSS and the urban and upper caste bias of its

organizationhavemeantthattheBJPhasnotsucceededtodevelopastronglocalpresence

in rural areas. The party did mobilize in the late 1980s and 1990s through large-scale

politicalandreligiouscampaigns,buttheeffectquicklyfadedout.Afteraperiodofpeakof

popularityintheearly2000,theBJPdeclinedinUttarPradesh.Itsorganizationretracted

itselfinurbanareasandtheylostmuchofthegroundpresencetheyhadinruralareas.

Inrecentyearsandparticularlyinthe2014GeneralElections,theBJPcompensatedforits

lack of local presence by saturating the public space with party images and sounds.

Heavily centralized campaigns relying on modern forms of communication tend to

supplanttraditionalformsofgroundmobilization,andtendtodiminishtheimportanceof

candidatesthemselves(Jaffrelot2015b).

336IamthankfultoShivamVij,whopointedthistome.

305

6.1.3.TheBahujanSamajParty:outsourcingcandidates

The BSP has by far the most centralized candidate selection process of the four main

parties of Uttar Pradesh. The allotment of tickets is determined by Mayawati herself,

assistedbyahandfulofaideswhoprovideherwithinformationonthecandidatesandon

possible alternative candidates. The principle guiding the nomination is the spending

capacityofthecandidate,whoisrequiredtomakeanupfrontdonationtothepartybefore

evenbeingconfirmedforacandidacy(FarooquiandSridharan2014).Theamountofthat

‘donation’isfixedbytheparty,likeascale,andnearlydoubleseveryelection337.

Thereasoning is that individualsabletoraisetherequiredamountandinadditionfund

theirowncampaignswillbeinagoodpositionto‘delivertheseat’.Thisismadepossible

bythepeculiarityoftheBSP’ssupportbase,largelycomposedofDalitvoters,andbythe

factthatwinningthresholdsinU.P.tendtobelow(seechapter3).

Guhahasshownthatingeneralseats,committedDalitBSPsupportersarewillingtotrade

descriptive representation and its benefits in order to ensure the party’s victory (Guha

2011).GiventhefactthattheprobabilityofDalitcandidatesbeingelectedingeneralseats

isverylow(Jensenius2012),itishardlyatradeatall.

Thisisthebaseofthemechanismdescribedasvotebanktransferability,ortheabilityofa

party to get its core supporters to transfer their votes to candidates belonging to other

ethnicgroups,forthesakeofpartyvictoryandprogrammaticgains.

IfoneassumesthatDalitsrepresentonaverage20percentoftheelectorateinanygiven

constituency,andassumingthattheBSP’svoteshareamongDalitvotersremainshigh,the

BSP candidate only need to mobilize a residual share of the electorate in order to win,

given the low victory thresholds. This is why the party can make the assumption that

nearlyanyindividualwiththerightcombinationofcasteidentityandresources“cando”

asacandidate.

337FarooquiandSridharanquotethefigureof5to10millionrupeesintheir2014article.Inthe2016 elections, the ‘ongoing rate’ start at 25 million. That figure also increases as the time ofnominationapproaches.

306

Asaresult,candidateselectionintheBSPisquiteopenamongthosewillingtopayfora

ticket. In the seats where several candidates offer to contribute, tickets are simply

auctionedamongaspiringcandidates.Itisalsoquitefrequentthatconfirmednominations

getcancelledaheadofthecampaign,andtransferredtoahigherbidder.

While thenominationprocess is indeedhighlycentralizedandreliantonmoney, itdoes

involve the participation of the party’s ground organization. Ticket allocation does not

onlydependonthespendingcapacityofthecandidates.Localcasteconfigurations–both

castedemographicsandthelocalhistoryofinter-casterelations–areconsideredfirst,on

thebasisofassessmentsproducedbythelocalbranchesoftheparty.

Theselocalbranches,essentiallycomposedofJatavDalits,sendonaregularbasistothe

party high command information regarding the local caste balance, local political

alignmentsandsocialalliances,aswellasany informationregardingsignificantpolitical

events,orincidencesofactsofatrocitiescommittedagainstDalits.Itisonthebasisofthat

informationthatthepartydetermineswhichconstituency-levelcastealliancetofoster(it

needsnotbeasinglecastealliance.Alternativealliancesarealsoconsidered).Thequest

fortheindividualwhowillcontestinthenameofthepartycomesthussecond.

ThisishowtheBSPachievesinclusiveness,bylocalizingitsticketdistributionprocessand

adapting it to local circumstances. There can be instances where certain alliances are

pushed–likethegreaterdistributionofticketstoBrahmincandidatesin2007–butonly

inconstituencieswhereitmakesrationalsensetodoso.

TheBSP’spyramid

TheBSP iscertainlyU.P.’smostcentralizedand leastdemocraticparty. Itsstructure isa

strictpyramid,whereinformationflowsupwardandordersdownward.

ThePartyPresidentconcentratesmostofthedecision-makingpowerwithintheparty.She

issurroundedbyahandfulofcloseaidesandseniorpartyofficebearers,entirelydevoted

toher.

307

Revealingly, the “leadership” section of the party’s officialwebpage provides only three

names,besidesMayawati’s:SatishChandraMishra,theBrahminfaceofthepartyandthe

personinchargeofstrategyandorganization;Dr.SureshMane,aBombay-basedacademic

andunionist338,inchargeofdevelopingtheparty’spresenceinSouthIndiaandtheNorth-

East;andNaseemuddinSiddiqui,GeneralSecretaryofthePartyandnumbertwoMinister

inMayawati’sCabinets.

Theseare theGeneralSecretaries,alsomemberof theparty’sNationalExecutive,which

includes between thirty to forty people, all appointed by Mayawati. Its main functional

roleistogiveapretenseofinternaldemocracy(theNationalExecutive“elects”theParty

President,whoneverhadtofaceachallengersofar).ThesecondpurposeoftheExecutive

Committeeistogivevisibleordescriptiverepresentationtotherangeofcastestheparty

aimstorepresent.Postsarethusdistributedtocastefigureheads,followingKanshiRam’s

precept of “jiski jitni sankhyabhari,uskiutnibhagidari”, that is the distribution of party

positionsproportionatetogroups’demographics.

The transformationof theBSP intoa catchallpartyhasmeant that theparty leadership

has created space for the representation of more groups within the organization. Data

collectedby Jaffrelot shows that in1996and2000,about45per centof the stateparty

officebearerswereOBCs,mostlyMBCs.Post-2000,theratioofuppercasteincreased,with

the induction of new office bearers, mostly Brahmins. The inclusion of a significant

numberofBrahminsandofafewYadavofficebearerscirca2007createdtensionsamong

theothermembersoftheCommittee.

Thesetensionsalsoreflect thegameofpersonaland individualambitionsthatplaysout

within the higher echelon of the party’s organization. Caste representation is literally

nominalandnocastecanavailofanycollectivestrengthwithintheparty.Besides,every

memberowesitspositiontoMayawatiandthereforeremaindependentfromher.

Casterepresentationwithinthepartyisthushighlyindividualized.Juniorpartymembers

frequently accuse senior members of abusing of their position and of promoting their

338HeisthePresidentoftheMumbaiPortTrustWorkersUnion.

308

individualsandfamilies’interests.Thus,theNationalExecutivecountsanumberofnewly

createdpoliticalfamilies.

Asaresultofthesetensions,thereisahighturnoveramongtheofficebearers.Manyofthe

prominentMBCfiguresofthepartyhaverecentlyleftthepartyorhavebeendismissed339.

Thesamelogicofcasteecumenismandpowerplayappliestotheparty’ssectorialbodies,

such as the women wing, the youth wing or the various caste-based platforms that the

party has created in order to mobilize various caste segments of the electorate. These

sectorialbodiesareweakandserveessentiallytoco-optlocalleadersandindividualswho

competewithintheseorganizationsforaccesstofavorsandappointments.

These sectorial bodies play a satellite role compared to the local branches of the party,

which remains controlled by Jatav Dalits, and which covers each district and each

constituency in a pyramidal structure, following both political and administrative

boundaries.

Therearefourlevelsoforganization,interwovenwitheachother.Thefirstlevelisatthe

zilla (district) level, where the party organization replicates the Zilla Parishad

organization. The party appoints an Adyaksh (“Chairman”) who supervises the entire

district organization. He is seconded by 4-5 members, who form a Zilla executive

(“Satyush”).Eachmemberofthatexecutiveisdrawnfromoneofthemajorcastespresent

inthedistrict(thethumbruleisthateachcastenumberingabove50,000individualsover

five seats within a district gets to be represented in the party’s Zilla executive)340. The

same organizational structure applies to Lok Sabha and Vidhan Sabha constituencies,

whereeachmajorcastegetsrepresentationwithintheorganization.Eachconstituencyis

thendividedintosectors,coveringcities,townsandvillages.TheconstituencyofJalalpur

forexample,isEasternU.P.,isdividedinto37sectors,including29Panchayats.

339 One can mention Babu Singh Kushwaha, Swami Prasad Maurya, Romi Sahmi and BrajeshVerma, all non-Yadav OBCs, who have left the party between 2012 and 2016. Mayawati alsoregularlyfiresnon-performingpartyleaders,particularlyafterelectoralsetbacks.340Interview with R. Pandey, defeated BSP candidate in the 2012 state elections, at the OberoiHotel,NewDelhi,2ndFebruary2013.

309

Each sector has an appointedadyaksh and before 2012 eachadyaksh was a JatavDalit.

Thus,whilerepresentationisorganizedandevenguaranteedforthemajorgroupswithin

a constituency andwithin a district, the structure is almost entirely controlled by Jatav

party cadre, trained and fully dedicated to their party work. They serve two main

functions.TomobilizetheDalitbaseaheadofelectionsandmaintainacommunicationline

with thebaseof theparty–atground level–and toprovide theparty’shighcommand

with informationon localevents,oncasteandpoliticaldynamics,aswellas information

onpotentialcandidates.

During elections, the party appoints 10 local party members to each polling booth, to

canvass for the party’s candidates and report on eventual incidents during polling. The

party is thus able tomobilize four to five thousand cadres andworkers in eachVidhan

Sabhaelection (which in2012 countedonaverage3.16 lakhselectors). In recent years,

this effort if complemented with intensive mobile campaigns, through SMS and other

socialnetworkapplications(JeffreyandDoron2012).

At this levelof theorganization,party control isweaker, given the scaleandnumberof

booths requiring supervision. A BSP MLA from Akbarpur district told me once that the

localorganizationusedtobestrongerunderKanshiRam,andthatthepartysufferedfrom

theintrusionofrent-seekers.

WhiletheZillaandlocaladyakshsendinformationonpotentialcandidates,theyplaylittle

if not any role in the actual candidate nomination process. This role is devoted to the

higherstrataoftheparty,tightlycontrolledbyMayawatiwhomakesalltheappointments.

Thetopleadershipofthepartydividesthestateterritoryintozones,dividedintoMandals.

Mandal leaders are appointed by the party high command. Each Mandal leader is

responsible for the ticket distribution across four or five Mandals. Ten to fifteen zonal

coordinatorsworkunderthem.Theyareinchargeof“collection”,ofcontributionstothe

party’s coffers from candidates and aspiring candidates. They are also in charge of

organizing local ticketauctions,when thepartydecides to follow thatmethod.The first

two layers of that zonal organization are directly appointed by Mayawati. They are all

JatavDalits.AccordingtoaBSPMLA:

“Mandalleadersarepowerfulfigureswiththeparty.Theyoperateinproximitywith

the topcadresof theparty, the localpowerholders, theMinisters.Thecadre runs

310

like a parallel government. They are in charge of collection but they were also

makingmoney”341.

“After the2012elections,Mayawaticoncededthat thepartyorganizationwastoo

Dalitdominatedandsheauthorizedthecreationofnon-Dalitlocalexecutives,called

Baicharas.In2013,therewerefivenon-DalitBhaicharas(Brahmin,Kurmis,Thakur,

Muslim andMaurua/Koeri). They triedwith Yadav but it’s not defunct. They [the

Baicharas]arenotaspowerful.EventheMuslimsarenottakenseriously”

Theseconcessionshavehardlyalteredpower relationswithin thepartyand theholdof

Jatav cadre.One reason is that thepowerbalance remains strongly tilted in their favor.

The second is that thesepartypositions arenot attractive for aspiringpoliticians, since

membersofthepartyorganizationdonothavethevocationofbecomingcandidates.

Thus, contrary to most parties in India, party work in the BSP does not constitute an

antechambertothenomination.Theydoexertinfluenceandthebusinessofticketauction

enables them tomakemoney, butotherwise, theyarepoliticaldeadends for individual

whoaspiretobecomeelectedrepresentatives.

ThisisacrucialaspectoftheBSP’sorganization,whichmaintainsastrictdivisionoflabor

between its organization, composed and controlled by Jatav cadres, and the pool of

candidates and elected representatives, who are largely drawn from outside. Thus, the

partydoesnotrequireitscandidatestoadhereorcommittotheparty’sideology.Theyare

pragmaticallyselectedaccordingtoanestimationoftheirabilitytodeliverenoughvotes

to the party’s local Dalit base in order to win the seat. With an average SC population

oscillatingbetween20to22percentinnearlyeveryseat,thesecandidatesonlyneedto

deliver a fragment of the vote – 13 to 15 per cent, in order to secure victory.Hence, it

becomespossibleforthesecandidatestobeelectedonthebasisofsmallnumbers.Usually,

candidatesareexpectedtobringinafractionofthevoteoftheircommunity.Oneofthe

evaluationcriteriaisthecapacityofthecandidatetomobilizewithinitsowngroups.But

inseatswheredemographicallydominantgroupsaredividedintofaction,thecapacityof

thecandidatetomobilizeacrosshisorhercasteisalsotakenintoaccount.

341InterviewwithaBSPMLAinDelhi,February2013.

311

This systemoffersmany advantages to theparty. The first is that it candepend froma

base cadre devoted to the party’s work rather than to the advancement of their own

electivepoliticalcareers.Thisautonomyofthelocalorganizationisfurtherreinforcedby

thehomogeneityofitssocialcomposition.LocalJatavBSPleadersareusuallywelltrained

disciplined,educatedandideologicallyawarefull-timepartyworkers,entirelydevotedto

thecauseofthepartyandofitsleader.

The second advantage is that autonomy confers a certain degree of reliability to the

informationthatthebasesendstothepartyhighcommand.Ihavequicklyabandonedthe

ideaof collecting constituency castedemographic compositiondata frompartyworkers

andmembersofotherpartiesastheywouldfrequentlydeliberatelyover-estimatethesize

of their own community, or minimize the size of their opponent’s social base, which is

equivalentlybad.Mostofmyinterlocutorsacrossthethreeotherpartiesadmittedthatthe

BSP’s organization is by far the most ‘scientific’, the most disciplined, and that the BSP

holdsthebestdataonlocalpoliticalcontexts.

After the2012defeat,however, crackshavestarted toappearat thebaseof theedifice.

Theexperienceofinductingoutsiders,fromanarrayofnon-Dalitcastes,withtheparty’s

fold and the maintenance of a wall between the party’s local organization and the

representativesmeantthattheserepresentationsfeltlittleobligationtoworkforthemor

topursueanagendaotherthantheirown.Themainissuewiththeauctioningorsellingof

tickets is that it creates little senseofobligationof theMLAsorMPs towards theparty,

sincetherelationshipisfundamentallytransactional.

Thisdivisionoflaboralsoappliestoreservedconstituencies,wherelocalpartycadresare

discouragestocontestthemselves–forthesakeofmaintainingtheautonomyofthelocal

organization,avoidconflictsofinterests,butalsofromtherealizationthatthevoterswho

decidewhichDalitgetselectedarepreciselythenon-Dalitvoters,whomaynotvotefora

overtlymilitantambedkaritecandidate.Asaresult,theBSPdoesnotnecessarilypickJatav

candidates in reserved seats. Once again, local circumstances prevail over the Jatav

representationagenda.

312

Therehasbeenthusagrowingsenseofdiscontentwithintheparty’srankandfile.Inthe

2012 elections, the Congress and the SP fielded a number of Jatav candidates who had

previouslyworkedwithintheBSP.Someofthemwonseats,underSPtickets(nonewith

Congress).

ThisdiscontentmentshowsthatthesymbolicvalueofhavingMayawatiasChiefMinister

and theBSP inpower inLucknowhas its limitations.ManyBSPworkersprotested that

theirplighthadnot improved locallydespite themajority that they conquered in2007.

Thisdiscontentmentalsopushedsomeoftheparty’scadretousetheirpartypositionfor

their own private benefit. Thus, the commodification of ticket distribution is a recent

phenomenon,aspointedbymyBSPinterlocutor.

“In the early days, the candidate selection was done by Kanshi Ram. The role of

adyakshwasmerely to pinpoint at possible candidates. Now, people runwith the

ticketmoney”342.

Beitasitmay,thisparticularorganizationofticketdistributionandthedivisionoflabor

betweencandidatesandorganizations, illustrateshowlocalizedtheprocessofcandidate

selection is, basedonpragmatic assessmentof the local casteandpower configurations

rather than pre-determined caste dosage. The BSP will literally distribute a ticket to

anyonewhocanconvincethepartythattheycandelivertheseat.

TheBSP’sorganizationconfirms theparty’speculiarity in the Indianpolitical landscape.

Thepartyishyper-centralizedandisledinaquiteauthoritarianfashion,butitalsodraws

itsstrengthfromitslocalimplantationandfromthedevotionofastrongmilitantbase.

6.1.4.TheSamajwadiParty:factionalismandlocalelitesintegration

TheSamajwadiPartyhasadifferent systemof candidatenomination,whichreflects the

factional character of the party. Formally speaking, the party’s Central Parliamentary

Board is responsible for the nomination of all candidates. This board comprises the

NationalPresident (asChair),aGeneralSecretaryandup to fivemembersappointedby

342Ibid.

313

theParty’sPresident343.StateParliamentaryBoardsareconstituted in thesamemanner

and contain a higher number of members (up to seven). State Boards form a panel of

candidatesthattheysubmittotheCentralBoard,whotakesthefinaldecision.

In effect, the ruling family completely canvasses the nomination process. As noted by

Chandra,severalmembersoftheNationalParliamentaryBoardandoftheStateBoardare

Mulayam’skin(hisbrother,hisson).

The family however does not constitute a cohesive block. Each of them leads its own

faction thatcompetes for ticketsandpositions.Mulayam’sbrothersShivpalSinghYadav

andRamGopalYadavmeasuretheirstrengthbythenumberofticketstheycandistribute

totheirfollowers.ThesamegoesforotherGeneralSecretariesoftheparty,whoalsolead

their own faction. For example, Azam Khan reigns over ticket distribution in much of

WesternU.P.andRohilkhand.

Thepartywasnotalwaysthatcentralized.Theconcentrationofpowerwithinthehandsof

the Yadav family took place as the old guard of the party withered or passed away. In

recentyears,MulayamSingh lost four longstandingcompanions.RamSaranDas,aclose

associate of Lohia and former UP President of the Samajwadi Party, died in 2008.

Janeshwar Mishra, co-founder of the Samajwadi Party, passed away in 2010344. Mohan

Singh, three-timeMP fromDeoria and aGeneral Secretary of theParty, passed away in

September2013345.BrijBhushanTiwari,afive-timeSamajwadiPartyMP,passedawayin

2012. These were historic figures of the party who could hold their ground vis-à-vis

Mulayamandhiskin,andensuredthatthemanagementofpartyaffairswasmorecollegial

thanitisnow.

343ConstitutionoftheSamajwadiParty.Quotedin(Chandra2016b,230)344Nicknamed ‘ChhoteyLohia’forhisassociationwithLohia,JaneshwarMishrawasamemberofwhat was known as the Quartet from Ballia, a group of four socialist activists who would beinstrumental in the building of the PSP in Eastern UP and play a role in national politicssubsequently.TheotherthreememberswereGauriShankarRai,KashiNathMishraandChandraShekhar. Rai was Lohia’s secretary, Mishra shifted to Congress in 1984 and Chandra ShekharbecamePrimeMinisterin1991.345Hehadleftthepartylessthanayearbefore,overaspataboutthenominationofD.P.Yadav,acontroversialpoliticalfigure,inthe2012elections.

314

A consequence of power concentration is thatmanymembers of the ruling family have

been inducted into politics, reinforcing thus the family-holding character of the party.

Today16membersoftheYadavfamilyareactiveinpolitics,atvariouslevels346.

Fig.:6.1CaricaturebySurendra,TheHindu.

