Women Prison

Theoretical Theoretical Criminology Crimi nology http://tcr.sagepub.com/ 

Policing Criminological Knowledge : The Hazards of Qualitative Research on Women in Prison Joane Martel Theoretical Criminology  2004 8: 157 DOI: 10.1177/1362480604042242

The online version of this article can be found found at: http://tcr.sagepub.com/content/8/2/157

Published by: http://www.sagepublications.com

Additional Additional services and information for Theoretical Criminology  Criminology  can be found at: Email Alerts: http://tcr.sagepub.com/cgi/alerts Subscriptions: http://tcr.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://tcr.sagepub.com/content/8/2/157.refs.html

>> Version of Record - May 1, 2004 What is This?

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Theoretical Criminology  © 2004 SAGE 2004 SAGE Publications London, Thousand Oaks and New Delhi.  Vol. 8(2): 157–189; 1362–4806 DOI: 10.1177/1362480604042 10.1177/1362480604042242 242

Policing criminological criminological knowledge: The hazards of qualitative research on women in prison JOANE MARTEL

University of Alberta, Canada Abstract This article draws on the sociology of knowledge literature to bring to light several social mechanisms used to police the production and circulation of criminological knowledge. The discussion draws  from actor-network theory and feminist analyses in science and technology studies to make visible mechanisms of construction,  filtering, negotiation and appropriation. appropriation. By way of illustration, the article draws on an actual case of scientific, political and media marginalization of disquieting research findings on prison conditions in Canada. Taking these reactions as a starting point, the analysis discusses, among other things, the hierarchy of paradigms and the gendered structures that affect researchers’ admission into a scientific community and their access to discursive resources.

Key Words gender • knowledge • network • policing • prison

To those interested in the sociology of knowledge, it will come as no surprise that information comes in diverse forms that are not accorded equal recognition or credibility. Here, though, I aim to explore this observation by illustrating several social mechanisms at work in policing the production and circulation of knowledge within criminology. For, apart 157 Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Theoretical Criminology  © 2004 SAGE 2004 SAGE Publications London, Thousand Oaks and New Delhi.  Vol. 8(2): 157–189; 1362–4806 DOI: 10.1177/1362480604042 10.1177/1362480604042242 242

Policing criminological criminological knowledge: The hazards of qualitative research on women in prison JOANE MARTEL

University of Alberta, Canada Abstract This article draws on the sociology of knowledge literature to bring to light several social mechanisms used to police the production and circulation of criminological knowledge. The discussion draws  from actor-network theory and feminist analyses in science and technology studies to make visible mechanisms of construction,  filtering, negotiation and appropriation. appropriation. By way of illustration, the article draws on an actual case of scientific, political and media marginalization of disquieting research findings on prison conditions in Canada. Taking these reactions as a starting point, the analysis discusses, among other things, the hierarchy of paradigms and the gendered structures that affect researchers’ admission into a scientific community and their access to discursive resources.

Key Words gender • knowledge • network • policing • prison

To those interested in the sociology of knowledge, it will come as no surprise that information comes in diverse forms that are not accorded equal recognition or credibility. Here, though, I aim to explore this observation by illustrating several social mechanisms at work in policing the production and circulation of knowledge within criminology. For, apart 157 Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

158

Theoretical Criminology 8(2)

from a few earlier but notable contributions to the sociology of (criminological) knowledge (e.g. Moyer, Moyer, 1986; Bowker et al., 1988; Windmayer and Rabe, 1990; Eigenberg and Baro, 1992), criminology has usually not been the focus of academic work with an epistemological orientation. To help remedy this absence, this article relies on an actual case study where disquieting research results were marginalized in the scientific, political and media arenas. The research setting was Canada, the object of  study prison segregation (solitary confinement) and the research subjects women (Martel, 1999). The study clearly used engaged feminist approaches, provided data on women’s experiences and was conducted by a female researcher. Arguably, these characteristics put the study in an epistemological framework far from the mainstream tradition of universalistic and objectivistic ‘science’. In spite of sustained efforts to disseminate the study’s findings about prison practices and human suffering, the study—along with the women’s experiences it gave voice to—was ignored and delegitimized by several of the audiences it sought to influence. Specifically, following the report’s 2000 publication, it was trivialized by the media, ignored by prison authorities and disqualified by administrative criminologists. The first section of the article presents a cursory overview of the major findings of this study—a segregation study—to provide readers with an appreciation of the reactions that followed its findings’ release. This is followed by a succinct discussion of the literature on the policing of  knowledge that serves as a backdrop for sections three and four; these establish the article’s theoretical framework, drawing on actor-network theory, feminist analyses of science, and technology studies. Section five, combining case study and theory, is divided into sub-sections corresponding to the different stages of the segregation study: the elaboration and realization of the study; the release of findings in the scientific community and subsequently in the media; and, finally, the elaboration of a broader proposal for further research. The narrative of this article is organized along conventional epistemological precepts of linearity and deduction because these precepts remain the standard method of knowing and adjudicating suitability for publication. Thus, one might say the article is written strategically, using traditional categories of logic to convey the relevance of alternative epistemological frames of reference from within the system of communication with which they are most familiar.

Canadian women in prison segregation Prison segregation is used to separate and isolate a prisoner from the general prison population for reasons of protection (protective custody) or punishment (disciplinary segregation). This practice stems from a model of  imprisonment that developed in the 18th century, aimed at isolating a

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

prisoner completely for a period of 23 hours a day, leaving 1 hour daily for solitary exercise in the prison courtyard. Since then, segregation has become a standard practice in North American prisons. More importantly, its continuous use since the 18th century has turned it into a taken-forgranted method of internal regulation within prisons. Bruno Latour (1987) concludes in this regard that, historically, prison segregation has been ‘translated’ (or blackboxed) into a ‘fact’, as if the practice of imprisonment could no longer be imagined without segregation cells. The very character of prison segregation may account for the sparse knowledge currently available about its realities in Canada and elsewhere, with the notable exception of a few recent studies (Canada, 1997; Motiuk and Blanchette, 1997; Shaylor, 1998; Zinger and Wichmann, 1999). Literature on the role played by segregation in prison governance, and on how segregation is experienced by carceral subjects is particularly meagre with the experiences of women prisoners especially and sorely unknown (Shaylor, 1998 notwithstanding). However, we know that the organization of  prison regimes for women directly implicates an ‘epistemology of female crime’ that is over-determined by value-laden, psychologically based assumptions about women’s ‘deviance’ (Genders and Player, 1986: 368–70). Indeed, various forms of social control of women are steeped in patriarchal assumptions about, for instance, ‘familial’ expectations that involve ideologically based constructions of gender-appropriate roles for women (Howe, 1994: 125–6). In the wider society, women are also constructed within frames of sociability, femininity, motherhood, adulthood and domesticity. Each of these frames has developed its own regulatory powers over time; taken together, an intricate web of interrelated powers aimed at controlling women has evolved. During the 19th and 20th centuries, prisons for women likewise developed on the basis of analogous patriarchal assumptions and gender-based constructions about women’s sexuality, domesticity and supposed pathology. We also know that institutionalized race, class and gender relations in western societies have been expressed through actors’ interpretations of their social experiences (Messerschmidt, 1997). Thus, prisons can be said to be revealing sites for analysing power relations in society, especially vis-`  a-vis their role in the reproduction of classed, raced and gendered relations (e.g. Howe, 1994; Hannah-Moffat, 2001). Feminist scholarship on women’s imprisonment is recent and, in terms of  coherent theoretical and epistemological perspectives, still emerging. Although already more sophisticated than traditional positivistic and pathologizing approaches, critical social studies of the punishment of women are still often marginalized within social analyses of penality. To this effect, Howe observes that feminist approaches to penality ‘languish behind the locked doors of masculinist (albeit revisionist masculinist) frameworks’ (1994: 166). It is within this context of limited knowledge that the study on women’s experiences of segregation was undertaken.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

159

160

Theoretical Criminology 8(2)

