http://tcr.sagepub.com/
Theoretical Criminology
http://tcr.sagepub.com/content/6/4/433
The online version of this article can be found at:
DOI: 10.1177/136248060200600403
2002 6: 433Theoretical Criminology
Jody Miller
The strengths and limits of 'doing gender' for understanding street crime
Published by:
http://www.sagepublications.com
can be found at:Theoretical CriminologyAdditional services and information for
http://tcr.sagepub.com/cgi/alertsEmail Alerts:
http://tcr.sagepub.com/subscriptionsSubscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://tcr.sagepub.com/content/6/4/433.refs.htmlCitations:
What is This?
- Nov 1, 2002Version of Record>>
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
The strengths and limits of
‘doing gender’ for
understanding street crime
JODY MILLER
University of Missouri-St. Louis, USA
Abstract
This paper is an engagement with Messerschmidt’s structured
action theory, and more generally with feminist criminologists’
applications of the concept ‘doing gender’ for understanding street
crime. Specifically, I investigate the ways in which the attribution of
gender difference and the near exclusive emphasis on normative
practices has limited our use of the doing gender model in
theorizing gender and crime. I discuss several avenues for
enhancing this approach, including the imperative to avoid
tautology, and suggestions for challenging gender dualism,
investigating the import of social hierarchies, and conceptualizing
the complexities of agency and social practice.
Key Words
feminist theory • crime as structured action • James
Messerschmidt • doing gender • gender dualism
In recent years, feminist criminologists have drawn on the concept of
situated action to help explain the gendered nature of crime. This ethnomethodological
approach (Garfinkel, 1967) was first applied critically to
gender by feminists in the late 1970s (Kessler and McKenna, 1978) and
gained widespread recognition and further refinement in the mid-1980s
(West and Zimmerman, 1987; see also Connell, 1987). James Messerschmidt
(1993) was largely responsible for the introduction of this approach
Theoretical Criminology
© 2002 SAGE Publications
London, Thousand Oaks
and New Delhi.
1362–4806(200211)6:4
Vol. 6(4): 433–460; 028592
433
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
to the field of criminology with his Masculinities and Crime. In particular,
his and other feminist developments of this approach have attempted to use
ethnomethodological insights while simultaneously theorizing the impact
of structural inequalities on the accomplishment of gender. Since Messerschmidt’s
landmark work, a number of feminist criminologists have
applied and refined ‘doing gender’ as a means of theorizing about crime.
Here my goal is to discuss both the strengths and limits of this approach,
especially as it is applied to young women’s involvement in crime; in its
capacity to deal with the intersections of race, class, gender, sexuality and
generation; and its utility in addressing gendered identities and gendered
social practices. In exploring these questions, I make use of illustrations
from my own and other relevant feminist works that deal with ‘doing
gender and crime’.1
Viewing gender as situated action or situated accomplishment means
recognizing that gender is ‘much more than a role or an individual
characteristic: it is a mechanism whereby situated social action contributes
to the reproduction of social structure’ (West and Fenstermaker, 1995).
According to this approach, women and men ‘do gender’ in response to
situated normative beliefs about masculinity and femininity. These actions
are the ‘interactional scaffolding of social structure’ (West and Zimmerman,
1987: 147), such that the performance of gender is both an indication
of and a reproduction of gendered (as well as raced, classed, generational,
and sexed) social hierarchies. This approach advances theoretical accounts
of gender in a number of significant ways.
One important facet of the approach is that it provides a means of
bridging the agency/structure divide in a way that allows theorists to go
beyond constructing women (and men) as simply passive victims of structural
conditions. Focusing narrowly on women’s victimization is problematic
because, as feminist philosopher Sandra Harding (1987: 5) notes, this
‘tend[s] to create the false impression that women have only been victims,
that they have never successfully fought back, that women cannot be
effective social agents on behalf of themselves or others’. Moreover, it
results in a static and deterministic view of social structure which many
social theorists have challenged (for recent discussions, see Emirbayer and
Mische (1998), McNay (2000) and Sewell (1992)). Specifically, recognizing
gender as situated action allows for recognition of agency, but does so in a
way thoroughly grounded in the contexts of structural inequalities such as
those of gender, sexuality, race, class and age. As social theorist Bob
Connell (1993: ix) notes, this approach insists that ‘social structure does
not exist “outside” everyday life’ and everyday practices. By grounding
social practices within the structural contexts in which they occur, the
approach makes logical sense out of actions that, decontextualized, might
appear illogical or senseless.
Focusing on situated social action also challenges the notion that ‘natural’
(bodily or biological or psychic) differences between women and men
Theoretical Criminology 6(4)434
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
account for gender, gender inequality or gendered action. As Connell
(1987: 77) points out, there is no:
logical sense in which the social processes of gender are exceptions to
principles that apply to other social processes. There is no reason to make
this exception. The body is implicated in the social processes of gender,
certainly; but then the body is involved in every kind of social practice.
Moreover, this approach provides several important critiques of ‘sex role’
or ‘gender role’ theory — a way of conceptualizing gender that remains
ever popular in criminology. First, while the concept of ‘gender roles’
assumes that ‘gender is logically prior to behavior, already settled, and can
be understood as [the cause of] behavior’ (Connell, 1993: x), the current
approach recognizes ‘gender as something constructed in social action, as
something done, accomplished in the everyday actions of social life’
(Connell, 1993: xi). Moreover, given its grounding in the agency/structure
nexus, viewing gender as situated action means recognizing that there are a
multitude of masculinities and femininities — each shaped by structural
positioning — rather than one static set of gender roles.
Doing gender and crime
As I indicated earlier, this approach has been applied most notably within
criminology in explanations of the links between masculinities and crime.
Messerschmidt has been at the forefront in this regard. Here, crime is
described as ‘a “resource” for accomplishing gender — for demonstrating
masculinity within a given context or situation’ (Simpson and Elis, 1995:
50). For example, robbery is described as epitomizing the use of crime to
construct masculine identity among urban young men.2
Messerschmidt
(1993: 107) argues:
The robbery setting provides an ideal opportunity to construct an ‘essential’
toughness and ‘maleness’: it provides a means with which to construct that
certain type of masculinity — hardman. Within the social context that
ghetto and barrio boys find themselves, then, robbery is a rational practice
for ‘doing gender’ and for getting money.
In Messerschmidt’s earlier work (1993, 1995, 1997), he primarily relies
on secondary analysis of previous research — an approach that, as I discuss
below and Messerschmidt (2002) himself later notes, makes a series of
problematic assumptions that can be better addressed through rigorous
empirical investigation. More recently, his investigation of gendered structured
action has instead been based on in-depth life-history narratives with
young men (see Messerschmidt, 2000), an approach that offers promise, as
it has allowed him to examine more thoroughly ‘the meanings boys and
men attach to their social actions and how these actions are related to
conscious choice and specific social structures in particular settings’ (Messerschmidt,
forthcoming).3
Miller—The strengths and limits of ’doing gender’ 435
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Where this theoretical approach has most stalled within criminology is in
attempts to apply the concept of ‘doing gender’ to explanations of female
participation in crime. Examining crime as masculine accomplishment can
help account for the gender ratio of offending; at least at first blush, it
offers an explanation of women’s involvement in crime in ways scripted by
femininities. But it often leaves women’s participation in presumably
‘masculine’ crime unexplained except as an anomaly. For this reason,
scholars have recently attempted to refine the approach to allow for the
explanation of women’s involvement in crime, especially violence. Typically,
this is done by emphasizing variations in normative femininity as they
emerge within different structural and situational contexts (see Messerschmidt,
1995; Simpson and Elis, 1995; but see also Bottcher, 2001; Jacobs
and Miller, 1998; Miller, 1998).
For example, Messerschmidt (1995, 1997) applied the concept of ‘doing
gender’ to young women’s involvement in street gangs, which he describes
as ‘doing difference’. He focused specifically on heterosexual relationships
within gangs and inter-gang conflict. With regard to the former, he notes
the following:
For both boys and girls . . . the street gang is ideal for ‘doing gender’ in
terms of difference. Through maintenance of and emphasis on the ‘femaleness’
of girl gang members — for example, through specific heterosexual
meanings and practices — gender difference is preserved and specific types
of masculinities and femininities are both validated and strengthened.
Consequently, girl gang members are not simply passive recipients of
‘patriarchy’, but actively participate in the construction of gender relations
and orchestrate the various forms of heterosexuality that results in varieties
of femininity. Indeed, these girls do difference differently.