The factional distribution of tickets also applies at the district level. The party’s main

factionsarethemselvesdividedintolocalstrongholdsandlocalfactions,wherepoweris

distributed among local party bosses who control the party’s apparatus over certain

territories.KanchanChandra is right tonote that therearenootherprominentpolitical

familieswithintheSP,otherthantheYadavfamily347.Butsheoverlooksthatfactthatthe

distributionofticketsamongthelowerlayersofthepartyisverymuchdoneonafactional

basis,withinwhichlocalpoliticalfamiliesfigureprominently.

The SP is in fact reputed for accommodating familymembers into the party ranks, and

local leadersoften fight inorder to get their relativesnominated.OneexampleareAtul

PradhanandShahidManzoor,twoSPlocalbossesinWesternU.P.,whofoughteachother

andlobbiedtogettheirwivesnominatedinMeerut,forthe2016elections.

346Thoughnotallatthenationalorstatelevel.RatanSinghYadav,brotherofMulayam,remaineda block Pramukh in Saifai, the family’s native village, all his life. Another brother, Rajpal, isChairman of the district Panchayat in Etawah. The latest entrant, Abhishek alias Anshul Yadav,contestedthe2015Panchayatelections.347Chandra,ibid.,p.232.

315

ShahidManzoorisathree-timeMLAfromKithore.HeisaleadinglocalMuslimfigureclose

toMulayam.AformerStateMinister,heisthesonofManzoorAhmad,aprominentMuslim

SSP figure who won five terms in Meerut and Kithore between 1967 and 1980. Atul

PradhanisaformerPresidentoftheSP’sstudentwingandlocalGurjarpoliticalfigure.He

is close to Akhilesh and contested in 2012 in Sardhana, unsuccessfully. Due to its

proximitywithAkhilesh,AtulPradhanoverseesmuchof theparty’snomination in local

elections –municipal, Panchayats, aswell as studentunions’ elections.Hiswife became

ChairoftheMeerutZilaParishadin2010.Despitebeingchargedinseveralcasesorrioting

andmurder,hebenefits fromanofficialpoliceprotection,grantedby theChiefMinister

(ENS2013).

Nominationsarethusdistributedonthebasisofthreemaincriteria.Thefirstoneiscaste.

Local caste configurations play a major role in the selection of candidates. Besides, the

party seeks to maintain a nominal descriptive representation to various groups. Its

organization contains a lot of caste-based branches, usually led by a single political co-

ethnicfigure,whoseaimistomobilizestrictlyoncastelines.

The second criterion is the individual strengthof the candidates,who like in theBSP is

expectedtofinditsowncampaign(withtheexceptionofsomepartyleadersbelongingto

theparty’sfirstcircle).Candidatesmustalsodemonstratetheirmobilizationcapacity,by

organizing rallies, road shows when party dignitaries are visiting their area. They also

mustbeabletodrawsupportfromlocaleconomicelitesandprivatecompaniesassociated

withtheparty,whichalsocontributetocampaignfunding.

The third criterion is factionalism. SP tickets are distributed clientelistically within

factionsthatintegratevariouslevelsofpoliticalpower.Thesefactionsalsoincludeparty

sponsors,individualswhofundcandidatesbelongingtotheircasteortheirlocality.

OnesuchsponsorisNarendraBhati,aGujjarpoliticianbasedinNOIDA.Hesupportsthe

careerofvariousotherGujjarpoliticiansinWesternUttarPradeshbycontributingtotheir

campaigns and organizing campaign events. A former congressman, he joined the SP in

1989andwonthreetimesinSikandrabad.Helosthisseatin2002andfailedtoregainitin

2007.Buthis statusof sponsorhelpshim tomaintain itspositionandstatuswithin the

party.Bhati’syoungerbrother,SurenderisanaccountantwiththeUttarPradeshHousing

316

and Development Board. Another brother, Bijender, served as SP President for Gautam

BudhhNagar.

There is no clear division of territories among factions, whose members often compete

with eachother in the samedistricts or localities. These factions aren’t alsobasedon a

single caste identity. Instead, a faction leader will induct representatives from different

castes, throughpatronage, inorder to expandhis supportbase. Faction leaders and the

casteleaderswithinfactionstendalsotopromotetheirfamily’sinterestwithintheparty

(andattimesoutsidetheparty).

It is true then that a lot of political families within the SP are horizontal rather than

vertical.Politicalfamiliesexpandtheirinfluencebygettingmemberselectedinavarietyof

positions,ratherthansimplytransmittingaseattoapoliticalheir.Inthatcontext,factions

trumptheformalorganizationoftheparty.

TheSPorganization:authoritarianyetdecentralized

Intermsoforganizationalstructure,theSPfollowsaclassicpyramidalmodelwithbodies

institutedfromthenationaltotheblocklevel.Aparticularityofthatorganizationisthat

contrarytootherpartieswhotendtohaveonlyonebodyforeachlevel,theSPhastwo–

anexecutiveandanassembly(Shafiuzzaman2003,80).Thus,thenationalorganizationis

divided between a National Conference and a National Executive; the state level

organization is divided between a State Conference and a State Executive. Similar

structures are replicated at the district level, at the city level (where they are called

cooperatives)andthenattheBlocklevel(organizationandexecutive).

Executivemembersareappointedfromthetopandthevariousassembliesaremeantto

oversee the working of the Executive. In reality, these assemblies function as forums

enablingpartymemberstodeveloptheirnetworksandtradeinfluenceratherthancheck

onpartyleaderswhodrawtheirauthorityandlegitimacyfromtheparty’shighcommand.

The most important layer in the party’s organization is the district level. This is where

partyofficeholdersinteractwiththebureaucracy,wherealargepartofpublicresources

317

are concentrated. Localbranchesof theparty areorganizedaround factionswho rarely

extendbeyondadistrict.Thus,theparty’szilaadyaksh(“DistrictPresident”)concentrates

a lotofpowerand influence.Theyarealso fairlyautonomous, in theirdaily functioning.

Thepartyledthemdeveloptheirownaffairsorbusinessaslongastheysupplyresources

andmanpowerforelections.

The party also has a series of affiliated bodies: The Mulayam Singh Youth Brigade, the

Mahila Vahini (Women’s Brigade), the Lohia Vahini, the Chatra Sabha (“Students’

Assembly”)whoareessentiallyusedformobilizationduringcampaigns.

Nominations to theparty’svariousbodiesandexecutives is topdown.Aparticularityof

theSamajwadiPartyisthepracticeofvastorganizationaloverhaulsbeforeandafterevery

election. Ahead of an election (usually a year before an election), the party leadership

assesseswhichlocalbodiesandwhichofficebearershaven’tperformedandarelikelyto

be more liabilities than assets in the incoming polls. It does the same with its elected

representatives.Localbodiesare thenreplenishedwithnewofficebearersandmilitant,

expectedtobreathinanewenergywithintheparty.

Aftertheelection,thepartyoftendissolveslocalbranchesintheseatstheydidnotwin.In

theprocess,cardsgetreshuffledbetweenfactions.AfterhisappointmentasChiefMinister

in2012,AkhileshYadavproceededtoseveralorganizationoverhaulsinordertoplace“his”

menwithintheorganization,andinordertosidelinesomeofthemorerebelliousfactions

associatedwithhisfatheranduncles.

Thepartypresentsthatprocessasanexerciseofinternaldemocracy,sincethelocalbase

ofthepartygetstoberegularlyrenewed.However,acursorylookattheselocalreshuffles

indicate thata lotofsacked local leadersgetasecondora thirdchanceandarequickly

reinstatedwithintheorganization.

Farfrombeingawell-oiledmachinery,liketheBSP,theSPorganizationisconstitutedasa

competitivespaceforaccesstopoliticalnetworksandresources.Factionleadersandtheir

followerpermanentlywrestleagainstrivalfactionsforpowerandinfluence.

318

Afterthe2007defeat,thepartytriedtoemulatetheBSPorganizationbutdidnotsucceed.

Factionalism and unruliness prevented the party to achieve the level of dedication and

discipline that characterizes the BSP’s organization. After the elections, the party

organizedaseriesoftrainingcampsforpartyworkers,aimedatteachingthemthetenets

ofsocialismandatreinforcingpartyloyalty.Theexperiencewasnotentirelysuccessful.A

party spokesperson confided: “Peoplewere disrespectful. Theywere not disciplined. They

wouldnotlistentotheleaders.Theywouldleavethemeetings”348.

TheSPisagoodreminderofthefactthatbuildingorganizationsiscostlyandcomplicated.

Loyaltyhastobeinspiredbythepartyanditsleadershipandcannotsimplybebought.So

ratherthanbuildingacomplexlocalorganization,theSPpreferstolet its localbranches

berunbypeopledrawnfromlocalelitegroups,andtrustthattheirsenseofself-interest

willtransmuteintoasenseofpartyinterest.

Conclusion

Thisdescriptionofhowpartiesrecruittheircandidatesisnotcomprehensiveasthereis

obviouslyadiversityofroutesleadingtothenominationineachparty.Partytransfersand

thepoachingofcandidatesforexampleconstituteanothermethod.

But there are interesting variations between parties and notably between national and

regionalparties. Innationalparties, the recruitment tends tobedone in closed circuits,

through a highly centralized process. It is easier for outsiders or individual political

entrepreneurstogettickets inregionalparties,particularlywiththeBSPwhorecruit its

candidates largely outside its organization.The SP is an intermediary casewhere ticket

distributioniscentralizedanddiscretionary,butwherefactionsandpoliticalfamiliesplay

animportantrole.

Theseorganizationalvariationsmattersincetheyaffecttheparties’abilitytoconnectwith

local elites – old and new – and therefore affect their overall electoral performance. I

discusssomeimplicationsofthesevariationsinthefinalsub-section.

348InterviewwithanunnamedSPspokesperson,inLucknow,April2012.

319

6.2.Implications

In this section, I enumerate a series of implication and consequences of party

organizationalvariations.The first implication concerns the importanceof studyingand

understandingthefunctioningofpartisanorganizationsinIndia,whotoooftenaretreated

asblocks,undertheassumptionthat theorganization followsabidesbythewillof their

centralizedleadership.Partyorganizationsmattersincetheyaffectthecapacityofparties

to connectwith voters aswell aswith local elites, anddetermine also themodalities of

these linkages. I argue that organizational variations contribute to andhelpunderstand

whypartiesoperatingunderthesamecontextfollowdifferenttrajectories.

The second implication has to do with the reasons behind the domination of regional

parties. Traditional explanations of the success of regional parties point at underlying

social transformations and movements, of which parties are either the political

beneficiaries or the political extension. While I do not deny the association between

regional parties and lower caste movements, I wish to offer another explanation – a

politicalone–thatsourcesthesuccessofregionalpartiestotheparticularwaytheyare

constitutedandorganizedandtotheirabilitytoconnectwithnewelites.

ThethirdpointIwishtodiscussishoworganizationalvariationsbetweentheSPandthe

BSPaffectthewaytheyconnectwithlocalelites.

Asmentionedearlier,theriseofbackwardshasbeenaccompaniedbyviolenceandbythe

criminalizationofthepoliticalsphere.Oneparty,theSP,embodiesthisphenomenonmore

thanothersandIwishtoofferanexplanationaboutwhythe“goondaraj”tagstickstothe

SP more than to its opponents, despite empirical evidence that all parties share the

responsibilityforthecriminalizationofpolitics.

Finally, the fifth implication that I wish to discuss is the meaning of these party-elite

linkages with regard to the democratization argument that characterizes much of the

literatureonthepoliticalriseoflowercastes.

320

6.2.1.Theimportanceofpartyorganizations

Political parties in India are known to be heavily centralized, personalized and often

dynastic (Chandra 2016a). As a result, much of the recent literature on parties focuses

either on party leadership or electoral strategy and overall performance, and tends to

overlook questions related to their internal organization (Chhibber, Jensenius, and

Suryanarayan 2014). There are several reasons to think that organizations matter,

includinginpartiesthatarehighlycentralized.

Chhibberetal.haverecentlyargued,withstatisticalevidence,thattheloworganizationof

parties incentivizes individual candidates to defect to other parties, thus increasing the

overall electoral volatility349. In a previous contribution, Chhibber observed that “the

absence of a party organization, of independent civil society associations that mobilize

support for the party and centralized financing of elections has led to the emergence and

sustenance of dynastic parties in India” (Chhibber 2011, 1). According to Chhibber, the

presenceofdynasticparties leads toa representationdeficitandmore instability in the

electoralsystem.

Wilkinsonpartlyrefutestheargumentthatweakpartyorganizationsnecessarilyweakens

democracy, citing counter-example and observing that indicators of party instability do

not correlate with outcome measures (Wilkinson 2015, 438). He further notes that

democracythrivesinIndiaevenwhenpartiesdonot.

Theconfigurationofpartyorganizationsalsodeterminesmodalitiesofparty-elitelinkages

andtheneedforcandidatestorelyonindividualpatronagenetworks.

Thedegreeoflocalpresenceofpartiesalsodeterminestheirmobilizationcapacityduring

andbetweenelections.Italsoaffectsthequalityoftheinformationtheyrelyontoreadthe

electoral map. The BSP is able to localize its electoral strategy effectively because its

groundworkersprovidethepartyleadershipwithreliableandrelevantinformationabout

local politics. So, despite its centralized character, the BSP relies on its organization, to

mobilizeitscoresupportbaseandtosourceinformation.

349Chhibberetal.,p.499.

321

6.2.2.Whyareregionalpartiesdominating?

Traditional explanations for the rise of regional parties insist on the history of social

movements and mobilizations that have purported regional parties dedicated to their

representation,aswellason theroleofcontextual transformations that facilitated their

development(Jaffrelot2000a,2003b,Pai2002a).Someauthorshavelinkedthesuccessof

regionalpartiestothepoliticizationofcasteandthedevelopmentofitsroleasvehicleof

mobilization(HeathandYadav2010,Jaffrelot2000a,b,2003b,Michelutti2008,Palshikar

2013).

Other sociological and cultural explanations include the role of regional identities,

explainingthesuccessofregionalistpartiesbytheirabilitytoincarnatearegionalidentity

in ways that transcends other social cleavages (Kohli 1997, Subramanian 1999, Wyatt

2013,Zavos,Wyatt,andHewitt2004).

There areolderpolitical explanations aswell, that situate theoriginof various regional

partiesintheprocessofscissionoftheCongressparty(Brass1977,1983,Burger1969).

In his book on regional parties in India, Adam Ziegfeld disputes the notion that the

electoral success of regional parties derives solely from their ability to harness popular

grievances and strong regional identities (Ziegfeld 2016). He offers an alternative

explanationoftheirsuccessbasedonclientelism,coalitiongovernmentsandelitefactional

alignments.

According toZiegfeld, regionalparties succeedbecause theyaremoreapt thannational

parties at connecting with voters through clientelistic ties. Decentralized clientelism

through brokers and fixers offers an effective substitute to the costly construction of a

locallyimplantedpartyorganization.

Secondly,regionalpartieshavealsobenefitedfromaprocessofpoliticaldecentralization

thatconsecratesthestateasamostimportantpoliticalunitforthearticulationofpolicies,

partypoliticsandvoters’preferences.Theadventofcoalitionpoliticsat thestateandat

thenationallevelhasfurtherraisedtheprofileofregionalparties,whohavebecomemore

attractivetoaspiringpoliticians.

322

The final factor is the factional alignments between parties and regional elites. Ziegfeld

statesthatregionalpartiesaremorelikelytoemergeinstateswherealargeportionofthe

elite isstronglyboundby ideologyorcannotconnectwith theeliteofadifferentregion

(Ziegfeld2016,206).Successfulregionalpartiescanalsobeboundbyacommonsenseof

opportunismthatbindspoliticalactorstogetherwithinasinglepartyumbrella.

Ziegfeld’s elite-centric approach finds a lot of resonance in this dissertation but suffers

fromtwoimportantlimitations.Thefirstoneislinkedtothebroadusageoftheterm‘elite’,

used broadly and interchangeably with the term ‘politician’. Ziegfeld therefore clubs all

elected representatives into one undifferentiated elite category, and discards all other

non-electedactorswhomaywieldpowerandinfluencewithinandoutsideparties.

Thesecondlimitation,relatedtothefirstone,isthattheuseoftheterm‘faction’appliesto

large groups who collectively decide to form or not a regional party. It conceals the

internaldiversityofregionalparties,andnotablythefactionalismthatcharacterizestheir

functioninginmanycases,regardlessofthefactthattheyhaveacentralizedorganization

ornot.

Thefactthatregionalpartiestendtobecentralized,personalized,weaklyinstitutionalized,

guided more by informal than formal rules and that they tend to place their elected

representatives in apositionofuncertainty regarding the longevityof their careerdoes

notmean that their organizations do notmatter. Quite the contrary, the actual internal

arrangements between leaders and factions are crucial to the performance of regional

parties.The SPmaynothave todealwith a complexwell-ramifiedorganization, but its

leadersdospendmostof their timeadjudicating(andat timesevennurturing)conflicts

withintheparty,betweenfactionleadersandtheirfollowers.

Theapproachadopted in thisdissertationalso focusesonpolitical elitesbut attempt to

unravelthevarietyofelitesandthedifferentiatedmodesofconnectionbetweenvarious

elites and various political parties. The explanation offered for the success of regional

partiesinUttarPradeshisthreefold.

323

First,thesuccessofregionalpartiesislinkedtotheirabilitytoattractandco-optmembers

of the new elites. Local elites and the new elites in particular are more incentivized to

contestonregionalparties’ticketsforthreemainreasons.Thefirst,quitesimply, isthat

successattractsstrongcandidates.Between1996and2012,theSPandBSPticketswere

objectivelystrongerthannationalparties’tickets.

ThesecondisthattheSPandBSPregimesarereputedtobemorepermissivetowardsthe

deploymentofclientelisticnetworksthatfeedfrompublicresources.Inshort,theSPand

the BSP provide more opportunities for ‘fundraising’ and a better protection from the

stateagainstpoliticalandbusinessmalpractices.

Thethirdreasonisthatwhilethecompetitionforticketsisharshandexpensive,itisalso

comparatively more open than with the BJP or the Congress, who tends to select their

candidateswithinrestrictedsociologicalpoolsorclosednetworks.

The second explanation for the success of regional parties in Uttar Pradesh is that,

contrarytoapopularassumption,theirorganizationisstrongerthanthenationalparties’.

Itisbetterdistributedacrosstheterritoryandbetterimplantedatthegroundlevel.This

organizationaldeploymentiscrucialtotheirabilitytoconnectwithlocalelites.Astronger

localorganizationmeansthatregionalpartiesalsobenefitfrombetterandmorereliable

informationaboutgroundrealities,castedemographicsanddynamics,andsoforth.

The third explanation finally consists in recognizing that the regional parties in Uttar

Pradesh,contrarytothenationalparties,benefitfromthesupportofacoreelectoratethat

gives them a head start advantage in elections (even if these core support bases are

eroding).Thereasonthesepartieshaveacoreelectoratecannotbelimitedtothepractice

ofclientelismalone.There isa logicofethnic identificationatworkbetweentheYadavs

and the SP, or between the Jatavs and the BSP, that goes beyond the tradeoff of votes

against material or even programmatic benefits. There are material considerations

nurturingtherelationbetweenregionalpartiesandtheirelectorate,butastheChhibber

andAhujapointout,patronagenetworkscannotreachenoughpeopletodeterminealone

electoral outcomes. Also, contrary to the assumption that patronage networks usually

benefittoco-ethnics,Ifindthatthepracticeofpatronageenablespartiesandcandidates

324

to expand their social base beyond the group they are traditionally identified with, by

tradingsupportagainstbenefitswithprominentmembersofthosegroups.

6.2.3.Differentiatedparty-elitelinkages

ElectoralpoliticsinUttarPradeshhasalwaysreliedontheinductionofcandidatesdrawn

from the elites into the competition. Elite candidates are deemed more legitimate and

effectivecandidatesthannon-elitecandidates.Partiesthemselvestendtobeorganizations

ledbyelites,eithertraditional,asinthecaseofthenationalparties,ornon-traditional,as

inthecaseoftheregionalparties.Partiescontrolledbytraditionalelitestendtobepartial

totheirownkindwhenitcomestodistributeticketsandultimatelysharepower.

Overtime,thedefinitionofwhotheelitesareinUttarPradeshhasbecomemorediverse

andcomplex.Vastprocessesofsocial,politicalandeconomictransformationshaveledto

the emergence of new local elites, diverse in their caste identity, and relatively

homogenousintheirinscriptionintothenewnetworksofeconomicandpoliticalinfluence.

Partiesvaryintheirabilitytoconnectwiththesenewelitesandinthemodalitiesofthese

connections.Thisispartlydeterminedbytheiroveralltrajectories.