I was initially approached by the Elizabeth Fry Society of Edmonton, Alberta: the society sought to commission a pilot study on women’s segregation to be undertaken by an academic, and feminist, researcher. The project received funding from Status of Women Canada, a federal government agency that funds gender-based research works and promotes equitable public policies. Data were collected via semi-structured interviews with four women who had been released in the community, four who were jailed at the Edmonton Institution for Women, two who were housed at the Regional Psychiatric Centre in Saskatchewan and two who were imprisoned at the Saskatchewan Penitentiary in Prince Albert (Saskatchewan). Most of the participants had experienced segregation as recently as 1996 and 1998. The interviews were two hours in length; they were taperecorded and conducted between March and April 1998. Quantitative data were also gathered through aggregation of participants’ individual characteristics, statistical summaries of their institutional records of segregation and segregation logs. The findings of the study reveal the still repressive nature of segregation in Canada in the late 1990s. Similar to 19th-century practices of segregation, shackles are used, and interviewees reported instances of filth, insects and diseases. In provincial prisons,1 segregated women have been denied underwear or toiletry articles such as a toothbrush or toilet paper. Others sleep on a mattress on the concrete floor with wet blankets and without a pillow. Women are often subjected to sexually humiliating practices such as showering with onlookers (frequently male staff), crossing a co-ed prison unit wearing a ‘baby doll’2 without underclothing underneath, or being refused a sufficient quantity of sanitary pads, forcing them into recurrent and degrading requests for more. Moreover, provincially sentenced women may be exposed to the sexualized behaviour of male prisoners (e.g. masturbation, indecencies) who are often segregated on the same unit. Such sexualized conditions can have significant effects on women since, as the literature amply demonstrates, a large proportion of women prisoners have past histories of abuse (e.g. Task Force on Federally Sentenced Women, 1990; Comack, 1993). Numerous women relive past abuse through these sexualized incidents. Repressive conditions also characterize the segregation of women in Canadian federal prisons. Socially helpful endeavours like schooling, treatment programmes or community contacts are seldom accessible to, or maintainable by, segregated women. As a consequence, social interactions go from sporadic to non-existent. Apart from shouting through the walls or under one’s cell door, women are literally silenced by the sheer lack of  quality or sustained human contact. Such social invisibility leads to dehumanization and generates feelings of abandonment. In both provincial and federal prisons, the racialized aspects of the practice of segregation further highlights the marginalization of segregated women. In Canada, segregation cells are predominantly occupied by prisoners of Aboriginal descent, and this disproportion is even more striking with Aboriginal women

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

(Solicitor General, 1998; Martel, 1999). Segregated Aboriginal women mentioned being subjected to race-laden comments and demeaningly gendered designations (e.g. ‘girl’); these women were often assigned poorer segregation cells than those allocated to non-Aboriginal women.3 They also reported being segregated because supposedly argumentative or ‘mouthy’. These problems would likely not have risen for women who were white and middle class. This last point echoes Mary Bosworth’s (1999) compelling argument about identity formation as a form of resistance to the controlling regimes of women’s prisons. Bosworth contends that contrary to men’s prisons which are organized along traditional (deterrence, retribution, incapacitation) or more ‘modern’ (efficiency, security) penal philosophies, the chief  goal of women’s prisons is to ‘care’ for prisoners and orient their conduct towards hetero-normative ideals of femininity. Thus, women—especially Aboriginal women—are encouraged to adopt passive, traditionally gendered behaviour, and to take on pre-given social roles as dependent, responsible wives and mothers. Yet, as Bosworth (1999) showed, women prisoners revert to multifarious self-determined stratagems and performances of feminine identity precisely to resist regimes of power (such as ideals of femininity) imposed by the prison—even though these ‘enactments’ often lead to negative consequences (i.e. segregation). In some instances, women will ‘co-opt elements from the dominant notion of  femininity to reinforce their own sense of self, and to challenge aspects of  the penal regime’ (Bosworth, 1999: 156). In other instances, women may choose to present themselves as independent agents by constructing alternative meanings of femininity like adopting confrontational behaviours that are in direct opposition to prisons’ feminine ideals (as did the Aboriginal women mentioned earlier). On a final note, the study indicates that apart from being segregated for understandable reasons like fighting, setting fires or attacking fellow prisoners or staff, women are also forced into isolation for more nebulous motives such as ‘being in a state other than normal’,4 being in possession of  contraband (i.e. a deck of cards, small packs of salt and pepper), swearing, tattooing and self-mutilating. In contrast, discussions that I have had with several wardens of prisons for men in Western Canada reveal the unlikelihood of men being segregated for similarly nebulous reasons (with the exception of self-mutilation). This suggests that the practice of segregation is used in decidedly gendered ways. Echoing Bosworth (1999) yet again, gender therefore seems to bring about the reproduction of asymmetrical power relations through the construction of socially defensive selves. Apart from particular circumstances where segregation seems reasonable, the dire experiences of women in isolation led to the conclusion that segregation was often misused, and that this misuse had two major consequences. First, segregation’s effects on women were significant and often long lasting; second, segregation tended to legitimize the prison rationale and its array of assumptions. These findings and conclusions were

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

161

162

Theoretical Criminology 8(2)

ultimately published in book form and released in a press conference, after which the media, prison authorities and mainstream criminologists engaged in a generalized retreat from these conclusions, as described in greater detail later.

The policing of knowledge A well-established contemporary literature shows that knowledges are subjected to gatekeeping scrutiny leading to their legitimization or to their marginalization (e.g. Knorr-Cetina, 1981; Code, 1991; Crossen, 1994). One of the key processes of this legitimization/marginalization is what Mitchell et al. have termed the ‘policing of knowledge’ (2000: 1). As they suggest, academia is one sphere of social influence where such policing occurs despite commitments, at least in theory, to diversifying intellectual thought. It should not be news to anyone that academic institutions, connected as they often are to private and governmental sources, tend to orient knowledge and marginalize particular structures of thought. Certainly Pierre Bourdieu (e.g. Bourdieu, 1984; Bourdieu and Passeron, 1990) has convincingly shown, both in sociological terms and in terms of the sociology of  education, that knowledge is framed by the social context of schooling. Bourdieu argues that the ways in which society selects, classifies, assesses and disseminates public knowledge leads to the cultural reproduction of  social ‘fields’ (Bourdieu, 1969) based on reputational structures and hierarchically stratified knowledge systems. Such structures are at work when academics try to access various material, normative and rhetorical resources necessary to extend their knowledge claims in time and space. These structures are also interwoven with access to discursive resources like award distributions and leading networks of scientific publications. Within academia, too, knowledge is policed through the ideological propensities of  scientific journals, the gatekeeping role played by editors of and reviewers for journals (e.g. Spender, 1981), the selectivity of publishers of books and the funding of research. Another example of how knowledge is policed in academia is simply through the tendency of professors to publicize certain perspectives to their students (usually their own) while marginalizing others. The books, scientific journals and reference works that proliferate in, or are absent from, library shelves likewise direct students’ intellectual attention towards particular universes of thought. Finally, ‘policing knowledge’ occurs through academia’s hiring and promotional mechanisms in so far as endless debates sometimes emerge about the ‘usefulness’ of a candidate’s research and whether that research has been disseminated in scientific journals or in academic presses deemed prestigious (O’Neill and Sachis, 1994). Strikingly, the way knowledge is presently ‘policed’ often involves a favourable bias towards deductive

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

rather than inductive epistemologies (Law, 1983; van den Hoonaard, 2001). In turn, this bias may place qualitative research at a disadvantage. Running through each and all of these examples, at present, is a process of ‘corporatization’ that has tended to realign academic research towards commercial or political objectives. In 1994, Gibbons et al. were among the first to document and analyse this instrumentalization of academic research stemming from demands and pressures in the political and economic fields. Part of this trend is increasingly to condition funding on the relevance and applicability of research projects to social policies. In the case at hand, the national funding agency for the social sciences in Canada has recently suffered significant budgetary cutbacks,5 forcing it to assess policy-related proposals more favourably than ‘pure research’. Combined with the preference towards quantitatively oriented epistemologies that this agency has also traditionally expressed, the end result is that funding for non‘mainstream’ research has become more and more difficult to obtain. Of course, failing to obtain research funding itself has significant consequences. Since the 1990s, according to Albert and Bernard (2000: 80–1), peer-reviewed research funding has gained considerable weight in the overall promotional assessment of academic researchers. In their study of  recent changes in the production of knowledge in sociology departments in Qu´ebec, these authors note growing institutional pressures placed on the faculty to secure research grants from recognized public funding agencies. These pressures may be especially difficult for researchers whose epistemological and methodological choices do not fit (or only marginally fit) with the standards of adjudication of funding agencies. In return, career structures are affected by reward systems that compensate particular researchers and forms of research that comply with particular norms of scientific behaviour; simultaneously these have tended to involve advocacy for a transcendent objectivity, dislocated or ‘society-free’ knowledge and tightly defined evidence standards. At this point, knowledge becomes an object on which power is exercised (Foucault, 1980); moreover, such exercises of  power create ‘outsiders’ who find little ‘legitimate’ space for their forms of  knowing (Rosenblum and Rosenblum, 1990). In this regard, feminist research and writing may have found itself even more pushed to the margin of scientific legitimacy in recent years due to reflexive styles that upset dominant social, political, cultural and scientific categories and question ‘objective’ or overly abstracted knowledge claims. In turn, feminists who occupy relatively more marginal positions in scientific communities are less likely to be noticed, cited and to have their work published. This discussion illustrates how scientific knowledge is a distinctive type of social practice; as such, it needs to be envisaged in its political, economic and moral dimensions without implying that this particular form of  knowledge is just an epiphenomenon of politics or morality (Giddens, 1979). Viewing scientific knowledge in this light, one can argue that certain discursive spaces rather than others become hegemonic via structural and

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

163

164

Theoretical Criminology 8(2)

cognitive mechanisms. But precisely because such social mechanisms often operate covertly and in scattered social spaces, they are more easily grasped using a broad conceptual apparatus. One methodological approach relevant to analysing hierarchies of knowledge paradigms is actor-network theory (ANT), popularized by Bruno Latour, Michel Callon and John Law.

