from ‘glass ceilings’ to ‘firewalls’ — different … ‘glass ceilings’ to...

From ‘Glass Ceilings’ to ‘Firewalls’— Different Metaphors forDescribing Discriminationgwao_520 612..634

Regine Bendl* and Angelika Schmidt

The glass ceiling metaphor, a framework of the 1980s, constructs discrimi-nation processes in a particular way in particular organizational frame-works. Using a procedure of metaphor evaluation, we examine the glassceiling metaphor to determine whether it continues to be useful in con-temporary social and economic contexts. We analyse the recently intro-duced firewall metaphor for its usefulness for constructing discriminatoryprocesses in organizations, which remain hidden in the glass ceiling meta-phor. Our analysis suggests that both metaphors are useful for construct-ing diverse aspects of discrimination. In the contemporary context,however, firewalls may have greater utility due to their complexity, fluid-ity, heterogeneity and possibilities for permeability.

Keywords: gender, career, organizational paradigm, metaphor evaluation,heuristic value

Introduction

The metaphor of the glass ceiling (Federal Glass Ceiling Commission,1987) has been applied for more than two decades to explore organiza-

tional discriminative processes inhibiting the advancement of women andother discriminated groups into higher management jobs. According toMurrell and Hayes James (2001, p. 244):

most well-known illustrations of discrimination in the workplace are cap-tured by the concept of glass ceiling, which defines the invisible barrier thatprevents many women and minorities from advancing into senior andexecutive management positions within organizations.

Address for correspondence: *Institute of Gender and Diversity in Organizations, Augasse 9,A-1090 Vienna, Austria, e-mail: [email protected]

Gender, Work and Organization. Vol. 17 No. 5 September 2010

© 2010 Blackwell Publishing Ltd

In other words, the glass ceiling metaphor is considered to be a device forcapturing and explaining the symbolic dimensions of discrimination againstwomen and other groups in organizations (for example, Altman et al., 2005;Belle et al., 2001; Cortina, 2008; Davidson and Cooper, 1992; Dreher, 2003;Insch et al., 2008; Morrison et al., 1987; Ryan and Haslam, 2005; Stroh et al.,2004). However, Meyerson and Fletcher (2000, p. 136) believe that

it is time for new metaphors to capture the subtle, systemic forms ofdiscrimination that still linger. It’s not the ceiling that’s holding womenback; it’s the whole structure of the organizations in which we work: thefoundation, the beams, the walls, the very air. The barriers to advancementare not just above women, they are all around them.

Also Altman et al. (2005, p. 76) state that current conceptualizations of theglass ceiling as an invisible, single barrier fail to capture the complexity of thesituation.

In fact, the social, demographic, legal and economic contexts have changedsince the 1980s. Drivers of change like globalization, virtualization, demo-graphic developments and value changes have led to new forms of organi-zations and organizing as well as to new forms of private and occupationallife concepts of individuals (e.g., Hooker et al., 2007; Ruigrok et al., 1999;Sennett, 1998). Also shifts in the meaning and understanding of career andadvancement have occurred. The upcoming forms of career have been givenpartly flashy labels such as boundaryless (Arthur and Rousseau, 1996; Elligand Thatchenkery, 1996; Nicholson, 1996), protean (Hall, 1996), nomad(Cadin et al., 2000), chaotic (Peterson and Anand, 2002; see also Gunz et al.,2002), spiral (Brousseau et al., 1996), post-corporate (Peiperl and Baruch, 1997)or chronically flexible (Iellatchitch et al., 2003). These forms have occurred,since stable career structures, which represent a modern understanding oforganizations (specialization, bureaucracy, hierarchy, inflexibility, see, Clegg,1992; Willmott, 1993) and produce and stabilize hierarchies, fit into neithertoday’s rapidly spreading horizontal and co-operation-oriented organiza-tional relationships, nor the current trend for extremely fast job rotation.Flexibility prevails in a neoliberal and postmodern way of arranging organi-zations (Alvesson and Billing, 1997; Clegg, 1992; Cummings, 1996; Knights,1997). Although both perspectives are present in today’s organizations, thetrend towards postmodern organizations has triggered a redefinition ofcareer and advancement and professional careers have become more diverse.The traditional model of a lifetime career starting with specific training in theearly stages of one’s career and leading to a quite stable career path in thesame profession or area of expertise, sometimes even in the same (kind of)organization for the rest of one’s life is currently being replaced by moreflexible types of careers that may lead individuals to different professions inor outside different organizations in different places all over the world.

FROM ‘GLASS CEILINGS’ TO ‘FIREWALLS’ 613

© 2010 Blackwell Publishing Ltd Volume 17 Number 5 September 2010

Protean or patchwork careers are just two of many examples (Hall andMirvis, 1996).

Despite all these changes, women are still disadvantaged in theworkplace and underrepresented in leadership positions (e.g., Adler, 1993;Catalyst, 2009; Davidson and Bruke, 2000; Ryan and Haslam, 2005, 2007),whereas men continue to advance into management positions. Even thoughrecent research reveals that women are beginning to break through theglass ceiling (e.g. Dreher, 2003; Ryan and Haslam, 2005; Van Vianen andFischer, 2002), evidence suggests that, once women attain these leadershiproles, their performance is often placed under close scrutiny (Ryan andHaslam, 2005).

However, apart from all these changes, the glass ceiling has been rein-forced as a metaphor for describing discrimination in organizations over theyears, but its heuristic value as a metaphor for describing discrimination hasbeen explored only marginally (see Belle et al., 2001; Catalyst, 1994, 2000;Morrison et al., 1987). Research has only indirectly referred to the metaphori-cal value of the glass ceiling and has not been a comprehensive account ofthe mechanisms and validity of the concept as a metaphor for the subjectthat it supposedly illuminates. Adhering to the perspective that metaphorsrepresent ‘conceptual windows which help the organizational analyst gainbetter access to rich avenues of meaning’ (Chia, 1996, p. 128) and as weassume that the glass ceiling metaphor also constructs discrimination pro-cesses against women and other groups in a particular way within particularmodern organizational frameworks, in this article we examine to whatextent and in what respect the glass ceiling metaphor still offers usefulinsights for describing and explaining discriminatory processes againstwomen and other groups in contemporary postmodern organizations (e.g.organizations as networks or virtual organizations: Daft, 2001; Sydow, 1999).We therefore pose the question: what additional perspectives does therecently introduced firewall metaphor offer for understanding the discrimi-natory processes in organizations that remain hidden by the glass ceilingmetaphor?

At this point, readers might ask ‘what problem does an evaluation of ametaphor solve?’ or ‘how does the introduction of a new metaphor help toovercome actual discrimination against women in organizations, since thefigures have not changed considerably over the years even though womenare better educated?’ (for example, Davidson and Cooper, 1992; Dreher, 2003;Morrison et al., 1987; Ryan and Haslam, 2005). We want to answer thosecritics by referring to Morgan (1996) and Weick (1997), who pointed out thatcertain metaphors play a heuristic role in organization theory by opening upnew ways and perspectives for understanding and enquiry. By drawing on apoststructuralist perspective that considers reality to be reproduced throughlanguage, we argue that our linguistic-oriented examination of the value of awell-established and a comparatively new metaphor offers a new impetus to

614 GENDER, WORK AND ORGANIZATION

Volume 17 Number 5 September 2010 © 2010 Blackwell Publishing Ltd

the discussion on the (re)production of discrimination in organizations byrevealing as yet undiscovered features and dynamics of such discrimination.Inkson (2006, p. 50) pointed out that metaphors are a powerful tool of rheto-ric and we need, in evaluating metaphors, to go beyond consideration ofconcrete testable propositions. This is especially true if we consider thatmetaphors used in science work by evoking associations among ‘speciallyconstructed systems of implications’ and the metaphor’s suitability forscientific purposes comes precisely from this ‘ability to be suggestive ofnew sets of implications, new hypotheses, and therefore new observations’(Cornelissen, 2002, p. 260).

In order to demonstrate whether the glass ceiling metaphor captures therecent shifts in organizational and career paradigms and what additionalperspectives the firewall metaphor captures, we apply Cornelissen’s (2004,2005, 2006a, 2006b) model of domain interaction for evaluating the heuristicstatus of the two metaphors. To begin with, we describe his model of meta-phor evaluation. Next, we apply this domain-interaction model to the glassceiling metaphor and then we proceed to the firewall metaphor in order toexamine its heuristic value. Finally, we discuss implications of our metaphorevaluation on research and practice.