(Messerschmidt, 1995: 177)
With regard to inter-gang conflict, Messerschmidt focused on what he
calls ‘bad girl’ femininity. He argues that group processes within gangs
provide situations (such as the protection of ‘the ‘hood’) in which there is
a ‘path for similarity in behavior’ (Messerschmidt, 1995: 182) between
males and females. Though he recognizes gang conflict as a site in which
gender differences are less salient than gang differences (for example,
rivalries across gangs), here Messerschmidt nonetheless characterizes girls’
participation in gang violence as a means of constructing a ‘bad girl’
femininity:
For girls in the gang, doing femininity means occasionally, and in appropriate
circumstances, doing violence. However, because participation in
violence varies depending upon the setting, girls are assessed and held
accountable as ‘bad girls’ differently. Given that gang girls realize that their
behavior is accountable to other girls and boys in the gang, they construct
their actions in relation to how those actions will be interpreted by others in
the same social context. These girls are doing femininity in terms of activities
appropriate to their sex category and in specific social situation. AccordTheoretical
Criminology 6(4)436
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
ingly, violence by young women in youth gangs should not be interpreted as
an attempt to be ‘male’ and ‘pass’ as the other gender.
(Messershmidt, 1995: 183)
Thus, he suggests that girls in gangs engage in activities, including violence,
with the goal of enacting normatively appropriate femininity — a femininity
situated within the social context of the gang.
For reasons I detail below, I suggest that this formulation offers a limited
account of girls’ (and women’s) gendered social action (and by implication,
men’s and boys’ as well). Here I argue that serious engagement with the
limitations of the approach offers promising directions for the reformulation
and expansion of evaluating structured action. Specifically, through
rigorous empirical investigation, we can better conceptualize and examine
the complexities of agency and social practices, and can better map the
structural and ideological hierarchies of gender, race, class, generation and
sexuality and their reciprocal affects on gendered social action.
‘Doing gender and crime’: evaluation and critique
I focus here on four specific, but overlapping, issues that I believe can
strengthen our use of the situated/structured action approach within feminist
criminology: (1) avoiding tautology; (2) challenging gender dualism;
(3) accounting for stratification, hierarchy and power; and (4) conceptualizing
the complexities of agency and social practice. In elaborating on these
issues, my goal is to propose ways to enrich our examinations and
theorizing about ‘doing gender and crime’, and to suggest avenues for
doing so that deal explicitly with the intersecting nature of various social
hierarchies and identities, avoid giving primacy to gender as normative
practice, and can account for the dynamic nature of inequality.
Avoiding tautology
In order for the concept of ‘doing gender’ to be conceptually useful, we
must theorize and apply it in ways that avoid tautology. This means, first,
that our analyses cannot employ circular reasoning; and, second, that they
must not be static, but instead capable of accounting for transformation
and social change. Circular reasoning can occur in several ways: by
presuming gendered action occurs primarily in response to gendered
norms, and through the assumption of gender duality. A static (and
ultimately deterministic) approach to ‘doing gender’ occurs when we
suggest that behaviour is governed by social structural position. Here
change resulting from individual or collective action is not accounted for,
but instead results from responses to external (structural) change. Instead,
our conceptualization of ‘doing gender’ must address the transformative
potential of agency. Changing our conceptualization of structure, and
Miller—The strengths and limits of ’doing gender’ 437
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
integrating the intersections with gender of such structuring positions as
race, class, sexuality and generation, provides a means by which to avoid
tautology and stasis.
Often the ways in which ‘doing gender’ is applied to explanations of
crime are problematic because they involve circular reasoning. Drawing
from West and Zimmerman’s (1987) seminal work, Messerschmidt (1995:
171) argues that ‘in all social situations we attempt to adorn ourselves with
culturally appropriate “female” or “male” fashion’ (my emphasis), because
we are accountable to others for our gendered actions. By definition, this
statement does not provide for individuals’ reinterpretation of, resistance
to, or subversion of culturally appropriate patterns (see Butler, 1990;
Thorne, 1993). Instead, variations in gendered actions are accounted for
exclusively by variations in cultural definitions of masculinity and femininity,
as emerging from variations in structural position (see Hood-Williams
(2001) for a similar critique).
As already noted, one important contribution of viewing gender as
situated action is that it allows for recognition that there are multiple
masculinities and femininities, shaped by structural positioning. Nonetheless,
it is tautological to assert that every action is taken with a goal of
accomplishing normative femininity, whereby different variations of social
action are simply reflective of different variations of normative femininity.
Inadvertently, this becomes a rather static view of gendered action. Moreover,
stated and applied in such a way, the propositions of the theory ‘are
so open-ended that any contradictory empirical evidence can be interpreted
or re-interpreted to support the theory’ (Tittle, 1995: 8).
One of the problems here lies with the primacy given to normative
aspects of gendered action. The emphasis on norms limits our ability to
grapple fully with power and inequality, that is, how gendered actions are
a response to structural or situation exclusion from other modes of action
rather than necessarily based on adherence to norms about masculinity and
femininity.4
Instead, a more complex concept of agency — including
recognition that action can be routinized and largely taken for granted,
strategic and goal oriented, norm oriented, resistant or negotiative, and
temporally based — provides one means of avoiding tautology in examining
gender as situated action (see Emirbayer and Mische, 1998; McNay,
2000).
Applying such a concept in the context of empirically rigorous research
provides a means of enriching the theoretical import of ‘doing gender’. For
example, Messerschmidt’s (1999, 2000) most recent study moves away
from tautology in just these ways as he blends specific empirical investigation
with theorizing. Evidence presented in his previous articulation of this
perspective was based on interpretation of data not collected with
the explicit goal of addressing this theoretical perspective. This made the
interpretative process more problematic because he was left with the task
of imbuing meanings and motives to behaviour without clear evidence of
the gendered intents of the actors under examination.5
Theoretical Criminology 6(4)438
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
In contrast, Messerschmidt’s (2000) analysis in Nine Lives is based on
comparative life-history interviews with three groups of young men: boys
who sexually abuse children, boys involved in physical violence, and a
comparison group of non-violent boys from the same white working-class
communities. The interview data collected for this analysis emphasized the
boys’ constructions and beliefs about masculinity (and the linkages of these
to the body as a resource for enacting masculinity), and these youths’
perceptions of the role hegemonic masculinity played in the commission of
their crimes. Such rigorous examination of accounts provides an important
means of ‘arriving at meanings or culturally embedded normative explanations
[for behavior. Accounts] represent ways in which people organize
views of themselves, of others, and of their social worlds’ (Orbuch, 1997:
455). Importantly, in his most recent work, Messerschmidt does not argue
that normative beliefs about masculinity are always primary in doing
gender, but instead focuses on instances of masculinity challenge, when
‘both body and gender became highly salient as organizing principles of
interaction’ (Messerschmidt, 2002). Moreover, his concern is with examining
these boys’ accounts of when and how masculine identity construction
was a goal of violent behaviour, rather than assuming this is always the
case.
The question of tautology still remains, however, and is linked specifically
to the problem of gender duality. This is a matter of selective attention
to accounts based on gender difference. Hood-Williams (2001: 45; see also
Spelman, 1988: 136), in a critique of Messerschmidt’s analysis of men’s
construction of specific forms of masculinity, poses and answers the
question:
And why is this masculine? Because men do it. The argument is clear: every
(criminal) thing that men do is masculine. But if everything that men do is
masculine then the concept of masculinity is an empty tautology: gender
collapses into sex.
By definition this approach reifies gender difference. Thus, returning to the
earlier example of the gendered actions of gang girls, Messerschmidt (1995:
180) argues that they behave in the ways they do in order to construct a
‘bad girl’ femininity in opposition to a ‘dud’ femininity, but always a
femininity in relation to other femininities. As Hood-Williams (2001: 42)
observes: ‘This raises . . . the question as to whether girls can ever be (or
“do”) masculine (or perform masculinity) and the tautological assumption
of this observation’. I examine the importance of challenging gender duality
in the following section. First, though, I return to the issue of stasis.
In order to address inequalities based on gender, race, class, sexuality and
generation, it is necessary to be attentive to the ways in which these
structural inequalities are reproduced through social action. The ‘doing
gender’ approach has potential precisely because it is a theory that situates
masculinities and femininities within hierarchies of structural position.
However, accounts that focus on the reproductive aspects of social action
Miller—The strengths and limits of ’doing gender’ 439
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
can speak to social change in only limited ways, typically by giving primacy
to structural changes that result in reconfigurations of agency and identity
(Emirbayer and Mische, 1998; Sewell, 1992). As noted, we need to theorize
structured action in ways that can sufficiently account for resistant, dynamic
and potentially transformative aspects of agency. One way to
improve our ability to do so is to ‘adopt a far more multiple, contingent,
and fractured conception . . . of structure’ (Sewell, 1992: 16).
Sewell (1992) proposes such a conceptualization of structure to account
for how the ordinary operation of structure through action can generate
change. His approach avoids a static model of the agency/structure nexus
and highlights potential avenues for transformative action, without ignoring
the import of structural inequalities in shaping agency. Moreover, it
provides a systematic means of addressing the intersecting and sometimes
competing nature of practices tied to such structuring positions as gender,
race, class, sexuality and generation.6
First, he highlights the multiplicity
of structures: ‘Societies are based on practices that derive from many
distinct structures, which exist at different levels, operate in different
modalities, and are themselves based on widely varying types and quantities
of resources’ (Sewell, 1992: 16). As a consequence, he argues that
individuals have the ability to draw from a wide array of schemas7
(including incompatible or contradictory ones) when engaging in social
action. For example, as I will illustrate below, one could consider the
schemas associated with gang membership as offering young women
the potential to engage in behaviours and construct identities that contradict
those schemas associated with a given definition of normative femininity.