The spread of knowledge claims and actor-network theory ANT is a branch of the sociology of science and technology that attempts to explain the processes by which a particular knowledge claim radiates in time and space, and comes to impose itself (or not) on various audiences (Latour, 1987). According to Callon (1986: 196), the legitimation of  knowledge claims occurs during a series of trials of strength as actors strive to enrol targeted audiences by convincing them of the validity of promoted claims (Latour, 1987: 108). ANT proponents argue that fact-building activities take place in three different sites: the laboratory (or any broadly defined locale where initial knowledge claims are constructed); the scientific community (where claims are debated in the light of ‘scientific’ criteria); and the larger community of non-scientists where trials of strength also take place (Latour, 1987). ANT also emphasizes that network-building activities are essential for extending scientific claims across time and space (Law, 1983; Callon, 1986). These activities involve the recruitment and enrolment of allies in an attempt to construct organizational networks of coalition. Such networkbuilding activities are at play in all three sites as knowledge claims are produced and legitimized. In the laboratory, the preferences of funding agencies (Latour, 1987) and the co-operation of the targets of the investigation (e.g. substances, populations) (Hacking, 1992) are primary basics of  network building. In the scientific community, the ‘evidence’ first produced in the laboratory must be reconstructed in a manner suitable to the editors of scientific journals (Latour, 1986). Here, evidence is flagged by cutting and pasting data together, while pointers directed in opposite directions minimize other features of the research findings. Graphing or tabling links findings to mathematical models, thus conferring measurability, universality and invariability (Amann and Knorr-Cetina, 1990). Referring to published articles can also make knowledge claims more persuasive as other authors are used as allies, thereby raising the cost of disagreeing with the article’s claims (Latour, 1987). However, once supportive networks are created within the first two sites of fact building, the success of knowledge claims is still highly dependent on the forging of alliances with the larger community of non-scientists (e.g. Callon and Law, 1989; Ward, 1996).

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

 Revisiting actor-network theory: insights from feminist  work

Although ANT sheds valuable light on the power relations that form the core of scientific knowledge making, to some extent, blind spots undermine this theory’s ability to provide a sophisticated account of scientific practice. In particular, ANT does not adequately address how ‘science’ is granted cultural legitimacy at the expense of alternative forms of knowing. In this section of the article, though, I argue briefly that access to discursive resources provides cultural legitimacy to certain forms of knowing, and that such access is largely dependent on researchers’ compliance with internal rules that oversee the conditions of participation in knowledge production. Although Latour (1987) discusses the issue of resources, he does so primarily in respect to scientists needing to enrol purveyors of  resources (e.g. funding agencies); however, he does not address biases constitutive of resource attribution itself. But two of these biases may well involve researchers’ gender and methodological preferences, both of which respectively affect researchers’ status in a scientific community and access to discursive resources. Many authors have noted, and criticized, how research done by and about women has often been marginalized (e.g. Hawkesworth, 1989; Harding, 1991; Fox, 1995). A significant body of literature has shown that women scholars are usually under-represented with respect to publication rates in a variety of academic disciplines,6 and that they tend to benefit less often from national funding programmes (Stebbins, 2001: 458). Moreover, influential criminological journals tend to show a stronger underrepresentation of women as published authors (Eigenberg and Baro, 1992), as well as an over-representation of women as co-authors (Eigenberg et al., 1992). While women are under-represented in editorial processes of sociological journals (Walker and Thompson, 1984; Ward and Grant, 1985), their under-representation is particularly high in major influential criminal justice journals (Eigenberg and Baro, 1992). Thus, in male-dominated disciplines like criminology and criminal justice, women often find themselves in situations where their access to the established reward structure— that is, funding, promotions, publications, reputation—is especially arduous. Given such limited access, it is not uncommon for women researchers to fall a long way behind in their ability to engage audiences and construct alliances supportive of, as is one important example, distinctively feminist knowledge claims. Usually, exclusionary tactics related to the marginalization of research done by feminist researchers and/or about women’s issues can only be inferred. However, in a complementary note to a feminist essay on child abuse, Lee Bowker relates how the essay was eventually rejected after an editor manipulated the review process (Bowker et al., 1988: 169–73). Following two divergent reviews, the editor requested a third review that came back in favour of publication. However, a fourth reviewer was also

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

165

166

Theoretical Criminology 8(2)

secretly sought, apparently intended to reverse the third review so that it more closely aligned with the editor’s own perspective. By a ‘lucky slip’ (Bowker et al., 1988: 172) Bowker was sent the third (positive) review along with the illegitimate fourth (negative) one with the editor’s rejection letter. According to Edward Said, an important role of academic intellectuals is to give voice to what, and who, has been repressed, ignored, forgotten or marginalized in society (Said, 1994). However, this ideal is rendered mythological to the extent that feminist research (often qualitatively oriented), along with other critical epistemologies, cannot occupy more than a peripheral space in the still scientifically oriented academy. Fuller (1997) observed that the gendered social division of labour that has historically attributed objectivity to men and subjectivity to women has been transposed into the field of knowledge production to create a ‘Great Divide’ (Latour, 1999: 19): an objective/subjective, quantitative/qualitative, hard/  soft, apolitical/political, masculinist/feminist set of dichotomies that still persist in many academic fields and subfields. In short, exclusionary practices make possible the dominance of certain epistemological approaches and the marginalization of others. Simultaneously, such practices often appear to promote the notion that ‘scientific’ knowledge has little to do with social divisions of race, class or gender.7 Yet while biases against studies of or by women, including those with a decidedly feminist orientation, certainly persist, these were not the primary obstacles encountered after releasing the segregation study. Rather, biases against qualitative methodologies per se constituted the main impediment against publicizing this study’s results, as I recount below.

Policing the segregation study  First site: the construction of initial knowledge claims

As seen in the previous discussion of actor-network theory, the first site of  fact building is the laboratory; following actor-network theory, I am defining this as any broadly defined locale where knowledge claims are first constructed. In relation to the segregation study, the first site of fact building entailed prison practices related to segregation, as described by interviewees and analysed by the researcher. At this first site, the segregation study found itself in a position to build supportive networks for the knowledge claims it was seeking to promote. Initial network-building activities included the participation of a local chapter of the internationally known Elizabeth Fry Society, a non-profit community-based agency that provides advocacy, services and programmes to marginalized women involved with the criminal justice system. The Elizabeth Fry Society not only instigated the research project but was actively involved in all parts of the study, from sampling and accessing segregated women to the publication of  the research report. This agency was also instrumental in securing a

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

research grant from Status of Women Canada. With the material and financial support of these two leading public entities, the Elizabeth Fry Society and Status of Women Canada, the study on women’s segregation developed a solid initial network. The strength of this network grew when the Canadian umbrella federation of Elizabeth Fry Societies was also successfully enrolled in the study. Another major player who became engaged, though to a lesser degree, was the Women’s Programme of Correctional Service Canada (CSC). This is the federal agency responsible for overseeing the administration of  correctional facilities and services that fall under federal jurisdiction. Since I sought to interview women segregated in several federal penitentiaries, permission to access these prisons was needed; for this purpose, the research project was submitted to CSC. After an initial review of the project, CSC commented that an experiential qualitative study had value ‘for studying women’s subjective experience of segregation’ but that for purposes of validity these data needed to be balanced with information regarding women’s ‘objective history of segregation’8 via quantitative data. To that effect, CSC was willing to provide statistical summaries of women’s histories of segregation in federal penitentiaries. The bipolarization of the segregation study into a subjective experience (qualitative) and an objective history (quantitative) provides a view into CSC’s method of adjudication of  credible and worthy research.9 The following excerpt is more clearly indicative of CSC’s use of mainstream standards of adjudication to assess methodologically the segregation study: This is a non-trivial methodological point since your study, as currently proposed, does not use random sampling (. . .) and you have not indicated any alternative method for ensuring that the sample obtained is representative of the population of incarcerated women. Some evidence that your sample is representative is required to make credible statements about the impact of segregation on incarcerated women in general. We strongly recommend that you request and use such information (quantitative data) during your research. 10