Metaphor and organization theory: models ofmetaphor application

Since the ground-breaking publication of Gereth Morgan’s Images of Organi-zation (1996), the debate in organization studies has focused on the value andpotential of metaphor for explaining organizational phenomena (for example,Black, 1962, 1979; Bono, 1990; Cornelissen, 2002, 2004, 2005; Davidson, 1978;Hesse, 1995; Hunt and Menon, 1995; Inns, 2002; Montuschi, 1995; Morgan,1980; Oswick et al., 2002; Tourangeau and Sternberg, 1981; Tsoukas, 1991,1993 and many others). Metaphor is often merely regarded as a device forembellishing discourse but its significance is much greater. The use of ametaphor implies a way of thinking and a way of seeing that determines howwe basically understand our world (Morgan, 1996, p. 4). A metaphor usesimplicit and explicit assertions that A is (or is like) B. In other words, ‘meta-phor is principally a way of conceiving of one thing in terms of another, andits primary function is understanding’ (Lakoff and Johnson, 1980, p. 36).Therefore, a metaphor provides fresh and previously non-existent insightsinto reality of organizational life (Weick, 1997).

There have been several models for examining the heuristic value of meta-phors. In contrast to those models of metaphor evaluation, which refer only tothe similarities between the domains of the metaphor (the similarity model[Schön, 1965], the transformational model [Tsoukas 1991, 1993], the compari-son model [Katz, 1992; Shen, 1997] and the interaction model [Black, 1962,



1979]), Cornelissen (2004, 2005) has introduced the domain-interactionmodel, which embraces the analogies as well as the differences between thetwo terms or domains in question. As an ‘informed, disciplined, and guidedprocess’ (Cornelissen, 2004, p. 706), two stages focus on assessing whether ametaphor has any heuristic value for the concept that it is supposed tocapture. In the first place, the ‘aspectual similarity’ between the two domainshas to be set up by applying three constitutive principles (the development ofa generic structure, the development and elaboration of the blend and theemergent meaning structure). Then, a matrix of governing rules, which refersto the degree of similarity and dissimilarity between the transferred words,determines the heuristic potential of the metaphor.

The domain-interaction model: constitutive principles of metaphor

Cornelissen (2004) assumes that there is an initial inherent structure betweenthe domains conjoined in a metaphor:

In metaphor, concepts in different domains are seen as occupying analo-gous positions (that is having analogous features) within their respectivedomains, which, being often specific to a domain, must be transformed(that is seen as a new way) if we are to find correspondences acrossdomains. In other words, the correspondence or similarity that is effec-tively constructed involves the aspectual shape of the concepts within theirdomains. (Cornelissen, 2004, p. 710)

This aspectual similarity refers to the fact that the target domain and thesource domain are both characterized by a specific notion. In both domainsactivities are carried out and statements are made that rely on the samenotions that represent the basis for the evaluation of the metaphor. Table 1presents Cornelissen’s three constitutive principles that are the basis for thisevaluation. After the development of a generic structure (step 1) and theelaboration of a blend (step 2) a new emergent structure (step 3) has to be setup that is translated back to the original concepts.

If this new perspective is indeed insightful or whether the metaphor hasthe heuristic value of providing theorists, researchers and practitioners witha fresh and previously non-existent understanding is based on the followingrules governing metaphors (Cornelissen, 2004, p. 713).

Domain-interaction model: the governing rules of the metaphor

As a metaphor consists of a similarity and a difference between the target andthe vehicle, Alvesson (1993, p. 116) points out that a good metaphor has theright mix of similarity and difference between the transferred word and thefocal one. Too much or too little similarity means that the point may not beunderstood and a successful metaphor will not have been created. The fit,



aptness or heuristic value of a metaphor depends on the conceived similarityand dissimilarity between concepts and their respective domains. As Table 2demonstrates, two dimensions (the exactness of the aspectual similaritybetween tenor and vehicle and the distance between the domains fromwith the concepts are drawn) are the basis for the governing rules thatfinally determine the heuristic value of the metaphor (Cornelissen, 2004,p. 717).

The first dimension refers to the exactness of the aspectual similaritybetween the conjoined concepts (or the aptness or meaningfulness of a meta-phor) and the second dimension represents the heuristic value of a metaphorindicated by the distance between the domains to which the conjoined con-cepts refer. Based on these two dimensions for evaluating the value of ametaphor the following two governing rules have to be applied (Cornelissen,

Table 1: Three constitutive principles of metaphor

Development of a generic structureThe terms of the metaphor are encoded, the relevant domains are inferred,the parallel structures are found and the correspondences between thesestructures are first mapped in a generic space. Metaphor processing works byconsidering the vehicle and topic on their (constructed) structural similarities,then making an inference about instance-specific relationships from this.Therefore, metaphor goes beyond analogy, which is about transferring onlythat information which is associated with the common structure of thedomains conjoined. In fact, the structure of the concepts in their respectivedomains and their aspectual shape and similarity provide the organizingframe for the further transfer and projection of specific information from theinput domain to the metaphoric domain.Development and elaboration of the blendThe constructed generic structure is transferred to a virtual ‘blended space’ or‘blend’. In this blend there is a new meaning which is neither the tenor northe vehicle. Thus, the emergent meaning cannot be reduced (and thereforeexplained) by referring to its constituent parts. Yet it is conceptuallyconnected to them and as such enables us to consider the way in which themeaning and image of the blend translate back to them and provide aninsightful perspective on the target subject.Emergent meaning structureThe emergent meaning structure that emerges from the blend and is linked tothe input concepts forces us to see a particular target subject and itsrespective domain in a new light. This means the image and meaning that inturn come from the blend that makes relations available that do not exist inthe separate tenor and vehicle concepts. The emergent meaning structurecontains a more specific structure not provided by the tenor or vehicle.

Source: Cornelissen, 2004, p. 710, 2005, p. 758.



2004, p. 719): (a) it is a necessary, though not sufficient, condition that in ametaphor the concepts are related by domains that are in the first instanceseen to be distant from one another; and (b) it is a necessary, though notsufficient, condition that a correspondence can be constructed between thetwo concepts conjoined in the metaphor.

However, Cornelissen (2004, p. 720) also points out that the notions ofexactness and distance are not only somewhat contextual and functional butare also discretionary because they depend on the relative perceptions ofthe comprehenders and their prior knowledge background and cognitiveabilities.

In this article, we apply the domain-interaction model since it offers a moreextensive evaluation of a metaphor than the aforementioned models thatfocus solely on the similarity between the domains. Thus, for our purposesthe domain-interaction model is the best suited model for shedding light onthe similarities and dissimilarities (distance) of the domains. This is importantbecause the glass ceiling metaphor has gained a prominent status in theliterature on women in or into management and the introduction of a newmetaphor represents a far-reaching step.

Table 2: Types of metaphor in organization theory

Dimension 1: Exactness of correspondence concepts

High/exact Low/inexact

Dis

tanc

eof

dom

anis Low Type 1 Type 2

Exact correspondence betweenthe concepts

Concepts from domains that areclose

Low distance No or less aspectual relationship orcorrespondence between them

Low heuristic value due to theits inability to offer any newand truly revealing insights

Produces weak and nonsensicalimagery

Dim

ensi

on2:

High Type 3 Type 4Distance of the domains Distance domains — hardly any

aspectual similarity between theconcepts and the figure in them

Provides conceptual advancesand clarifications

Inexact concepts

New insights and hence ahigh heuristic value

Nonsensical

Source: Cornelissen, 2004, p. 719.



Structural aspects: the evaluation of the‘glass ceiling’ metaphor

Applying Cornelissen’s (2004, 2005) domain-interaction model we find twometaphorical domains in the glass ceiling metaphor: the term, discrimination,as the tenor or target concept and the term, glass ceiling, as vehicle or sourceconcept. Table 3 presents the meaning of these terms given in the New OxfordDictionary, and the suggested generic space that can be derived from therespective domains.

As the aspectual similarity, which combines the tenor and vehicle andshapes the generic structure of the metaphor, we suggest the term, structure(step 1: the development of a generic structure), since we can find analogiesin both domains that refer to structure. The term, structure, is defined as

the arrangement of and relations between the parts or elements of some-thing complex; the organization of society or other group and the relationsbetween its members, determining its working; a building or other objectconstructed from several parts; the quality of being organized. (New OxfordDictionary, 1998, p. 1844)

The term, discrimination, on the one hand, pinpoints the ‘relation betweenparts’ by ‘the unjust or prejudicial treatment of different categories’ and ‘thedifference between one thing and another: discrimination between right andwrong’ and ‘to distinguish between different stimuli’ — all notions thatimplicitly denote structure. The term, glass ceiling, on the other hand, refersto structure (‘building or other object constructed from several parts’) by‘hard brittle substance’ in the definition of the term, glass, and by ‘the upperinterior surface or a room or other similar compartment’ in the definition ofthe term, ceiling. In other words, the communality of the two domains ‘dis-crimination’ and ‘glass ceiling’, or the generic space of the metaphor, drawsattention to the fact that individuals, groups or elements are constructed orarranged according to a given, in this case, discriminatory structure. This goesbeyond the analogy of each domain, since neither discrimination nor glassceiling mention the term, structure, directly in their definition.