The Congress Party after Independence was a well-institutionalized catch-all party,

present in every district and with an effective ground organization that co-opted local

notabilitiesintopatronagenetworks.Today,ithaslostitslocalorganizationandcapacity

toconnectwithlocalelites.ACongressticket,erstwhileanelectoralsesame,asbecomea

liability. Its social compositionof oldnotables and youngprofessionals is alsodetached

fromthesociologicalworldthathasproducedthesenewelites.

Therecenteffortsatrebuildinganorganizationfromthebottombyincludingmembersof

deprivedsegmentsofthepopulationhasfailedsofartodeliveranyresult,sinceittendsto

excludethosewhotendtowinelections,i.e.thelocalelites.

TheBJPrecruitsitscandidatesinrelativelyclosedcircuits.Thepartyremainsdominated

bytheuppercastesandthrivesinregionswheretheascendencyoftheuppercasteshas

325

not been challenged from below. Candidates from the backward classes and from the

lowercastestendtobeselectedwithintheambitoftheSanghParivar.Thefew‘newelites

candidates’,orthefigureofthebusinessmenpolitician,thatarefoundintheBJPtendtobe

urbanbased,ordefectorsfromotherparties.

TheBJPmobilizationinruralareasislargelydonebyexternalorganizations–theRSSand

itssubsidiaries–thataimtobuilddirecttieswithsegmentsofvotersthroughsocialand

politicalaction,ratherthansimplyrelyontheintermediationoflocalelitesbyco-optation.

TheBSP and the SP are both able to connectwith the local elites through their ground

organizations.However,themodalitiesoftheseconnectionsdifferwidely.

TheBSPhasastrongandcohesivemilitantbasethatisspreadovertheterritory.Thisbase

enables theparty to ‘read’ theelectoralmapwithgreatprecisionandtoaptlydesign its

local alliances. The party however maintains a strict division of labor between its

organization–primarily composedof Jatavs–and thecandidates,drawn from the local

elites. Inshort, theBSPexternalizes thebusinessofwinningseats to individualpolitical

entrepreneurswhoinvestinanelectiontofurthertheirprivateinterests.Whotheselocal

elitesareandtowhichgroupdotheybelongtoliterallyvariesfromoneseattoanother.

ButwhattheBSPcandidateshaveincommonisthattheytendtobelongtolocalnetworks

thatcontrolpartsofthelocaleconomicandpolitical institutions,oroccupyapositionof

strengthinthelocalpoliticaleconomyoftheirconstituency.

TheSamajwadiPartyisanintermediarycase.Thepartyhasaweakunstableorganization

butdrawsitsstrengthsfromtheintegrationofitslocalbrancheswithlocalelitegroups.It

is usually assumed that parties seek to rely on local elites since the cost of building a

strongandstableorganizationishigh.ThisappliestotheSP.

However, one should not dismiss entirely the party’s organization because of its

centralizationanddynasticcharacter.Nominations to the formalbodiesof thepartyare

tightlycanvassedandcontrolledbythepartyleadershipbutinreality,theirfunctioningis

far more fluid and autonomous than it first appears. Local elites compete within the

parties’ organizational layers forpositions, resources and influence.That competition is

organizedaround factions thatareunrulyandundisciplinedbutat the same timemake

326

thepartyattractivetoindividualsbelongingtothenewelites.Theorganizationreinforces

themintheirdealingswiththebureaucracyandvariouseconomicforces.

With Akhilesh at the helm, the party has regained some organizational capacity, by

enrollinglargenumbersofyoungpartyworkers–mostlystudents–whocanbemobilized

duringelectioncampaigns.But thesedonot formastablesolidcadre that thepartycan

relyonbetweenelections.Further,thisarmyofyoungmilitantsisnotyetpartofthelocal

powercirclesfromwherepoliticalinfluencecanbesourced.

6.2.4.WhydoestheGoondaRajtagsticktotheSPandnottheBSP?

ItiswidelyassumedthattheSPlostthe2007electionsbecauseoftheseveredeterioration

of lawandorder that hadmarked its 2003mandate.The state frequently topsnational

crime rankings and the evocation of “U.P. politics” itself has become a semantic signal

referringtothecriminalizationofpolitics.In2007,Mayawatiactivelycampaignedonthe

themeofprobity,promisingtoputtheGoondaRajoftheSPtoanend.

However,anexaminationofthecrimerecordoftheBSPcandidatesandMLAsrevealthat

BSPpoliticiansarenoless‘criminals’thantheirSPcounterparts.

In fact, while the SP and the BSP do speak and mobilize different segments of the

electorate,theydonotdiffermuchintermsofcandidates’profile.Therearevariationsin

terms of caste – more Yadavs, Thakurs and non-Jatav Dalits with the SP, more Kurmis,

BrahminsandJatavswiththeBSP.Butintermsofeconomicbackground,theircandidates

are strikingly undifferentiated. The reason is that both parties recruit their candidates

from the same sociological pool of local elites, and notably the new elites, who have

emergedfromsomeofthemostdynamicandcriminalizedsectorsofeconomicactivity.

One would also be hard pressed to distinguish the two parties in terms of corrupt

practiceswhile in office.Bothpartieshavebeenheld responsible (thoughnot guilty) of

large-scalescamsandscandals.BothMayawatiandMulayamareunderaCBIinvestigation

fordisproportionateassets.

327

Yet,the“goondatag”sticksmoretotheSPthantheBSP,whoremainsdefinedinpopular

imagination as a Dalit party rather than a party harboring criminals. One can think of

severalreasonsthatexplainthisvariationofopinion.

One obvious and valid explanation is that these parties have distinct identities and

political cultures, leading to differentiated images. LuciaMichelutti argues that violence

and a ‘masculine’ brand of politics is consubstantial to the political culture of the

SamajwadiParty,apartyoftendescribedbyitssupportersasa‘goondaparty’:apartyof

musclemenandfixers(Michelutti2008,48).Itslocalpartyworkers,especiallytheYadavs,

cultivate an image of ‘strongmen’ essential to maintain their social and political

ascendency.

TheBSPonitssidecultivatesthe imageofapartydedicatedtosocialchangeandtothe

emancipationofthemostdeprivedgroups.OnecouldpointthatBSPlocalrepresentatives

haveoftencontradictedthatimagebutgenerallyspeaking,theBSPtendstowinthebattle

oftheimage.

Asecondexplanation lieswith the twoparties’modeof interactionwith localelites.We

sawthattheBSPmaintainsadivisionoflaborbetweentheorganization,whosemainrole

istomobilizethecoresupportbaseofthepartyandprovidethepartyhighcommandwith

informationonconstituencies,candidatesandtheimplementationofschemesinfavorof

Dalits,andthecandidates,whoaredrawnfromlocalelitegroups,mostofthetimedistinct

fromthelocalDalitpopulation.

The separation between the candidates and the party makes the former expendable.

SackingthemandreplacingthemwithnewcandidatesorMLAsusuallyposenothreatto

theorganization.

In the 2007 BSP regime, criminal politicians no longer benefited from the kind of

protectionandimpunitytheyusedtoenjoyinthe1990s.TheBSPinparticularexplicitly

warns its cadre, MPs and MLAs to not cross certain yellow lines (which is a plead for

discretion in the conduct of their illegal activities rather than a thou-shalt-not kind of

order), at the risk of being expelled from the party. During her fourth tenure as Chief

328

Minister,MayawatiexpelledanumberofherownMLAsandMinistersfromthepartyfor

actsorallegationsofcriminality350.

In2011,shesackedherBijnorMLA,ShahnawazRana,forprotectingtwoofhisaideswho

hadattemptedtorapetwowomeninDelhi. InthemonthofDecember2011,shesacked

nineteenofherMinistersafterthestateLokayukta(“ombudsman”)foundthemguilty(or

was about todeclare themguilty) of embezzlement of public funds, land grab, abuseof

power and illegal earnings.One of them, in charge of animal husbandry, had attributed

publiccontractstobuild21veterinaryhospitalsinEtahdistricttohisownson.Another,

Minister for secondaryeducation,haddiverted funds fromhisministry to fundhisown

school.

Aheadof the2012elections,Mayawati suspended frompartymembership anumberof

prominent criminals-turned politicians, among which Dhananjay Singh, the MP from

Jaunpur,withadozenmurdercasesonhishead;JitendraSinghBablu(MLAfromBikapur),

another notorious killer and land grabber, who shot to fame in 2009, when he set the

houseof theCongressStatePresident,RitaBahuguna,on fire.OtherMLAschargedwith

violentandheinouscrimesweresuspendedduringthesameperiod,inAuraiya(Shekhar

Tiwari), Bilsi (Yogendra Sagar) and Bulandshahr (Bhagwan Sharma, alias Guddu

Pandit)351.

Theevictionofcriminalelementsfromthepartyissometimesvoluntarilydramatized.On

May29,2007,aBSPMPfromAzamgarh,UmakantYadav,wasaccusedoflandgrabbingin

hisdistrict.Twodayslater,MayawatiinvitedhimtoherresidenceinLucknow;onlytoget

himarrestedbyasquadofpoliceSpecialForces,andinthepresenceoftelevisioncrews.

YadavandhissonDinesh,whohadcontestedtheAssemblyelectiononaBSPticket,were

bothexpelledfromtheparty.

TheSamajwadiParty,ontheotherhand,isunabletodosoasthecriminalelementswithin

the party are very much part of the organization, or are protected by the factions they

belongto.RajaBhaiyaisafamousexample.TheMLAfromKundawasexpelledfromthe

350‘4expelledbyBSPforcriminalactivities’,IndianExpress,April26,2010.351Facingarrest,GudduPanditmigratedtoDubaiandjoinedatthesametimetheSamajwadiParty.

329

partybeforethe2012election,asasignalofdeparture fromoldpractices.Hecontested

andretainedhisseatasanIndependentcandidate.Immediatelyafterthevictory,hewas

appointed,againsttheChiefMinister’swill,MinisterofFoodandCivilSupplies–thesame

MinistrythathehadplunderedwhenhewasaBJPMinister–andofPrisons352.

Since party bosses, candidates and local elites are all integrated locally, the sacking of

MLAs poses the risk of splitting the party’s local organization, and to lose not just

individualsbutalsonetworks,partyworkersandresourcestoarivalparty.

LocalSPbrancheshaveagooddealoffunctionalautonomy,whichleadstothedifficultyof

controllingtheiractionandexcesses.

Thus,despitetheconcentrationofpowerinthehandsofthepartyleadership,thepartyis

farmoredependablefromitsbasethantheBSP.Localbossesarewellawareofthatfact,

whichencouragesastrongcultureofimpunitywithintheSP.

Thiscultureofimpunityisfurtherreinforcedbythefactthatthelocalelite’scontroland

influenceoftenextendtolocalpoliceforces,whoareoftenhelpless,andattimescomplicit,

toSPpoliticians’wrongdoings.

AnupamMishra, anAllahabadbased journalist sumsup thedifferencebetween the two

partiesasfollows:“UnderSP,thereisdemocratizationofcrime.EverySPworkerfeelshecan

takeonthelaw.UnderBSP,thereistotalcentralization,andsoeventhecadreiscareful”353.

6.2.5.Consequencesforgovernance

If the Samajwadi Party and the Bahujan Samaj Party don’t differ much in terms of the

sociologyoftheircandidates,dotheydiffer intermsofpolicieswheninpower?Dothey

servetheinterestsofthegroupswhoelectedtheminofficeinthefirstplace?

352HewasbrieflysuspendedayearlaterwhenhisnamecameupinthemurdercaseofapoliceofficerinKunda.Hegotclearedofthechargesandresumedhisoffice.353Quotedfrom(Jha2016).

330

MostcontributionsonthegovernanceordevelopmentaloutcomesinUttarPradeshtend

to focusonthestateaswholeandnotontheactionofaparticularparty.AtulKohli, for

example,clubstheSPandtheBSPinthesameneo-patrimonialistformationscategoryand

blamesthemequallyforthestateofderelictionofUttarPradesh(Kohli2012).According

to him, party centralization, caste-based electoral considerations and a generally weak

economyhavepushedbothpartiestoprivilegenarrowpolicyagendas,tostrayawayfrom

publicinterestpoliciesandtoseektofavortheirelectoralbasethroughthedistributionof

publicgoods,throughpreferentialpoliciesandthroughthepoliticsofsymbols354.

This description echoes the diagnostic that he had made in 1990, when he attributed

India’scrisisofgovernabilitytothedeclineoftheCongressandtheinaptitudeofregional

and communal parties to propose a viable alternative in terms of policy and state

authority(Kohli1990).

Inarecentcontribution,PrernaSingh,aformerstudentofKohli,assertsthattheabsence

ofastrongover-archingregional(subnational)identityhaspreventedtheformationofa

publicmindedethosamongthepoliticalclass(Singh2015).Shenotesthatinastatelike

Kerala, notwistanding the legacy of awelfare tradition inherited from former rulers, an

electoralcompetitionorganizedaroundtwoleftandcentre-leftleaningcoalitions,asense

of commonbelonging to a subnational spacehasbeen conducive to thedevelopmentof

generalistwelfarepolicies.Bycontrast,shenotesthat inUttarPradesh, theabsenceofa

strong subnational bond across social categories has encouraged the development of

divisive caste-based politics and the “conceptualization of welfare in narrow, sectional

terms”355.

In fairness, these authors recognize that there is a longer legacy of poor governance in

UttarPradesh.Buttheynotethatinthecontextofliberalization,thestateofUttarPradesh

hasbeenparticularlycrippledbyitspolitics.

Beyondthequestionofeconomicperformanceanddevelopment indicators,canwespot

differences in thepoliciesdesignedand implementedby theSPandtheBSP?Orcanwe

354Ibid.,p.172.355Ibid.,p.545.

331

state,asitisoftenassumed,thatbothpartiesworkfortheexclusivebenefitoftheircore

electoralbases,DalitsfortheBSP,YadavsandMuslimsfortheSP.

What I suggest here is that bothparties donot fundamentally differwith each other in

terms of policies. They both also practice what could be termed as segmentary

distribution,orthepreferentialaccesstopublicgoodsandwelfareschemesofparticular

segmentsofthepopulation,definedincasteandcommunitytermsordefinedintermsof

theirpoliticalpreferences.WheretheSPandtheBSPvaryisinthedeterminationofwho

benefitfromtheirpoliciesandlargesses.

Tobeginwith,therearestrongargumentsforlookingatthesetwopartiesasbeinglargely

policy-indifferentiated.Afteritsvictoryin2012,theAkhileshYadav’sgovernmentrapidly

announced the dismantling of 27 schemes and policies established under the previous

government. The main reasons cited were the lack of budget or the inefficacy of those

schemes.

A closer look reveals thatmost of the cancelled schemesbore thenames ofDalit icons,

suchastheSavitribaiPhuleBalikaMadadYojna,acash-for-girlschemeineducation,the

ManyawarKanshiramShahriGaribAwasYojna,anurbanpoorfreehousingscheme,orthe

BhimraoAmbedkartubewellscheme.

Some of these schemes suffered from strong limitations. The Kanshiram Shahri Garib

AwasYojnanevertookoffduetothelackofavailabilityoflandinurbanareas.Butinmost

cases, these policies were quickly re-enacted under new denominations. The Savitribai

PhuleBalikaMadadYojnawas replacedbyKanyaVidyaDhanYojna, inwhich a bicycle

wasaddedtothecashgiventofamilieswithgirlsenrolledinschool.Theurbanpoorfree

housingschemewasre-launchedbytheMinistryforUrbanDevelopmentunderthe‘Aasra’

scheme.

While re-naming the scheme, the government also changed the conditions of eligibility.

Dalitslostthe50percentquotatheyhadinthepreviousschemetothebenefitofMuslims

andpoorOBCs.Theschemewasre-launchedinRampurdistrict,thehomedistrictofthe

Ministerinchargeofurbanpovertyalleviationandurbandevelopment,AzamKhan.

332

In November 2012, the Akhilesh government announced the discontinuation of two

schemesdevisedbytheBSP,theBhimraoAmbedkartubewellschemeandtheAmbedkar

collective tube well scheme (for construction of tube wells). In the same breath, It

announced the creation of a unified Dr. Ram Manohar Lohia Community Tube Well

Scheme (later on Dr. Ram Manohar Lohia Collective Tube Well Scheme), which exactly

reproduced the same features as the BSP schemes, including higher and preferential

subsidiesforSCsandSTs.

One of the most popular and ambitious welfare policy of the BSP, the Ambedkar Gram

SabhaVikasYojna(AGSVY)–alsoknownas‘Ambedkarvillages’–wasalsodiscontinuedin

favor of a Dr. Ram Manohar Lohia Samagra Gram Vikas Yojana, a village development

scheme.MostvillagesfallingundertheAGSVY–Dalitdominated–wereexcludedfromthe

newschemewhichcouldtechnicallyapplytoanyvillage.TheSPtargetedthecreationof

1600 new ‘Lohia villages’, on the premise that the Ambedkar villages had already

benefitedfrompublicattentionandthattherewerenootherdalitvillagetofindtojustify

thecontinuationofthescheme.

The Ambedkar Village scheme had been one of BSP’s most successful policy, including

overitsvariousphasesofexistenceabout19,000villagescomprisingamajorityoranear-

majority of SC/ST population356. It included the construction of apartment complexes

reservedforDalits.Painteddarkblue(thecoloroftheparty),thesebuildingswereusually

locatedintheoutskirstsofvillages,nearcrossroadsornationalroads,wheretheywould

bevisibleandwhereDalitscouldliveamongthemselves,atsafedistancefromtheirlocal

oppressors.

Inalltheseexamples,theSPsimplyextendedexistingschemeswhilerenamingthemand

changing the beneficiaries. The BSP in its time did in part the same thing. The current

Kanya Vidya Dhan Yojna of the SP government was initially launched by the Mulayam

Singh Yadav government in 2004. Under that scheme, girls from families below the

356TheAmbedkarVillage schemewas launched in 1991 and initially targeted only villages thathad 50 per cent of Dalit population or more. Subsequently, Mayawati relaxed the criteria toinclude villages containing 22 to 30 per cent of Dalit population. Dalits in these villages get aprivilegedaccesstopublicgoodsandothergovernmentdevelopmentandwelfareschemessuchasroads,electrification,hand-pumps,housingschemes,etc.

333

poverty lines were entitled to a 20,000 Rs cheque upon higher secondary education

graduation,tohelpthempursuehighereducation.

With the recent inclusive turn of Uttar Pradesh politics, these schemes tend to be less

oriented towards specific groups, as compared to earlier. This fits with the necessary

imageofinclusivenessthatpartiesmustproject.Butthelanguageofgeneralinterestcan

beused toobfuscatepractices of discrimination in thedistributionof public goods.Not

specifying the beneficiaries of a policy enables those who handle those funds –

bureaucrats and elected representatives alike – to distribute them in an arbitrary or a

discriminatory fashion. While Dalits are usually not explicitely excluded from most SP

schemes, they are diluted into a broader definition of beneficiaries, based officially on

classaswellasoncaste357.

TheSPgovernmentusedthesameinclusiveargumenttodiscardthetwopercentquota

forDalitsingovernmentcontractsthattheBSPhadintroducedinJune2009358.

Most of the policies and schemes of recent Uttar Pradesh governments have been

redistributive in nature. On the one hand, few governments have been interested or

incentivized to engage with structural reforms policies, or to improve the provision of

basic public services such as education or health care.On the other hand, the room for

maneuvre inapoorendebtedstate is limited.Thegovernment cannot touch toplanned

expenditures or central funds to finance its policy and therefore mainly has the

instrumentsofdistributionandsubsidiestocomeupwithnewpolicies.

Parties use those instruments to garner support from their core support base. For

instance,in2012,theSPwaselectedonthepromisethatitwouldprovidefreeelectriticy

andfreewaterforirrigation(apromiseitcouldnotentirelyuphold).

Inthatregard,theSPandtheBSParelargelyundifferentiated.

357Thereisconsiderablesocialandpressureontopoliticalpartiestoincludepoormembersoftheuppercastesintothoseschemes.358Initiallyforsmallcontractsuptoamaximumoffivelakhs.Thebarwasraisedtotwenty-fivelakhssubsequently.

334

These two parties however differ in terms of expectactions. The BSP victory in 2007

createdthehope,fromitssupporters,thatitwouldfinallybeabletoworkintheirfavor,

nowthatiswasfreefromtheshacklesofcoalitionpolitics.

Some BSP schemes, such as the Ambedkar Village, have been resounding successes,

leadingtotheeffectivedistributionofaffordablehousetoSCcitizens.Otherschemes,and

thegeneralimplementationofcentralschemessuchastheNationalRuralHealthMission,

havebeenmoreintunewiththeregulargovernmentalmalpracticesthathaveplaguedthe

stateofUttarPradesh.