This excerpt arguably suggests an attempt to police knowledge production. CSC’s recommendation is to modify the study in order to incorporate a form of knowledge that ‘fits’ evidence standards through the use of  quantitative data. What this suggests is that the recommendation is based on an assumed hierarchy of paradigms where a universalistic, objectivistic and rationalistic view of social reality occupies a higher rank than alternative interpretations of social reality. What it further suggests is that this hierarchization of paradigms operates to reproduce a particular kind of  order within the scientific community where the truth status of forms of  knowing is chiefly measured by more traditional ideals of ‘good’ or ‘hard’ science.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

167

168

Theoretical Criminology 8(2)

The recommendation made by CSC may be seen as the outcome of a politically charged epistemological negotiation between proponents of a mainstream scientific tradition and those who seek to do science differently. This negotiation took place in two different places. It first occurred internally within CSC’s own research department among mainstreamers and qualitative researchers.11 CSC’s final recommendation, then, becomes the compromise at which these two intellectual communities arrived. For CSC’s qualitative researchers, the study could go ahead and make an ‘important contribution to the literature on women in prison’.12 For CSC’s mainstreamers, though, the requirement to consider objective data allowed them to perpetuate ideologies of control and dominance of a particular form of scientific knowledge. The initial hesitance of the mainstreamers, as suggested in the earlier quotation, may have stemmed from the view that the segregation study was a one-sided, and overtly politically engaged form of analysis on women prisoners—a category of women that, at least in the past, has been assumed to be emotionally unstable deceivers13—rather than first-hand knowledge bearers. To the mainstreamers, the incorporation of  quantitative material would insure a minimal validity, objectivity and political neutrality to the study. It would also guarantee their partial and fragile enrolment in the study. The veiled idea appears to be to add a little bit of ‘good’ science to ‘political’ feminist work. Following this first round of negotiations, CSC’s recommendation was then negotiated in a second place. Specifically, it was negotiated between CSC and myself. A key issue in this second round of negotiations was the condition linked to it, as the earlier quotation reveals. That is, permission to access the targeted penitentiaries rested upon my integration of CSC’s recommendation into the study’s design. This requirement was acceptable to me because the supplemental data supplied by CSC promised to be helpful in supplementing the aggregated data featured in the study.14 Overall, CSC’s data provided three types of supplemental information: data on generalized institutional practices of segregation, on staff’s informal perceptions of segregation units and on their perceptions of their female population. It also promised to provide comparative information on the women participants. Finally, CSC’s requirement is also worthy of a brief comment on its symbolic representation. Should the findings of the segregation study be to their liking, CSC’s requirement would serve symbolically to maintain the agency’s long-standing reputation as a scientifically based policy-making organization, for CSC could not give their approval to soft, unscientific or fuzzy research. In contrast, should CSC dislike the findings, it could discard the study on the basis of its qualitative methodology. In sum, the segregation study, in its proposal form, was successful in developing an initial supportive network of key players.15 Upon completion of the study, it was submitted to trials of strength in the scientific community, the second site of fact building.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

Second site: the enrolment of the scientific community

As Callon and Law emphasized in 1989, the support of the scientific community is crucial to the extension of supportive networks and the legitimation of knowledge claims. In terms of enrolment, I sought the support of two specific scientific audiences, the general scientific community (e.g. through publication in a peer-reviewed journal), and the more restricted scientific community of correctional authorities (i.e. employed researchers). The claims promoted in the segregation study found a favourable audience in the general social scientific community. An article on the findings of the study (Martel, 2001) was published by Social Justice, a progressive journal based in the USA known to be sympathetic to nonmainstream social analyses in criminology and to contemporary Marxist perspectives. Submission to a scientific journal enabled the knowledge claims to be debated within the scientific community (through the peer review), to become established as valid (through the publication) and to spread across time and space (through the journal’s readership). However, it is also important to note that this social scientific community wherein the segregation study found support was comprised of researchers already either familiar and sympathetic to alternative paradigms, and/or whose research interests and activities were already informed by alternative epistemologies. This obviously helped to garner acknowledgement. Not so, though, for the reception of the segregation study’s claims in a second scientific community. Indeed, the pilot study was disturbing enough to warrant the intervention of a number of other social processes to police its circulation. These processes became evident when I attempted to enrol the restricted scientific community of the federal and provincial correctional authorities. For example, soon after the release of the segregation study, high-ranking officials within the Ministry of Justice of Alberta cancelled a meeting with the Executive Director of the Elizabeth Fry Society; the purpose of this meeting had been to discuss (i.e. test and debate) the findings of the study. By cancelling this meeting, the Albertan authorities short-circuited yet another16 strategic trial of strength whose objective was to enrol particular audiences. This meeting still has not taken place more than four years after its cancellation. This action suggests that provincial authorities simply chose to overlook the segregation study, thereby restricting serious steps being taken to investigate (and remedy) the study’s findings. In addition, CSC declined to reflect publicly on its segregation practices, and avoided all contact that the research team repeatedly tried to establish with the organization. Moreover, as discussed earlier, while CSC initially accepted to collaborate in the segregation study (albeit in a restrained way), its support had vanished by the time the study was published. This change may have been motivated by CSC’s efforts to avoid angered public actions that the dissemination of the segregation study could generate, harming this

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

169

170

Theoretical Criminology 8(2)

agency’s public legitimacy by making visible one essentially punitive and alienating practice of the prison. Consequently, CSC’s apparent indifference can be viewed as a pretence, and as a way of controlling potentially disruptive knowledge. Yet, by its actions, CSC also effectively silenced the protesting voices of women whose experiences were at the heart of the segregation study. On a slightly different note that also illustrates the ongoing hegemony of  certain kinds of claims within correctional research communities—that is, traditional positivistic knowledge—a colleague reported the following story. She was privy to a conversation between two senior researchers employed by CSC in Ottawa wherein several demeaning comments were made about the segregation study. Some of these comments were to the effect that it barely takes grade four-level intelligence to do qualitative research, and that the main investigator of the study (myself) only studied twelve women17 anyway, so who cares! In ‘scientific’ terms, this comment amounts to saying that non-representative samples cannot possibly reveal anything worth knowing, and that the study was ‘scientifically’ invalid. Yet one could reply that such comments are therefore also, in effect, dismissive of much contemporary sociological and criminological work that also employs qualitative methods wherein small samples and openended questions generate (in)valuable insights about the particular, or ‘local’, character of our global world. In criminology alone, a rich tradition of admirable qualitative work has been influential and informative. Here, among other works, one could cite Pat Carlen’s seminal work on women in prison (1983) as well as Lisa Maher’s (1997) and Chris Bruckert’s (2002) studies on sex work and the strip trade; each of these works have given lived experiences methodological validity. Elizabeth Comack’s studies of  women in prison (1996) and past sexual abuse (1993), as well as Sylvie Frigon’s recent work on femicide (2003) have likewise strengthened the credibility of qualitative methodologies within criminology. Beyond feminist work, too, lies an established and growing group of sociologists—often men studying men—who use qualitative methods, not necessarily or usually based on ‘random samples’, to study a variety of subjects from the punk subculture (Baron, 1989) and street gangs (Sanch`  ez-Jankowski, 1991) through drug users (Bourgois, 1995) and African-American boxers (Wacquant, 2001). Thus the comments made by researchers at CSC, ones that seem arguably related to cancelling meetings and overlooking disturbing research results, provide insight into dominant universes that still dominate—and affect the dissemination of—criminological knowledge. Apparently, if the implications of these comments and actions are generalized, researchers should undertake mostly or only large sample studies and ignore principles of adequacy that may link types of sampling and objects of research in more diverse ways. Thus, in this case, CSC researchers dismissed the validity of semi-structured interviews that often endow qualitative research with a depth only achievable through detailed involvement with the