This form of arrangement or structure is specified in the blend (step 2:development and elaboration of the blend), which comprises elements of theinput spaces. The organizing structure which the metaphor evokes can bedescribed as ‘static, transparent and, thus, invisible, but firm and touchablespaces’, with, however, no reference to time. As this blend represents a newmeaning, which is not a composition of meanings that can be found in eitherthe tenor or vehicle concept, it describes discrimination as a static phenom-enon taking place in transparent but firm and tangible spaces. The OxfordDictionary (1998, p. 778) defines the term, glass ceiling, as an ‘unacknowl-edged barrier to advancement’ while our blend denotes this barrier to



Tab

le3:

Gla

ssce

iling

met

apho

r

Prim

ary

subj

ect

(old

)(t

enor

)Se

cond

ary

subj

ect

(new

)(v

ehic

le)

Asp

ects

DIS

CR

IMIN

AT

ION

isno

un:

1th

eun

just

orpr

ejud

icia

ltr

eatm

ent

ofd

iffe

rent

cate

gori

esof

peop

leor

thin

gs,

espe

cial

lyon

the

grou

nds

ofra

ce,a

gese

x:vi

ctim

sof

raci

ald

iscr

imin

atio

n/d

iscr

imin

atio

nag

ains

tho

mos

exua

ls2

reco

gnit

ion

and

und

erst

and

ing

ofth

ed

iffe

renc

ebe

twee

non

eth

ing

and

anot

her:

dis

crim

inat

ion

betw

een

righ

tan

dw

rong

/(c

ount

noun

)yo

ung

child

ren

have

dif

ficul

ties

wit

hm

akin

gfi

ned

iscr

imin

atio

nsth

eab

ility

tod

isce

rnw

hat

isof

high

qual

ity;

good

jud

gem

ent

orta

ste;

thos

ew

hoco

uld

affo

rdto

buy

show

edlit

tle

tast

eor

dis

crim

inat

ion

psyc

holo

gyth

eab

ility

tod

isti

ngui

shbe

twee

nd

iffe

rent

stim

uli;

(as

mod

ifier

)d

iscr

imin

atio

nle

arni

ng3

Ele

ctro

nics

:the

sele

ctio

nof

asi

gnal

,hav

ea

requ

ired

char

acte

rist

ic,s

uch

asfr

eque

ncy

oram

plit

ude,

bym

eans

ofa

dis

crim

inat

orw

hich

reje

cts

allu

nwan

ted

sign

als

(New

Oxf

ord

Dic

tion

ary,

1998

,p.5

28)

GL

ASS

CE

ILIN

G(n

oun)

(usu

.in

sing

.):an

unac

know

led

ged

barr

ier

toad

vanc

emen

tin

apr

ofes

sion

,es

peci

ally

affe

ctin

gw

omen

and

mem

bers

ofm

inor

itie

s(O

xfor

dD

icti

onar

y,19

98,p

.778

).Th

ete

rm,g

lass

refe

rsto

(Oxf

ord

Dic

tion

ary,

1998

,p.7

78):

noun

1(m

ass

noun

)a

hard

brit

tle

subs

tanc

e,ty

pica

llytr

ansp

aren

tor

tran

sluc

ent,

mad

eby

fusi

ngsa

ndw

ith

soda

,lim

ean

dso

met

imes

othe

rin

gred

ient

san

dco

olin

gra

pid

ly.I

tis

used

tom

ake

win

dow

s,d

rink

ing

cont

aine

rsan

dot

her

arti

cles

any

sim

ilar

subs

tanc

ew

hich

has

solid

ified

from

am

olte

nst

ate

wit

hout

cris

talli

zing

2a

thin

gm

ade

from

,or

part

lyfr

omgl

ass,

inpa

rtic

ular

aco

ntai

ner

tod

rink

from

:abe

ergl

ass

(mas

sno

un)

glas

sw

are

(mas

sno

un)

gree

nho

uses

orco

ldfr

ames

cons

ider

edch

iefly

Bri

t.a

mir

ror

date

dsh

ort

for

aw

eath

ergl

ass

arch

aic

anho

urgl

ass

3a

lens

,or

anop

tica

lins

trum

ent

cont

aini

nga

lens

orle

nses

,in

part

icul

ara

mon

ocle

ora

mag

nify

ing

lens

Ori

gin:

Old

Eng

lish

glae

s,of

Ger

man

Ori

gin;

rela

ted

toD

utch

glas

and

Ger

man

Gla

sTh

ete

rm,c

eilin

g,(O

xfor

dD

icti

onar

y,19

98,p

.293

)no

un:

1th

eup

per

inte

rior

surf

ace

ofa

room

orot

her

sim

ilar

com

part

men

t,fig

urat

ive:

anup

per

limit,

typi

cally

one

set

onpr

ices

,wag

es,o

rex

pend

itur

e,th

em

axim

umal

titu

de

that

apa

rtic

ular

airc

raft

can

reac

hth

eal

titu

de

ofa

base

ofa

clou

dla

yer

2th

ein

sid

epl

anki

ngof

ash

ip’s

bott

oman

dsi

des

Ori

gin:

Mid

dle

Eng

lish

(den

otin

gth

eac

tion

oflin

ing

the

inte

rior

ofa

room

wit

hpl

aste

ror

pane

lling

):fr

omC

EIL

+IN

G.

aspe

ctua

l sim

ilar

ity:

str

uctu

re

invi

sibl

esp

aces

,but

not

virt

ual

Ble

ndst

able

boun

dari

es—

stat

ic(c

onta

iner

,wal

l,ce

iling

)tr

ansp

aren

t(E

yrin

gan

dSt

ead,

1998

)fi

rm(J

acks

on,2

001)

touc

habl

eno

refe

renc

eto

gend

erat

all

gend

erne

utra

l(gl

ass

asa

mat

eria

l)(E

lliso

n,20

01)

conn

ecti

onto

real

bod

ies,

sinc

efo

cus

onm

ater

ialit

y(g

lass

)—

repr

esen

tati

ond

epen

din

gon

cont

exts

whi

char

equ

ite

stab

leE

mer

gent

mea

ning

Stru

ctur

eno

refe

renc

eto

the

pers

onw

hobu

iltit

viol

ent

act

tocr

ack

it(W

ilson

,200

2)up

per

limit

boun

dari

eson

alls

ides

(gla

ssw

alls

)(O

hlen

die

ck,2

003)

nore

fere

nce

toti

me,

but

basi

cally

only

slow

lych

ange

able

inte

rms

ofti

me

A

A C1

B1

A C B

A1

Gen

eric

Spa

ce =

Str

uctu

re

Ten

orD

iscr

imin

atio

n

Veh

icle

Gla

ss C

eili

ng

1. D

evel

opm

ent o

f a

gene

ric

stru

ctur

e

2. D

evel

opm

ent a

nd E

labo

rati

on o

f th

e B

lend

3. E

mer

gent

Mea

ning

Str

uctu

re Dis

crim

inat

ion

as G

lass

Cei

ling

Ble

nd



advancement as a transparent boundary which seems to be firm or static andin terms of metaphorical space also touchable. If we relate it to the term, glass,the boundary is also tangible, but mostly invisible. Glass, as a hard but also afragile medium, provides an invisible barrier (Gabriel, 2005, p. 22). Althoughit allows the insider to see outside and the outsider to see inside, the space islimited by glass walls and a glass ceiling. In other words, a ‘glass box’(Ohlendieck, 2003) hinders different groups from organizational advance-ment. Since these boundaries are transparent, the individuals are able to seethe next management level but the invisible hurdles prevent access to it.Many researchers have already referred to this blend (see, for example,Ellison, 2001; Ohlendieck, 2003; Wilson, 2002).