When elected in 2007, Mayawati found herself entangled between two contradictory

objectives: tomeet thepressingneedofher coreelectoralbase, and tonurtureabroad

social alliance not only through the instrument of representation but through the

instrumentsofpolicyandgovernance.In2012,theSPcampaignedonthethemethatthe

BSPwasprivilegingDalitsoverotherneedyindividualsandcategories,whicheventually

contributedtoherdefeat.

The functioningof thestate inUttarPradesh is rathernebulous,markedbyopacityand

arbitrariness.Thisleavesplentyofspaceandopportunitiesforthemisuseofpublicfunds

and various forms of abuse of power. Atul Kohli is right to underline that the state’s

politicalclassis“focusedoneverythingbutthestate’sdevelopment”.Headds,“…avarietyof

socioeconomic problems continue to accumulate: poor infrastructure; the decay of major

publicinstitutions,includinguniversities;thedeteriorationoflawandorder;andwidespread

corruption”359.

6.2.6.Consequencesfordemocratization

Finally, the process of integration of political and economic elites within the regional

parties raises question about the emancipatory promise that these parties embodied at

the time of their inception. The rules of the electoral game and party politics have

contributed tocreateaclassofnewpoliticianswhoget intopoliticsasaway to further

359Op.Cit.,p.178.

335

private interests, rather thanpursue theprogressive emancipatory agendapromisedby

politicalparties.

TheSPandtheBSPhavebeensuccessfulpartiesnotjustbecausetheyhavesucceedingin

capturinglargesegmentsofvoters’imaginationwiththeirdiscourseonsocialjustice,but

because they have aligned themselves with the new elites that dominate the social,

economic,andpoliticallandscapetheyoperatein.Thesenewelitestendtobeinabetter

positionthanotherstodeliverandmeettheexpectationsofvoters.

Therulesofthegamesandpartypoliticshavecontributedtocreateapredatorypolitical

class that invests inpoliticsasamean to furtherprivate interests.Electoralpoliticsand

representationareaninstrumentforthemaintenanceanddevelopmentofthecontrolthat

groupsorindividualsexertoverterritories.Thespiralingcostofentrytopoliticsexcludes

thosewhocannotaffordtocompete.Andpartieswhofieldsuchcandidatesareboundto

loseelections.

Theseevolutionshavehadanegative impacton the respectof conventionaldemocratic

norms by the political class. The nonchalant attitude of parties towards crime and

lawlessnessamongthepoliticalclasshascontributedto thedevelopmentofacultureof

impunity,whichinturnhasfurtherencouragedvariousformofpredatorybehaviour.

Also,whatisthemeaningofrepresentationiftherecruitmentofthepoliticalclassismade

onthebasisofcriteriathatexcludethemajorityofthepopulation?Whatistherelevance

of having caste diversity if the way to achieve caste-based inclusiveness is to recruit

preferentiallyamongthenewbusinesselites?

It hasbeendocumented that the elitist andbiased recruitment of candidatesbyparties

tends toexclude ‘weaker’ segmentsof thepopulation– thepoor,women,and the lower

castes.Analyzingthecandidatesfieldedbypartiessince1962,FrancescaJenseniusshows

thattherecentincreaseofthenumberofwomennominatedtocontestislargelylimitedto

the reserved constituencies, which “tend to be less competitive and less dominated

by’moneyandmuscle’politics”(Jensenius2016a,2-3).

336

Theprofileofwinners inU.P. election is consistentwithAdamZiegfeld’s findingsabout

thecandidatesandwinnersattributesinHaryana.Ziegfeldnotesthat“candidateswhoare

natives of their constituency, previously held local political office, had family who were

involved in local politics, are members of the state’s major political dynasties, and have

occupationsinbusinessalsotendtowinlargervoteshares”(Ziegfeld2016,244).

Various authors have insisted on the symbolic benefits of having backward and lower

casteleaderselectedtotheAssemblyandrulinggovernments(Jaffrelot2003b,Pai2000a,

2002a,Varshney2000a,b).ButwhileIdonotdenythetangibilityofthepoliticsofdignity,

I raise doubts that any structural change to social and economic inequalities will come

fromthepoliticalclass,whosecommitmenttodemocratizationremainscontingenttothe

pursuitoftheirinterests.

337

Conclusion

Thisdissertationsetouttoachievetwointer-relatedtasks.Thefirstwastodescribesome

of the more significant political changes that have occurred in Uttar Pradesh after the

MandalandMandirperiod.Twoofthesesignificantchangeswerethegrowingdomination

ofregionalpartiesandthegradualheterogeneisationofcasterepresentationwithinthem.

The second task consisted in placing electoral politics in a broader context of social,

politicalandeconomictransformations.Thestudyofsub-regionalvariationsforinstance

illustratedhoweconomicchangeinducestheemergenceofnewelites,whohavebeenco-

optedmostlybytheregionalparties.

Inordertoaddressthesequestions,Ihaveadoptedanelite-centricapproach,examiningin

chapter three the set of institutional and political constraints candidates and elected

representativesoperateunderandhow thoseconstraints impact their selectionprocess

andtheirbehaviouronceelected.

Inchapterfour,Ihaveexaminedtheevolutionofthesociologicalcompositionofthestate

assembly, in terms of castes and communities but also in terms of other socio-

demographic variables such as education and occupation. The data on the former two

variables being unsatisfactory, I proceeded in chapter five to examine the question of

representatives’ socio-economic background qualitatively, to illustrate how caste,

economicpositionandpoliticalstatusinterweaveinlocalcontexts.

Inchaptersix,Ihavecomparedthetrajectoryofthefourmainparties,withregardtotheir

candidates’selectionprocesses,theirorganizationalstrengthsandweaknesses,andtheir

abilitytoconnectwithlocalelites.

ThefirstconclusionthatIdrewfromthisstudywasthatelectoralpoliticshasundergonea

processoflocalization,owingtothelocalizationofparties’electoralstrategies.Partiesno

longerseek tomobilizeexclusivelyacoreelectorateat thecostof thesupportofothers

but instead seek to forge social alliances in order to broaden their support base. The

partieswhohavewonelectionsinthepasttwentyyearsorsoarethosewhoadaptedtheir

338

strategiestolocalsocial,politicalandeconomicconfigurations–coveringcasteandlocal

elitesdynamics.

The second conclusion is that even if state-wide caste-based mobilizations no longer

operatethewayitdidinthelate1980sandearly1990s,casteandpoliticsremaindeeply

intertwined.Theexpressionsof thatconnectionoccur locally,where theirconsequences

aretangible.

The conjunction of these two phenomenon, the localization of electoral politics and the

interlocking of local caste and political economy contexts, has led to a process of

integration of local political and economic elites, marked by the greater induction of

candidateswithabusinessbackgroundintotheelectoralfray.Thisprocessisstrongerin

sub-regionsthathaveundergonedeepeconomictransformations,suchasWesternUttar

Pradesh but is also noticeable in some under-developed parts of the state, such as

Bundelkhand,where themajorityof thepolitical class comes from theconstructionand

publiccontractingsectors.

Thisprocessofintegrationofpoliticalandeconomiceliteshasfarreachingconsequences,

in termsofelitecaptureof localpublic institutions, controlover theallocationofpublic

resources, probity in public life and criminalization of politics in general. This process

helpsalsotounderstandwhypoliticalchangehasnotcontributedtobringthestructural

change in terms of local caste-based domination that was expected from the rise of

backward. In other terms, there is a clear contradiction between the emancipatory

aspirationofbackwardandlowercastepartiesandtheirpracticesofpoliticalrecruitment.

Traditional elites have resisted to pressures from below in areas that have stagnated

economically.

Implications

ThisdissertationcontributestofiveacademicdiscussionsoncontemporaryIndianpolitics.

The first one is the identity politics literature. A significant change of the past twenty-

years is that the tropes of transversal caste mobilization have largely lost their effect.

Backwardcastevoterswhoweremobilizedonthethemeofreservationstwenty-yearsago

339

are less likely torespondto thiskindofcampaignor thiskindofgenericcasteappeal. I

have however argued that caste, as a vehicle of political mobilization, has not lost its

saliencybuthasbeenprogressivelyde-linkedfromstablepartyaffiliationonthebasisof

identity,tothebenefitofthelocalizedinterlockingofcaste,politicalandeconomicfactors.

Castepoliticsneedstobelocatedatthelevelwhereiteffectivelyoperates,inconjunction

withothersalientpoliticalvariables.

The second contribution concerns the discussion on clientelism, or patronage. This

dissertation adopts an intermediary position between those advocating that India is a

patronagedemocracyandthosewhoclaimthatpost-liberalizationprogrammaticpolicies

have trumped patronage. Patronage networks are an indispensable component to any

politician’s career but hardly ever cover the entirety of candidate-voters linkages and

party-voterslinkages.Inotherterms,theymaybeanecessaryconditiontobecompetitive

intheelectoralracebutcertainlynotasufficientone,otherfactorssuchaspartyappeal,

localdemographicsand inter-castepowerrelations,andcandidates’ individualqualities,

amongothers,alsoinfluencevoters’decisions.Inanycase,patronagedoesnotdetermine

electoraloutcomes.

Infact,Idonotclaimthatanelite-centricapproachcoversexhaustivelythepoliticalfield

orthatelite-centricexplanationssubsumeotherformsofsocialandpolitical factorsthat

affectvoters’preferencesandchoices.

Thisdissertation takesa similarpositionon thequestionof criminalizationofpolitics. I

have attempted to describe the mechanisms and incentives that attract individual

embeddedincriminalitytoinvestinpolitics.Ihavealsoattemptedtoshowthatcriminal

attributesarebutoneamongotherresourcesthatcandidatescanuseinordertowinan

election. “Muscle” is a resource thathelpswinning seatsbutnotaguaranteeof success.

Criminalpoliticianswhodonotmeettheexpectationsofvoters intermsofaccessibility,

distributionofresourcesforexamplecannotcountontheircriminalattributestosustain

themselvesinpolitics.Similarly,andevenifthedatashouldbeusedwithcaution,thereis

also evidence thatmany candidateswith criminal charges lose elections.Once again, all

theseattributesofsuccessfulcandidatestendtoworkinconjunctionwitheachotherand

notseparately.

340

Thefourthdiscussionthisdissertationcontributestoistheliteratureonpartiesandparty

politics. This literature has traditionally focused on party leadership and broad party-

voteralignments,buthasnotsufficientlypaidattentiontopoliticalactorsandtheiragency.

Similarly, not enough attention has been paid to party organizations, which are often

assumedtobeweakandhyper-centralized,andthereforenotrelevantbesidestheobvious

taskofmobilizingvoters.Ihaveshowedthatorganizationalvariationsimpactthecapacity

ofpartiestoconnectwithlocalelitesandtointegratethemwithintheirranks.Ihavealso

shown that even hyper-centralized parties do need an organization in order to gather

information on local caste and political contexts and configuration and to connect with

local elites. Parties that outsource information collection to external agencies, such as

private companies or external political operators usually fail to grasp these local social

andpoliticalconfigurations.

Finally,thisdissertationcontributestotheelectionanalysisliteraturebyinsistingonthe

need to considerdata analysis at intermediate levelsof observations, between the state

and the local. Looking at sub-regional variations for instance is a convenient way to

deconstruct electoral trends built on the basis of aggregate data. Similarly, following

patterns of individual representatives’ career trajectory helps to unearth a number of

insights about the competitiveness of elections and about the many hurdles politicians

face in their career. This helps to measure power concentration within parties and

assemblies, patterns of incumbents’ re-nomination and measure the phenomenon of

turncoats.Thereisscopehereformuchinter-statecomparativework.

Ultimately, the question that matters is what are the sources of political change in

contemporary India? Sociological explanations tend to assume too quickly that political

change and political actors are merely the reflection or the extension into the political

domain of deeper social transformation dynamics. I have shown that by controlling the

politicalsupply,partiesdeterminewhogetstostandforelectionsandthereforewhogets

representedinthefirstplace.Theunder-representationofMostbackwardClassesandof

thenon-JatavDalitsisacaseinpoint.Ultimately,votersareconstrainedtochooseamong

thecandidatesthatpartieschoosetofield.

341

Finally,thisdissertationcoversaperiodofUttarPradeshpoliticsthatbroadlystartsIthe

late1980sandendswiththe2012stateelections.Manychangeshaveoccurredsincethen,

someofthemchallengingsomeofthefindingsenunciatedhere.

The first major change is the new rise of the BJP, on terms that vary from its earlier

ascension in the late 1980s. The 2014 General Elections and the BJP campaign in

particularusherednewmethodsofmasscampaigning,personalizedofelectoralcampaign,

added to older practices of caste-based ticket distribution, religious symbolism and

communalpolarization(Jaffrelot2015b).TheBJPwon71ofthe80LokSabhaseatswith

42.3percentofthevotes. Italsostoodsecondinthesevenconstituenciesthat it lost360,

andsurpassedthecombinedvoteshareofitsthreeopponents(Congress,BSPandSP)in

twenty-twoconstituencies(JaffrelotandVerniers2015,31-32).

ThestrengthoftheNarendraModi-ledcampaignwassuchthatwhothecandidateswere

didnotmattermuch.TheBJP’sopponentsandtheindividualstrengthoftheircandidates

couldnotmatchtheappealoftheBJPcampaign.

It remains tobeseenwhether2014wasacriticalelection(Key1955),markingamajor

anddurableelectoralrealignmentbetweenpartiesorifitwasmerelyanexception,oran

anomaly, linkedto theparticularcontextofastrongrejectionof theCongresspartyand

thesimultaneouspoliticalascensionofNarendraModi.

While these events andwhatwill followarebeyond the scopeof this project, I hope to

have made the case that political transformations ought to be scrutinized both

quantitatively and qualitatively, using various forms of empirical evidence and that the

findingsthatIhavesummarizedinthisconclusionprovideausefulanalyticalframework

forthecomparativestudyofstatepoliticsanddemocratization.

360ThetworemainingsetsweregiventotheBJP’spartner,theApnaDal.

342

Bibliography

Acharya,Avidit,JohnE.Roemer,andRohiniSomanathan.2015."Caste,CorruptionandPoliticalCompetitioninIndia."ResearchinEconomics69(3):336-352.

Adeney,Katharine,andLawrenceSáez.2005.CoalitionpoliticsandHindunationalism.1sted.NewDelhi:Routledge.

Ahuja,Amit,andPradeepChhibber.2010.Akarsh,Arora.2015."PovertyacrossSocialandReligiousGroupsinUttarPradesh."

EconomicandPoliticalWeekly:100-109.Alm,Bjorn.2010."CreatingFollowers,GainingPatrons:LeadershipStrategiesina

TamilNaduVillage."InPowerandInfluenceinIndia.Bosses,LordsandCaptains,editedbyPamelaPriceandRuudAlridEngelsen,1-19.NewDelhi:Routledge.

Amaresh,Misra.1997."UttarPradesh:WeakResponsefromAnti-BJPCamp."EconomicandPoliticalWeekly.

Anirudh,Deshpande.1997."UttarPradesh:Casteism,CommunalismandPolitics."EconomicandPoliticalWeekly.

Ansari,Ghaus.1960.MuslimCasteinUttarPradesh.Lucknow:TheEthnographicandFolkCultureSociety.

Ansari,IqbalA.2006.PoliticalrepresentationofMuslimsinIndia,1952-2004.1sted.NewDelhi:ManakPublications.

Ansolabehere,Stephen,andAlanGerber.1997."IncumbencyAdvantageandthePersistenceofLegislativeMajorities."LegislativeStudiesQuarterly22(2):161-178.doi:10.2307/440380.

Ansolabehere,Stephen,JohnMarkHansen,ShigeoHirano,andJamesM.Snyder.2007."TheincumbencyadvantageinU.S.primaryelections."ElectoralStudies26(3):660-668.doi:10.1016/j.electstud.2007.06.003.

Aron,Sunita.2016.TheDynasty:BorntoRule.NewDelhi:HayHouse.Awasthi,Dilip.1989."UttarPradesh:ABitterBeginning."IndiaToday,31December.Bailey,F.G.1969.Stratagemsandspoils;asocialanthropologyofpolitics,ThePavilion

seriesSocialanthropology.NewYork,:SchockenBooks.Bailey,F.G.1988.Humbuggeryandmanipulation:theartofleadership.Ithaca,N.Y.:

CornellUniversityPress.Bailey,F.G.1991.Theprevalenceofdeceit.Ithaca,N.Y.:CornellUniversityPress.Bailey,F.G.1998.Theneedforenemies:abestiaryofpoliticalforms.Ithaca:Cornell

UniversityPress.Bailey,F.G.2001.Stratagemsandspoils:asocialanthropologyofpolitics.Boulder,Colo.:

WestviewPress.Bakshi,Sanchita,ArunishChawla,andMihirShah.2015."RegionalDisparitiesinIndia:

AMovingFrontier."Economic&PoliticalWeekly50(1):44-104.Banerjee,Mukulika.2007.SacredElections.EconomicandPoliticalWeekly(April28):

1556-1562.Banerjee,Mukulika.2010."LeadershipandPoliticalWork."InCreatingFollowers,

GainingPatrons:LeadershipStrategiesinaTamilNaduVillage,editedbyPamelaPriceandRuudAlridEngelsen,20-43.NewDelhi:Routledge.

Banerjee,Mukulika.2014.WhyIndiavotes?,ExploringthepoliticalinSouthAsia.NewDelhi:Routledge.

Bardhan,Pranab.1982."AgrarianclassformationinIndia."TheJournalofPeasantStudies10(1):73-94.doi:10.1080/03066158208438190.

343

Barnett,MargueriteRoss.1975.ElectoralpoliticsintheIndianstates:partysystemsandcleavages,StudiesinelectoralpoliticsintheIndianstates.Delhi:ManoharBookService.

Baxter,Craig.1971.TheJanaSangh;abiographyofanIndianpoliticalparty.Philadelphia,:UniversityofPennsylvaniaPress.

Bentall,Jim,andStuartCorbridge.1996."Urban-RuralRelations,DemandPoliticsandthe'NewAgrarianism'inNorthwestIndia:The

BharatiyaKisanUnion."TransactionsoftheInstituteofBritishGeographers21:27-48.Berenschot,Ward.2014."MuzaffarnagarRiots.PerilsofaPatronageDemocracy."

Economic&PoliticalWeekly49(12):15-18.Bhalla,Abhishek.2012."HowtheyMadetheNHRMSick."Tehelka,March17.Bhatt,VirendraNath.2013."Bhadohicarpetindustryfacingtoughtimes."Tehelka,Sept.

3.Bhavnani,RikhilR.2009."DoElectoralQuotasWorkafterTheyAreWithdrawn?

EvidencefromaNaturalExperimentinIndia."AmericanPoliticalScienceReview103(01):23.doi:10.1017/s0003055409090029.

Blair,HarryW.1979.Voting,caste,community,society:explorationsinaggregatedataanalysisinIndiaandBangladesh.NewDelhi:YoungAsia.

Brass,PaulR.1964a."TheCongressPartyOrganizationinUttarPradesh:thetransformationfrommovementtopartyinanIndianstate."PhD,UniversityofChicago.

Brass,PaulR.1964b."FactionalismandtheCongressPartyinUttarPradesh."AsianSurvey4(9):1037-1047.

Brass,PaulR.1965.FactionalpoliticsinanIndianstate;theCongressPartyinUttarPradesh.Berkeley,:UniversityofCaliforniaPress.

Brass,PaulR.1974.Language,religionandpoliticsinNorthIndia.London;NewYork:CambridgeUniversityPress.

Brass,PaulR.1977."PartySystemsandGovernmentStabilityintheIndianStates."TheAmericanPoliticalScienceReview71(4):1384-1405.

Brass,PaulR.1980a."ThepoliticizationofthepeasantryinaNorthIndianstate:I."TheJournalofPeasantStudies7(4):395-426.doi:10.1080/03066158008438115.

Brass,PaulR.1980b."ThepoliticizationofthepeasantryinaNorthIndianstate:II."TheJournalofPeasantStudies8(1):3-36.doi:10.1080/03066158008438124.

Brass,PaulR.1981.Congress,theLokDal,andtheMiddle-PeasantCastes:AnAnalysisofthe1977and1980ParliamentaryElectionsinUttarPradesh.InPacificAffairs:PacificAffairs,UniversityofBritishColumbia.

Brass,PaulR.1983.Caste,faction,andpartyinIndianpolitics.2vols.Delhi:ChanakyaPublications.

Brass,PaulR.1984a.Caste,Faction,andPartyinIndianpolitics.Delhi:ChanakyaPublications.