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

 particular. For since sociology (and, by extension as one area of inquiry within it, criminology as well) cannot grasp its social object in its entirety, studying the particular can and does shed light on the global (Hamel, 1993: 52, 72). Applied to the segregation study, this suggests that it is problematic indeed simply to dismiss in-depth interviews—even only a small number of  such interviews—as a way of learning, and coming to know, about the general character of social life for women in prison. None the less paradigms and methods not grounded in statistical theory may be often relegated to the margins, especially by administrative-oriented criminologists who are likely to cater to the managerial and legitimizing needs of  criminal justice agencies. CSC’s denial of the segregation study’s ‘scientific’ status also illustrates social processes of demarcation. As Gieryn argues (1983), scientific ideologies about professionalism also serve to construct boundaries that cordon off some intellectual activities as non-science (or pseudo-science). The contrasting of the segregation study with the alleged axiological neutrality of science indicates, indeed, that CSC’s assessment was made from an epistemological position that borrows from the ‘hard’ sciences (e.g. physics, chemistry). A similar assessment of the study was made by the Alberta leader of the New Democratic Party (left-wing political party) who was quoted in the Edmonton Journal 18 as saying that the findings of the study would have more weight if Canadian statistical accounts of the number of  women in segregation were included. I concur that these data would have helped to legitimate the findings by emphasizing the scope of the problem. Reporting available statistical information would not have undermined or in any way lessened the disturbing qualitative findings of the study. Yet by multiplying these disturbing findings’ scope, quantitative data would have served my research purposes by showing that the issues identified by interviewees potentially applied to a much larger group of women. However, these criticisms of the dispute were in practice moot since the Canadian Centre for Justice Statistics (CCJS), the national producer of  aggregate data on criminal justice practices, does not publish data on the use of segregation anyway. Nor, for that matter, do the Canadian provinces; in and of itself, then, this absence of data warrants discussion. One identifiable reason is bureaucratic in character. The mandate of the CCJS is to provide statistics on comprehensive trends within the criminal justice apparatus (spending, use of courts, community corrections, imprisonment and the like) but not to report on specific practices in which particular agencies—like prisons—engage. Unfortunately, though, the effect of such bureaucratically based decisions is to render invisible many harsh realities of incarceration that include segregation, self-mutilation and mental health problems. In sum, it seems that two stakes were involved in this recounted clash between CSC’s recommendation and the research proposal as originally designed. One was the validity of two paradigms of knowledge, qualitative ‘versus’ quantitative, that were polarized in reactions to this study as

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

171

172

Theoretical Criminology 8(2)

though mutually exclusive. The second entailed whether or not women would be recognized as valid claimsmakers—that is, as subjects and researchers—if their ‘ways of knowing’ differed from more traditionally recognized ‘scientific’ data. For women, especially feminists, have often been perceived as what Whelan (2001: 560) calls boundary challengers, actively seeking to show biases built into hegemonic forms of knowing that have historically excluded gender and sexuality as important subject matters to ‘scientifically’ oriented communities. This point will be developed further later in the article. At present, though, suffice it to say that at the second site, the segregation study extended its support network to one social scientific community of journal readers while losing ground in another, that is, with correctional authorities. However, the segregation study met its most serious challenge at the third and last site of networkbuilding activities, namely, when turning to the larger community of nonscientists. Third site: the enrolment of the larger community

As Callon and Law (1989) emphasized, the support of the larger community of non-scientists is frequently central to the spread of supportive networks and to the legitimation of knowledge claims. In terms of enrolment, I specifically sought the support of the media. Communication scholars have contended that research evidence and theoretical logic are rarely sufficient to ensure the viability of knowledge claims in non-scientific communities (e.g. Gieryn, 1999); rather the mass media offer a particularly powerful cultural space where claims’ validity and researchers’ credibility are played out (e.g. Lewenstein, 1995; Winch, 1997). Since I expected the research report would raise eyebrows, I expected that enrolling the media would be challenging and adopted tactics aimed at arousing their interest in the research report released in January 2000. Our research team’s thinking about whether, where and when to call a press conference was all strategically motivated by hopes of generating extensive media coverage for the study.19 Press packages were distributed one week prior to the eventually arranged press conference to representatives of the three official political parties of the Province of Alberta, representatives of the five prisons located in the vicinity of the city of  Edmonton and to members of the written and electronic local media, French as well as English. The Elizabeth Fry Society led the press conference along with myself (as the main researcher), and one of the participants who voluntarily relinquished her anonymity to tell the media her story. Although a total of fifty media representatives were invited, only three appeared on the day of the press conference. Other than these three, in the audience was also one provincial member of the Legislative Assembly for the New Democratic Party (left-wing political party) as well as several supporters of the Elizabeth Fry Society. As with the media, federal and provincial correctional service representatives were also conspicuously

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

absent. Due to this small audience, then, the research team had little opportunity to negotiate—through what Callon and others have called ‘trials of strength’—the study’s legitimacy as a critique of prison segregation practices. As many observers have noted, the media determine events’ newsworthiness through well-established practices that usually prioritize broadly covered topics like politics and crime (Hackett and Gruneau, 2000). Moreover, traditional media practices have often involved relying on official authorities—in this case, prison authorities—as primary sources of  news. Yet these practices are also often influenced by economic criteria like advertising sales and ratings. By these last criteria, many members of the media have often viewed prison stories as potentially sensationalistic and therefore potentially producing boosts in sales and higher ratings. For instance, not uncommonly, prison riots or prison escapes make news headlines in and outside of Canada. None the less, the press conference about the results of the segregation study obviously did not foster much media interest. What explains this lack of interest? Sometimes events assessed as newsworthy early in the day are relegated to a lower priority position, later that same day, after other events unfold that are more likely to capture media interests. Hypothetically, then, the press conference on the segregation study could have been demoted after the surprise resignation of a political leader, a deadly multiple car crash or an environmental accident, had occurred. Yet, no such news events made media headlines on the day of the press conference. In fact, two local television stations headlined the release of another research report that recommended the immediate improvement of living conditions for caged animals in all of the provincial zoos as well as the closure of several of them due to the atrocious conditions in which many animals were forced to live.20 That day, the misery of caged animals was deemed more newsworthy than the misery of caged women. Another possibility is that, when pilot studies are exploratory in nature, media coverage may be minimal at best. Indeed, in the case of the segregation study, media coverage was limited to one short article in each of two newspapers: the first article was one of three full columns,21 and the second one of two short columns. 22 The second short article, though, is worth analysing in greater detail. In it, the study’s findings were filtered down to an exclusively financial argument that relied on the Canadian Taxpayers Federation for ‘expert’ information. The first two sentences of  the newspaper article read as follows: The Elizabeth Fry Society has spent $13,000 of taxpayer money to discover that women in solitary confinement are lonely. The study, released yesterday, was panned by Mitch Gray, Alberta director of the Canadian Taxpayers Federation, as a waste of cash. (Edmonton Sun, 28 January 2000, p. 20)

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

173

174

Theoretical Criminology 8(2)

Here, then, is a telling illustration of the precedence, for the Canadian Taxpayers Federation as well as for the Edmonton Sun, of financial matters over broader social concerns. Since the resurgence of neo-liberalism in the 1980s, public spending has occupied a top position in the world of the Canadian media (Hackett and Gruneau, 2000). Moreover, as the influence of neo-liberalism has spread, lobby groups like the Canadian Taxpayers Federation have emerged as one of the media’s favourite spokespersons. While this particular lobby group uses a populist title, its board of  governors are mostly lawyers, bankers and high corporate officials (Hackett and Gruneau, 2000). In other words, the Federation is more of a conduit for corporate public relations than a platform that represents the interests of the average taxpayer. Moreover, this article appeared on page 20 of a newspaper that habitually comprises approximately 75 pages, and wherein sports and classified ads are noticeably predominant. On the basis of traditional journalistic practices, the article’s location indicates that the segregation study was considered by the newspaper to be far removed from more newsworthy stories. Similarly, the first newspaper article appeared in the B section of the Edmonton Journal , a section where journalists traditionally insert news items of lesser newsworthiness. Specifically, the newspaper article appeared alongside obituaries. Moreover, the piece relied largely on correctional spokespersons’ obviously authorized views that the information contained in the segregation report was contrary to actual practices of segregation, thereby further delegitimizing the study’s findings. Still, the study did receive some support from one radio media outlet— the Canadian Broadcast Corporation (CBC)—that had sent one representative to the press conference. Two brief radio interviews were done subsequently with CBC radio, one of which was broadcasted nationally later that evening. Yet, these two interviews did not generate further media inquiries or attention. Overall, then, the media’s (lack of) response to the press conference meant that the research team was largely unable to engage the media, especially the newspapers, in publicizing the study’s results. For a brief  moment, though, this disinterest was reversed: one week after the press conference, a woman committed suicide while segregated in a men’s prison in the neighbouring province of Saskatchewan. This tragedy did lead to several radio interviews (but only within the CBC network), and to another short newspaper article that mentioned the segregation study. Yet the tone of this particular newspaper article23 was again demeaning, illustrating again the particular kind of ‘knowledge policing’ in which the media can engage: The study is anchored by [. . .] anecdotal evidence [and] personal stories from unidentified women in unnamed Prairie prisons [which] makes responding to specific concerns difficult. It’s a report involving self-reported