As the emergent meaning structure (step 3: emergent meaning structure)is very efficient for the purpose of transferring the intended inferences backto the target input. It forces us to see the target concept in a new light if theelaborated blend is not merely compositional. In other words, the glassceiling metaphor cannot be reduced to or explained by referring to its initialdomains (‘discrimination’ and ‘glass ceiling’). Since the metaphor works withmateriality (glass), we assume that it refers to material bodies (sex, race — seethe definition of discrimination) that can be considered as being able to bechanged only with immense effort, as some examples have shown (e.g.,transgender individuals, changing skin colour — e.g. the singer MichaelJackson, the French performance artist Orlan in the ‘Reincarnation of StOrlan’ (see Thanem, 2006).

Interestingly, the glass ceiling metaphor, however, does not refer to theindividuals who built these glass ceilings or glass walls. There is no indicationof the builders or maintainers either in the target concept nor in the vehicleconcept, thus implying that people were not responsible for creating bound-aries in order to define and defend their interests and the boundaries came tobe on their own.

Furthermore, the glass ceiling metaphor fits with Schein’s (1971) ‘space ofcareer’, since it focuses on the person in a special limited ‘career space’. Thedark shaded areas in Figure 1 represent the glass ceiling and the glass walls,which construct a glass box and must be transgressed in order advance acareer up the hierarchy.

However, since neither aspectual similarity, the blend nor the emergentmeaning structure refer to time or process, this metaphor triggers the under-standing of discrimination as a static or stable notion which cannot bechanged but only displaced.

If the glass ceiling metaphor is indeed meaningful and insightful this isonly indicated by its heuristic value which, according to Cornelissen (2004,2005, 2006a, 2006b), is based upon the governing rules mentioned earlier. Theapplication of the two governing rules, which refer to the exactness ofthe aspectual similarity between the tenor and vehicle concepts as well asto the distance between the domains from which the concepts are drawn,



provides the following results. As far as the aptness or, in other words,relevance, fitness and suitability of metaphor is concerned, which is indicatedby the conceptual similarity between the two domains, we conclude there is ahigh degree of similarity between ‘discrimination’ as the target concept and‘glass ceiling’ as the source concept. In this respect, we are in line with themanifold literature on the glass ceiling phenomenon (e.g., Altman et al., 2005;Dreher, 2003; Stroh et al., 2004). Accordingly, the metaphor has a high level ofaptness. The distance between the domains, which correlates with the heuris-tic value of the metaphor, can also be concluded to be very high: the definitionof discrimination refers to behavioural, psychological and electronic aspects,whereas the term, glass ceiling, indicates substance, materiality, architectureand physical notions (see Table 3). With respect to the four types of metaphorin Table 2 we can therefore conclude that the glass ceiling metaphor refers toType 3. This means that the glass ceiling metaphor has a high heuristic valueand offers deep insights into discrimination. It also provides conceptualadvances and clarifications, as has already been demonstrated in the literatureon the glass ceiling metaphor; (for example, Davidson and Cooper, 1992;Dreher, 2003; Morrison et al., 1987; Stroh et al., 2004; Ryan and Haslam, 2005).

However, our process of metaphor evaluation raises two questions. Inwhat (organizational) contexts does the glass ceiling metaphor offer insights,if ‘structure’ is the generic space that forms the basis for this metaphor?Where are the boundaries or limits of this metaphor? Firstly, as far as thegeneric space, structure, is concerned, we can see that the metaphor corre-sponds with a hierarchical orientation and highly static notions of career intraditional modern organizations as previously described (e.g. Ellig andThatchenkery, 1996). Secondly, in terms of boundaries, we also want to pointout that neither of the two definitions, discrimination or glass ceiling, refer to

vertical

logisticshuman resources

marketingcontrolling

centripetal

horizontal

production

glass ceiling

glass wall

Figure 1: The glass ceiling metaphor according to ScheinSource: derived from Schein, 1971



time. As the static glass box (Ohlendieck, 2003) suggests, discrimination isconstructed as a given as well as an unchangeable fact. This, however, wouldimply that any attempts to overcome discriminatory situations do not makeany sense, since, as the metaphor indicates, discrimination is linked to static,invisible and unmovable frontiers.

As far as processual aspects are concerned, they occur only in the definitionof the term, discrimination (e.g., ‘treatment’, ‘recognition and understanding’as well as ‘discrimination learning’). The definition of the term, glass ceiling,however, lacks dynamic or process aspects. Thus, if we had chosen the term,process, as aspectual similarity for the examination of the heuristic value ofthe glass ceiling metaphor the result would have been that the metaphor isnot apt. According to the governing rules, even though there would havebeen a high distance between the domains, the level of exactness of thecorrespondence would have been low. Therefore, the glass ceiling metaphorwith a generic space in terms of process would produce weak and nonsensi-cal imagery (see Type 2 in Table 2). Consequently, we deduce that the glassceiling metaphor does not provide a reference to processual aspects of dis-crimination but offers the possibility of describing discrimination from astructural perspective.

Next, we want to draw attention to states of being (stable reality) andbecoming (fluid reality) (see e.g., Bakken and Hernes, 2005; Chia, 1996). Inthis respect, the glass ceiling metaphor refers to ‘being’, which emphasizesthe primacy of fact over the process. Therefore, the glass ceiling metaphorprovides a picture of discrimination as static and discrete but also as to someextent unidentifiable. With no reference to processes of discrimination andchange, the glass ceiling metaphor blanks out contemporary developments inorganizational research and practice as well as neglects discrimination asprocess that fluctuates contextually and is based on fluid phenomena andimages (e.g., Linstead and Brewis, 2004).

Processual aspects: evaluation of the ‘firewall’ metaphor

Recent texts (Bendl and Schmidt, 2003, 2004a and 2004b; Franke and Simöl,2002) have proposed the metaphor of the ‘firewall’ for describing discrimina-tion against women and other minority groups. The term, firewall, does notonly have a literal meaning, as presented in Table 4, but in computer scienceit is also used as a technical term for preventing outsiders from entering thesystem. The term, firewall, as considered from the technological point of view,was originally defined as

software that sits between you and the outside world. A firewall is config-ured on a need to know basis, i.e., all incoming connections are refusedunless you explicitly enable them. Firewall software is usually also small



Tab

le4:

Gla

ssce

iling

met

apho

r

Prim

ary

subj

ect

Seco

ndar

ysu

bjec

tA

spec

ts

DIS

CR

IMIN

AT

ION

isno

un:

1th

eun

just

orpr

ejud

ical

trea

tmen

tof

dif

fere

ntca

tego

ries

ofpe

ople

orth

ings

,esp

ecia

llyon

the

grou

nds

ofra

ce,a

gese

x:vi

ctim

sof

raci

ald

iscr

imin

atio

n/d

iscr

imin

atio

nag

ains

tho

mos

exua

ls2

reco

gnit

ion

and

und

erst

and

ing

ofth

ed

iffe

renc

ebe

twee

non

eth

ing

and

anot

her:

dis

crim

inat

ion

betw

een

righ

tan

dw

rong

/(c

ount

noun

)yo

ung

child

ren

have

dif

ficul

ties

wit

hm

akin

gfin

ed

iscr

imin

atio

nsth

eab

ility

tod

isce

rnw

hat

isof

high

qual

ity;

good

jud

gem

ent

orta

ste;

thos

ew

hoco

uld

affo

rdto

buy

show

edlit

tle

tast

eor

dis

crim

inat

ion

psyc

holo

gyth

eab

ility

tod

isti

ngui

shbe

twee

nd

iffe

rent

stim

uli;

(as

mod

ifier

)d

iscr

imin

atio

nle

arni

ng3

Ele

ctro

nics

:the

sele

ctio

nof

asi

gnal

havi

nga

requ

ired

char

acte

rist

ic,

such

asfr

eque

ncy

oram

plit

ude,

bym

eans

ofa

dis

crim

inat

orw

hich

reje

cts

allu

nwan

ted

sign

als

(New

Oxf

ord

Dic

tion

ary,

1998

,p.

528)

FIR

EW

AL

L.