Likewise, generational schemas (such as the acceptance of adolescence
as a time for ‘role experimentation’) can be adopted as accounts for the
abandonment of gang activities.
Sewell (1992: 18) defines agency as ‘entailing the capacity to transpose
and extend schemas to new contexts’. Here, for example, a schema that
shifts definitions of gendered behaviour in one context may be brought to
bear, in another context, with the potential for transformation. Importantly,
Sewell notes that whether and how schema are brought to bear in
various situations remains an empirical question that requires specific
investigation. In this process, he highlights the unpredictability of resource
accumulation: the enactment and application of schemas require validation
in the form of accumulated resources.8
Whether and how such resource
accumulation that occurs affects the reproduction or modification of
schema application in any given setting is, again, a question for empirical
investigation.
Such an approach provides a means of addressing the intersections of
gender, race, class, sexuality and generation, and allows for a broadened
and dynamic conceptualization of agency that can address the impact of
differences in power, access to and exclusion from resources, as well as the
role of potentially competing or contradictory schemas. As Sewell (1992:
20–1) argues:
Theoretical Criminology 6(4)440
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Agency is formed by a specific range of cultural schemas or resources
available in a person’s particular social milieu. . . . What kinds of desires
people can have, what intentions they can form, and what sorts of creative
transpositions they can carry out vary dramatically from one social world to
another depending on the nature of the particular structures that inform
those social worlds. Occupancy of different social positions — as defined,
for example, by gender, wealth, social prestige, class, ethnicity, occupation,
generation, sexual preference, or education — gives people knowledge of
different schemas and access to different kinds and amounts of resources
and hence different possibilities for transformative action.
Importantly, such an approach does not necessarily give primacy to
gender. Returning to the argument that ‘in all social situations we attempt
to adorn ourselves with culturally appropriate “female” or “male” fashion’
(Messerschmidt, 1995: 171), we see that even when attending to structural
and situational differences (for example, ‘white working-class boys’ or
‘urban African-American girls’), the primary concern remains the construction
of gender identities (and specifically identities constructed in reference
to gender norms). Primacy is given to the accomplishment of gender rather
than enactments founded on other bases of identity. This is not to suggest
that constructions of femininities and masculinities do not vary across
structural positions such as race, class, sexuality and generation, but to
suggest that primary attention to gender ultimately has the potential both
to elide other bases of identity and power, and, as noted earlier, to reify
gender difference (see Spelman, 1988). To examine these issues in detail, I
turn now to the dichotomous treatment of gender.
Challenging gender dualism
As noted, the problem of tautology is linked with the maintenance of
gender duality. In theorizing ‘doing gender’, we need to ensure that our
approach does not inadvertently continue to reify the dichotomous treatment
of gender by analysing gender exclusively through the ‘dynamics of
varied same-gender groups or styles’ (Thorne, 1993: 107). Selective attention
to difference results in the argument that women’s actions are always
an articulation of ‘femininity’ and men’s of ‘masculinity’, even when
oppositional or diverse femininities and masculinities are being described
and even when there is behavioural similarity across gender. In this
approach, as feminist sociologist Barrie Thorne (1993: 107) notes:
While the groups and subcultures are multiple, a sense of deep division
between boys and girls persist; how far such divisions may vary by situation
or context is not made clear. Dualistic assumptions poke through the
multiplicity. (Emphasis added.)
Miller—The strengths and limits of ’doing gender’ 441
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Thorne, in contrast, insists that while ‘gender categories, gender identities,
gender divisions, gender-based groups, gender meanings — all are
produced, actively and collaboratively, in everyday life’ (1993: 4), it is
nonetheless the case that gender ‘may be more or less relevant, or relevant
in different ways, from one social context to the next. . . . Gender is not
always at the forefront of . . . interactions’ (1993: 5, 29). I return to the
implications of this statement with regard to integrating race, class, sexuality
and generation into the analysis below. First, through an illustration
from my own work, I examine why the relevance and significance of gender
— and the existence of gender differences — cannot be assumed at the
outset. If we only seek to uncover gender differences — and even if we are
examining race, class or culturally based differences within femininities and
masculinities — we will not be able to attend to commonalities across
gender, or instances of what Thorne (1993: 121–34) calls ‘the continuum of
[gender] crossing’ — seeking access to and participation in the activities
and groups of the other gender.
In a seemingly controversial move, I titled my recent (2001) book on
young women in gangs One of the Guys. Some may find it antithetical for
a feminist scholar to describe gang girls as constructing a masculine
identity, as it would appear to echo the antiquated ‘tomboy’ stereotype that
feminist research has repeatedly challenged (see Campbell (1984) for an
overview). This was not my intent when I embarked on the research. I
hoped to find those gang girls other feminist scholars had described: girls
who expressed the importance of ‘sisterhood’ and close familial-like bonds
with other young women in their gangs (see Joe and Chesney-Lind, 1995;
Lauderback et al., 1992). But empirical discovery led me elsewhere. The
phrase ‘one of the guys’ highlights the gender crossing that occurs in some
instances within gangs, and underscores the gender and gang identities
prevalent among the young women I spoke with. With a few notable
exceptions, these young women heavily identified with the young men in
their gangs and described these groups as masculinist enterprises. To be
sure, ‘one of the guys’ is only one part of a complex tapestry of gender
beliefs and identities held by the gang girls I spoke with — and is rarely
matched by gendered actions — but it remains significant nonetheless.
The vast majority of gang-involved girls in my study were in integrated,
mixed-gender gangs in which the majority of members were males. They
were not, then, in single-gender groups affiliated with one another, or in
independent female gangs. All of the gangs I studied were predominantly
African-American, though a small proportion of the girls I spoke with were
white members of these groups. With regard to gender ideologies, I found
a significant collection of contradictions. On the one hand, most girls
vigorously held to a belief that gender equality was a normative feature of
their gangs, but they also described a distinct gender hierarchy within their
gangs (which young women themselves often upheld) that included male
leadership, a double standard with regard to sexual activities, the sexual
Theoretical Criminology 6(4)442
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
exploitation of some girls, and most girls’ exclusion from serious gang
crime (Miller, 2001).
With regard to the norm of gender equality, girls’ discussions were not
about suggesting that all women should be treated equally, but about
differentiating themselves from other young women through a construction
of ‘one of the guys’. It was specifically girls’ success at gender crossing that
designated them as equals. Even within their gangs, status hierarchies
among girls were evident and dictated in part by how successfully girls
could resist gender typecasting and cross into boys’ terrain. Clearly, part of
what young women were doing was constructing an identity in opposition
to other normative constructs of femininity. But they were ‘crossing’ gender
to do so. As Latisha explained: ‘I was the girl who done everything the
dudes done. I wasn’t scared of nothing. I was just like, I was just like a dude
in a girl’s body’ (emphasis added). Likewise, Trina explained, ‘They [gang
members] just treated me like a little boy. . . . I got to do what dudes
usually do. I got to be like dudes.’ These girls’ accounts do not simply
reflect the construction of a ‘bad girl’ femininity that is differentiated from
other femininities; instead, they reflect gender crossing, embracing a masculine
identity that they view as contradicting their bodily sex category (that
is, female).
Young women’s identification of gangs as masculinist enterprises also
came through in their discussions of all-female gangs. Veronica, for instance,
described an all-female gang she was familiar with as ‘stupid’, and
said the boys referred to it as ‘pussy-infected’. She explained, ‘They try to
have their own little girl group goin’ on. [Laughs] It was silly.’ LaShawna
said she had not heard of any female gangs, and if she did, she would
‘probably laugh or something’. Asked why she would laugh, she explained,
‘Cause! What they gonna do? They can’t do nothin’ about it. Nothin’
about nothin’! They probably could be hard or whatever, but they wouldn’t
have no props. They wouldn’t get no props.’ While for at least a moment
LaShawna was torn and admitted that an all-female gang could be ‘hard’,
her general reaction, like Veronica’s, was to laugh. Part of the reason that
both found the notion of an all-female gang ‘silly’, to use Veronica’s term,
was because they believed that without males, the group would not be
respected. This is because, for the most part, gangs in their communities
(and their own gangs) were primarily male and were male-dominated in
structure, composition and status orientation. Thus, young women’s efforts
at identity construction as ‘one of the guys’ made sense within the specific
gang contexts in which they were involved.