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge  information from 12 offenders [. . .] talking about their opinions of segregation, that’s basically what the report is. (Star Phoenix, 2 February 2000, p. A7)

Obviously, a radical and notable change had occurred in the study’s knowledge claims once the media got wind of them: from social science, the study was now transformed into ‘anecdotal’. In this regard, Baudrillard (1981: 82, 86) might contend that in effecting this qualitative shift, members of the media were perpetuating their own kind of deterrence— that is, they were engaged in cooling down and neutralizing the meaning and energy of events. Specifically, Baudrillard (1981: 123) suggests that mediated information loses the meaning attached to events through its gigantic process of simulation (of reality) that short-circuits a priori all possibilities of communication. Through its involvement in a circular logic of absorption of meaning, the media participate in the gradual reduction of  the dialectics of communication, thereby creating a ‘mediated’ reality that can have little relationship to events initially reported. What we are left with is a simple imprint of a faceless study watered down through the mass processes of media banalization of images and ideas (Baudrillard, 1993: 9). As such, media information is a form of knowledge that does not go beyond itself. In further exploring the study’s weak media dissemination, it should be noted that certain types of stories tend systematically to be absent from news coverage (Hackett and Gruneau, 2000). Among the most frequent omissions in the Canadian press, Hackett and Gruneau (2000: 166) list topics centring on women. Observers have long demonstrated women’s under-representation in the media either as journalists or as subjects of  news (e.g. Voumvakis and Ericson, 1984; MediaWatch, 1993; McCormick, 1995). Moreover, when the media do prioritize topics about women, underlying social and political causes are likely to be overlooked, or obfuscated, in resulting accounts. This filtering notably stems from a still-frequent lack of solicitation of  women as ‘experts’ and as sources of information (Hackett and Gruneau, 2000). This gender imbalance may arise in part from the media’s propensity to rely on institutional and political ‘expertise’; in practice, this means that even after a quarter of a century of feminist struggles, men’s perspectives are still likely to be accorded hugely disproportionate expression. Simultaneously, alternative types and contents of knowledge developed by women in relation to other women are often relegated to the margins of public debate. Thus, the media play a considerable part in reproducing perceptions of women’s issues as peripheral, and somehow less than ‘real’, social problems; women, and often people of African ancestry as well, thereafter come to be seen as occupying marginal discursive spaces. Several lessons can be drawn from this discussion of the media’s role in disseminating and legitimating knowledge. One concerns the press release and subsequent press conference just described. According to Gieryn

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

175

176

Theoretical Criminology 8(2)

(1999: 188), these must feature three things to be deemed, by the media, successful: first, scientific (or social scientific) claims must look more like fact than fiction; second, important consequences must seem to be entailed; and, third, original and unique research findings must be presented. In two respects, the press release about the segregation study met these criteria by presenting serious research claims and by offering unusual (and therefore seemingly ‘catchy’) findings. However, the claims may not have been received as ‘important’ in so far as not boding wide ramifications except for women and correctional authorities. This third point, then, may have contributed to lack of media interest in the press conference that announced the segregation study’s results. But Gieryn (1999: 189–91) also argues that key to a successful press conference announcing research results is establishing the ‘mood of normal science’ such that claims are conveyed in mathematical discourse (the emblem of science) and perplexing technical language. In addition, Gieryn suggests, claims need to be set within controlled experiments and calibrated measurements and must come to the only logical conclusion after eliminating alternative explanations. Finally, they must come across as resulting from strict adherence on the part of the researcher to the normative conventions of the scientific community. In this respect, though, our press conference had virtually the opposite characteristics. The research team had made deliberate efforts to exclude complex methodological issues and analyses and to ‘tone down’ discourse in order—so we thought—to present findings in a jargon-free manner that would attract the media’s attention. Obviously this strategy proved to be inadequate as the delivery of the message did not produce extensive (or serious) coverage in the next edition or the nightly news. Yet scholarly advocates for individuals whose voices are often rendered silent, and whose perspectives are overlooked, need to be politically savvy. So would we do it again? Perhaps the press conference’s avoidance of complex methodological issues too easily allowed the mainstream press to dismiss the study. A more elaborated setting of the ‘mood of normal science’ might have compelled the attendance of more media representatives, ensured better coverage of the findings and broadcast a more favourable image of the research team and its research practices (Lievrouw, 1990). On the other hand, as noted earlier, quantitative data about prison segregation in Canada overall that would have helped to lend the study a more ‘scientific’ air are not kept, and therefore were not available for presentation at the press conference. In this respect, the situation left researchers in a frustratingly ‘catch 22’, and indeed itself  marginalized, position. Another valuable lesson, though, relates to cultural boundaries between the media and science. In the natural sciences, this boundary is rather impervious and diligently maintained—the media in one space, science in another—through the active participation of scholarly stakeholders. For example, in the now infamous cold fusion24 matter, Gieryn (1999) shows how in 1989 two American physicists—claiming they had successfully

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

sustained nuclear fusion at low temperatures—circumvented the conventional claims-adjudicating process of science (i.e. the peer review), and brought their claims directly to the media. He then proceeds to show how other physicists assembled at an American Physical Society meeting in Baltimore one month later painstakingly restored the traditional boundary between scientists and the media by delegitimizing the cold fusion claim by excluding the latter as an adjudicator of the truth or falsity of natural claims. As with other social sciences, though, one might say that boundaries between criminologists and the media are much less discrete. Historically, criminologists have derived a good deal of credibility from transporting claims-making and claims-adjudication processes into the public sphere. Criminologists and journalists depend on one another; criminologists are often ‘talking heads’ for crime news, and the media are often conduits for spreading the latest criminological policy-oriented theories. But these conduits have been closed for some stakeholders if not others, an observation that has significant ramifications for some subjects of criminological knowledge—especially women. In fact, when knowledge claims do not garner support and are quickly discredited, the individuals or social groups from whom the knowledge originates—that is, actual research subjects— are often affected. So what might we have done differently with regard to this second lesson? Given blurred boundaries between the media and criminologists, perhaps our research team might have been more successful in enrolling media interest had we waited to release our research report simultaneously with a likely-to-be-publicized ‘prison incident’. Although this would have posed logistic (and arguably ethical) difficulties, ironically enough, resultant media interest would likely have been greater. While it is well known that social groups may be negatively affected by knowledge claims that turn into scientific ‘truths’,25 less discussed is how they can also suffer from discredited knowledge. This poses particular problems for historically marginalized groups, like women, whose political clout has been kept at bay by dominant interests in society. And this is even more so in the case of women in prison, especially women in prison segregation. Failure to recognize and validate the knowledge claims promoted in the segregation study has rendered the experiences of this group of incarcerated women invisible, further disenfranchising an already excluded social group. Indeed, experiences of oppression are particularly blatant in prison contexts where women are constructed as outsiders, and as particularly illegitimate claimsmakers; this marginal ‘status’ is only exacerbated when, as here, it is echoed by a study of segregated women’s experiences becoming marginalized as well. Precisely because the segregation study failed to enrol sufficiently supportive actors to produce social scientific legitimacy—especially among authorities with the power to alter oppressive segregation practices—women in prison continue to suffer repressive and punitive treatment.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

177

178

Theoretical Criminology 8(2)