The

term

,fire

wal

l,re

fers

tono

un:(

New

Oxf

ord

Dic

tion

ary,

1998

,p.6

90):

1a

wal

lor

part

itio

nd

esig

ned

toin

hibi

tor

prev

ent

from

spre

adof

fire

Com

puti

nga

part

ofa

com

pute

rsy

stem

orne

twor

kw

hich

isd

esig

ned

tobl

ock

unau

thor

ized

acce

ssw

hile

perm

itti

ngou

twar

dco

mm

unic

atio

nan

othe

rte

rmfo

rC

hine

seW

all

The

term

,fire

,ref

ers

tono

un(m

ass

noun

):1

Com

bust

ion

orbu

rnin

g,in

whi

chsu

bsta

nces

com

bine

chem

ical

lyw

ith

oxyg

enfr

omth

eai

ran

dty

pica

llygi

veou

tbr

ight

light

,hea

tan

dsm

oke:

his

hous

ew

asd

estr

oyed

byfir

eon

eof

the

four

elem

ents

inan

cien

tan

dm

edie

valp

hilo

soph

yan

din

astr

olog

y(c

onsi

der

edes

sent

ialt

oth

ena

ture

ofth

esi

gns

Ari

es,L

eoan

dSa

gitt

ariu

s):(

asm

odifi

er)

afir

esi

gnco

unt

noun

ad

estr

ucti

vebu

rnin

gof

som

ethi

ng:a

fire

ata

hote

lco

unt

noun

aco

llect

ion

offu

el,e

spec

ially

coal

orw

ood,

burn

tin

aco

ntro

lled

way

topr

ovid

ehe

ator

am

eans

for

cook

ing:

we

had

aba

thin

ati

ntu

bby

the

fire

coun

tno

unB

rit.

shor

tfo

rE

LE

CT

RIC

FIR

EO

RG

AS

FIR

Eco

unt

noun

abu

rnin

gse

nsat

ion

inth

ebo

dy:t

hew

hisk

ylit

afir

ein

the

back

ofhi

sth

roat

ferv

ent

orpa

ssio

nate

emot

ion

oren

thus

iasm

:the

fire

ofth

eir

relig

ious

conv

icti

onpo

etic

,lit

erar

ylu

min

osit

y;gl

ow:t

heir

soft

smile

slig

htth

eai

rlik

ea

star

’sfir

e2

the

shoo

ting

ofpr

ojec

tile

sfr

omw

eapo

ns,e

spec

ially

bulle

tsfr

omgu

ns:a

burs

tof

mac

hine

-gun

fire

ast

rong

crit

icis

mor

anta

goni

sm:h

ed

irec

ted

his

fire

agai

nst

polic

ies

prom

otin

gA

mer

ica’

sca

pita

lflig

htO

rigi

n:O

ldE

nglis

hfy

r(n

oun)

,fyr

ian

‘sup

ply

wit

hm

ater

ialf

orfir

e’,o

fW

est

Ger

man

icor

igin

;rel

ated

toD

utch

vuur

and

toG

erm

anFe

uer

(New

Oxf

ord

Dic

tion

ary,

1998

,pp.

688f

.)Th

ete

rm,‘

wal

l’,re

fers

to(n

oun)

1a

cont

inuo

usve

rtic

albr

ick

orst

one

stru

ctur

eth

aten

clos

esor

div

ides

anar

eaof

land

:aga

rden

wal

lor

farm

land

trav

erse

dby

dry

ston

ew

alls

asi

de

ofa

build

ing

orro

om,t

ypic

ally

form

ing

part

ofth

ebu

ildin

g’s

stru

ctur

ean

yhi

ghve

rtic

alsu

rfac

eor

faca

de,

espe

cial

lyon

eth

atis

impo

sing

insc

ale:

the

east

ern

wal

lof

the

valle

yor

figur

ativ

ea

wal

lof

soun

da

thin

gpe

rcei

ved

asa

prot

ecti

veor

rest

rict

ive

barr

ier:

aw

allo

fsi

lenc

eSo

ccer

alin

eof

def

end

ers

form

ing

aba

rrie

rag

ains

ta

free

kick

take

nne

arth

epe

nalty

area

shor

tfo

rC

LIM

BIN

GW

AL

LM

inin

gth

ero

cken

clos

ing

alo

de

orse

amA

nato

my

and

zool

ogy

the

mem

bran

ous

oute

rla

yer

orlin

ing

ofan

orga

nor

cavi

ty:t

hew

all

ofth

est

omac

hB

iolo

gyse

eC

EL

LW

AL

L(t

heW

all)

shor

tfo

rB

ER

LIN

WA

LL

Ori

gin:

Old

Eng

lish,

from

Lat

inva

llum

‘ram

part

’,fr

omva

llus

‘sta

ke’(

New

Oxf

ord

Dic

tion

ary,

1998

,p.2

078)

as

pect

ual s

imil

arit

y: p

roce

ss

virt

uals

pace

s(F

rank

ean

dSi

möl

,200

2)

Ble

nddy

nam

icin

term

sof

spac

ein

visi

ble

(Fra

nke

and

Sim

öl,2

002)

unto

ucha

ble

flex

ible

nore

fere

nce

toti

me,

but

basi

cally

quic

kly

tobe

chan

ged

inte

rmof

tim

eun

clea

rbo

unda

ries

beca

use

offl

exib

ility

(Ben

dl

and

Schm

idt,

2004

a)no

tge

nder

neut

ral(

cod

esar

eno

tap

plie

dge

nder

neut

ral)

conn

ecti

onto

real

bod

ies,

beca

use

focu

son

mat

eria

lity

(fire

),ho

wev

erd

isem

bod

ied

thro

ugh

cod

es

Em

erge

ntm

eani

ngst

ruct

ure

virt

uals

pace

sre

pres

enta

tion

dep

end

ing

onco

ntex

tsw

hich

chan

gequ

ickl

yre

fere

nce

toth

epe

rson

who

built

it,(g

atek

eepe

r,ad

min

iste

rof

the

firew

all

(Ben

dla

ndSc

hmid

t,20

03)

Non

-phy

sica

lvio

lent

act

tocr

ack

itop

enan

dcl

osed

atth

esa

me

tim

em

ovin

g

A

A C1

B1

A C B

A1

Gen

eric

Sp

ace

= P

roce

ss

Tenor

Dis

cri

min

ation

Veh

icle

Fir

ew

all

1.

De

velo

pm

ent

of

a g

eneri

c s

tructu

re

2.

De

velo

pm

ent

and E

labora

tion o

f th

e B

lend

3.

Em

erg

en

t M

eanin

g S

tructu

re

Dis

crim

inat

ion

as

Fir

ewal

l

Ble

nd



and carefully written, so that the incidence of security loopholes in thesoftware is vastly lower than a complex computer operating system that hasto do many other things as well as guard the door. [BT Net E1 lines, n.d.]

Like the glass ceiling metaphor, the firewall metaphor refers to a wall. TheNew Oxford Dictionary (1998, p. 690) defines a wall as ‘a continuous verticalbrick or stone structure that encloses and divides an area of land, a side of abuilding or room, typically forming a part of the building’s structure’, ‘a thingperceived as a protective or restrictive barrier’. However, there is one maindifference: in information technology the term, firewall, represents a virtualwall, whereas the glass ceiling, even if it is transparent, can, literally speaking,be touched. The firewall protects a ‘virtual intangible space from inside’ anddenies access to the outsiders, whom the insiders regard as not belonging tothe system. Here, in terms of access, the question immediately arises: whodefines, in whose interests, who may be considered to belong to the system ornot. Or, in other words, which space is safeguarded by whom for whoseinterests?

The term, firewall, is a ‘part of a computer system or network which isdesigned to block unauthorized access while permitting outward communi-cation’ (New Oxford Dictionary, 1998, p. 690). Who is allowed to enter thesystem is defined by those who manage or administer the firewall (e.g., the‘gatekeepers’, ‘firewall administrators’). In order to apply Cornelissen’sdomain-interaction model in the same way as for the glass ceiling metaphor,Table 4 presents the original meaning of the term, discrimination as tenor ortarget concept and the term, firewall, as vehicle or source concept accordingto the New Oxford Dictionary. Table 4 also suggests the aspectual similaritybetween tenor and target and it shows the generic space of the firewallmetaphor.

By following Cornelissen’s model again, we suggest the term, process, fordescribing aspectual similarity between the terms, discrimination and fire-wall, (step 1: development of a generic structure): as this term is definedamong others as

a series of actions or steps taken in order to achieve a particular end; anatural or involuntary series of changes; a systematic series of mechanizedor chemical operations that are performed in order to produce or manu-facture something. (New Oxford Dictionary, 1998, p. 1477)

we find processual aspects in both the tenor and the vehicle. In the tenor theprocessual notions can be considered as being represented on the one handby the terms, race, age and sex. Obviously, there is no doubt that age changesover time but, as already pointed out, sex and skin colour can be changed aswell, although it takes greater effort. On the other hand, discrimination islinked with ‘recognition and understanding’, ‘stimuli’ and ‘learning’ — termsthat signal progression, evolution and the capacity to change. The vehicle



‘firewall’ conveys operation and thus, process in different settings, whichimplies a course of action and movement with different effects: ‘spread offire’, ‘combustion or burning’, ‘a destructive burning’, ‘to light a fire’, ‘theshooting of projectiles from weapons’ and ‘a line of defenders forming abarrier’. Altogether, these notions show that process-orientation as actionrepresents the generic structure of the firewall metaphor.