In examining gender as situated action, then, it is necessary to be
attentive to the possibilities of ‘gender crossing’ and similarities across
gender, in addition to gender difference. This case illustrates several
additional noteworthy issues: the significance of organizational structures
in shaping gender dynamics and identities, and the intersections of race and
gender in constructions of gang girl identity (see below). In an important
Miller—The strengths and limits of ’doing gender’ 443
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
article on how organizational structure shapes gender dynamics, sociologist
Rosabeth Moss Kanter (1977: 967) suggests:
In both macroscopic and microscopic analysis, sex and gender components
are sometimes confounded by situational and structural effects. . . . Conclusions
about ‘women’s behavior’ or ‘male attitudes’ drawn from such situations
may sometimes confuse the effect of situation with the effect of sex
roles [sic]; indeed such variables as position in opportunity and power
structures account for a large number of phenomena related to . . . behavior
that have been labeled ‘sex differences’.
Kanter’s research is particularly relevant to my findings with regard to
girls in gangs (see Peterson et al., 2001), because she emphasizes that the
relative numbers of males and females ‘are critical in shaping interactional
dynamics’ within groups (Kanter, 1977: 965; see also Konrad et al., 1992).
In skewed groups, where females are a minority and males a majority,
Kanter suggests that women typically adopt two strategies for assimilation:
overachievement according to the masculine standards of the group; and
attempting to become ‘socially invisible’, to ‘minimize their sexual attributes
so as to blend unnoticeably into the predominant male culture’
(1977: 974).
Likewise, in her analysis of gender crossing among elementary students,
Thorne (1993) notes that when youths successfully cross into the activities
of the other gender:
Gender remains relatively low in salience; the gender tokens . . . participate
on the terms of the majority, and not as ‘the other sex’. This can only be
accomplished if gender-marking is minimized and heterosexual meanings are
avoided. (Emphasis added.)
Thus, it is not surprising that many of the young women I spoke with —
particularly those in skewed groups with a majority of male members
— identified with what they described as the masculine orientation of the
gang and strove, in certain circumstances (but by no means all), to be ‘one
of the guys’. Our interviews with male gang members offer similar evidence
(Miller and Brunson, 2000). Young men in majority male gangs with one
or a handful of female members described these young women as essentially
‘token’ or ‘honorary’ males. For instance, Doug described one girl as
garnering a lot of respect because ‘she’s got a male’s mentality. She acts like
a male, she just be down for whatever’. Likewise, Robert explained: ‘Tia’s
not a regular girl, she like a boy for real. . . . She just like the dudes for
real.’ As with the girls’ accounts, these young men did not view the girls in
their gangs as enacting a ‘bad girl’ femininity, but a masculinity that was
incongruent with their sex. West and Zimmerman (1987: 139) note that
there are circumstances in which ‘parties reach an accommodation
that allows women to engage in presumptively masculine behavior’. These
findings highlight the salience of gender crossing, and illustrate some of the
spaces that exist for commonalities across gender.9
Theoretical Criminology 6(4)444
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Accounting for stratification, hierarchy and power
Messerschmidt’s (1993) original formulation in Masculinities and Crime
was developed specifically to describe the relationship between ‘doing
masculinity’ and ‘doing crime’. Part of the difficulty feminist scholars have
had in translating this theoretical perspective to explanations of female
offending is in theorizing women’s participation in presumably ‘masculine’
crime, except through constructs such as ‘oppositional’ or ‘bad girl’
femininity (Messerschmidt, 1995; Portillos, 1999). As noted earlier, giving
primacy to normative aspects of gendered action limits our ability to
grapple fully with power and inequality, and address how agency can be a
response to structural or situational exclusion rather than adherence to
norms. Because ‘gender relations and constructs of masculinity and femininity
are not symmetrical but are based on an organizing principle of
men’s superiority and social and political-economic dominance over
women’ (Daly and Chesney-Lind, 1988: 504), it is unlikely that the same
patterns and meanings of gender construction will apply, in a generalized
way, across gender.
With regard to ‘gender crossing,’ for instance, the asymmetry of constructs
of femininity and masculinity mean that there are greater rewards
and incentives for women to ‘cross’ into culturally defined masculine
terrain than there are for men to cross into feminine terrain (though
certainly this occurs as well; see Chauncey (1994) and Kulick (1998) for
examples). Thorne’s (1993) work again is instructive. Comparing ‘tomboy
lore’ with ‘sissy lore’, she highlights the ways in which girls’ gender crossing
is more likely to result in status and autonomy, whereas boys’ is more likely
to result in stigma and censure. This is precisely because of the asymmetry
of the cultural categories of ‘masculine’ and ‘feminine’ (even recognizing
that there are masculinities and femininities). In her study of elementary
students, it was only boys with extensive masculine social resources
(popularity or excellence in sports) who could ‘cross’ into girls’ activities
without facing denigration. Few were motivated to do so. Dominant, even
situationally specific, notions of masculinity are more heavily policed for
males than dominant notions of femininity are policed for females, precisely
because of the devaluation of femininity (Thorne, 1993: 120; see also
Segal, 1990).
The devaluation of cultural constructs of femininity is readily apparent
within the majority-male, mixed-gender youth gangs I investigated. Given
this, it makes sense that young women would strive to adopt a gang
identity as ‘one of the guys’, particularly given the status and respect
available within gangs to youths who exhibit characteristics typically
associated with the highly valued cultural construction of gang masculinities.
In doing so, they attempt to resist others’ attribution to them of
devalued feminine characteristics. Young women’s heavy policing of one
another’s sexuality within these gangs makes sense in this context as well —
not as a means of constructing a gang-specific femininity, but as a way to
Miller—The strengths and limits of ’doing gender’ 445
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
minimize or downplay gender-marking and heterosexual meanings. These
functioned to perpetuate girls’ devaluation and subordination in the gangs
I studied.
For example, when gang girls attempt to enforce a gendered sexual code
that requires serial monogamy for girls, this does in fact result in gender
difference, as Messerschmidt (1995, 1997) notes, because gang boys have
considerably wider latitude in acceptable sexual behaviours. Ironically,
however, this was not the goal of such policing among the young women I
spoke with. Instead, they were attempting to minimize gender difference by
limiting the extent to which boys could apply derogatory sexual labels to
the girls in the gang. Thus girls’ policing of one another’s sexuality — and
their vilification of girls they deemed to be ‘hos’ and ‘sluts’ — allowed them
to distance themselves from a denigrated sexual identity and maintain an
identity as a ‘true’ member. To illustrate, let me return to Latisha, who
described herself as ‘like a dude in a girl’s body’. She continued:
We just like dudes to them [male gang members]. We just like dudes, they
treat us like that ’cause we act so much like dudes they can’t do nothing.
They respect us as females though, but we just so much like dudes that they
just don’t trip off of it. (Emphasis added.)
Latisha’s meaning of ‘respect’ here is specifically that boys do not overemphasize
gender-marking and heterosexual meanings — they do not
interact with girls who are ‘one of the guys’ in overly sexualized ways. In
this way, the girls’ orientation towards sexuality is a response to gender
inequalities rather than simply a situationally-specific normative construction
of femininity.
Thus far, I have focused specifically on one example to illustrate the
import of challenging gender duality. These findings provide evidence of the
need to examine group structures and gender composition when investigating
girls’ constructions of gender identities within gangs10
(see also JoeLaidler
and Hunt, 1997; Peterson et al., 2001). As further evidence, it is
worth noting that the few girls I interviewed who were in all-female gangs
or gangs with a sizeable proportion of female members were more likely to
describe valuing their friendships and relationships with other girls in their
gangs. This is in keeping with previous feminist scholarship on girls
in gangs, which describes female gang members providing mutual support
and sharing familial-like bonds with one another (Joe and Chesney-Lind,
1995; Lauderback et al., 1992). Moreover, these young women did not
situate themselves as ‘one of the guys’ in their gangs, but had gender
identities that are perhaps more in line with Messerschmidt’s ‘bad girl’
femininity.11
As noted earlier, our examination of ‘doing gender and crime’
is further strengthened by examining the intersecting nature of structural
positions in shaping both constructions of gender and access to resources
and power. As Messerschmidt (2002: 16) notes:
To understand crime, we must comprehend how gender, race, and class
relations are part of all social existence, and not view each relation as
Theoretical Criminology 6(4)446
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
extrinsic to the others. Because crime operates through a complex series of
gender, race, and class practices, crime usually is more than a single
activity.
Continuing my focus on young women, gangs and gang structure,
research suggests that both gang identities and the gendered structures of
gangs are tied to race and ethnicity. For example, Chicana and Latina gang
members are those most likely to describe their gangs as female groups
affiliated with male gangs, while African-American young women are more
likely to describe their gangs as gender integrated in structure. The handful
of all-female gangs documented by scholars have largely been AfricanAmerican
as well (Curry, 1997; Joe-Laidler and Hunt, 1997; Lauderback et
al., 1992; Miller, 2001; Venkatesh, 1998). There also is evidence that race,
ethnicity and gender intersect in shaping gang girls’ identities. Whereas the
girls in my study viewed violence as normative and status-enhancing
aspects of their gang involvement, Joe and Chesney-Lind’s (1995) study of
Samoan and Filipino gang girls in Hawaii suggests that participation in
violence is a stronger normative feature of male gang involvement than it is
for gang girls in these ethnic groups. They argue that for girls, ‘violence
(gang and otherwise) is not celebrated and normative; it is instead more
directly a consequence of and a response to the abuse, both physical and
sexual, that characterizes their lives at home’ (Joe and Chesney-Lind, 1995:
428).