Brass,PaulR.1984b."NationalPowerandLocalPoliticsinIndia-ATwenty-YearPerspective."ModernAsianStudies18(1):89-118.

Brass,PaulR.1984c."ThePoliticizationofthePeasantryinaNorthIndianState."InCaste,FactionandPartyinIndianPolitics,65-161.Delhi:ChanakyaPublications.

Brass,PaulR.1985.Caste,FactionandPartyinIndianPolitics.Vol.Vol.11.Brass,PaulR.1986."The1984ParliamentaryElectionsinUttarPradesh."AsianSurvey

26(6):653-669.Brass,PaulR.1990.ThePoliticsofIndiasinceIndependence.

344

Brass,PaulR.1997a."GeneralElections,1996inUttarPradesh:DivisiveStrugglesInfluenceOutcome."EconomicandPoliticalWeekly32(38):2403-2421.doi:10.2307/4405870.

Brass,PaulR.1997b.TheTheftofanIdol.Brass,PaulR.2003.TheproductionofHindu-MuslimviolenceincontemporaryIndia,

JacksonSchoolpublicationsininternationalstudies.Seattle:UniversityofWashingtonPress.

Brass,PaulR.2004."DevelopmentofanInstitutionalisedRiotSysteminMeerutCity,1961to1982."EconomicandPoliticalWeekly.

Brass,PaulR.2006.Formsofcollectiveviolence:riots,pogroms,&genocideinmodernIndia.1sted.NewDelhi:ThreeEssaysCollective.

Brass,PaulR.2011.AnIndianpoliticallife:CharanSinghandCongresspolitics,1937to1961.ThousandOaks:SAGEPublications.

Burger,AngelaSutherland.1969.Oppositioninadominant-partysystem;astudyoftheJanSangh,thePrajaSocialistParty,andtheSocialistPartyinUttarPradesh,India.Berkeley,:UniversityofCaliforniaPress.

Burger,AngelaSutherland,andUniversityofCaliforniaBerkeley.CenterforSouthandSoutheastAsiaStudies.1969.Oppositioninadominant-partysystem;astudyoftheJanSangh,thePrajaSocialistParty,andtheSocialistPartyinUttarPradesh,India.Berkeley,:UniversityofCaliforniaPress.

Butler,David,AshokLahiri,andPrannoyRoy.1995.Indiadecides:elections1952-1995.3rded.NewDelhi:Books&Things.

Campbell,JamesE.1983."TheReturnoftheIncumbents:TheNatureoftheIncumbencyAdvantage."TheWesternPoliticalQuarterly36(3):434-444.doi:10.2307/448401.

Carey,JohnM.,RichardG.Niemi,andLyndaW.Powell.2000."IncumbencyandtheProbabilityofReelectioninStateLegislativeElections."TheJournalofPolitics62(3):671-700.doi:10.2307/2647956.

Carter,AnthonyT.1974.ElitepoliticsinruralIndia:politicalstratificationandpoliticalalliancesinWesternMaharashtra,Cambridgestudiesinsocialanthropology.London;NewYork:CambridgeUniversityPress.

Chakraborty,Pinaki.2010.intra-regionalinequalityandtheroleofpublicpolicy:LessonsLearntfromKerala.InEconomic&PoliticalWeekly.

Chandra,Kanchan.2000."EliteIncorporationinMultiethnicSocieties."AsianSurvey40(5):836-865.

Chandra,Kanchan.2004a.ElectionsasAuctions.Seminar:1-7.Accessed22/05/1210.09PM.

Chandra,Kanchan.2004b."Whyethnicpartiessucceed:patronageandethnicheadcountsinIndia."Cambridgestudiesincomparativepolitics:xxi,343p.

Chandra,Kanchan.2004c.Whyethnicpartiessucceed:patronageandethnicheadcountsinIndia,Cambridgestudiesincomparativepolitics.Cambridge,UK;NewYork:CambridgeUniversityPress.

Chandra,Kanchan.2009."Whyvotersinpatronagedemocraciessplittheirtickets:Strategicvotingforethnicparties."ElectoralStudies28(1):21-32.doi:10.1016/j.electstud.2008.06.006.

Chandra,Kanchan.2016a.Democraticdynasties:state,party,andfamilyincontemporaryIndianpolitics.NewYork:CambridgeUniversityPress.

Chandra,Kanchan.2016b."Why"ForwardCastes"haveadynasticadvantage:intra-partypoliticsandco-ethnicfavoritism."InDemocraticDynasties.State,Partyand

345

FamilyinContemporaryIndianPolitics,editedbyKanchanChandra,207-237.NewDelhi:CambridgeUniversityPress.

Chatterjee,Partha.1997.StateandpoliticsinIndia,Themesinpoliticsseries.Delhi;NewYork:OxfordUniversityPress.

Chaudhuri,Kausik,FriedrichSchneider,andSumanaChattopadhyay.2006."Thesizeanddevelopmentoftheshadoweconomy:anempiricalinvestigationfromstatesofindia."JournalofDevelopmentEconomics80(2):428-443.

Chawla,Prabhu.2014."PAC:Acultureofbrutality."IndiaToday,January6.Chhibber,Pradeep.2011."Dynasticparties:Organization,financeandimpact."Party

Politics.doi:10.1177/1354068811406995.Chhibber,Pradeep,FrancescaRefsumJensenius,andPavithraSuryanarayan.2014.

"PartyorganizationandpartyproliferationinIndia."PartyPolitics20(4):489.Chishti,Seema.2013."WesternUPontheedge."TheIndianExpress,October13.

Accessed6August,2016.Chopra,VirK.1996.Marginalplayersinmarginalassemblies:theIndianMLA.New

Delhi:OrientLongman.Collie,MelissaP.1981."Incumbency,ElectoralSafety,andTurnoverintheHouseof

Representatives,1952-76."TheAmericanPoliticalScienceReview75(1):119-131.doi:10.2307/1962163.

Coppedge,Michael.1993."PartiesandSocietyinMexicoandVenezuela:WhyCompetitionMatters."ComparativePolitics25(3):253-274.doi:10.2307/422245.

Cox,GaryW.,andJonathanN.Katz.1996."WhyDidtheIncumbencyAdvantageinU.S.HouseElectionsGrow?"AmericanJournalofPoliticalScience40(2):478-497.doi:10.2307/2111633.

Cox,GaryW.,andMathewD.McCubbins.1986.ElectoralPoliticsasaRedistributiveGame.InTheJournalofPolitics.

Cox,GaryW.,andScottMorgenstern.1993."TheIncreasingAdvantageofIncumbencyintheU.S.States."LegislativeStudiesQuarterly18(4):495-514.doi:10.2307/439852.

Cox,GaryW.,andScottMorgenstern.1995."TheIncumbencyAdvantageinMultimemberDistricts:EvidencefromtheU.S.States."LegislativeStudiesQuarterly20(3):329-349.doi:10.2307/440224.

Dahl,RobertAlan.1961.Whogoverns?DemocracyandpowerinanAmericancity,Yalestudiesinpoliticalscience,.NewHaven,:YaleUniversityPress.

Deshpande,Rajeshwari.2004."HowGenderedWasWomen'sParticipationinElection2004?"Economic&PoliticalWeekly39(4):5431-5436.

Doron,Assa.2014."ThePoliticsofIdentityandthePeopleLeftBehind."InDevelopmentFailureandIdentityPoliticsinUttarPradesh,editedbyRogerJefferey,CraigJeffreyandJensLerche,188-210.NewDelhi:Sage.

Dubey,Amaresh.2010.intra-statedisparitiesinGujarat,Haryana,Kerala,Orissaandpunjab.InEconomic&PoliticalWeekly.

Duncan,Ian.1988."PartypoliticsandthenorthIndianpeasantry:TheriseoftheBharatiyaKrantiDalinUttarPradesh."TheJournalofPeasantStudies16(1):40-76.doi:10.1080/03066158808438382.

Duncan,Ian.1997."NewPoliticalEquationsinNorthIndia-Mayawati,Mulayam,andGovernmentInstabilityinUttarPradesh."AsianSurvey37(10):979-996.

Duncan,Ian.1999."DalitsandpoliticsinruralNorthIndia:TheBahujanSamajpartyinUttarPradesh."TheJournalofPeasantStudies27(1):35-60.doi:10.1080/03066159908438724.

346

Elder,JosephW.1962."LandConsolidationinanIndianVillage:ACaseStudyoftheConsolidationofHoldingsActinUttarPradesh."EconomicDevelopmentandCulturalChange11(1):16-40.

Elliott,Carolyn.2011."Movingfromclientelistpoliticstowardawelfareregime:evidencefromthe2009assemblyelectioninAndhraPradesh."Commonwealth&ComparativePolitics49(1):48-79.doi:10.1080/14662043.2011.541111.

Engineer,AsgharAli.1984.Communalriotsinpost-independenceIndia.HyderabadIndia:SangamBooks:DistributedbyOrientLongman.

ENS.2013."SPstudentwingchief,otherworkersbookedforrioting,murderattempt."IndianExpress,Nov.15.Accessed15August2016.

Erdman,HowardL.1967.TheSwatantraPartyandIndianconservatism,CambridgeSouthAsianstudies.London,:CambridgeU.P.

Erikson,RobertS.1971."TheAdvantageofIncumbencyinCongressionalElections."Polity3(3):395-405.doi:10.2307/3234117.

Farooqui,A.,andE.Sridharan.2014."Incumbency,internalprocessesandrenominationinIndianparties."Commonwealth&ComparativePolitics52(1):78-108.doi:10.1080/14662043.2013.867690.

Fickett,LewisP.,Jr.1973."ThePrajaSocialistPartyofIndia--1952-1972-AFinalAssessment."AsianSurvey13(9):826-832.

Fickett,LewisP.,Jr.1993."TheJanataDalintheNinthIndianGeneralElectionof1989andItsFutureProspects."InIndiaVotes.AlliancePoliticsandMinorityGovernmentsintheNinthandTenthGeneralElections,editedbyHaroldA.GouldandSumitGanguly,81-107.Boulder,Colorado:WestviewPress.

Field,JohnOsgood,andMarcusF.Franda.1974.ElectoralpoliticsintheIndianstates:theCommunistpartiesofWestBengal,StudiesinelectoralpoliticsintheIndianstates.Delhi:ManoharBookService.

Field,JohnOsgood,andMyronWeiner.1975.ElectoralpoliticsintheIndianstates:threedisadvantagedsectors,StudiesinelectoralpoliticsintheIndianstates.Delhi:ManoharBookService.

Field,JohnOsgood,andMyronWeiner,eds.1977.ElectoralpoliticsintheIndianstates:theimpactofmodernization,StudiesinelectoralpoliticsintheIndianstates.NewDelhi:ManoharBookService.

Franda,MarcusF.1971.RadicalpoliticsinWestBengal,Studiesincommunism,revisionism,andrevolution.Cambridge,Mass.,:M.I.T.Press.

Frankel,Francine.1977."ProblemsofCorrelatingElectoralandEconomicVariables:AnAnalysisofElectoralBehaviourandagrarianModernizationinUttarPradesh."InElectoralpoliticsintheIndianstates:theimpactofmodernization,editedbyJohnOsgoodFieldandMyronWeiner,149-181.NewDelhi:ManoharBookService.

Frankel,Francine,FrancineFrankel,andM.S.A.Rao.1990.DominanceandStatePowerinModernIndia:DeclineofaSocialOrder.Vol.Vol.11.

Frankel,FrancineR.1969."India'sNewStrategyofAgriculturalDevelopment:PoliticalCostsofAgrarianModernization."TheJournalofAsianStudies28(4):693-710.doi:10.2307/2942406.

Frankel,FrancineR.1971.India'sgreenrevolution;economicgainsandpoliticalcosts.Princeton,N.J.,:PrincetonUniversityPress.

Frankel,FrancineR.,andM.S.A.Rao.1989.DominanceandstatepowerinmodernIndia:declineofasocialorder.2vols.Delhi;NewYork:OxfordUniversityPress.

Friese,Kai.1990."PeasantCommunitiesandAgrarianCapitalism."EconomicandPoliticalWeekly25(39):A135-A143.

347

Galanter,Marc.1984.Competingequalities:lawandthebackwardclassesinIndia.Delhi:Oxford.

Galonnier,Juliette,andVioletteGraff.2013.Hindu-MuslimCommunalRiotsinIndiaII(1947-2011).editedbyOnlineEncyclopediaofMassViolence.

Gayer,Laurent,andChristopheJaffrelot.2012.MuslimsinIndiancities:trajectoriesofmarginalisation.NewYork(N.Y.):ColumbiaUniversityPress.

Gould,HaroldA.1963."TheAdaptiveFunctionsofCasteinContemporaryIndianSociety."AsianSurvey3(9):427-438.

Gould,HaroldA.1969."Towarda'JatiModel'forIndianPolitics."EconomicandPoliticalWeekly4(5):291-297.

Gould,HaroldA.,andSumitGanguly.1993.Indiavotes:alliancepoliticsandminoritygovernmentsintheninthandtenthgeneralelections.Boulder,Colo.:WestviewPress.

Graham,B.D.1986."Thecandidate�selectionpoliciesoftheIndiannationalcongress,1952–69."TheJournalofCommonwealth&ComparativePolitics24(2):197-218.doi:10.1080/14662048608447494.

Graham,BruceDesmond.1993.Representationandpartypolitics:acomparativeperspective,Comparativepolitics.Oxford,UK;Cambridge,Mass.,USA:B.Blackwell.

Guha,Sohini.2008."EthnicParties,MaterialPoliticsandtheEthnicPoor:TheBahujanSamajPartyinNorthIndia."Ph.D.PoliticalScience,DepartmentofPoliticalScience,McGillUniversity.

Guha,Sohini.2011."FromEthnictoMultiethnic:TheTransformationoftheBahujanSamajPartyinNorthIndia."Ethnopolitics:1-29.doi:10.1080/17449057.2011.601850.

Gupta,Akhil.2012.Redtape:bureaucracy,structuralviolence,andpovertyinIndia,AJohnHopeFranklinCenterbook.Durham:DukeUniversityPress.

Gupta,Dipankar.2016."Biharpollsnailinthecoffinforcastepolitics:DipankarGupta."BusinessStandard,Nov.14.Accessed24July2016.

Gupta,Dipankar,andYogeshKumar.2007."WhentheCasteCalculusFails:AnalysingBSP'sVictoryinUP."EconomicandPoliticalWeekly42(33):3388-3396.

Hansen,ThomasBlom.1999.Thesaffronwave:democracyandHindunationalisminmodernIndia.Princeton,NJ:PrincetonUniversityPress.

Hansen,ThomasBlom.2001.Wagesofviolence:namingandidentityinpostcolonialBombay.Princeton,N.J.:PrincetonUniversityPress.

Hansen,ThomasBlom,andChristopheJaffrelot.2001.TheBJPandthecompulsionsofpoliticsinIndia.2nded,OxfordIndiapaperbacks.NewDelhi;NewYork:OxfordUniversityPress.

Harriss,John.2011."ReflectionsonAgrarianChangeinIndiasinceIndependance:Does'Landlordism'StillMatter?"SimonFraserUniversity.

Hasan,Zoya.1989.DominanceandMobilisation:RuralPoliticsinWesternUttarPradesh1930–1980.

Hasan,Zoya.1993."PowerandMobilization:PatternsofResilienceandChangeinUttarPradeshPolitics."InDominanceandStatePowerinModernIndia,editedbyFrancineFrankelandM.S.A.Rao,133-203.NewDelhi:OxfordUniversityPress.

Hasan,Zoya.1994."Shiftingground:Hindutvapoliticsandthefarmers’movementinUttarPradesh."TheJournalofPeasantStudies21(3-4):165-194.doi:10.1080/03066159408438558.

Hasan,Zoya.1998.QuestforPower.OppositionalMovementsandPost�CongressPoliticsinUttarPradesh.

349

Jaffrelot,Christophe,andSanjayKumar.2009.RiseofthePlebeians?:theChangingFaceofIndianLegislativeAssemblies,ExploringthepoliticalinSouthAsia.NewDelhi:Routledge.

Jaffrelot,Christophe,andGillesVerniers.2011."Re-nationalizationofIndia’sPoliticalPartySystemorContinuedPrevalenceofRegionalismandEthnicity?Evidencefromthe2009GeneralElections."AsianSurvey51(6):1090-1112.

Jaffrelot,Christophe,andGillesVerniers.2012."Castes,CommunitiesandPartiesinUttarPradesh."EconomicandPoliticalWeekly37(32):89-93.

Jaffrelot,Christophe,andGillesVerniers.2014a."InvisibleintheHouse."TheIndianExpress,28May.AccessedJuly1,2016.

Jaffrelot,Christophe,andGillesVerniers.2014b."Lawmakersandlaw-breakers."IndianExpress,July2.Accessed21April2016.

Jaffrelot,Christophe,andGillesVerniers.2015."TheResistanceofRegionalism:BJP’sLimitationsandtheResilienceofStateParties."InIndia's2014Elections.AModi-ledSweep,editedbyPaulWallace.NewDelhi:Sage.

Jalan,Bimal.2005.ThefutureofIndia:politics,economics,andgovernance.NewDelhi;NewYork,NewYork:Penguin,Viking.

Jaoul,Nicolas.2010."Politicaland'Non-Political'MeansintheDalitMovement."InPoliticalProcessinUttarPradesh.Identity,EconomicReformsandGovernance,editedbySudhaPai,191-220.NewDelhi:PearsonLongman.

Jayal,NirajaGopal.2006.RepresentingIndia:ethnicdiversityandthegovernanceofpublicinstitutions,Ethnicity,inequality,andpublicsectorgovernance.BasingstokeEngland;NewYork:PalgraveMacmillan.

Jeelani,Mehboob.2013."UndertheInfluence"TheCaravan,1November.Jeffery,Patricia,andRogerJeffery.2006.Confrontingsaffrondemography:religion,

fertility,andwomen'sstatusinIndia.Gurgaon:ThreeEssaysCollective.Jeffery,Roger,CraigJeffrey,andJensLerche.2014.Developmentfailureandidentity

politicsinUttarPradesh.NewDelhi:SAGEPublicationsIndiaPvtLtd.Jeffrey,Craig.2001."‘Afistisstrongerthanfivefingers’:casteanddominanceinrural

northIndia."TransactionsoftheInstituteofBritishGeographers26(2):217-236.doi:10.1111/1475-5661.00016.

Jeffrey,Craig.2002."Caste,Class,andClientelism:APoliticalEconomyofEverydayCorruptioninRuralNorthIndia."EconomicGeography78(1):21-41.doi:10.2307/4140822.

Jeffrey,Craig.2003."SoftStates,HardBargains:RichFarmers,ClassReproductionandtheLocalStateinRuralNorthIndia."InSocialandPoliticalChangeinUttarPradesh.EuropeanPerspectives,editedbyRogerJefferyandJensLerche,225-246.NewDelhi:Manohar.

Jeffrey,Craig.2009."FixingFutures:EducatedUnemploymentthroughaNorthIndianLens."ComparativeStudiesinSocietyandHistory51(1):182-211.doi:10.2307/27563735.

Jeffrey,Craig.2010a."ContradictoryYouthPolitics:StudentMobilisationinUttarPradesh."InPowerandInfluenceinIndia.Bosses,LordsandCaptains,editedbyPamelaPriceandRuudAlridEngelsen,96-118.NewDelhi:Routledge.

Jeffrey,Craig.2010b.Timepass:youth,class,andthepoliticsofwaitinginIndia.Stanford,Calif.:StanfordUniversityPress.

Jeffrey,Craig.2013."NoCountryforYoungMen?"Economic&PoliticalWeekly48(24):33-35.

350

Jeffrey,Craig,PatriciaJeffery,andRogerJeffery.2005."ReproducingDifference?Schooling,Jobs,andEmpowermentinUttarPradesh,India."WorldDevelopment33(12):2085-2101.doi:10.1016/j.worlddev.2005.07.006.

Jeffrey,Craig,PatriciaJeffery,andRogerJeffery.2008a."DalitRevolution?NewPoliticiansinUttarPradesh,India."TheJournalofAsianStudies67(4):1365-1396.doi:10.2307/20203489.

Jeffrey,Craig,PatriciaJeffery,andRogerJeffery.2008b.DegreeswithoutFreedom?Education,Masculinities,andUnemploymentinNorthIndia.Stanford,Calif.:StanfordUniversityPress.

Jeffrey,Robin,andAssaDoron.2012."Mobile-izing:Democracy,OrganizationandIndia'sFirst“MassMobilePhone”Elections."TheJournalofAsianStudies71(01):63-80.doi:doi:10.1017/S0021911811003007.