As far as the elaboration of the blend is concerned (step 2), it involves animage of discrimination as ‘virtual flexible spaces’ that cannot be touched. Theblend of virtual flexible spaces refers to the tenor and the target, since bothterms also cover technical aspects (discrimination: ‘the selection of a signalhas a required characteristic, such as frequency of amplitude, by means of adiscriminator which rejects all unwanted signals’; firewall: ‘a part of a com-puter system or network which is designed to block unauthorized accesswhile permitting outward communication’). Both concepts, discriminationand firewall, also hint at a discriminator who administers this ‘intangibleseemingly borderless virtual space’ (‘creator’, ‘gatekeeper’, ‘firewall develop-ers’ [Frantzen et al., 2001] and ‘a discriminator which rejects unwantedsignals’ [New Oxford Dictionary, 1998, p. 528]). Since by definition this meta-phor allows a reference to a gatekeeper or discriminator, the boundaries canbe considered as being made up intentionally based on (personal and orga-nizational) interests managed through structural characteristics.

Regarding the established meaning structure (step 3), the source concept(firewall) sheds new light on the target concept (discrimination) in six ways:

Firstly, a firewall imports the notion of individuals exerting discriminationby referring to ‘firewall developers’ and so on. More precisely, it hints atcreators and administrators of discrimination who decide who is allowed toenter their system. It is a decision between ‘yes’ and ‘no’.

Secondly, by referring to gatekeepers, an intentional modus operandi ofpreventing unwanted people from entering the system comes into the pictureand allows us to imagine login codes applied for selection and the processesand results of their being shifted around. In order to enter the system one hasto know and decipher the meaning of those codes (e.g., sex, age, religion,ethnic background, education, wealth, language, social capital and manyother notions of diversity) which are contextually performed differently inorganizations all the time.

Thirdly, we claim that cracking codes is not just a violent act, as is largelyunderstood in terms of information technology, since there are differentdefinitions of the term, hacker.1 We exclusively refer to the non-criminaldefinition. We assume that in order to receive social benefits as a basic humanright, individuals use their skills to employ them in a non-criminal butperhaps unconventional manner, which could be considered hostile by thosewho are within the system. If we considered the transgression and change ofsocietal conventions, which per se discriminate against women and othergroups (as constituted by the term, not-A), as a hostile or even criminal act



then this metaphor would not make sense for the possibility of emancipation.Rather, the argument of hostility serves to maintain the closedness of theparticular system or organization, which, according to the firewall metaphor,preserves its ‘exclusiveness’ by applying selected codes (compare the darkcircles in Figure 2). The solution or key for overcoming discrimination in thismetaphor is to decipher the right passwords. In fact, there is hardly anyfirewall that cannot be passed through with the appropriate knowledge,creativity, persistence and network connections. In contrast to the glassceiling, which suggests the need to violently crack the glass to progress, theprocess of overcoming discrimination by finding codes only appears to be lessviolent or physical, but, in fact, it is not less difficult.

Fourthly, in contrast to the glass ceiling metaphor, the firewall metaphorrepresents invisible boundaries that can be changed quickly by altering thecodes of the firewall. However, as a consequence of this flexibility of thesystem, the codes and the boundaries remain quite obscure for those who areoutside the system.

Fifthly, the firewall metaphor shows more clearly the importance of par-ticipating in networks and mirrors the discursive reproduction of organiza-tions, such as gender and knowledge through codes. In fact, codes2 or keys3

cannot be applied gender neutrally as gender represents a basic structuringprinciple of society (for example, Harding, 1986, 1987).

Sixthly, by referring to networking, the firewall metaphor symbolizes morerelational and flexible rather than hierarchical and stable phenomena. Thus,as Figure 2 demonstrates, the centripetal aspect of Schein’s model is signifi-cant in the firewall metaphor.

The firewall metaphor highlights several points of discrimination missedby the glass ceiling metaphor. However, according to Cornelissen (2004), this

vertical

logistics

marketing

human resourcescontrolling

centripetal

horizontal

production

Figure 2: The firewall metaphor according to ScheinSource: derived from Schein, 1971



fact does not determine the firewall’s heuristic value. In order to evaluate itsvalue the governing rules have to be applied again (see Table 2). In terms ofaptness, which is indicated by the aspectual similarity (‘process’) between thetwo concepts of discrimination and firewall, their exactness is high. Thedistance between the two concepts can also be rated as high. Thus, accordingto the governing rules the firewall metaphor, like the glass ceiling metaphor,is of high value.

As in the glass ceiling metaphor, contextual conditions must also be con-sidered for the firewall metaphor. Process organization was one of the morepopular topics in organization studies between 1992 and 2002 and networkand process organizations were among the most prominent forms (Zaugg,2003, p. 11). In contrast to the glass ceiling metaphor, the firewall’s focus ison the process of discrimination and not on structures; therefore, it corre-sponds better with postmodern organizations and careers, which are char-acterized not only by diffusion, democracy and empowerment but alsoby flexibility (for example, Ellig and Thatchenkery, 1996; Gunz et al., 2002;Iellatchitch et al., 2003; Nicholson, 1996). Like the concept of businessre-engineering, which revolutionized organization theory (e.g., Hammerand Champy, 1993; Hammer and Stanton, 1994) and focuses on businessprocesses, the firewall metaphor refers to ‘doing’ more than ‘being’. Thus,business re-engineering and the firewall metaphor have three centralaspects in common (Osterloh and Frost, 1996, p. 87): Firstly, the idea ofprocess stresses the dominance of processes over structure; secondly, theidea of triage takes into account different processes (functional segment, aproblem-oriented perspective and a customer perspective) and third, theidea of informal networks points to boundaries which are kept up inten-tionally. Particularly the trends towards networks, fast moving openness tochange and co-operation as well as adaptability seem to be mirrored moreprecisely in the firewall metaphor than in the glass ceiling metaphor, whichfocuses on structure. Also, as we have pointed out, considering ‘process’ asaspectual similarity between ‘discrimination’ and ‘glass ceiling’ would onlyproduce a weak or even nonsensical imagery. But, if we choose ‘structure’as generic space for the firewall metaphor it provides different results(Table 4). We find structural aspects in the primary subject as well as in thesecondary subject (firewall — a wall: typically forming part of the buildingstructure). In this case, when applying the governing rules, the aptness ofthe two concepts and the distance of the domains are high. Hence, also withthe generic space of ‘structure’, the firewall metaphor provides insights dueto its high heuristic value (type 3). In this regard, the firewall metaphorseems to have an even higher heuristic value than the glass ceiling meta-phor for pointing at discrimination in organizations because it reflects thecomplementarity of structure and process in discrimination, whereasthe glass ceiling metaphor offers insights only into structural aspects ofdiscrimination.



Discussion: from ‘having discrimination’ to‘doing discrimination’

While both metaphors have a high heuristic value, they convey metaphoricaldifferences. The glass ceiling serves as an appropriate means for describing thestatus quo of discrimination in organizations by pointing to its structuralaspects, which refers to ‘having discrimination’. The firewall metaphor, how-ever, stresses a processual view of discrimination as ‘doing discrimination’.Here the metaphorical value is that discriminatory processes are considered tobe contextually performed and reproduced as well as creating structure. As wepointed out, a shift from structure to process has already occurred in bothorganizational theory and organizational practice (from a traditional to aboundaryless career paradigm, from modern to postmodern organizations). Inthis sense, the firewall metaphor corresponds better than the glass ceilingmetaphor to developments in these fields. The former offers more elaborateinsights into the reproduction of discrimination in contemporary organiza-tions. However, we do not advocate abandoning the glass ceiling metaphor.Rather, we propose that greater care should be given to choosing betweenthese metaphors and in using them in ways best suited to different kinds andsettings of discrimination in order to unveil and reduce discrimination.

Whereas the firewall portrays discrimination as complex, fluid, incoherentand heterogeneous, the glass ceiling reflects an image of space as a geometricstructure but transparent. However, both metaphors convey the notion thatdiscrimination is an intentional and long-term phenomenon (either throughstable glass walls or codes that can be changed more quickly). In contrast to theglass ceiling metaphor, the firewall metaphor points directly to the decision-making domain with the potential to reinforce or eliminate discrimination.