Portillos’ (1999) study of Chicana gang girls indicates that gender
identities in these groups are constructed in part around difference. For
instance, girls sometimes dress in ways that distinctively highlight their
sexuality, and also use their sexuality for gang purposes — to set up rival
gang members for confrontations with males. These types of behaviours
were avoided by the young women I spoke with, for whom such activities
contradicted their identity as ‘one of the guys’. Thus, in contrast to my
portrait of African-American girls in majority-male gangs, Portillos describes
Chicana gang girls constructing an oppositional femininity that is
clearly differentiated from (and subordinate to) male members’ gang
masculinity.
Schemas of gender duality, in fact, appear to be at least partially shaped
by racial identity. Owing not just to historical legacies of slavery, but also
contemporary conditions in urban African-American communities, black
women have occupied social positions that require independence — economic
and otherwise — for survival. Moreover, they face entrenched
controlling images of black female sexuality as promiscuous (Collins,
1990). The young women I interviewed bore witness to the sexual and
other abuses of women around them in their communities. Thus, the
rejection of gender duality and construction of gender identity as ‘one of
the guys’ was an attempt to insulate themselves from these practices and
resist negative consequences of their sex category, and was a more available
schema given larger practices and acceptance of economic self-sufficiency
Miller—The strengths and limits of ’doing gender’ 447
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
and independence among poor and working-class African-American
women.
On the other hand, gender dualism is a schema that has been entrenched
in white discourses on womanhood, offering white women both protection
and privilege, and structuring dependency and subservience (see Spelman,
1988). Thus, the passive feminine ideal linked to gender dualism is ‘much
more relevant (and restrictive) for white’ than African-American females
(Simpson and Elis, 1995: 71). White women can call upon this particular
schema to deflect responsibility for their participation in crime. For example,
in a study of gender and the accomplishment of robbery, I found that
the handful of white women interviewed for the project described themselves
as mere accomplices in armed robberies and argued that they would
not participate were it not for their African-American boyfriends (Miller,
1998). This was despite the fact that they wielded guns on victims and
engaged in these acts on multiple occasions.
It is also perhaps notable that the only young woman I interviewed in the
gang study who articulated a ‘gender difference’ ideology in describing her
gang was white. Describing herself as a ‘Lady Crip’, Diane compared gang
organization to traditional patriarchal family organization: ‘Like in a
family: in a regular family there might be the dad and four brothers, and
the mom and three sisters. . . . When mom says do somethin’ but dad’s over
mom and dad says “no you do this,” then it all goes back to dad, you see?’
Ironically, she was in fact one of the most hardcore gang members I
encountered — she and the few other girls in her gang routinely engaged in
serious offending. She was also the only white member of her gang. She
said people who did not know her often attempted to make race a salient
issue, but she rejected these arguments and believed her gang identity had
more primacy than her racial identity on the streets:
When I first started comin’ here I really had to fend for mine ’cause I never,
I mean, I’ve know that I’m white and I’m proud that I’m white and like all
that stuff. I don’t act black or don’t wanna be black or none of that stuff. . . .
When I first came here I always had to defend myself ’cause I’m white. But
I’m not a punk. I’m not gonna go out like a little, a little white perky girl or
whatever they wanna call it. . . . I really don’t see anything now because ever
since I was younger I’ve always been in black neighborhoods and been
around black people and sometimes people, sometimes people be like, ‘aw,
she just a little punk’. But they find out. I’m different. Sometimes people like
to stereotype. You do this, you’re actin’ black. If you’re doin’ this you’re
actin’ white. I think that’s a stereotype. I don’t think there is no actin’ black
or actin’ white. I’m a Crip.
The findings highlighted here are indicative of the interconnections of
race, ethnicity and gang structure in shaping girls’ gender identities within
gangs. They also illustrate why it is important to be attentive, not just to
gender difference, but also gender similarities and instances of ‘gender
crossing’, as these findings suggest that such gender overlap is linked to
Theoretical Criminology 6(4)448
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
multiple sites of positioning (see also Moore and Hagedorn, 1996). Thus,
as Cerulo (1997: 386) argues, it is necessary to:
Challenge the dualistic, oppositional nature by which gender is traditionally
framed. . . . Elements such as race and social class produce multiple
variations of ‘women’ and ‘men,’ distinctions that many societies use to
build complex hierarchical stratification systems. The existence of these
multiple categories alerts us to the flaws of binary gender conceptualizations,
focusing us instead on the ways in which multiple identity affiliations
qualitatively change the nature of human experience.
Such hierarchical stratification systems are also constructed and reproduced
in the urban street milieu, narrowing women’s options for social
action and placing limitations on their available choices and activities
(Anderson, 1999; Maher, 1997). Attention to the factors that structure
opportunities for involvement in street networks highlights further why
normative aspects of femininity are insufficient to account for women’s
involvement in crime. Messerschmidt (1995: 178), for example, suggests
that prostitution is a ‘principal criminal resource’ for young women, not
just for making money, but for ‘doing difference’ — constructing a
situationally specific femininity. While the result of this situated action is
the perpetuation of ‘gender difference’, it is more problematic to argue that
women engage in street-level sex work as a means of constructing a
feminine identity, particularly given the dangers and stigma attached to
such activities. Rather, gender narrows those options available to women
on the streets, making sex work one of the few money-generating activities
open to women in certain contexts (see Maher, 1997). To examine the
significance of this for understanding ‘doing gender and crime’, I consider
how we may better conceptualize the complexities of agency so as to
specify better the nature of the situated action.
Conceptualizing the complexities of agency and social
practices
Lisa Maher’s (1997) Sexed Work provides a nuanced account of the
interplay of race and gender in structuring women’s activities in a drug
economy. Maher specifically examines how women’s actions take place
within complex relations of gender and race and in the context of a rigid
division of labour. Important for the current discussion, Maher’s work
provides a model for linking micro-processes — in this instance, of a local
drug economy, gender and race — to their larger structural underpinnings.
She highlights both the reproductive aspects of social action by documenting
how ‘gender and race are conceived and structured within broader
social, cultural and economic spaces’ (Maher, 1997: 206), but also documents
sites of resistance that play themselves out, often in small ways,
in women’s action. For example, Maher adopts the term ‘viccing’ to
describe women’s robbery of clients in the sex trade, and documents its
Miller—The strengths and limits of ’doing gender’ 449
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
proliferation as a form of resistance against women’s greater vulnerability
to victimization and against cheapening sex markets within the drug
economy.
Maher’s examination of the impact of structures of racial and gender
exclusion and differential allocation of resources in the drug economy
highlights that ‘any theory of agency must be placed in the context of
structural, institutional or intersubjective constraints’ (McNay, 2000:
20–1). In addition, in our investigations of doing gender, we must strive to
disaggregate agency into its component parts and varied dimensions. Social
theorists have conceptualized individual action in a range of ways. These
include a focus on rational action that is goal-seeking, purposeful and
deliberate; on norm-oriented action or that which is taken in response to
normative expectations; on actions that may be viewed as ‘resistance,’
negotiation or rebellion against norms or inequalities; as well as routinized
actions that are largely unreflective or taken for granted (see Emirbayer and
Mische, 1998). Recent attempts to theorize agency suggest the necessity to
offer ‘a more precise and varied account of agency’ (McNay, 2000: 4) that
distinguishes its various dimensions. Such an approach helps ‘explain the
differing motivations and ways in which individuals and groups struggle
over, appropriate and transform cultural meanings and resources’ (McNay,
2000: 4) and helps ‘account for variability and change in actors’ capacity
for imaginative and critical intervention in the diverse contexts within
which they act’ (Emirbayer and Mische, 1998: 970).
Integrating the features of agency noted above, Emirbayer and Mische
(1998) argue that a focus on the temporal dimensions of agency provides a
means of unpacking and understanding agency’s various dimensions. They
distinguish between three general temporal elements. First, the iterational
element, built upon past patterns, includes ‘habitual, unreflected, and
mostly unproblematic patterns of action by means of which we orient our
efforts in the greater part of our daily lives’ (Emirbayer and Mische, 1998:
975). This aspect of agency typically involves the invocation of classificatory
schemes (such as those based on gender, race and so forth) and
‘cultural competences’ informed by structural locations. Second is the
projective element: ‘the imaginative generation by actors of possible future
trajectories of action in which received structures of thought and action
may be creatively reconfigured in relation to actors’ hopes, fears, and
desires for the future’ (Emirbayer and Mische, 1998: 971). Here the various
schemas available to the actor based on structural positions may be
brought to bear for the purpose of creative reconfiguration of action, as
illustrated above in the case of girls in gangs. Finally, Emirbayer and
Mische (1998: 971) describe the practical-evaluative element of agency:
‘the capacity of actors to make practical and normative judgments among
alternative possible trajectories of action, in response to the emerging
demands, dilemmas, and ambiguities of presently evolving situations.’