Jenkins,Rob.2004.Regionalreflections:comparingpoliticsacrossIndia'sstates.NewDelhi;NewYork:OxfordUniversityPress.

Jenkins,Rob.2005.WashingtonDC.Jensenius,FrancescaR.2016a."CompetingInequalities?OntheIntersectionofGender

andEthnicityinCandidateNominationsinIndianElections∗."GovernmentandOpposition51(3).Jensenius,FrancescaR.2016b."ASignogBackwardness?WhereDynasticleadersare

electedinIndia."InDemocraticdynasties:state,party,andfamilyincontemporaryIndianpolitics,editedbyKanchanChandra,83-104.NewYork:CambridgeUniversityPress.

Jensenius,FrancescaRefsum.2012."PoliticalQuotasinIndia:PerceptionsofConstituentPoliticalRepresentation."AsianSurvey52(2):373-394.

Jensenius,FrancescaRefsum.2013."WastheDelimitationCommissionUnfairtoMuslims?"StudiesinIndianPolitics1(2):213-229.

Jha,MithileshKumar.2010."GyaneshKudaisya,Region,Nation,‘Heartland’:UttarPradeshinIndia’sBodyPolitic(SageSeriesinModernIndianHistory–X).NewDelhi:SagePublications,2006.xxvi+471pp.Figures,plates,maps,notes,glossary,bibliography,index.Rs595(paperback)."ContributionstoIndianSociology44(3):462-464.

Jha,Prashant.2016."MissionUP2017:Howpartiesarepavingtheirwaytothetop."HindustanTimes,11July.Accessed11July2016.

Jhangiani,MotilalA.1967.JanaSanghandSwatantra;aprofileoftherightistpartiesinIndia.Bombay,:Manaktalas.

Jodhka,SurinderS.2015.CasteincontemporaryIndia.NewDelhi:Routledge.Jones,PhilipEdward.2012."TheEffectofPoliticalCompetitiononDemocratic

Accountability."PoliticalBehavior35(3):481-515.Kapoor,Mudit,andShamikaRavi.2014.WhySoFewWomeninPolitics?Evidencefrom

India.NewDelhiBrookingsIndia.Kapur,Devesh,ChandraBhanPrasad,LantPritchett,andD.ShyamBabu.2010.

"RethinkingInequality:DalitsinUttarPradeshintheMarketReformEra."EconomicandPoliticalWeekly45(35):39-49.

Kapur,Devesh,andMilanVaishnav.2011.Washington.Kashyap,SubhashC.1969.ThePoliticsofDefection:AStudyofStatePoliticsinIndia.

Delhi:NationalPublishingHouse.Key,V.O.1955."ATheoryofCriticalElections."TheJournalofPolitics17(1):3-18.doi:

10.2307/2126401.Khetan,Ashish.2012."TheRajawhostolefromthepoor."Tehelka,April.

351

Kitschelt,Herbert,andStevenWilkinson.2007.Patrons,clients,andpolicies:patternsofdemocraticaccountabilityandpoliticalcompetition.Cambridge,UK;NewYork:CambridgeUniversityPress.

Kochanek,StanleyA.1966."TheIndianNationalCongress-TheDistributionofPowerbetweenPartyandGovernment."TheJournalofAsianStudies25(4):681-697.

Kochanek,StanleyA.1968.TheCongresspartyofIndia;thedynamicsofone-partydemocracy.Princeton,N.J.,:PrincetonUniversityPress.

Kohli,Atul.1987.ThestateandpovertyinIndia:thepoliticsofreform,CambridgeSouthAsianstudies.CambridgeCambridgeshire;NewYork:CambridgeUniversityPress.

Kohli,Atul.1990.Democracyanddiscontent:India'sgrowingcrisisofgovernability.Cambridge;NewYork:CambridgeUniversityPress.

Kohli,Atul.1997."CanDemocraciesAccommodateEthnicNationalism-RiseandDeclineofSelf-DeterminationMovementsinIndia."TheJournalofAsianStudies56(2):325-344.

Kohli,Atul.2012.PovertyamidplentyinthenewIndia.Cambridge;NewYork:CambridgeUniversityPress.

Kothari,Rajni.1964."TheCongress'System'inIndia."AsianSurvey4(12):1161-1173.Kothari,Rajni.1967."India-TheCongressSystemonTrial."AsianSurvey7(2):83-96.Kothari,Rajni.1970a.CasteinIndianpolitics.NewDelhi:OrientLongman.Kothari,Rajni.1970b.PoliticsinIndia.NewDelhi:OrientLongman.Kothari,Rajni.1974."TheCongressSystemRevisited-ADecennialReview."Asian

Survey14(12):1035-1054.Kothari,Rajni.1990."CasteandPolitics:TheGreatSecularUpsurge."InMandal

CommissionControversy,editedbyAsgharAliEngineer.NewDelhi:AjantaPublications.

Krashinsky,Michael,andWilliamJ.Milne.1993."TheEffectsofIncumbencyinU.S.CongressionalElections,1950-1988."LegislativeStudiesQuarterly18(3):321-344.doi:10.2307/439829.

Krishna,Anirudh.2003."WhatIsHappeningtoCaste?AViewfromSomeNorthIndianVillages."TheJournalofAsianStudies62(4):1171-1193.

Kudaisya,Gyanesh.2006.Region,nation,"heartland":UttarPradeshinIndia'sbody-politic,SageseriesinmodernIndianhistory.NewDelhi;ThousandOaks,Calif.:SagePublications.

Kumar,Arun.1997.Theturningpoint,1996pollstory.Delhi:PressTrustofIndiainassociationwithKonarkPublishers.

Kumar,Sanjay.2009."PatternsofPoliticalParticipation:TrendsandPerspective."Economic&PoliticalWeekly44(39):47-51.

Kumar,Sanjay,andPraveenRai.2013.MeasuringvotingbehaviourinIndia.LosAngeles:SAGE.

Kumar,Sanjay,ShreyasSardesai,andPranavGupta.2013."TheWeakeningofElectoralAnti-Incumbency."Economic&PoliticalWeekly48(13):128-131.

Kumar,Satendra.2014."RuraltransformationandoccupationaldiversificationinWesternUttarPradesh:Economicanddemographicchangeinavillage."InDevelopmentFailureandIdentityPoliticsinUttarPradesh,editedbyRogerJeffery,CraigJeffreyandJensLerche,18-46.NewDelhi:Sage.

Kumar,Satendra.2016."AgrarianTransformationandtheNewRuralityinWesternUttarPradesh."EconomicandPoliticalWeekly:62-71.

Kushner,Daniel.2015."HowIndianPartiesConnectVoterstoLeaders."AsianSurvey55(5):969-990.

352

Lerche,Jens.1999."Politicsofthepoor:AgriculturallabourersandpoliticaltransformationsinUttarPradesh."TheJournalofPeasantStudies26(2-3):182-241.doi:10.1080/03066159908438707.

Lieten,GeorgesKristoffel,andRaviSrivastava.1999.Unequalpartners:powerrelations,devolution,anddevelopmentinUttarPradesh,Indo-Dutchstudiesondevelopmentalternatives.NewDelhiThousandOaks,Calif.:SagePublications.

Lindberg,Staffan,andStigToftMadsen.2003."ModellingInstitutionalFate:TheCaseofaFarmer'sMovementinUttarPradesh."InSocialandPoliticalChangeinUttarPradesh.EuropeanPerspectives,editedbyRogerJefferyandJensLerche,199-224.NewDelhi:Manohar.

Lohia,RamManohar,andMastramKapoor.2011.CollectedworksofDr.RammanoharLohia.9vols.NewDelhi:PublishedforSamajwadiSahityaSansthanbyAnamikaPublishers&Distributors.

Mahapatra,Dhananjay.2012."Ex-MLCunderlensfor10,000cr'blackmoney'"TheEconomicTimes,February9.AccessedJuly13,2016.

Malhotra,Inder.2004.DynastiesofIndiaandbeyond:Pakistan,SriLanka,Bangladesh.NewDelhi:HarperCollinsPublishersIndia.

Manor,James.2000."Small-TimePoliticalFixersinIndia'sStates:"ToweloverArmpit"."AsianSurvey40(5):816-835.

Manor,James.2003."OrganisationalRenewal."Seminar.Manor,James.2005."InPartaMyth:TheBJP'sOrganizationalStrength."InCoalition

PoliticsandHinduNationalism,editedbyKatharineAdeneyandLawrenceSaez,55-74.NewDelhi:Routledge.

Manor,James.2010."Epilogue:casteandpoliticsinrecenttimes."InCasteinIndianPolitics,editedbyRajniKothariandJamesManor.NewDelhi:OrientBalckswan.

Mathur,M.L.2004.Encyclopaediaofbackwardcastes.4vols.Delhi:KalpazPublications.Mayer,Albert.1958.Pilotproject,India:thestoryofruraldevelopmentatEtawah,Uttar

Pradesh.Berkeley:UniversityofCaliforniaPress.Mayhew,DavidR.2008."IncumbencyAdvantageinU.S.PresidentialElections:The

HistoricalRecord."PoliticalScienceQuarterly123(2):201-228.doi:10.2307/20203009.

McMillan,Alistair.2000."Delimitation,Democracy,andEndofConstitutionalFreeze."Economic&PoliticalWeekly(April8):1271-1275.

McMillan,Alistair.2005.Standingatthemargins:representationandelectoralreservationinIndia.NewDelhi:OxfordUniversityPress.

Mehrotra,Santosh.2011.IndiaHumanDevelopmentReport.TowardsSocialInclusion.NewDelhi:PlanningCommission.

Mendelsohn,Oliver.1993."TheTransformationofAuthorityinRuralIndia."ModernAsianStudies27(4):805-842.doi:10.2307/312832.

Mendelsohn,Oliver,andMarikaVicziany.1998.TheUntouchables:Subordination,PovertyandtheStateinModernIndia.

Metcalf,ThomasR.1967."LandlordsWithoutLand-TheU.P.ZamindarsToday."PacificAffairs40(1/2):5-18.

Meyer,RalphC.1969."ThePoliticalEliteinanUnder-developedSociey:TheCaseofUttarPradesh."UniversityofPennsylvania.

Michelutti,Lucia.2004."Review:India'sSilentRevolution:TheRiseoftheLowerCastesinNorthIndia."ReviewofIndia'sSilentRevolution:TheRiseoftheLowerCastesinNorthIndia,CristopheJaffrelot.TheJournaloftheRoyalAnthropologicalInstitute10(2):473-474.

353

Michelutti,Lucia.2007."TheVernacularizationofDemocracy:PoliticalParticipationandPopularPoliticsinNorthIndia."TheJournaloftheRoyalAnthropologicalInstitute13(3):639-656.

Michelutti,Lucia.2008.Thevernacularisationofdemocracy:politics,caste,andreligioninIndia.NewDelhi:Routledge.

Michelutti,Lucia.2010."Wrestlingwith(Body)Politics:Understanding'Goonda;PoliticalStylesinNorthIndia."InPowerandInfluenceinIndia.Bosses,LordsandCaptains,editedbyPamelaPriceandRuudAlridEngelsen,44-69.NewDelhi:Routledge.

Michelutti,Lucia.2014."KingshipwithoutkingsinnorthernIndia."InPatronageasPoliticsinSouthAsia,editedbyAnastasiaPiliavsky,283-302.Cambridge:CambridegUniversityPress.

Mustafa,Seema.1995.Thelonelyprophet:V.P.Singh,apoliticalbiography.NewDelhi:NewAgeInternational.

Narayan,Badri.2006."Memories,SaffronisingStatuesandConstructingCommunalPolitics."EconomicandPoliticalWeekly41(45):4695-4701.

Narayan,Badri.2014."CommunalRiotsinUttarPradesh."EconomicandPoliticalWeekly:29-32.

Neale,WalterC.1962.EconomicchangeinruralIndia;landtenureandreforminUttarPradesh,1800-1955,Yalestudiesineconomics,.NewHaven,:YaleUniversityPress.

Neale,WalterC.1970.LandreforminUttarPradesh,India.WashingtonD.C.;:AgencyforInternationalDevelopment.

Oldenburg,Philip.1974."BigcitygovernmentinIndiacouncilor,administratorandcitizeninDelhi."PhD,UniversityofChicago.

Oldenburg,Philip.1987."MiddlemeninThird-WorldCorruption:ImplicationsofanIndianCase."WorldPolitics39(4):508-535.doi:10.2307/2010290.

Oldenburg,PhilipK.1988."Pollsters,PunditsandaMandatetoRule:InterpretingIndia's1984ParliamentaryElections."JournalofCommonwealth&ComparativePolitics26(3):296-317.

Pai,Sudha.1986.ChangingagrarianrelationsinU.P.:astudyofthenorth-easternareas.NewDelhi,India:Inter-IndiaPublications.

Pai,Sudha.1993.UttarPradesh,agrarianchangeandelectoralpolitics.Delhi,India:Shipra.

Pai,Sudha.1994."CasteandCommunalMobilizationintheElectoralPoliticsofUttarPradesh."TheIndianJournalofPoliticalScience55(3):307-320.doi:10.2307/41855702.

Pai,Sudha.1997."DalitAssertioninUP:ImplicationsforPolitics."EconomicandPoliticalWeekly32(37):2313-2314.

Pai,Sudha.1998a."TheIndianPartySystemunderTransformation:LokSabhaElections1998."AsianSurvey38(9):836-852.

Pai,Sudha.1998b."NewPoliticalTrendsinUttarPradesh:TheBJPandtheLokSabhaElections,1998."EconomicandPoliticalWeekly33(28):1841-1845.

Pai,Sudha.1999."BSP’sNewElectoralStrategyPaysOff."Economic&PoliticalWeekly(October30):3099-3101.

Pai,Sudha.2000a."ChangingSocio-EconomicandPoliticalProfileofScheduledCastesinUttarPradesh."JournalofIndianSchoolofPoliticalEconomy12(3&4):405-422.

Pai,Sudha.2000b.StatePolitics.NewDimensions(PartySystem,LiberalisationandPoliticsofIdentity).NewDelhi:Shipra.

354

Pai,Sudha.2002a.Dalitassertionandtheunfinisheddemocraticrevolution:theBahujanSamajPartyinUttarPradesh,CulturalsubordinationandtheDalitchallenge.NewDelhi;ThousandOaks,Calif.:SagePublications.

Pai,Sudha.2002b."ElectoralIdentityPoliticsinUttarPradesh:HungAssemblyAgain."EconomicandPoliticalWeekly37(14):1334-1341.

Pai,Sudha.2007."PoliticalprocessinUttarPradesh:identity,economicreforms,andgovernance."xlviii,415p.

Pai,Sudha.2009."NewSocialEngineeringAgendaoftheBahujanSamajParty:ImplicationsforStateandNationalPolitics."SouthAsia:JournalofSouthAsianStudies32(3):338-353.doi:10.1080/00856400903374277.

Pai,Sudha.2013."UttarPradesh.Newpatternsofmobilizationinthe1990sandbeyond."InRoutledgeHandbookofIndianPolitics,editedbyAtulKohliandPrernaSingh,261-269.Oxon:Routledge.

Pai,Sudha,andJagpalSingh.1997."PoliticisationofDalitsandMostBackwardCastes:StudyofSocialConflictandPoliticalPreferencesinFourVillagesofMeerutDistrict."EconomicandPoliticalWeekly32(23):1356-1361.

Paley,Julia.2002."TowardAnanthropologyOfdemocracy."AnnualReviewofAnthropology31(1):469-496.doi:10.1146/annurev.anthro.31.040402.085453.

Paley,Julia.2008.Democracy:anthropologicalapproaches.1sted,SchoolforAdvancedResearchadvancedseminarseries.SantaFe:SchoolforAdvancedResearchPress.

Palshikar,Suhas.2007."TheImaginedDebatebetweenPollstersandEthnographers."EconomicandPoliticalWeekly.

Palshikar,Suhas.2013."RegionalandCasteParties."InRoutledgeHandbookofIndianPolitics,editedbyAtulKohliandPrernaSingh,91-104.Abingdon,UK:Routledge.

Palshikar,Suhas,K.C.Suri,andYogendraYadav.2014.PartycompetitioninIndianstateselectoralpoliticsinpost-congresspolity.NewDelhi:OxfordUniversityPress.

Parikh,Manju.1993."TheDebacleatAyodhya:WhyMilitantHinduismMetwithaWeakResponse."AsianSurvey33(7):673-684.

Parmar,Chandrika.1996."PartyStrategiesintheUttarPradeshAssemblyElections,1996."

Piliavsky,Anastasia.2014.PatronageaspoliticsinSouthAsia.NewYork:CambridgeUniversityPress.

PlanningCommission,ofIndia.2010.UttarPradeshDevelomentReport.editedbyAcademicFoundation.NewDelhi:PlanningCommissionofIndia.

Powell,G.Bingham.2000.Electionsasinstrumentsofdemocracy:majoritarianandproportionalvisions.NewHaven,CT:YaleUniversityPress.

Price,PamelaG.,andArildEngelsenRuud.2010.PowerandinfluenceinIndia:bosses,lords,andcaptains,ExploringthepoliticalinSouthAsia.NewDelhi:Routledge.

Przeworski,Adam.2000.Democracyanddevelopment:politicalinstitutionsandwell-beingintheworld,1950-1990,Cambridgestudiesinthetheoryofdemocracy.Cambridge:CambridgeUniversityPress.

Quraishi,S.Y.2014.Anundocumentedwonder:thegreatIndianelection.NewDelhi:Rainlight,RupaPublications.

Raghavan,E.,andJamesManor.2009.Broadeninganddeepeningdemocracy:politicalinnovationinKarnataka,ExploringthepoliticalinSouthAsia.NewDelhi:Routledge.

Rashid,Omar.2016."47policemengetlifetermforfakeencounterkillingsinUttarPradesh."TheHindu,April5.

355

Reddy,G.Ram,andG.Haragopal.1985."ThePyraveekar:"TheFixer"inRuralIndia."AsianSurvey25(11):1148-1162.

Reeves,Peter.1991.LandlordsandgovernmentsinUttarPradesh:astudyoftheirrelationsuntilzamindariabolition.Bombay;NewYork:OxfordUniversityPress.

Retzlaff,RalphHerbert.1959.AcasestudyofpanchayatsinanorthIndianvillage.Berkeley,:CenterforSouthAsiaStudies,InstituteofInternationalStudies.

Retzlaff,RalphHerbert.1962.VillagegovernmentinIndia;acasestudy.Bombay,NewYork,:AsiaPub.House.

Richter,WilliamL.1977."ElectoralPatternsinPost-PrincelyIndia."InElectoralpoliticsintheIndianstates:threedisadvantagedsectors,editedbyJohnOsgoodFieldandMyronWeiner,1-77.Delhi:ManoharBookService.

Robin,Cyril.2009."Bihar:TheNewStrongholdofOBCPolitics."InRiseoftheplebeians?:thechangingfaceofIndianlegislativeassemblies,65-102.NewDelhi:RoutledgeIndia.

Robins,RobertS.1967."PoliticalEliteFormationinRuralIndia:TheUttarPradeshPanchayatElectionsof1949,1956,and1961."TheJournalofPolitics29(4):838-860.

Roy,Ramashray.1966."SelectionofCongressCandidatesI-Theformalcriteria."EconomicandPoliticalWeekly1(20):833-840.

Roy,Ramashray.1967a."SelectionofCongresscandidatesII:Pressuresandcounter-pressures."Economic&PoliticalWeekly2(1):17-24.Roy,Ramashray.1967b."SelectionofCongressCandidatesIII-Claimsandcounter-

claims."EconomicandPoliticalWeekly2(2):61-66.Rudolph,SusanneHoeber,andLloydI.Rudolph.2010."Anintellectualhistoryofthe

studyofIndianpolitics."InTheOxfordCompaniontoPoliticsinIndia,editedbyPratapBhanuMehta,555-586.NewDelhi:OxfordUniversityPress.

Ruparelia,Sanjay.2015.Dividedwegovern:coalitionpoliticsinmodernIndia.Oxford;NewYork:OxfordUniversityPress.

Sachar,Rajinder.2006.Social,EconomicandEducationalStatusoftheMuslimCommunityofIndia.NewDelhi:PrimeMinister’sHighLevelCommittee,Cabinet

Secretariat,GovernmentofIndia.Sanchez,Andrew.2016.CriminalCapital.Violence,CorruptionandClassinIndustrial

India.EditedbyMukulikaBanerjee,ExploringthePoliticalinSouthAsia.NewDelhi:Routledge.

Sartori,Giovanni.1987.Thetheoryofdemocracyrevisited.Chatham,N.J.:ChathamHousePublishers.