First site again: a second round for the segregation study Two reasons prompted me to undertake further research on this topic even though the pilot study garnered only small support from target audiences, and was effectively marginalized once media interest failed to materialize. The first was that a new research project would provide much ‘stronger’ knowledge claims; this time, data would be culled from a combination of  sources and from varied methodological approaches. The second reason was that as an exploratory and relatively localized26 piece of research into women’s experiences of prison segregation, the pilot study produced insights begging additional investigation. Thus, an extensive project was put together and submitted to the Social Sciences and Humanities Research Council of Canada, the most prominent national agency that funds social academic research. The research proposal was submitted to the Criminology Evaluation Committee since the subject matter of the research was evidently criminological. As Latour (1987) underscores, funding agencies are primary sources of  network building. Yet, in the case of the Canada-wide research project on segregation, the proposal did not find takers for the knowledge claims it brought forth. The project failed to get funding in two consecutive attempts. Despite changes made in the second year to meet assessors’ comments, the negative reviews took precedence over more positive ones. These negative reviews largely focused on methodological components judged to be weak or lacking and, both times, critiques presumably faulted the project’s inductive mode of obtaining knowledge. For example, reviews have criticized a lack of clear a priori research questions posed to interviewees. Yet, for many qualitative researchers, knowing in advance the exact questions interviewees will be asked is, in practice, impossible. This is because the character and direction of in-depth interviews may be dictated at least in part, and often in large part, by the interviewee herself rather than by researchers whose role thereby becomes limited to maintaining the interview’s broad research goals (Strauss, 1987: 27). But this relatively ‘unmediated’ approach to interviews may make ‘mainstreamers’ uncomfortable; these approaches tend to be perceived as inductive and, therefore, flawed. In contrast, presenting research as the ‘product of the machine and the computer’ is more likely to appeal to funding agencies that may ‘fear the controversial’, viewing research that does not yield clear answers as suspect (Bosworth, 2001: 438). Clearly, this problem affects a wide array of  academic disciplines in addition to criminology including sociology, psychology, economics, accounting, nursing, education and history. After two unsuccessful submissions to the Criminology Evaluation Committee, the research proposal was sent to a different committee that seemed closer in its epistemological orientation to the proposal. This third attempt ended up in the Women’s Studies Evaluation Committee. Drawing on actornetwork theory, it may be said that the proposal was ultimately submitted

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

to a different network of assessors, one with a potentially different view of  what defines ‘good’ research. Indeed, without minimizing the often deeprooted politics involved in decision making within research funding agencies, acceptance this time of the proposal may have stemmed, in part, from the proposal being more kindred with the new assessors’ community of  ideas. This re-orientation, if sufficiently repeated, can produce what Bruce Arrigo calls a ‘chilling effect’ (1999: 10): researchers undertaking critically inspired scholarship may eventually give up submitting funding proposals to criminology evaluation committees, engaging albeit inadvertently in selfmarginalization.

Conclusion In this article, I strove to illustrate how the ‘policing of knowledge’ can cause knowledge claims to become relatively unsuccessful. Specifically, I recounted how a study of women’s experiences of prison segregation in Canada—a largely undocumented prison practice—was released to governmental entities, certain scientific communities and media outlets generally unsupportive of the claims the study aimed to legitimize. Neither criminological researchers working for Correctional Service Canada (CSC), nor media representatives who relied on these researchers’ authorized claims, supported the particular social scientific bases of knowing expressed through the segregation study. By the same token, these parties did not support the study’s call for social justice for women segregated under unacceptable conditions. More generally, though, this article also contributes to a larger discussion of how the sociology of knowledge—especially the policing of certain forms of knowledge through covert mechanisms—applies to criminology as an academic subfield wherein epistemological insights have been relatively overlooked. As Latour argues (1987), science has both a formal and a hidden profile; to date, though, few criminologists have systematically studied either. Consequently, I hope through this article both to have stimulated further inquiry and to have contributed to the scant extant literature by highlighting three key mechanisms through which knowledge still tends to be ‘policed’ in criminological research. The first mechanism is bipolarization. Knowledge still tends to be divided into paradigmatic universes, especially ‘qualitative’ versus ‘quantitative’ and ‘subjective’ versus ‘objective’ dichotomies, that are then habitually conceived as mutually exclusive. Such was the case when CSC commented about the pilot study that the subjective (i.e. qualitative) experiences of  women in segregation needed to be counter-balanced by ‘objective’ (i.e. quantitative) histories of segregation as aggregated in CSC’s logs and statistical sets—even though, it should be added, the latter data were not even available. Bipolarization also came into play when the Canada-wide research project was submitted to the national funding agency and twice

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

179

180

Theoretical Criminology 8(2)

rejected due to a perceived split between inductive and deductive reasons that fails to apprehend the complementarity of both approaches in many research endeavours. As evidenced in this article, though, a sobering consequence of bipolarization is that it maintains methodologies, conceptual frameworks and scholarly communities oriented around two different poles gradually perceived as dissimilar and unequal in the value of their contributions to social science. The second mechanism is banalization. By this, I mean social processes that trivialize alternative methodologies and knowledge claims. This mechanism, more than any other, was evident in this article’s recounting of  reactions to the segregation study. From cancelled meetings with governmental authorities to anecdotal comments about the level of education needed to perform qualitative research, the study became an insignificant event unconnected with the magisterial scope of ‘science’ and apparently unworthy of being taken seriously in its policy implications. From the positioning of the newspaper articles to the trifling writing style of journalists, knowledge claims were turned on their head and made to seem ridiculous. Likely, in and of itself, this process of banalization is enough to consign research results to the margins. However, this article also suggests that banalization can work in conjunction with bipolarization in so far as the latter, too, often entails trivialization of ‘qualitative’ and ‘subjective’ paradigmatic universes. (Note, though, that the reverse proposition is not as obvious: banalization does not necessarily result in bipolarization.) The third mechanism is the inter-penetration of criminological knowledge and the media that blurs boundaries between the two. As previously discussed, this inter-penetration is especially salient to criminology since this area of knowledge—involving social problems of broad public concern—derives a good portion of its credibility from interaction with the media. Moreover, this mechanism works in a circular and symbiotic fashion as media logics enter theory production (e.g. through theorization that may result from media hype) and social scientific logics enter crimerelated news (e.g. through solicitation of ‘experts’). In the process, alternative ideas that could weaken this co-dependency—for example, critiques of political economy or of the crime industry—are cordoned off. Thus, this inter-penetration of the cultural spaces of the media and criminology plays a sure part in the marginalization of knowledge. In themselves, these three mechanisms are certainly not unique to criminology per se. Science and technologies studies and feminist critiques suggest that similarly biased reactions also occur in other disciplines. However, the theoretical conclusions drawn in this article may prove especially relevant to criminologists in the following ways.27 For one thing, as I have also tried to show, actor-network theory can be used to investigate the legitimization (or lack thereof) of knowledge claims in criminology, an academic area of study that currently does not use ANT extensively. But ANT is an analytic framework that was developed in the area of science

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge 

and technology studies. Therefore it may be particularly applicable to criminology that has traditionally adopted the normative standards of the scientific method and presented itself as a science. Again, though, ANT also has ‘blind spots’ that necessitate resorting to other approaches such as (in this case) feminist perspectives to illuminate specific structures of inequality that permeate knowledge production. Thus, I have argued that substantial benefits arise from cross-fertilization of varied theoretical approaches. The theoretical conclusions of this article are also relevant to criminologists in another way: to some extent, they challenge the history of criminological knowledge itself. For the emphasis on ‘scientific explanation’ that has traditionally characterized many criminological approaches can protect the discipline from internal scrutiny of its own epistemological orientations. As Lynch argues, understanding alternative histories ‘is not part of  the pursuit of “criminology as science”’ (2000: 147). Indeed, following Bosworth’s (1999, 2001) appeal to move beyond criminology’s entrenched methodological traditions, I would contend that the field can greatly benefit from incorporating diverse critical and self-reflective perspectives. Again, though, criminology is not the only discipline that has participated in the objectification and marginalization of populations through the use of  scientific discourses (Foucault, 1975). Yet arguably too, criminological knowledge is increasingly produced in response to social demands rather than internally generated research interests. Nowhere, perhaps, is the production of knowledge more enmeshed with social policies and political agendas than in criminology. Criminological research either feeds criminal justice policies and practices, participating in their exponential growth, or it critiques them. One way or another, though, criminological research is inseparable from what Nils Christie (1994) refers to as the industry of crime control. Thus criminological ‘science’ is embedded in political realities that tend to steer knowledge construction away from epistemological orientations that could threaten the field’s very existence and relevance to punitive contemporary policies about crime and punishment. Consequently, while this article is applicable to other academic disciplines, it has particular meaning for criminology in so far as the latter has often been immune to self-reflexivity and introspection over the course of  its history. Moreover, this analysis bears on theories of criminological knowledge more specifically by exemplifying how feminist ‘oppositional knowledge’ (Comack, 1999: 300) is still sometimes subordinated within and outside this academic area. As the segregation study of women’s experiences in Canadian prisons demonstrated, oppressive mechanisms contribute to certain perspectives being overlooked altogether while others are ‘sanitized’ and relegated to the basement of ‘scientific’ criminology. Thus it is high time that dominant epistemological perspectives shift, and in so doing allow alternative knowledges and practices to become influential as well.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