With regard to gendered action, Emirbayer and Mische’s iterational
element is primarily a site of reproduction, in which identities are naturalTheoretical
Criminology 6(4)450
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
ized in individual action. Consider the original ethnomethodological studies
on gender, which focused specifically on gender accomplishment among
transsexual individuals (see Garfinkel, 1967; Kessler and McKenna, 1978).
These studies focused on transsexuals’ need to learn and practice doing
gender so that their gender displays would be read by others in ways that
would allow appropriate gender attribution. Kessler and McKenna (1978:
157) highlight the interactive nature of this process:
The gender attribution process is an interaction between displayer and
attributor, but concrete displays are not informative unless interpreted in
light of the rules which the attributor has for deciding what it means to be
a female or male. As members of a sociocultural group, the displayer and the
attributor share a knowledge of the socially constructed signs of gender.
They learn these signs as part of the process of socialization (becoming
members).
But they also argue that once transsexual individuals succeed in learning
how to display gender, their gendered actions are less likely to involve a
heightened awareness in doing so: ‘It is not the particular gender which
must be sustained, but rather a sense of its “naturalness”. . . . In ongoing
interactions, once a gender attribution has been made, it is no longer
necessary to keep “doing male” or “doing female”’ (Kessler and McKenna,
1978: 159). Thus Kessler and McKenna raise a critical issue concerning the
extent that women’s and men’s gendered actions are often taken for
granted, rather than enacted as a means of accomplishing gender.
In fact, Messerschmidt takes this issue into account with his refinement
of the concept of ‘masculinity challenges’. He notes that ‘the taken-forgranted
masculinity of a man or boy can be challenged in certain contexts’
(Messerschmidt, 2000: 13), and describes these challenges as ‘contextual
interactions that result in masculine degradation’ (Messerschmidt, 2000:
13). Because such actions make gender explicitly salient, they are likely to
‘motivate social action toward masculine resources . . . that correct the
subordinating social situation, and various forms of crime can be the result’
(Messerschmidt, 2000: 13). It is clear nonetheless that ‘doing gender’ can
reproduce social structure via individual action, even when the accomplishment
of gender does not occur at the conscious level or as an explicit goal.
The example of sex work is one example. There is little evidence women
choose to engage in sex work as a means of accomplishing femininity,
but it remains the case that their actions help to reproduce gendered social
structures. This results from patterns of behaviour informed by
social position and based on available resources for action.
One means of getting at this facet of gender and gendered reproduction
is to focus, not on individual action and its goals and motives, but instead
on the gendering of social practice. Bottcher’s (2001) analysis of gender and
delinquency provides an example of such an approach. Based on comparative
interviews with male and female siblings, her unit of analysis is social
practice. She shows that the practices of everyday life reveal gendered
Miller—The strengths and limits of ’doing gender’ 451
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
patterns ‘that intertwined with delinquent activities, constraining female
delinquency while enabling and rewarding male delinquency’ (Bottcher,
2001: 893). Moreover, she avoids the limitations of gender duality by
demonstrating that these gendered patterns are not universally applicable
to all males or all females:
Some male-typed social practices appear to encourage or enable delinquent
activity for either sex. Conversely, some female-typed social practices appear
to discourage delinquent activity for both sexes. Thus, the social practices of
gender disclose social conditions and activities that influence delinquent
involvement, regardless of sex.
(Bottcher, 2001: 904)
This approach offers a promising avenue for the study of doing gender,
particularly when coupled with analyses of agency and grounded in how
gendered practices may be shaped by other social positions such as race,
class and generation.
Likewise, Emirbayer and Mische’s (1998) practical-evaluative element of
agency reveals that women’s situated action is as much a response to and
negotiation with gender inequality as it is a resource for accomplishing
gender. To capture this phenomenon, sociologist Deniz Kandiyoti (1988)
coined the phrase ‘bargaining with patriarchy’, highlighting women’s strategies
of action as they arise within particular sets of gendered constraints.
She notes: ‘Different forms of patriarchy present women with distinct
“rules of the game” and call for different strategies to maximize security
and optimize life options with varying potential for active or passive
resistance in the face of oppression’ (Kandiyoti, 1988: 274). Kandiyoti’s
work suggests that women’s situated actions can be examined as gender
strategies for navigating within male-dominated terrains. Applying this to
doing gender and crime means being attentive to the reciprocal relation of
gender and crime. For example, rather than crime simply being a resource
for accomplishing gender, the converse may also be true: gender may be
used as a resource for women to accomplish their participation in and
avoidance of crime.
One type of gender strategy is to draw from cultural beliefs about
femininity in order successfully to engage in or avoid particular types of
crime. I found this to be the case with the gang-involved girls I spoke with.
For instance, Vashelle suggested that when she sold drugs, she purposely
avoided doing so with groups of young men on the street corner. This is the
approach routinely adopted by male street-level drug sellers, and young
women such as Vashelle were aware that such an approach intensified
police scrutiny (Jacobs, 1999; Jacobs and Miller, 1998). In contrast, she
described using more discreet methods, drawing from police officers’ lack
of suspicion concerning female sellers to accomplish her drug sales suc-
cessfully:
The police, they don’t be on the girls for real, females, but if they see a
whole crowd of niggers sitting out, they gonna get down on them. But I’m
Theoretical Criminology 6(4)452
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
saying if there are niggers out there and I’m with them too they gonna
shake me too. If I’m walking up the street by myself they ain’t gonna trip
off me cause I’m a gal but a crowd of niggers just walking, they gonna get
on them.
Likewise, girls in my study recognized that they could use their presence to
shield boys in the gang from police suspicion. Tonya explained:
Like when we in a car, if a girl and a dude in a car, the police tend not to trip
off of it. When they look to see if a car been stolen, police just don’t trip off
of it. But if they see three or four niggers in that car, the police stop you
automatically, boom. . . . [Girls have] little ways that we got to get them out
of stuff sometimes, we can get them out of stuff that dudes couldn’t do, you
know what I’m saying.
Thus, gender — and gender stereotypes — were often resources that
young women drew from to facilitate the success of gang members’ crimes.
Likewise, young women routinely articulated the position that male members
tend to be ‘harder’ than females. Despite the incompatibility of this
position with their assertion of equality and status as ‘one of the guys’, it
nonetheless provided a number of advantages for gang girls. It meant
accepting protection from male members of their gangs in recognizably
dangerous environments; it furnished a justification for avoiding or limiting
participation in those aspects of gang involvement that were dangerous or
morally troubling; and lastly, it allowed young women to view the gang as
less central to their long-term life plans and, instead, to define their gang
involvement as a primarily adolescent commitment. Thus girls were able to
cull from various schemas to construct contingent identities (‘one of the
guys’ versus traditional gender schemas such as ‘boys are harder’) as the
situation warranted and as best benefited them in a given situation.12
The utility of examining women’s situated actions as gender strategies in
male-dominated contexts is also illustrated by a comparative study of
women’s and men’s participation in street robbery I completed several years
ago (Miller, 1998). While respondents’ discussions of their motives for
engaging in robbery reflected similarity across gender (and thus did not
appear to be based on normative features of masculinity or femininity),
there were clear differences in how women and men accomplished robberies.
These differences resulted from women’s strategic choices in the context
of gender-stratified environments. Men in the sample described committing
robberies in a strikingly uniform manner, using guns and physical violence
or its threat in close proximity to their victims. In contrast, women’s
techniques for committing robberies varied considerably depending upon
whether they were targeting male or female victims, and whether they were
working with male accomplices.
The most obvious illustration of using gender as a criminal resource was
women’s robberies of men. These events involved the use of a gun and
feigned sexual interest in the victim. They used men’s assumptions that
Miller—The strengths and limits of ’doing gender’ 453
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
women could be taken advantage of sexually in order to accomplish the
robbery. Quick explained:
They don’t suspect that a girl gonna try to get ’em. You know what I’m
saying? So it’s kind of easier ’cause they like, she looks innocent, she ain’t
gonna do this, but that’s how I get ’em. They put they guard down to a
woman. . . . Most of the time when girls get high they think they can take
advantage of us so they always, let’s go to a hotel or my crib or some-
thing.