Sasikumar,Karthika,andGillesVerniers.2013."TheIndia-U.S.NuclearCooperationAgreement:ExplainingtheContentiousIndianDebate."AsianSurvey53(4):679-702.

Schoenfeld,BenjaminN.1965."TheBirthofIndia'sSamyuktaSocialistParty."PacificAffairs38(3/4):245-268.

Seshan,TirunellaiNarayanaIyer.1995.Aheartfullofburden.NewDelhi:UBSPD.Shafiuzzaman.2003.TheSamajwadiParty:astudyofitssocialbase,ideology,and

programme.NewDelhi:APHPub.Corp.Shankar,Kripa.1996."LokSabhaElections:BalanceSheetinUP."Economicand

PoliticalWeekly31(21):1243-1244.Sharma,Rita,andThomasT.Poleman.1994.TheNewEconomicsofIndia'sGreen

Revolution:IncomeandEmploymentDiffusioninUttarPradesh.

356

Sheth,Dhirubhai.1996."ChangesinSocialStratification."InIndiaBriefing:LookingBack,LookingAhead,editedbyPhilipOldenburgandMarshallBouton.Armonk,N.Y.:M.E.Sharpe.

Singer,Wendy.2012."ASeatattheTable:ReservationsandRepresentationinIndia'sElectoralSystem."ElectionLawJournal11(2):202-215.

Singh,AjitKumar.2009."UttarPradeshEconomy:PastPerformanceandFutureChallenges."InUttarPradesh.TheRoadAhead,editedbyVenkiteshRamakrishnan,69-83.NewDelhiAcademicFoundation.

Singh,AjitKumar,andSantoshK.Mehrotra.2014.Landpoliciesforequityandgrowth:transformingtheagrarianstructureinUttarPradesh.LosAngeles:SAGEPublications.

Singh,Charan.1947.Abolitionofzamindari,twoalternatives.Allahabad,:Kitabistan.Singh,Charan.1959.Jointfarmingx-rayed;theproblemanditssolution.1sted.Bombay,:

BharatiyaVidyaBhavan.Singh,Charan.1964.India'spovertyanditssolution.2drev.ed.NewYork,:AsiaPub.

House.Singh,Jagpal.1992.Capitalismanddependence:agrarianpoliticsinwesternUttar

Pradesh,1951-1991.NewDelhi:ManoharPublishersandDistributors.Singh,Nirmal.2014."BSPinPunjab.AnalysingItsFailure."Economic&PoliticalWeekly

49(48):17-19.Singh,Nirvikar,JakeKendall,R.K.Jain,andJaiChander.2014."RegionalInequalityin

Indiainthe1990s.ADistrict-levelView."Economic&PoliticalWeekly49(15):71-76.

Singh,Prerna.2015."SubnationalismandSocialDevelopment:AComparativeAnalysisofIndianStates."WorldPolitics67(03):506-562.

Singh,Rajendra.1974."AgrarianSocialStructureandPeasantUnrest:AStudyofLand-GrabMovementinDistrictBasti,EastU.P."SociologicalBulletin23(1):44-70.

Singh,Shrinath.1976.ModernisationofAgriculture:ACaseStudyinEasternUttarPradesh.

Singh,Sunit.2002.CastereconfigurationandU.P.elections.NewDelhi:KanishkaPublishers.

Sircar,Neelanjan,andGillesVerniers.2016."Title."CPRWorkingPapers,NewDelhi.Smith,TimothyHallam.2013."Areyousittingcomfortably?Estimatingincumbency

advantageintheUK:1983–2010–Aresearchnote."ElectoralStudies32(1):167-173.doi:10.1016/j.electstud.2012.12.002.

Spencer,Jonathan.2007.Anthropology,politicsandthestate:democracyandviolenceinSouthAsia,Newdeparturesinanthropology.Cambridge;NewYork:CambridgeUniversityPress.

Sridharan,E.2014."ClassVotinginthe2014LokSabhaElections."EconomicandPoliticalWeekly49(39):72-76.

Srinivas,MysoreNarasimhachar,andA.M.Shah.2007.Thegrassrootsofdemocracy:fieldstudiesofIndianelections.Delhi:PermanentBlack:DistributedbyOrientLongman.

Srivastava,Saraswati.1976."UttarPradesh.Politicsofneglecteddevelopment."InStatePoliticsinIndia,editedbyIqbalNarain,323-369.Meerut:MeenakshiPrakashan.

Statistics,Special.2012."SixteenthAssemblyElectionsinUttarPradesh."Economic&PoliticalWeekly47(14):80-86.

Still,Clarinda.2014.DalitsinNeoliberalIndia:mobilityormarginalisation,ExploringthepoliticalinSouthAsia.NewDelhi:Routledge.

357

Stokes,Eric.1975."AgrarianSocietyandthePaxBritannicainNorthernIndiaintheEarlyNineteenthCentury."ModernAsianStudies9(4):505-528.

Stone,BrewerS.1988."InstitutionalDecayandtheTraditionalizationofPolitics:TheUttarPradeshCongressParty."AsianSurvey28(10):1018-1030.doi:10.2307/2644704.

Subas,Ram.1984."DynamicsofAgrarianRelationsinSultanpurEastUttarPradesh."SocialScientist12(7):57-63.doi:10.2307/3517059.

Subramanian,Narendra.1999.Ethnicityandpopulistmobilization:politicalparties,citizens,anddemocracyinSouthIndia.Delhi:OxfordUniversityPress.

Suresh,AppuEsthose.2014a."Expressinvestigationpart-I:Over600‘communalincidents’inUPsinceLSresults,60%nearbypollseats."IndianExpress,August9.AccessedApril17,2016.

Suresh,AppuEsthose.2014b."Expressinvestigationpart-II:Myloudspeakerversusyourloudspeaker."IndianExpress,August9.AccessedApril17,2016.

Suryanarayana,M.H.2009."intra-stateeconomicdisparities:KarnatakaandMaharashtra."EconomicandPoliticalWeekly44(26/27):215-223.

TawaLama-Rewal,Stéphanie.2009."StudyingElectionsinIndia-ScientificandPoliticaldebates."SAMAJ(3).

Thachil,Tariq.2014.Eliteparties,poorvoters:howSocialServiceswinvotesinIndia,Cambridgestudiesincomparativepolitics.NewYork:CambridgeUniversityPress.

Thelma,Paris,SinghAbha,LuisJoyce,andMahabubHossain.2005."LabourOutmigration,LivelihoodofRiceFarmingHouseholdsandWomenLeftBehind:ACaseStudyinEasternUttarPradesh."EconomicandPoliticalWeekly40(25):2522-2529.

Thorat,Sukhadeo,andAmareshDubey.2012."HasGrowthBeenSociallyInclusiveduring1993-94–2009-10?"EconomicandPoliticalWeekly42(10):43-54.

TNN.2006."Gangwarsragestill,inwestUP."TimesofIndia,August18.Accessed9August,2016.

Tripathi,PurnimaS.2003."L'AffaireAmarmani."Frontline,11-24October.Uppal,Yogesh.2009."TheDisadvantagedIncumbents:EstimatingIncumbencyEffects

inIndianStateLegislatures."PublicChoice138(1-2):9-27.doi:http://link.springer.com/journal/volumesAndIssues/11127.

Uppal,Yogesh.2011."Doeslegislativeturnoveradverselyaffectstateexpenditurepolicy?EvidencefromIndianstateelections."PublicChoice147(1/2):189-207.

Uppal,Yogesh,andThushyanthanBaskaran.2014.Vaishnav,Milan.2011.Vaishnav,Milan.2012."TheMeritsofMoneyand“Muscle”:EssaysonCriminality,

ElectionsandDemocracyinIndia."Ph.D,GraduateSchoolofArtsandScience,ColumbiaUniversity.

VanDyke,Virginia.1997."GeneralElections,1996:PoliticalSadhusandLimitstoReligiousMobilisationinNorthIndia."EconomicandPoliticalWeekly32(49):3149-3158.

Varshney,Ashutosh.1995.Democracy,development,andthecountryside:urban-ruralstrugglesinIndia,Cambridgestudiesincomparativepolitics.CambridgeEngland;NewYork,NY,USA:CambridgeUniversityPress.

Varshney,Ashutosh.1996.Hindu-Muslimriots1960-93newfindings,possibleremedies,RGICSproject.NewDelhi:RajivGandhiInstituteforContemporaryStudies.microform.

Varshney,Ashutosh.2000a."IsIndiaBecomingMoreDemocratic?"TheJournalofAsianStudies59(1):3-25.

358

Varshney,Ashutosh.2000b."WhyHavePoorDemocraciesNotEliminatedPoverty?ASuggestion."AsianSurvey40(5):718-736.

Varshney,Ashutosh.2001."EthnicConflictandCivilSociety:IndiaandBeyond."WorldPolitics53(3):362-398.

Varshney,Ashutosh,andStevenWilkinson.2006.Varshney-WilkinsonDatasetonHindu-MuslimViolenceinIndia,1950-1995,Version2.Inter-universityConsortiumforPoliticalandSocialResearch(ICPSR)[distributor].

Verma,A.K.2002a."PoliticalProspectsinUPPointersfromaPre-PollSurvey."EconomicandPoliticalWeekly(February9):525-527.

Verma,A.K.2002b."UPAssemblyElections:CasteDominatesIdeology."Economic&PoliticalWeekly(May25):1975-1980.

Verma,A.K.2003."ShiftingPoliticalEquationsinUP."Economic&PoliticalWeekly(June7):2244-2246.

Verma,A.K.2004a."SamajwadiPartyinUttarPradesh."EconomicandPoliticalWeekly39(14/15):1509-1514.

Verma,A.K.2004b."UttarPradesh:CasteandPoliticalMobilisation."EconomicandPoliticalWeekly39(51):5463-5466.

Verma,A.K.2005a."BSP’sStrategyinUttarPradesh.WooingtheBrahminsforaNewAlliance."EconomicandPoliticalWeekly(June25):2647-2648.

Verma,A.K.2006."UttarPradesh.FormationofMuslimPoliticalFronts."EconomicandPoliticalWeekly(October7):4241-4243.

Verma,A.K.2007a."Mayawati’sSandwichCoalition."EconomicandPoliticalWeekly(June2):2039-2043.

Verma,A.K.2007b."ReverseSocialOsmosisinUttarPradesh."EconomicandPoliticalWeekly(March10):817-820.

Verma,A.K.2012a."MuslimResurgenceinUrbanLocalBodiesofUttarPradesh."Economic&PoliticalWeekly47(40):30-31.

Verma,A.K.2012b."WhydidMayawatiLose?"EconomicandPoliticalWeekly47(18):17-19.

Verma,A.K.2014a."UttarPradesh-CommunalPolarisationvsCasteCalculus."Economic&PoliticalWeekly49(18).

Verma,A.K.2014b."DevelopmentandGovernanceTrumpCasteIdentitiesinUttarPradesh."EconomicandPoliticalWeekly49(39):89-94.

Verma,Arvind.2005b."PolicingElectionsinIndia."IndiaReview4(3-4):354-376.doi:10.1080/14736480500302217.

Verniers,Gilles.2015.HowtheGrandAlliancesweptBihar:18chartsthatexplaintheelectoralverdict.Scroll.in.AccessedApril16,2016.

Veyne,Paul.1992.Breadandcircuses:historicalsociologyandpoliticalpluralism.London:Penguin.

Vora,Rajendra,andSuhasPalshikar.2004.Indiandemocracy:meaningsandpractices.NewDelhi;ThousandOaks:SagePublications.

Wallace,Paul,andRamashrayRoy.2003.India's1999electionsand20thcenturypolitics.NewDelhi;ThousandOaks,CA:SagePublications.

Weiner,Myron.1967.Partybuildinginanewnation;theIndianNationalCongress.Chicago:UniversityofChicagoPress.

Weiner,Myron.1978.Indiaatthepolls:theparliamentaryelectionsof1977,AEIstudies202.Washington:AmericanEnterpriseInstituteforPublicPolicyResearch.

Weiner,Myron.1997."StruggleforEquality."IndiaatFifty,PrincetonUniversity,31October-1November.

359

Weisberg,HerbertF.2002."PartisanshipandIncumbencyinPresidentialElections."PoliticalBehavior24(4):339-360.doi:10.2307/1558378.

Wilkinson,Steven.2006.Votesandviolence:electoralcompetitionandethnicriotsinIndia,Cambridgestudiesincomparativepolitics.Cambridge;NewYork:CambridgeUniversityPress.

Wilkinson,StevenI.2004."TheElectoralIncentivesforEthnicViolence."InVotesandViolence:ElectoralCompetitionandEthnicRiotsinIndia.WestNyack,NY:CambridgeUniversityPress.

Wilkinson,StevenI.2005."CommunalRiotsinIndia."EconomicandPoliticalWeekly40(44/45):4768-4770.

Wilkinson,StevenI.2013."ElectoralCompetition,theState,andCommunalViolence:AReply."IndiaReview12(2):92-107.doi:10.1080/14736489.2013.788316.

Wilkinson,StevenI.2015."Where’stheParty?TheDeclineofPartyInstitutionalizationandWhat(ifAnything)thatMeansforDemocracy."Government&Opposition50(3):420.

Witsoe,Jeffrey.2009."TerritorialDemocracy:Caste,DominanceandElectoralPracticeinPostcolonialIndia."PoLAR:PoliticalandLegalAnthropologyReview32(1):64-83.doi:10.1111/j.1555-2934.2009.01024.x.

Witsoe,Jeffrey.2012."EverydayCorruptionandthePoliticalMediationoftheIndianState

AnEthnographicExplorationofBrokersinBihar."EconomicandPoliticalWeekly42(6):47-54.

Witsoe,Jeffrey.2013.Democracyagainstdevelopment:lower-castepoliticsandpoliticalmodernityinpostcolonialIndia,SouthAsiaacrossthedisciplines.Chicago;London:TheUniversityofChicagoPress.

Wright,Joseph.2008."PoliticalCompetitionandDemocraticStabilityinNewDemocracies."BritishJournalofPoliticalScience38(2):221-245.

WrightJr.,TheodoreP.1964."MuslimLegislatorsinIndia:ProfileofaMinorityÉlite."TheJournalofAsianStudies23(2):253-267.

WrightJr.,TheodoreP.1966."MuslimEducationinIndiaattheCrossroads-TheCaseofAligarh."PacificAffairs39(1/2):50-63.

Wyatt,Andrew.2013."PopulismandpoliticsincontemporaryTamilNadu."ContemporarySouthAsia21(4):365-381.doi:10.1080/09584935.2013.803036.

Yadav,Yogendra.1996a."TheMaturingofIndianDemocracy."IndiaToday,31August.Yadav,Yogendra.1996b."ReconfigurationinIndianPolitics:StateAssemblyElections,

1993-95."EconomicandPoliticalWeekly31(2/3):95-104.Yadav,Yogendra.2000."UnderstandingtheSecondDemocraticUpsurge."In

TransformingIndia:socialandpoliticaldynamicsofdemocracy,editedbyFrancineFrankel,ZoyaHasan,RajeevBhargavaandBalveerArora,120-145.NewDelhi:OxfordUniversityPress.

Zavos,John.2000.TheemergenceofHindunationalisminIndia.NewDelhi;NewYork:OxfordUniversityPress.

Zavos,John,AndrewWyatt,andVernonMarstonHewitt.2004.ThepoliticsofculturalmobilizationinIndia.NewDelhi;NewYork:OxfordUniversityPress.

Zerinini,Jasmine.2009."TheMarginalisationoftheSavarnasinUttarPradesh?"InRiseofthePlebeians?TheChangingFaceofIndianLegislativeAssemblies,27-64.NewDelhi:Routledge.

Zerinini-Brotel,Jasmine.1998."TheBJPinUttarPradesh:FromHindutvatoConsensualPolitics?"InTheBJPandtheCompulsionsofPoliticsinIndia.,editedbyThomas

360

BlomHansenandChristopheJaffrelot,72-100.Oxford,UK:OxfordUniversityPress.

Ziegfeld,Adam.2016.Whyregionalparties?:clientelism,elites,andtheIndianpartysystem.NewYork,NY:CambridgeUniversityPress.

ANNEXURES

361

Annexure1:TenureofUttarPradeshChiefMinisters(1950–Present)

Assem

bly

Notyetcreated

1stA

ssem

bly(195

2–57

)

2ndAssembly(195

7–62

)

3rdAssembly(196

2–67

)

4thAssembly(196

7–68

)

Dissolved

5thAssembly(196

9–74

)

Dissolved

6thAssembly(197

4–77

)

Dissolved

7thAssembly(197

7–80

)

Dissolved

8thAssembly(198

0–85

)

Tenu

releng

th

4years,335

days

5years,344

days

2years,298

days

3years,162

days

19days

328da

ys

1year,1day

356da

ys

225da

ys

17days

167da

ys

2years,69da

ys

148da

ys

2years,21da

ys

52days

1year,99da

ys

54days

1year,249

days

354da

ys

113da

ys

2years,39da

ys

2years,14da

ys

Term

ofo

ffice 20-M

ay-52

27-Dec-54

09-Apr-57

06-Dec-60

14-M

ar-62

01-Oct-63

13-M

ar-67

02-Apr-67

25-Feb

-68

26-Feb

-69

17-Feb

-70

01-Oct-70

18-Oct-70

03-Apr-71

12-Ju

n-73

08-Nov-73

04-M

ar-74

29-Nov-75

21-Ja

n-76

30-Apr-77

23-Ju

n-77

27-Feb

-79

17-Feb

-80

09-Ju

n-80

18-Ju

l-82

02-Aug-84

26-Ja

n-50

20-M

ay-52

28-Dec-54

10-Apr-57

07-Dec-60

14-M

ar-62

02-Oct-63

14-M

ar-67

03-Apr-67

25-Feb

-68

26-Feb

-69

18-Feb

-70

01-Oct-70

18-Oct-70

04-Apr-71

13-Ju

n-73

08-Nov-73

05-M

ar-74

30-Nov-75

21-Ja

n-76

30-Apr-77

23-Ju

n-77

28-Feb

-79

17-Feb

-80

09-Ju

n-80

19-Ju

l-82

Party

INC

INC

INC

INC

BKD

INC

BKD

INC

INC

INC

INC

INC

JP

JP

INC

Constituen

cy

BarreillyM

unicipality

Varana

siCitySou

th

RanikhetSou

th

Men

dhwal

Ranikhet

Chap

rauli

Ranikhet

Chap

rauli

Chan

dauli

Bara

Kashipur

Azam

garh(LokSab

ha)

Hapu

r

Tind

war

Nam

e

GovindBa

llabh

Pan

t

Sampu

rnan

and

Chan

draBh

anuGup

ta

SuchetaKriplani

Chan

draBh

anuGup

ta

Charan

Singh

Presiden

t'sru

le

Chan

draBh

anuGup

ta

Charan

Singh

Presiden

t'sru

le

Tribhu

vana

Narayan

aSingh

Kamlapa

tiTripathi

Presiden

t'sru

le

HemwatiN

anda

nBa

huguna

Presiden

t'sru

le

N.D

.Tiwari

Presiden

t'sru

le

RamNareshYada

v

Bana

rsiD

as

Presiden

t'sru

le

V.P.Singh

SripatiM

ishra

No 1

2

3

4 3 5

3 5

6 7

8

9

10

11

12

13

362

Annexure1:TenureofUttarPradeshChiefMinisters(1950–Present)(continued) A

ssem

bly

9thAssembly(198

5–89

)

10thAssem

bly(198

9–91

)

11thAssem

bly(199

1–92

)

Dissolved

12thAssem

bly(199

3–95

)

Dissolved

13thAssem

bly(199

6–20

02)

14thAssem

bly(200

2–07

)

15thAssem

bly(200

7–12

)

16thAssem

bly(201

2–17

)

Source:G

overnm

ento

fUttarPrade

shofficialweb

site:http://up

.gov.in

/upe

xcms.aspx.Lasta

ccessedon

March9,201

6.

*MLC

:Mem

bero

fthe

Leg

isla

tive

Coun

cil,

Utta

rPra

desh

’sUp

perH

ouse

.