In conclusion, we propose both a deeper analysis of the reigning frames ofthe two metaphors which we have examined as well as other metaphors whichrepresent gender topics in organizations (such as gender mainstreaming as ametaphor for gender equality). Lastly, from a practical perspective, firewallsoffer human resource professionals and diversity managers a new metaphor,a tool perhaps, to work with in conducting workshops to identify catalysts andsources of discrimination and strategies for addressing or solving them. More-over, the firewall has the advantage of elasticity and permeability. It is appli-cable for many places and manifestations of discrimination, from hiringprocesses to the appointment of members to work teams and boards, and itoffers the possibility of organizational reprogramming for anti-discrimination.

Notes

1. A hacker: 1. Someone who uses a high degree of computer skill to carry outunauthorized acts within a network: for example, a hacker might have no access



rights to a network, yet might bypass all the access controls and carry out some actsuch as leaving a file in a public directory which describes what they have done.The action of hackers can range from relatively harmless ones to malicious actswhere system resources are incapacitated. 2. Someone who is possessed of a highdegree of computer skills and who employs them in a conventional, non-criminalway. Because of this duality of meaning, hackers who are in the second categoryuse the word Cracker to describe those in the first category. The second meaningpredates the first (Ince, 2001).

2. Code (Pearsall, 1998, p. 354): a system of words, letters, figures, or symbols usedto represent others, especially for the purposes of secrecy; a word, phrase, orconcept used to represent another in a euphemistic or indirect way; a series ofletters, numbers or symbols assigned to something for the purpose of classifica-tion or identification; a sequence of numbers dialled to connect a telephone withthe exchange of the telephone being called; program instructions (computing).

3. Key (Pearsall, 1998, p. 1004): basically to solve something. A word or system forsolving a cipher or code.

References

Adler, N. (1993) An international perspective on the barriers to the advancement ofwomen managers. Applied Psychology: An International Review, 42,4, 289–300.

Altman, Y., Simpson, R., Baruch, Y. and Burke, R.J. (2005) Reframing the ‘glass ceiling’debate. In Burke, R.J. and Mattis, M.C. (eds) Supporting Women’s Career Advance-ment: Challenges and Opportunities, pp. 58–81. Cheltenham: Edward Elgar.

Alvesson, M. (1993) The play of metaphors in postmodernism and organization. In:Hassard, J. and Parker, M. (eds) Postmodernism and Organizations, pp. 114–31.London: Sage.

Alvesson, M. and Billing, Y.D. (1997) Understanding Gender and Organizations. London:Sage.

Arthur, M.B. and Rousseau, D.B. (eds) (1996) The Boundaryless Career. A New Employ-ment for a New Organizational Era. New York and Oxford: Oxford University Press.

Bakken, T. and Hernes, T. (2005) Organizing is both verb and noun: Weick meetsWhitehead. Paper read at the First Organization Studies Summer Workshop onTheorizing Process in Organizational Research, 12–13 June, Santorini.

Belle, L., Edmondson, J. and Nkomo, S.M. (2001) Our Separate Ways: Black and WhiteWomen and the Struggle for Professional Identity. Boston, MA: Harvard UniversityPress.

Bendl, R. and Schmidt, A. (2003) Gender mainstreaming — can it really break downthe ‘firewall’? In Morley, M.J., Gunnigle, P., Heraty, N., Pearson, J., Sheikh, H. andTiernan, S. (eds) Exploring the Mosaic, developing the Discipline. Proceedings of the 7th

Conference on International Human Resource Management, Limerick, pp. 18–19.Dublin: Interesource Group Publishing.

Bendl, R. and Schmidt, A. (2004a) ‘Firewall’: a new metaphor for the underrepresen-tation of women in management. Proceedings of SAM/IFSAM World CongressGothenburg, 5–7 July 2010.

Bendl, R. and Schmidt, A. (2004b) Firewalls und gender mainstreaming. Durchbruch oderAusschluss von Frauen in Führungspositionen? In Heitzmann, K. and Schmidt, A.(eds): Wege aus der Frauenarmut. Schriftenreihe Frauen, Forschung und Wirtschaft,Band 14m, pp. 137–70. Frankfurt: Peter Lang Verlag.



Black, M. (1962) Models and Metaphors. Ithaca, NY: Cornell University Press.Black, M. [1979](1993) More about metaphor. In Ortony, A. (ed.) Metaphor and Thought,

pp. 10–43. Cambridge: Cambridge University Press.Bono, J.J. (1990) Science, discourse and literature: the role/rule of metaphor in science.

In Peterfreund, S. (ed.) Literature and Science, pp. 59–89. Boston, MA: NortheasternUniversity Press.

Brousseau, K.R., Driver, M.J., Eneroth, K. and Larsson, R. (1996) Career pandemo-nium: realigning organizations and individuals. The Academy of Management Execu-tive, 10,4, 52–66.

BT Net E1 lines, n.d. Available online at http://www.dslreports.com/information/kb/firewall Last accessed 28 May 2010.

Cadin, L., Bender, A.-F., de Saint Giniez, V. and Pringle, J. (2000) Carrières nomades etcontextes nationaux. Revue de Gestion des Ressources Humaines, 37, 76–96.

Catalyst (1994) On the Line: Women’s Career Advancement. New York: Catalyst.Catalyst (2000) Cracking the Glass Ceiling. New York: Catalyst.Catalyst (2009) Women in US Management 2009. New York, available online at http://

www.catalyst.org/publication/206/women-in-us-management Last accessed 30May 2010.

Chia, R. (1996) The problem of reflexivity in organizational research: towards a post-modern science of organization. Organization, 3,1, 31–59.

Clegg, S.R. (1992) Postmodern management? Journal of Organizational Change, 5,2,31–49.

Cornelissen, J.P. (2002) On the ‘organizational identity’ metaphor. British Journal ofManagement, 13,3, 259–68.

Cornelissen, J.P. (2004) What are we playing at? Theatre, organization, and the use ofmetaphor. Organization Studies, 25,5, 705–26.

Cornelissen, J.P. (2005) Beyond compare: metaphor in organization theory. Academy ofManagement Review, 30,4, 751–64.

Cornelissen, J.P. (2006a) Metaphor in organization theory: progress and past. Academyof Management Review, 31,2, 485–8.

Cornelissen, J.P. (2006b) Metaphor and the dynamics of knowledge in organizationtheory: a case study of organizational identity metaphor. Journal of ManagementStudies, 43,4, 683–709.

Cortina, L.M. (2008) Unseen injustice: incivility as modern discrimination in organi-zations. Academy of Management Review, 33,1, 55–75.

Cummings, S. (1996) Back to the oracle: postmodern organization theory as a resur-facing of premodern wisdom. Organization, 3,2, 249–66.

Daft, R.L. (2001) Organization theory and design. Ohio: South-Western CollegePublishing.

Davidson, D. (1978) What metaphor means. Critical Inquiry, 5, 31–48.Davidson, M.J. and Bruke, R. (2000) Women in management: current research issues.

In Davidson, M.J. and Burke, R. (eds), Women in Management: 1–7. London: PaulChapman.

Davidson, M.J. and Cooper, C.L. (1992) Shattering the Glass Ceiling: The WomenManager. London: Paul Chapman Publishing.

Dreher, G.F. (2003) Breaking the glass ceiling: the effects of sex ratios and work–life programs on female leadership at the top. Human Relations, 56,5, 541–62.

Ellig, J. and Thatchenkery, T.J. (1996) Subjectivism, discovery, and boundarylesscareers: an Austrian perspective. In Arthur, M.B. and Rousseau, D.M. (eds) TheBoundaryless Career — A New Employment Principle for a New Organziational Era, pp.171–86. New York: Oxford.



Ellison, L. (2001) Senior management in chartered surveying: where are the women?Women in Management Review, 16,6, 264–78.

Eyring, A. and Stead, B.A. (1998) Shattering the glass ceiling: some successful corpo-rate practices. Journal of Buisness Ethics, 17,3, 245–51.

Federal Glass Ceiling Commission (1987) The glass ceiling. In Dunn, D. (ed.)Workplace/Women’s Place: an Antohology, pp. 226–33. Los Angeles, CA:Roxbury.

Franke, M. and Simöl, I. (2002) Teilzeit im Management von Organisationen. Firewalls fürden Aufstieg von Frauen ins Top-Management. In Kramer, C. (ed.) FREI-Räume undFREI-Zeiten: Raum-Nutzung und Zeit-Verwendung im Geschlechterverhältnis, pp. 165–78. Heidelberg: Nomos.

Frantzen, M., Kerschbaum, F., Schultz, E. and Fahmy, S. (2001) A framework forunderstanding vulnerabilities in firewalls using a data model of firewall internals.Computer and Security 20,3, 263–0.

Gabriel, Y. (2005) Glass cages and glass palaces: images of organization in image-conscious times. Organization, 12,1, 9–27.