Men’s actions, and their attitudes about women, thus made them vulnerable
targets. It is also notable that only one woman described committing
a robbery by herself in the prototypical masculine style. Significantly, she
engaged in ‘gender crossing’ to do so, not just by adopting a masculine
robbery style, but by dressing and comporting herself in such a way as to
conceal her gender from the victim. These findings suggest that it is not
sufficient to examine women’s crime as a means of accomplishing femininity;
instead, women also react to and strategically draw from normative
beliefs about femininity in order to accomplish crime.
In examining gender as situated action and theorizing its relationship to
participation in violence and crime, the recent attempts social theorists
have made to conceptualize the complexities of agency offer intellectual
promise for expanding our investigations. Though I have presented individual
examples of studies that illustrate various facets of gendered social
action, a fruitful approach for future research will involve the integration
and unpacking of the various features highlighted here. As Emirbayer
and Mische (1998: 973, 1005) suggest:
A disaggregated conception of agency . . . allows us to locate more precisely
the interplay between the reproductive and transformative dimensions of
social action and to explain how reflectivity can change in either direction,
through the increasing routinization or problematization of experience. . . .
The empirical challenge becomes that of locating, comparing and predicting
the relationship between different kinds of agentic processes and particular
structuring contexts of action.
Concluding remarks
In Gender: An Ethnomethodological Approach, Kessler and McKenna
(1978) argue that the gender attributor plays just as important a role in
doing gender as the gender displayer. More recently, theorists have discussed
this in terms of accountability:
Because individuals realize their behavior is accountable to others, they
configure and orchestrate their actions in relation to how they might be
interpreted by others in the particular social context in which they occur.
(Messerschmidt, 1995: 172)
But Kessler and McKenna (1978: 157) go further in their analysis of
Theoretical Criminology 6(4)454
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
the role of those who attribute gender: ‘the attributor contributes to the
accentuation of gender cues by selective perception’. Moreover, they argue
that social researchers are not exempt from this process: ‘Our seeing of two
genders leads to the “discovery” of . . . social differences’ (Kessler and
McKenna, 1978: 163). Consequently, in addition to the need to recognize
gender as situated accomplishment, they argue that we need to recognize
the primacy of gender attribution, including in our own research.
My goal in this paper has been to suggest some of the ways that our
attribution of gender difference has limited our uses of the doing-gender
model for understanding and theorizing about gender and crime. Ultimately,
I suggest that a dualistic model of gender limits our ability to
address the ways in which social positioning based on such factors as
gender, race, class, sexuality and generation both intersect in the construction
of identity and also offer contradictory schemas for identity construction
and action. This means that the salience of various facets of
identity — based, for example, on gender or race — can take primacy in
some circumstances and not in others. In addition, I have suggested that the
doing-gender approach has the danger of slipping into tautology when
theorizing is not explicitly grounded in empirical investigations of individuals’
constructions and beliefs about the role gender plays in their
activities, and when we presume actions are always undertaken with
specific reference to accomplishing normative gender. This is exacerbated
when evidence of gender crossing is downplayed or is interpreted only in a
dualistic way, such as the assignment of femininity only to women and
masculinity only to men.
In addition, because gender is an asymmetric construct in culture and
social structure, we cannot assume that ‘doing gender’ will occur in a
symmetrical way across gender. Kandiyoti’s concept of ‘bargaining with
patriarchy’ illustrates the need to examine gendered actions as they are
situated within particular types of gendered constraints. This provides a
useful means of examining gender as situated action in a reciprocal way. It
allows us to explore not just how action is sometimes guided by the goal of
enacting gender, but also how women accommodate and adapt to specific
forms of gender stratification and inequality in their enactment of crime.
Moreover, even when ‘doing gender’ becomes naturalized, it nonetheless
provides evidence of the reproduction of gender inequality in practice (see
Maher, 1997).
I suggest several avenues out of this conundrum, including reconceptualizing
social structure to recognize that it is ‘multiple, contingent, and
fractured’ (Sewell, 1992: 16). This approach allows us to address social
action, not just in its reproductive capacity, but also in its transformative
aspects. In addition, I draw from recent social theory on agency to argue
the benefits that emerge from adopting a dynamic conceptualization of
agency that disaggregates various facets of social action. Such an approach
provides promise for expanding our investigation of gendered social action
and the accomplishment of crime. Ultimately, our best prospects will come
Miller—The strengths and limits of ’doing gender’ 455
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
from rigorously grounding our theorizing in comparative scholarship (see
Simpson, 2000) that includes both observational work and the investigation
of actors’ accounts of their activities and identity construction. Such an
approach allows us to ‘leav[e] open the issue of empirical gender difference
. . . [and] attend to multiple differences and sources of commonality’
(Thorne, 1993: 108), while grounding our studies in the intersecting
positions of gender, race, class, sexuality and generation. Attention to these
issues maximizes the significant insights to be garnered from the structured
action approach.
Notes
A version of this paper was presented at the 2000 annual meeting of the
American Society of Criminology. I appreciate the detailed feedback I received
at that time from Jim Messerschmidt and Nancy Jurik. Thanks also to
Christopher Mullins and the anonymous reviewers at Theoretical Criminology
for their comments and suggestions.
1. This work does not attempt to summarize and synthesize the large and
diverse body of feminist criminological investigations of gender and crime.
The goal is to assess and refine one facet of feminist theorizing —
structured action theory — and thus I draw primarily on qualitative
investigations of this approach.
2. While this approach is a general theory used to describe a variety of types
of crime, including white collar and corporate crime and sexual crimes
(Messerschmidt, 1997: 2000), my focus here will be on street crime.
3. See Cerulo (1997) and Orbuch (1997) on the utility of accounts-based
analyses of identity and action.
4. Consider, for example, the concept of ‘doing race’. An emphasis on
normative features of racialized behaviour clearly is lacking when it is not
explicitly informed by structural inequalities, power relations, differences
in access to resources, and privileges resulting from whiteness. The same
should be true when examining gender: an overemphasis on normative
features of behaviour has the potential to eclipse issues of power and
inequality.
5. Messerschmidt (2002) later argues this point himself in a critique of
Jefferson’s work on masculinities: ‘without empirical verification, literally
anything could be defined as discourse depending on how the theorist
chooses to interpret it’.
6. While it is the case, as feminists have argued (see Collins, 1990; Spelman,
1988), that structural positions such as race and gender are intersecting —
such that there are racialized gender identities and gendered racial identities
— it is also the case that, in a given situation, a particular axis of
identity may take primacy. I discuss this further below.
7. In defining ‘schemas’, Sewell (1992: 8) draws from Giddens’ (1984)
conception of rules, and refers to these as ‘fundamental tools of thought,
Theoretical Criminology 6(4)456
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
but also the various conventions, recipes, scenarios, principles of action,
and habits of speech and gesture built up with these fundamental tools’.
With regard to gender, for instance, we could consider these as conventional
or normative expectations associated with masculinities and
femininities.
8. Resources here include material objects, social capital, as well as forms of
interpersonal status that result in the maintenance or enhancement of
power.
9. Nonetheless, it is worth noting that young women’s ‘crossing’ in this
instance does not challenge gender boundaries or dualisms. And, despite
girls’ identity constructions as ‘one of the guys’, there remained vast
differences by gender in participation in the most serious forms of gang
violence (Miller and Decker, 2001), as well as other gang activities: ‘one of
the guys’ is more ideology than practice in these girls’ gangs. As such, it is
typically not the actualization of ‘masculinity’ in practice, but a contingent,
situational masculine gender orientation. Nonetheless, as Thorne (1993:
134) notes: ‘incidents of crossing may chip away at traditional ideologies
and hold out new possibilities’.
10. Though my focus here is on gangs, it is nonetheless the case that the
approach I call for — avoiding selective attention to gender difference, and
seeking explanations for gender difference in multiple arenas, rather than
assuming it is based on gender norms — is widely applicable in the study
of gender and crime.
11. Notably, the few case studies of all-female gangs available reveal that these
groups are also often involved in fairly organized economic endeavours
such as drug sales and distribution — presumptively ‘male’ undertakings
(Lauderback et al., 1992; Venkatesh, 1998; but see Peterson et al.,
2001).
12. Similarly, in a recent analysis of gender constructions among active burglars,
Mullins and Wright (2002) found that women drew from traditional
gender schemas such as family obligations, shame and informal social
controls to explain why they would desist from offending, while they did
not describe gender as having an overarching influence on their motives for
offending.
References
Anderson, Elijah (1999) Code of the Street. New York: W.W. Norton & Co.
Bottcher, Jean (2001) ‘Social Practices of Gender: How Gender Relates to
Delinquency in the Everyday Lives of High-Risk Youths’, Criminology 39:
893–932.
Butler, Judith (1990) Gender Trouble: Feminism and the Subversion of Identity.
New York: Routledge.
Campbell, Anne (1984) The Girls in the Gang. New York: Basil Blackwell.
Miller—The strengths and limits of ’doing gender’ 457
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Cerulo, Karen A. (1997) ‘Identity Construction: New Issues, New Directions’,
Annual Review of Sociology 23: 385–409.