Tenu

releng

th

1year,52da

ys

2years,274

days

1year,163

days

1year,201

days

1year,165

days

363da

ys

1year,181

days

137da

ys

1year,154

days

184da

ys

2years,52da

ys

351da

ys

1year,131

days

56days

1year,118

days

3years,257

days

4years,307

days

3years,344

days

Term

ofo

ffice

10-M

ar-85

24-Sep

-85

24-Ju

n-88

05-Dec-89

24-Ju

n-91

06-Dec-92

04-Dec-93

03-Ju

n-95

18-Oct-95

17-Oct-96

21-M

ar-97

21-Sep

-97

12-Nov-99

28-Oct-00

08-M

ar-02

03-M

ay-02

29-Aug-03

13-M

ay-07

15-M

ar-12

Incumbe

nt

03-Aug-84

11-M

ar-85

24-Sep

-85

25-Ju

n-88

05-Dec-89

24-Ju

n-91

06-Dec-92

04-Dec-93

03-Ju

n-95

18-Oct-95

17-Oct-96

21-M

ar-97

21-Sep

-97

12-Nov-99

28-Oct-00

08-M

ar-02

03-M

ay-02

29-Aug-03

13-M

ay-07

15-M

ar-12

Party

INC

INC

INC

JD

BJP

SP

BSP

BSP

BJP

BJP

BJP

BSP

SP

BSP

SP

Constituen

cy

Kashipur

Paniara

Jasw

antnagar

Atrauli

Jasw

antnagar

MLC*

Harora

Atrauli

Haidargarh

Harora

Gun

naur

MLC

MLC

Nam

e

N.D

.Tiwari

VirB

ahad

urSingh

N.D

.Tiwari

Mulayam

SinghYad

av

Kalyan

Singh

Presiden

t'sru

le

Mulayam

SinghYad

av

Mayaw

ati

Presiden

t'sru

le

Presiden

t'sru

le

Kalyan

Singh

RamPrakashGup

ta

RajnathSingh

Presiden

t'sru

le

Mayaw

ati

Mulayam

SinghYad

av

Mayaw

ati

AkhileshYad

av

No

9

14

9 15

16

15

17

16

18

19

17

15

17

20

363

Annexure2:CoalitiongovernmentsinUttarPradesh(1977-Present)No Coalition ChiefMinister Tenure MainParty’s

SingleMajority

1 JNP RamNareshYadav 23.06.1977–27.02.1979 Yes

2 JNP BanarasiDas 28.02.1979–17.02.1980 Yes

3 SP-BSP MulayamSinghYadav

04.12.1993–03.06.1995 No

4 BSP-BJP Mayawati 03.06.1995–17.10.1995 No



7 SP+* MulayamSinghYadav

29.08.2003–12.05.2007 No

Source: Government of Uttar Pradesh official website: http://up.gov.in/upexcms.aspx. LastaccessedonMarch9,2016.*TheSamajwadiParty formed thegovernment in2003with143seats,with thesupportofCongress(16seats),theRashtriyaLokDal(14seats),theRashtriyaKrantiParty(2seats),theCPI-M (2 seats), smaller parties and Independents (19) and 13 defectors from the BahujanSamajParty,foratotalof209seats.

364

Annexure3:Representationofcastesandcommunitiesinmainparties,1989-1996(%

)

1996

SP

23.15%

1.85

%

0.93

%

2.78

%

- -

15.74%

0.93

%

0.93

%

-

0.93

%

0.93

%

37.04%

- - - - - - -

0.93

%

0.93

%

- - - -

BSP

14.93%

-

1.49

%

4.48

%

1.49

%

-

5.97

%

1.49

%

- -

1.49

%

1.49

%

37.31%

1.49

%

- - -

1.49

%

1.49

%

- - - - - - -

BJP

48.04%

3.91

%

1.68

%

15.08%

1.68

%

2.23

%

21.79%

0.56

%

1.12

%

-

2.23

%

2.23

%

22.35%

- -

1.12

%

- -

0.56

%

-

2.23

%

-

0.56

%

- - -

INC

65.63%

3.13

%

-

25.00%

- -

34.38%

3.13

%

- -

6.25

%

6.25

%

12.50%

- - - - - - -

3.13

%

- - - - -

1993

JD

22.58%

- -

3.23

%

- -

19.35%

- - -

16.13%

16.13%

29.03%

- - - - - - -

6.45

%

- - - - -

BSP

4.48

%

-

1.49

%

1.49

%

- -

1.49

%

- - - - -

43.28%

- - - - - - - - - - - - -

BJP

52.78%

5.00

%

0.56

%

15.56%

1.67

%

1.67

%

25.56%

0.56

%

1.11

%

1.11

%

3.33

%

3.33

%

20.56%

- - - - - - -

1.67

%

0.56

%

0.56

%

0.56

%

- -

INC

55.56%

3.70

%

-

25.93%

- -

25.93%

- - -

3.70

%

3.70

%

14.81%

- - - - - - -

11.11%

- - - - -

1991

JD

23.33%

1.11

%

1.11

%

1.11

%

1.11

%

-

18.89%

- - -

10.00%

10.00%

32.22%

- -

1.11

%

- -

1.11

%

-

5.56

%

- - - - -

BSP

8.33

%

-

8.33

%

- - - - - - - - -

83.33%

- - - - - - - - - - - - -

BJP

51.57%

3.59

%

0.90

%

20.18%

0.90

%

1.35

%

21.08%

0.45

%

2.24

%

0.90

%

1.79

%

1.79

%

19.28%

- -

0.45

%

- - - -

0.45

%

-

0.45

%

0.45

%

- -

INC

48.00%

2.00

%

-

26.00%

- -

18.00%

2.00

%

- - - -

16.00%

- - - - - - -

2.00

%

- - - - -

1989

JD

36.63%

-

1.49

%

12.87%

1.98

%

0.50

%

19.31%

0.50

%

- -

6.93

%

6.93

%

27.72%

- -

0.50

%

- - - -

4.46

%

- - - - -

BSP

- - - - - - - - - - - -

21.43%

- - - - - - - - - - - - -

BJP

56.14%

12.28%

1.75

%

19.30%

1.75

%

1.75

%

17.54%

-

1.75

%

- - -

19.30%

- -

1.75

%

- - - - - -

1.75

%

1.75

%

- -

INC

41.18%

-

0.98

%

19.61%

- -

18.63%

-

0.98

%

0.98

%

0.98

%

0.98

%

16.67%

- - - - - - -

2.94

%

- - - - -

UC

Ba

nia/Jain

Bh

umihar

Brah

min

Kayastha

Kh

atri

Ra

jput

Tyagi

Va

ishya

Others

IC

Jat

OBC

Ba

ghel

Bishno

i

Ch

auha

n

Ch

aurasia

Gad

ariya

Gared

ia

Goswam

i

Gujjar

Jaisw

al

Kapu

r

Kashyap

Katiy

ar

Ko

eri

365


)(continued).

1996

SP

-

4.63

%

- -

0.93

%

- -

0.93

%

-

0.93

%

0.93

%

- -

25.93%

0.93

%

17.59%

- -

0.93

%

- - - -

2.78

%

- - - -

BSP

1.49

%

8.96

%

1.49

%

4.48

%

-

1.49

%

1.49

%

4.48

%

2.99

%

-

1.49

%

- -

1.49

%

2.99

%

29.85

% - - -

1.49

%

- - -13

.43

% -

1.49

%

- -

BJP

-

3.35

%

0.56

%

4.47

%

-

0.56

%

-

0.56

%

0.56

%

0.56

%

0.56

%

- -

4.47

%

2.23

%

21.23%

- - - - - - -

3.91

%

-

1.68

%

- -

INC - - - - - -

3.13

%

- - - - - -

6.25

%

-

9.38

%

- - - - - - - - - - - -

1993

JD

-

9.68

%

- - - - - - - - - - -

9.68

%

3.23

%

6.45

%

- - - - - - - - - - - -

BSP

-

4.48

%

2.99

%

5.97

%

- - -

1.49

%

2.99

%

- - - -

11.94%

13.43%

34.33%

- - - - - - -

7.46

%

-

1.49

%

- -

BJP

-

5.56

%

-

5.00

%

- - - - -

0.56

%

0.56

%

0.56

%

-

1.11

%

3.89

%

20.00%

- - - - - - -

5.56

%

-

0.56

%

- -

INC -

3.70

%

- - - - - - - - - - - - -

22.22%

3.70

%

-

3.70

%

- - - -

3.70

%

- - - -

1991

JD

-

6.67

%

- - -

1.11

%

- - - - - - -

13.33%

3.33

%

22.22%

-

1.11

%

- - - - -

5.56

%

- - - -

BSP

8.33

%

16.67%

-

8.33

%

- - -

8.33

%

8.33

%

- - - -

16.67%

16.67%

- - - - - - - - - - - - -

BJP

-

4.48

%

-

2.69

%

- - -

0.45

%

0.45

%

0.45

%

0.45

%

0.90

%

-

3.59

%

4.04

%

25.11%

- - - - - - -

5.83

%

-

2.69

%

0.45

%

-

INC -

2.00

%

2.00

%

- - - -

2.00

%

- - - - -

6.00

%

2.00

%

16.00%

- - - - - - -

4.00

%

-

2.00

%

- -

1989

JD

-

5.45

%

0.50

%

0.99

%

0.50

%

- - - - -

0.50

%

- -

11.88%

2.97

%

24.75

%

0.99

%

- - - - -

0.50

%

8.42

%

-

0.50

%

- -

BSP

-

7.14

%

7.14

%

- - - - - - - - - -

7.14

%

-

35.71%

- - - - - - - - - - - -

BJP

-

5.26

%

-

5.26

%

- - - - - - - - -

1.75

%

1.75

%

22.81%

- - - - - - -

7.02

%

-

3.51

%

- -

INC -

1.96

%

-

0.98

%

0.98

%

- - - -

0.98

%

- - -

8.82

%

-

18.63%

0.98

%

-

0.98

%

- - - -

1.96

%

-

1.96

%

-

0.98

%

Ku

mha

r

Ku

rmi

Ku

shwah

a

Lodh

i

Mallah

Mau

rya

Murao

n

Nish

ad

Ra

jbha

r

Saini

Shakya

Soni

Teli

Yada

v

Unide

ntified

SC

Arya

Ba

iragi

Ba

rwar

Ba

udh

Be

lldara

Chau

han

Be

ria

Jatav

Dh

anuk

Dh

obi

Dh

rikar

Dh

usia

366


)(continued)

1996

SP

- - - - -

1.85

%

- - - -

3.70

%

-

0.93

%

-

7.41

%

0.93

%

18.52%

- -

1.85

%

100

N=1

08

Source:A

utho

r’sfieldw

ork

BSP

- - - - -

1.49

%

-

1.49

%

- -

2.99

%

- - -

7.46

%

-

16.42%

- - - 100

N=6

7

BJP

- - - -

0.56

%

2.23

%

0.56

%

2.23

%

- -

0.56

%

0.56

%

- -

8.94

%

-

2.79

%

-

1.12

%

2.23

%

100

N=1

79

INC - - - - - - -

3.13

%

- -

3.13

%

- - -

3.13

%

-

6.25

%

- - - 100

N=3

2

1993

JD

- - - - -

3.23

%

- - - - - - - -

3.23

%

-

25.81%

- - - 100

N=3

1

BSP

- - - - - - - - - -

1.49

%

- - -

23.88%

1.49

%

16.42%

- - - 100

N=6

7

BJP

-

0.56

%

- - -

1.11

%

-

1.11

%

0.56

%

-

1.11

%

-

0.56

%

0.56

%

8.33

%

-

2.22

%

-

0.56

%

0.56

%

100

N=1

80

INC - - - - - - -

3.70

%

- -

3.70

%

- - -

3.70

%

3.70

%

- - - - 100

N=2

7

1991

JD

- - -

1.11

%

-

1.11

%

- -

1.11

%

-

1.11

%

-

1.11

%

-

10.00%

1.11

%

11.11%

- - - 100

N=9

0

BSP

- - - - - - - - - - - - - - - -

8.33

%

- - - 100

N=1

2

BJP

0.45

%

0.45

%

- - -

1.35

%

-

0.90

%

0.45

%

0.45

%

1.35

%

0.45

%

0.45

%

0.45

%

9.42

%

-

0.90

%

-

0.45

%

0.90

%

100

N=2

23

INC - - - - - - - - - -

10.00%

- - - - -

16.00%

- -

4.00

%

100

N=5

0

1989

JD

- - - - -

0.99

%

- -

0.50

%

-

3.47

%

-

0.99

%

-

8.42

%

-

3.47

%

- -

0.50

%

100

N=2

02

BSP

- - - - - - - - - - - - - -

35.71%

-

42.86%

- - - 100

N=1

4

BJP

- - - - - - - - -

1.75

%

- - -

1.75

%

8.77

%

- - -

1.75

%

- 100

N=5

7

INC - - - - - - - - - -

5.88

%

0.98

%

- -

4.90

%

0.98

%

18.63%

- -

2.94

%

100

N=1

02

Gon

d

Gou

d

Katheria

Katik

Katoriy

a

Kh

atik

Ko

l

Ko

ri

Ku

reel

Lonia

Pasi

Shilpkar

Va

lmiki

Visvar

Unide

ntified

ST

Muslim

s

Christian

Sikh

Unide

ntified

Total

367


)

2012

SP

26.79%

2.23

%

0.45

%

9.82

%

0.45

%

0.45

%

12.95%

-

0.45

%

- - -27

.23%

- - -

0.45

%

0.45

%

- - -

0.45

%

- - - -

BSP

20.48%

1.20

%

-

12.05%

- -7.23

%

- - -

1.20

%

1.20

%

33.73%

- - - -2.41

%

- -3.61

%

- - - - -

BJP

55.10%

10.20%

2.04

%

20.41%

2.04

%

4.08

%

16.33%

- - -

8.16

%

8.16

%

22.45%

-

2.04

%

- - - - -4.08

%

2.04

%

- - - -

INC

48.39%

-

3.23

%

12.90%

6.45

%

6.45

%

19.35%

- - -

3.23

%

3.23

%

9.68

%

- - - - - - -3.23

%

3.23

%

- - - -

2007

SP

29.47%

4.21

%

1.05

%

10.53%

- -13

.68%

- - -

1.05

%

1.05

%

36.84%

- - -

1.05

%

- - - -

1.05

%

-1.05

%

- -

BSP

27.40%

0.96

%

0.48

%

16.83%

0.48

%

-8.65

%

- - -

2.40

%

2.40

%

24.04%

- -0.48

%

-0.96

%

-

0.48

%

2.40

%

0.48

%

- -

0.48

%

-

BJP

58.82%

13.73%

3.92

%

13.73%

3.92

%

5.88

%

17.65%

- - -

3.92

%

3.92

%

23.53%

-

1.96

%

1.96

%

- - - -3.92

%

- - - - -

INC

58.33%

4.17

%

-

12.50%

4.17

%

8.33

%

29.17%

- - -

4.17

%

4.17

%

8.33

%

- - - - - - - - - - - - -

2002

SP

26.81%

1.45

%

2.17

%

8.70

%

- -13

.04%

-

1.45

%

-

0.72

%

0.72

%

33.33%

- - -

0.72

%

0.72

%

- - -

0.72

%

-0.72

%

- -

BSP

24.74%

2.06

%

2.06

%

6.19

%

- -14

.43%

- - -

1.03

%

1.03

%

34.02%

1.03

%

- - -1.03

%

1.03

%

-1.03

%

- -1.03

%

-1.03

%

BJP

61.80%

5.62

%

3.37

%

15.73%

2.25

%

3.37

%

30.34%

-

1.12

%

-

1.12

%

1.12

%

16.85%

-

1.12

%

2.25

%

- - - -2.25

%

- - - - -

INC

50.00%

6.67

%

-

13.33%

3.33

%

3.33

%

20.00%

3.33

%

- -

3.33

%

3.33

%

10.00%

- - - - - - -3.33

%

- - - - -

UC

Ba

nia/Jain

Bh

umihar

Brah

min

Ka

yastha

Kh

atri

Ra

jput

Tyagi

Va

ishya

Others

IC

Jat

OBC

Ba

ghel

Bishno

iCh

auha

n

Ch

aurasia

Gad

ariya

Gared

ia

Goswam

iGujjar

Jaisw

al

Ka

pur

Ka

shyap

Ka

tiyar

Ko

eri

368


)(continued)

2012

SP

0.45

%

3.13

%

0.45

%

3.13

%

0.45

%

- -0.89

%

- -0.89

%

- -16

.52%

-

25.89%

0.45

%

- - - -0.45

%

5.36

%

-0.89

%

- -

BSP

-7.23

%

4.82

%

4.82

%

-2.41

%

-2.41

%

-1.20

%

3.61

%

- -1.20

%

-19

.28%

- - - - - -16

.87%

- - - -

BJP

-2.04

%

2.04

%

4.08

%

-2.04

%

- - - - - - -4.08

%

-6.12

%

- - - - - -2.04

%

- - - -

INC - - -

3.23

%

- - - - - - - - - - -12

.90%

- - - - - -6.45

%

-3.23

%

- -

2007

SP

-5.26

%

1.05

%

4.21

%

1.05

%

- -1.05

%

- -2.11

%

- -18

.95%

-

12.63%

- -1.05

%

-1.05

%

-2.11

%

2.11

%

- - -

BSP

-5.77

%

1.92

%

0.96

%

-1.44

%

-0.96

%

1.44

%

0.48

%

0.96

%

- -4.81

%

-29

.81%

- - - - - -21

.15%

-

1.44

%

- -

BJP

-7.84

%

-3.92

%

- - - - - - - - -3.92

%

-13

.73%

- - - - - -3.92

%

-3.92

%

- -

INC -

4.17

%

-4.17

%

- - - - - - - - - - -20

.83%

- - - - - -4.17

%

-4.17

%

- -

2002

SP

-5.80

%

-0.72

%

- - -1.45

%

0.72

%

-1.45

%

-0.72

%

19.57%

-

26.09%

0.72

%

-0.72

%

- -0.72

%

10.14%

0.72

%

2.90

%

- -

BSP

1.03

%

5.15

%

2.06

%

3.09

%

-2.06

%

-2.06

%

3.09

%

2.06

%

1.03

%

- -6.19

%

-25

.77%

- - - - - -17

.53%

- - - -

BJP

-7.87

%

- - - - - - - - - - -3.37

%

-19

.10%

- - - - - -10

.11%

-

3.37

%

- -

INC - - - - - - - - - - - - -

6.67

%

-6.67

%

- - - - - - - -3.33

%

- -

Ku

mha

rKu

rmi

Ku

shwah

aLodh

iMallah

Mau

rya

Murao

nNish

ad

Ra

jbha

rSaini

Shakya

Soni

Teli

Yada

vUnide

ntified

SC

Arya

Ba

iragi

Ba

rwar

Ba

udh

Be

lldaraCh

auha

nBe

ria

Jatav

Dh

anuk

Dh

obi

Dh

rikar

Dh

usia

369


)(Continued)

2012

SP

0.45

%

-

0.45

%

0.89

%

-

1.34

%

0.45

%

3.13

%

-

0.45

%

8.93

%

-

1.79

%

-

0.89

%

-

19.20%

- -

0.89

%

100.00

%

N=2

24

Source

:Aut

hor’s

fiel

dwor

k.

BSP

- - - - - - - - - -

2.41

%

- - - - -

21.69%

- -

3.61

%

100.00

%

N=8

3

BJP

- - - - - - - - - -

2.04

%

-

2.04

%

- - -

4.08

%

- -

4.08

%

100.00

%

N=4

9

INC - - - - - - - - - - - -

3.23

%

- - -

22.58%

- -

3.23

%

100.00

%

N=3

1

2007

SP

- - - - -

1.05

%

- - - -

4.21

%

- - -

1.05

%

-

20.00%

- - -

100.00

%

N=9

5

BSP

- - - - - - - - - -

4.81

%

-

0.48

%

-

1.92

%

0.48

%

14.90%

- -

0.96

%

100.00

%

N=2

08

BJP

1.96

%

1.96

%

- - - - - - - -

1.96

%

- - - - - - - - -

100.00

%

N=5

1

INC - - - - -

4.17

%

- - - -

8.33

%

- - - - -

8.33

%

- - -

100.00

%

N=2

4

2002

SP

0.72

%

- - - -

2.17

%

-

0.72

%

0.72

%

-

5.07

%

-

0.72

%

- - -

13.04%

- - -

100.00

%

N=1

38

BSP

- - - - -

1.03

%

- - - -

5.15

%

- - -

2.06

%

1.03

%

13.40%

- - -

100.00

%

N=9

7

BJP

-

1.12

%

- - - - - - - -

3.37

%

- - -

1.12

%

-

1.12

%

- - -

100.00

%

N=8

9

INC - - - - -

3.33

%

- - - - - - - - - -

26.67%

3.33

%

- -

100.00

%

N=3

0

Gon

d

Gou

d

Katheria

Katik

Katoriy

a

Khatik

Kol

Kori

Kureel

Lonia

Pasi

Shilpkar

Valm

iki

Visvar

Unide

ntified

ST

Muslim

s

Christian

Sikh

Unide

ntified

Total

the localization of caste politics in uttar pradesh after .../2441/5gpts2mlf09rtau5...the...

Documents