Gunz, H.P., Lichtenstein, B.M.B. and Long, R. (2002) Self-organization in careersystems: a view from complexity science. M@n@gement, 5,1, 63–88.

Hall, D.T. (1996) Protean careers of the 21st century. Academy of Management Executive,10,4, 8–16.

Hall, D.T. and Mirvis, P. (1996) The new protean career: psychological success and thepath with a heart. In Hall, D.T. (ed.) The Career is Dead — Long Live the Career, pp.1–12. San Francisco, CA: Jossey-Bass.

Hammer, M. and Champy, J. (1993) Re-engineering the Corporation. New York: HarperBusiness.

Hammer, M. and Stanton, S.A. (1994) The Reengineering Revolution. A Handbook. NewYork: Harper Business.

Harding, S. (1986) The Science Question in Feminism. Ithaca, NY and London: CornellUniversity Press.

Harding, S. (1987) Feminism and Methodology. Bloomington, IN: Indiana UniversityPress.

Hesse, M. (1995) Models, metaphors and truth. In Radman, Z. (ed.) From a MetaphoricalPoint of View: A Multidisciplinary Approach to the Cognitive Content of Metaphor, pp.351–72. Berlin: De Gruyter.

Hooker, H., Neathey, F., Casebourne, J. and Munro, M. (2007) the third work–lifebalance employee survey: main findings. Employment Relations Research Series No.58. London: Department of Trade and Industry.

Hunt, S.D. and Menon, A. (1995) Metaphors and competitive advantage: evaluatingthe use of metaphors in theories of competitive strategy. Journal of BusinessResearch, 33,2, 81–90.

Iellatchitch, A., Mayrhofer, W. and Meyer, M. (2003) Career fields: a small steptowards a grand career theory? International Journal of Human Resource Management,14,5, 728–50.

Ince, D. (2001) A Dictionary of the Internet. Oxford Reference Online: Oxford and Vienna:Oxford University Press and Wirtschaftsuniversitat, available by subscription athttp://www.oxfordreference.com/views/ENTRY.html last accessed 16 April2004.

Inkson, K. (2006) Protean and boundaryless careers as metaphors. Journal of VocationalBehavior, 69,1, 48–61.

Inns, D. (2002) Metaphor in the literature of organizational analysis: a preliminarytaxanomy and a glimpse at a humanities-based perspective. Organization, 9,2,305–30.



Insch, G.S., McIntyre, N. and Napier, N.K. (2008) The expatriate glass ceiling: thesecond layer of glass. Journal of Business Ethics, 83,1, 19–28.

Jackson, J.C. (2001) Women middle managers’ perception of the glass ceiling. Womenin Management Review, 16,1, 30–41.

Katz, A. (1992) Psychological studies in metaphor processing: extensions to the place-ment of terms in semantic space. Poetics Today, 13,4, 607–32.

Knights, D. (1997) Organization theory in the age of deconstruction: dualism, genderand postmodernism revisited. Organization Studies, 18,1, 1–19.

Lakoff, G. and Johnson, M. (1980) Metaphors We Live by. London: University ofChicago Press.

Linstead, A. and Brewis, J. (2004) Editorial: beyond boundaries: towards fluidity intheorizing and practice. Gender, Work & Organization, 11,4, 355–62.

Meyerson, D.E. and Fletcher, J.K. (2000) A modest manifesto for shattering the glassceiling. Harvard Business Review, 78,1, 127–36.

Montuschi, E. (1995) What is wrong with talking of metaphors in science? Radman, Z.(ed.) From a Metaphorical Point of View: A Multidisciplinary Approach to the CognitiveContent of Metaphor, pp. 309–27. Berlin: De Gruyter.

Morgan, G. (1980) Pardigms, metaphors and puzzle solving in organizational theory.Administrative Science Quarterly, 25,4, 605–22.

Morgan, G. (1996) Images of Organization. Thousand Oaks, CA: Sage.Morrison, A.M., White, R.P. and Van Velsor, E. (1987) Breaking the Glass Ceiling:

Can Women Reach the Top of America’s Largest Corporations? Reading, MA:Addison-Wesley.

Murrell, A.J. and Hayes James, E. (2001) Gender and diversity in organizations: past,present, and future directions. Sex Roles, 45,4–5, 243–57.

New Oxford Dictionary (1998) Oxford: Oxford University Press.Nicholson, N. (1996) Career systems in crisis: change and opportunity in the infor-

mation age. Academy of Management Executive, 10,4, 40–51.Ohlendieck, L. (2003) Die Anatomie des Glashauses: Ein Beitrag zum Verständnis des

Glass-Ceiling-Phänomens. In Pasero, U. (ed.) Gender — from costs to benefits, pp.183–93. Wiesbaden: Westdeutscher Verlag.

Osterloh, M. and Frost, J. (1996) Prozessmanagement als Kernkompetenz: wie Sie BusinessReenginieering strategisch nutzen können. Wiesbaden: Gabler.

Oswick, C., Keenoy, T. and Grant, D. (2002) Metaphor and analogical reasoning inorganizational theory: beyond orthodoxy. Academy of Management Review, 21,2,294–303.

Pearsall, J. (ed.) (1998) New Oxford Dictionary. Oxford: Oxford UniversityPress.

Peiperl, M. and Baruch, Y. (1997) Back to square zero: the post-corporate career.Organizational Dynamics, 25,4, 6–22.

Peterson, R.A. and Anand, N. (2002) How chaotic careers create orderly fields. InPeiperl, M.A., Arthur, M. and Anand, N. (eds) Creative Careers, pp. 257–89. NewYork: Oxford University Press.

Ruigrok, W., Pettigrew, A., Peck, S.I. and Whittington, R. (1999) Corporate restructur-ing and new forms of organizing: evidence from Europe. Management InternationalReview, Special Issue, 2,2, 41–64.

Ryan, M.K. and Haslam, S.A. (2005) The glass cliff: evidence that women are over-represented in precarious leadership positions. British Journal of Management, 16,2,81–90.

Ryan, M.K. and Haslam, S.A. (2007) The glass cliff: exploring the dynamics surround-ing the appointment of women to precarious leadership positions. Academy ofManagement Review, 32,2, 549–72.



Schein, E.H. (1971) The individual, the organization and the career: a conceptualscheme. Journal of Applied Behavior Science, 7,4, 401–26.

Schön, D.A. (1965) Displacement of Concepts. London: Tavistock.Sennett, R. (1998) The Corrosion of Characters. New York: Norton.Shen, Y. (1997) Metaphors and conceptual structure. Poetics, 25,1, 1–16.Stroh, L.K., Langlands, C.L. and Simpson, P.A. (2004) Shattering the glass ceiling in the

new millennium. In Stockdale, M.S. and Crosby, F.J. (eds) The Psychology andManagement of Workplace Diversity, pp. 147–67. Malden, MA: Blackwell.

Sydow, J. (1999) Management von Netzwerkorganisationen. Wiesbaden: Gabler.Thanem, T. (2006) Living on the edge: towards a monstrous organization theory.

Organization 13,2, 163–93.Tourangeau, R. and Sternberg, R. (1981) Aptness in metaphor. Cognitive Psychology

13,1, 27–55.Tsoukas, H. (1991) The missing link: a transformational view of metaphors in organi-

zational science. Academy of Management Review, 16,3, 566–85.Tsoukas, H. (1993) Analogical reasoning and knowledge generation in organization

theory. Organization Studies, 14,3, 323–46.Van Vianen, A.E. and Fischer, A.H. (2002) Illuminating the glass ceiling: the role of

organizational culture preferences. Journal of Occupational and Organizational Psy-chology, 75,3, 315–37.

Weick, K.E. (1997) Sensemaking in Organizations. Thousand Oaks, CA: Sage.Willmott, H. (1993) Strength is ignorance; slavery is freedom: managing culture in

modern organizations. Journal of Management Studies, 30,4, 515–52.Wilson, F.M. (2002) Management and the professions: how cracked it that glass

ceiling? Public Money and Management, 22,1, 15–20.Zaugg, R.J. (2003) Organisation — quo vadis? Entwicklungstendenzen und Zukun-

ftsperspektiven einer Disziplin im Spannungsfeld einer paradigmatischen Prolif-eration. Zeitschrift für Organisation, 72,1, 4–12.



Copyright of Gender, Work & Organization is the property of Wiley-Blackwell and its content may not be

copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written

permission. However, users may print, download, or email articles for individual use.

from ‘glass ceilings’ to ‘firewalls’ — different … ‘glass ceilings’ to...

Documents