Chauncey, George (1994) Gay New York: Gender, Urban Culture and the
Making of the Gay Male World, 1890–1940. New York: Basic Books.
Collins, Patricia Hill (1990) Black Feminist Thought: Knowledge, Consciousness,
and the Politics of Empowerment. Boston: Unwin Hyman.
Connell, R.W. (1987) Gender and Power. Stanford: Stanford University
Press.
Connell, R.W. (1993) ‘Foreword’, in James W. Messerschmidt (ed) Masculinities
and Crime, pp. vii–xvi. Lanham, MD: Rowman & Littlefield.
Curry, G. David (1997) ‘Selected Statistics on Female Gang Involvement’,
paper presented at the Fifth Joint National Conference on Gangs, Schools,
and Communities, Orlando.
Daly, Kathleen and Meda Chesney-Lind (1988) ‘Feminism and Criminology’,
Justice Quarterly 5: 497–538.
Emirbayer, Mustafa and Ann Mische (1998) ‘What is Agency?’, American
Journal of Sociology 103: 962–1023.
Garfinkel, Harold (1967) Studies in Ethnomethodology. Englewood Cliffs, NJ:
Prentice Hall.
Giddens, Anthony (1984) The Constitution of Society: Outline of the Theory
of Structuration. Berkeley: University of California Press.
Hagedorn, John M. and Mary Devitt (1999) ‘Fighting Female: The Social
Construction of Female Gangs’, in Meda Chesney-Lind and John Hagedorn
(eds) Female Gangs in America: Essays on Girls, Gangs and Gender, pp.
256–76. Chicago: Lakeview Press.
Harding, Sandra (1987) ‘Introduction: Is There a Feminist Method?’, in Sandra
Harding (ed) Feminism and Methodology, pp. 1–14. Bloomington: Indiana
University Press.
Hood-Williams, John (2001) ‘Gender, Masculinities and Crime: From Structures
to Psyches’, Theoretical Criminology 5(1): 37–60.
Jacobs, Bruce A. and Jody Miller (1998) ‘Crack Dealing, Gender, and Arrest
Avoidance’, Social Problems 45: 550–69.
Joe, Karen A. and Meda Chesney-Lind (1995) ‘“Just Every Mother’s Angel”:
An Analysis of Gender and Ethnic Variations in Youth Gang Membership’,
Gender & Society 9: 408–30.
Joe-Laidler, Karen A. and Geoffrey Hunt (1997) ‘Violence and Social Organization
in Female Gangs’, Social Justice 24: 148–69.
Kandiyoti, Deniz (1988) ‘Bargaining with Patriarchy’, Gender & Society 2:
274–90.
Kanter, Rosabeth Moss (1977) ‘Some Effects of Proportions of Group Life:
Skewed Sex Ratios and Responses to Token Women’, American Journal of
Sociology 82: 965–90.
Kessler, Suzanne J. and Wendy McKenna (1978) Gender: An Ethnomethodological
Approach. Chicago: University of Chicago Press.
Konrad, Alison M., Susan Winter and Barbara A. Gutek (1992) ‘Diversity in
Theoretical Criminology 6(4)458
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Work Group Sex Composition: Implications for Majority and Minority
Members’, Research in the Sociology of Organizations 10: 115–40.
Kulick, Don (1998) Travesti: Sex, Gender and Culture among Brazilian
Transgendered Prostitutes. Chicago: Chicago University Press.
Lauderback, David, Joy Hansen and Dan Waldorf (1992) ‘“Sisters Are Doin’ It
For Themselves”: A Black Female Gang in San Francisco’, The Gang Journal
1: 57–70.
Maher, Lisa (1997) Sexed Work: Gender, Race and Resistance in a Brooklyn
Drug Market. Oxford: Clarendon Press.
McNay, Lois (2000) Gender and Agency: Reconfiguring the Subject in Feminist
and Social Theory. Cambridge: Polity Press.
Messerschmidt, James W. (1993) Masculinities and Crime. Lanham, MD:
Rowman & Littlefield.
Messerschmidt, James W. (1995) ‘From Patriarchy to Gender: Feminist Theory,
Criminology and the Challenge of Diversity’, in Nicole Hahn Rafter and
Frances Heidensohn (eds) International Feminist Perspectives in Criminology:
Engendering a Discipline, pp. 167–88. Philadelphia: Open University
Press.
Messerschmidt, James W. (1997) Crime as Structured Action: Gender, Race,
Class, and Crime in the Making. Thousand Oaks, CA: Sage.
Messerschmidt, James W. (1999) ‘Making Bodies Matter: Adolescent Masculinities,
the Body, and Varieties of Violence’, Theoretical Criminology 3(2):
197–220.
Messerschmidt, James W. (2000) Nine Lives: Adolescent Masculinities, The
Body, and Violence. Boulder, CO: Westview Press.
Messerschmidt, James W. (2002, forthcoming) ‘Men, Masculinities, and
Crime’, in Michael Kimmel, R.W. Connell and Jeff Hearn (eds) Handbook
on Men and Masculinities. Thousand Oaks, CA: Sage.
Miller, Jody (1998) ‘Up It Up: Gender and the Accomplishment of Street
Robbery’, Criminology 36: 37–66.
Miller, Jody (2001) One of the Guys: Girls, Gangs and Gender. New York:
Oxford University Press.
Miller, Jody and Rod K. Brunson (2000) ‘Gender Dynamics in Youth Gangs: A
Comparison of Male and Female Accounts’, Justice Quarterly 17(3):
801–30.
Miller, Jody and Scott H. Decker (2001) ‘Young Women and Gang Violence:
An Examination of Gender, Street Offending and Violent Victimization in
Gangs’, Justice Quarterly 18(1): 115–40.
Moore, Joan and John Hagedorn (1996) ‘What Happens to Girls in the
Gang?’, in C. Ronald Huff (ed) Gangs in America, 2nd edition, pp. 205–18.
Thousand Oaks, CA: Sage Publications.
Mullins, Christopher W. and Richard T. Wright (2002) ‘Gender, Social Networks
and the Accomplishment of Residential Burglary’, Unpublished manuscript,
Department of Criminology and Criminal Justice, University of
Missouri-St. Louis.
Miller—The strengths and limits of ’doing gender’ 459
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from
Orbuch, Terri L. (1997) ‘People’s Accounts Count: The Sociology of Accounts’,
Annual Review of Sociology 23: 455–78.
Peterson, Dana, Jody Miller and Finn-Aage Esbensen (2001) ‘The Impact of
Sex Composition on Gangs and Gang Member Delinquency’, Criminology.
Portillos, Edwardo Luis (1999) ‘Women, Men and Gangs: The Social Construction
of Gender in the Barrio’, in Meda Chesney-Lind and John M. Hagedorn
(eds) Female Gangs in America: Essays on Girls, Gangs and Gender, pp.
232–44. Chicago: Lake View Press.
Segal, Lynne (1990) Slow Motion: Changing Masculinities, Changing Men.
New Brunswick, NJ: Rutgers University Press.
Sewell, William H., Jr (1992) ‘A Theory of Structure: Duality, Agency, and
Transformation’, American Journal of Sociology 98: 1–29.
Simpson, Sally (2000) ‘Gendered Theory and Single Sex Research’, DivisioNews.
Newsletter of the Division of Women and Crime, American Society of
Criminology, August, http://www.ou.edu/soc/dwc/newsletter.htm.
Simpson, Sally and Lori Elis (1995) ‘Doing Gender: Sorting out the Caste and
Crime Conundrum’, Criminology 33: 47–81.
Spelman, Elizabeth V. (1988) Inessential Woman: Problems of Exclusion in
Feminist Thought. Boston: Beacon Press.
Thorne, Barrie (1993) Gender Play: Girls and Boys in School. New Brunswick,
NJ: Rutgers University Press.
Tittle, Charles R. (1995) Control Balance: Toward a General Theory of
Deviance. Boulder, CO: Westview Press.
Venkatesh, Sudhir Alladi (1998) ‘Gender and Outlaw Capitalism: A Historical
Account of the Black Sisters United “Girl Gang”’, Signs 23: 683–709.
West, Candace and Sarah Fenstermaker (1995) ‘Doing Difference’, Gender &
Society 9: 8–37.
West, Candace and Don H. Zimmerman (1987) ‘Doing Gender’, Gender &
Society 1: 125–51.
Wright, Richard T. and Scott H. Decker (1997) Armed Robbers in Action:
Stickups and Street Culture. Boston: Northeastern University Press.
JODY MILLER is Associate Professor of criminology and criminal justice at
the University of Missouri-St. Louis. She specializes in feminist theory and
qualitative research methods. Her research focuses on gender, crime and
victimization, particularly in the contexts of youth gangs, urban
communities, and the commercial sex industry.
Theoretical Criminology 6(4)460
at LUND UNIVERSITY LIBRARIES on March 19, 2012tcr.sagepub.comDownloaded from