181

182

Theoretical Criminology 8(2)

Notes 1 In Canada, provinces manage prison sentences of less than two years while the federal government has jurisdiction over sentences of more than two years. 2 Baby dolls are heavy non-flammable gowns usually barely tied to the sides and that offer little privacy to the body. 3 The Canadian Commission of Inquiry into Certain Events at the Prison for Women in Kingston, recorded a similar practice in its 1996 report on the ill treatment of several Aboriginal women in segregation following a violent confrontation with staff. 4 Expression used by a participant. 5 Social Sciences and Humanities Research Council’s Update and Report on the March Council Meeting , http://www.sshrc.ca/english/resnews/presdesk/  march2002.html, posted 11 April 2002. 6 See Eigenberg and Baro (1992) for an exhaustive review of this particular literature. 7 Bosworth (1999: ch. 3) provides an exhaustive review of criticisms made of  traditional research methodologies in terms of their gender-blindness. 8 Letter from Ms Nancy Stableforth, Deputy Commissioner for Women, CSC, 26 February 1998, p. 1, emphasis added. 9 Recent feminist studies have demonstrated that such bipolarization is unnecessary, and that quantitative data do not hinder feminist epistemologies and political strategies (e.g. Bowker et al., 1988; Baro and Eigenberg, 1993; Gaarder and Belknap, 2002). In actual fact, one may argue that the incorporation of both ‘styles of research’ (i.e. qualitative and quantitative) is desirable and necessary as both have rewards and pitfalls. For example, by leaning towards objectivity standards, quantitative data tend to strip human beings of their reality and dignity while qualitative methods, by leaning towards subjectivity, tend to engender a selective inattention to the generalizable. When reconciled, qualitative and quantitative methodologies may begin to contribute to develop a strong foundation upon which credible research may be built. The structure, measurement and consideration of the general (provided by quantitative data) ought to combine with the checks of lived experiences and the significance of the particular (characteristic of qualitative methods) to produce a more ‘faithful representation’ (Liebling, 1999: 164) of feminist concerns. Moreover, the use of  quantitative data in feminist research may prove to be a wise option strategically as it could propel feminist knowledge sufficiently out of the periphery to be recognized and legitimated. To engage in the theoretical debate around the perceived exclusionary nature of the quantitative/  qualitative divide is beyond the scope of this article. However, it is this author’s opinion that this debate is a moot discussion, given that the quality of a research depends neither on the type of sampling nor on the nature of 

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge  empirical evidence (quantitative or qualitative) but rather on whether it is ‘well constructed’ (Bourdieu, 1992: 57, author’s translation). Furthermore, the traditionally assumed distinction between (statistically) probabilistic and non-probabilistic sampling—with its adjoining impression that quantitative procedures are more rigorous—tends to operate a slippage of meaning that disqualifies qualitative sampling. Thus, this author concurs with  Jayaratne and Stewart’s (1991) and Pir`  es’ (1993, 1997) argument for a general conception of methodology that does not dictate fixed rules of  savoir-faire—therefore ruling out quantitative or qualitative strategies of  sampling in the process—but rather allows researchers to adapt methods, sampling and data as much as possible to their research object. 10 Letter from Ms Nancy Stableforth, Deputy Commissioner for Women, CSC, 26 February 1998, p. 3. 11 This writer was aware that at least two of CSC’s researchers were inclined towards qualitative epistemologies. 12 Letter from Ms Nancy Stableforth, Deputy Commissioner for Women, CSC, 26 February 1998, p. 3. 13 During these rounds of ‘negotiations’, CSC researchers commented several times to this writer that nothing guaranteed that women prisoners would be telling the truth as they tend to have a propensity towards lying and manipulation. 14 The pilot study opted for a combination of methodological approaches. Although qualitative interviews comprised the essence of the analytical material, quantitative data were used to establish the significance of  segregation as a topic of inquiry (e.g. data on women in Canadian prisons, on the over-representation of Aboriginal women prisoners, on the statistical profiles of those likely to be segregated, on the personal and institutional profiles of the participants). 15 One of these players, however, was only weakly supportive of the study. 16 On previous occasions, trials of strength had been attempted with correctional authorities to test and debate the claims promoted in the segregation study. 17 More than 12 women were lined up for interviews but the data started to saturate exceptionally early, around the 12th interview. 18 ‘Segregation Harms Female Prisoners. Report Exposes Difficulty of Conditions’, Edmonton Journal , 28 January 2000, p. B9. 19 The press conference was scheduled on a Monday because Sundays are usually down days where not much happens which leaves the media with little to report on Mondays. It was also planned early Monday morning, before journalists actively start developing their story of the day—usually by mid-day when probes disseminated in the morning are starting to pay off. Also, the press conference took place in a downtown building, close to the local media’s head offices. 20 First story on the 6 o’clock news on ITV television station, a privately owned global network, and on the publicly owned Canadian Broadcast

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

183

184

Theoretical Criminology 8(2)

21 22 23 24 25

26

27

Corporation (CBC). This story also made headlines the following morning on a local radio news show. ‘Segregation Harms Female Prisoners. Report Exposes Difficulty of Conditions’, Edmonton Journal , 28 January 2000, p. B9. ‘Bulletin: Women in Solitary Confinement are Lonely. $13000 in Tax Funds Used to Reach this Conclusion’, Edmonton Sun, 28 January 2000, p. 20. ‘U of A Study Details Hardships of Women in Solitary Confinement’, Star Phoenix, 2 February 2000, p. A7. The sustained nuclear fusion reaction obtained at room temperature (rather than at horrendously high temperatures). For example, the successful ‘blackboxing’ (Latour, 1987) of urinalysis as a well-established scientific technology for detecting illegal drug consumption has encouraged random testing of ‘high-risk’ populations such as athletes, prisoners and parolees. Such mandatory randomization of social control has had significant consequences on the lives of thousands of individuals. The pilot study investigated segregation in two Canadian provinces although interviewees referred to their experiences in several prisons in other provinces. I thank one anonymous reviewer for raising this important point.

References Albert, Mathieu and Paul Bernard (2000) ‘Faire Utile ou Faire Savant? La “Nouvelle Production de Connaissances” et la Sociologie Universitaire Qu´eb´ecoise’, Sociologie et Soci´  et´  es 32(1): 71–92. Amann, Klaus and Karin Knorr-Cetina (1990) ‘The Fixation of (Visual) Evidence’, in Michael Lynch and Steven Woolgar (eds) Representation in Scientific Practice, pp. 85–122. Cambridge, MA: MIT Press. Arrigo, Bruce A. (1999) ‘Critical Criminology’s Discontent: The Perils of  Publishing and the Call to Action’, The Critical Criminologist  (newsletter) 10(1): 10–13. Baro, Agnes and Helen Eigenberg (1993) ‘Images of Gender: A Content Analysis of Photographs in Introductory Criminology and Criminal Justice Textbooks’, Women and Criminal Justice 5(1): 3–36. Baron, Stephen W. (1989) ‘The Canadian West Coast Punk Subculture: A Field Study’, Canadian Journal of Sociology 14(1): 289–316. Baudrillard, Jean (1981) Simulacres et Simulation. Paris:  ´Editions Galil´ee. Baudrillard, Jean (1993) The Transparency of Evil. Essays on Extreme Phenomena. London: Verso. Bosworth, Mary (1999) Engendering Resistance: Agency and Power in Women’s Prisons. Aldershot: Ashgate Press. Bosworth, Mary (2001) ‘The Past as a Foreign Country? Some Methodological Implications of Doing Historical Criminology’, British Journal of Criminology 41(3): 431–42.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge  Bourdieu, Pierre (1969) ‘Intellectual Field and Creative Project’, Social Science Information 8(9): 89–119. First published in French in 1966. Bourdieu, Pierre (1984) Distinction: A Social Critique of Taste . Cambridge: Harvard University Press. Bourdieu, Pierre (1992) R´  eponses. Pour une Anthropologie R´  eflexive. Paris: Seuil. Bourdieu, Pierre and Jean-Claude Passeron (1990) Reproduction in Education, Society, and Culture. London: Sage Publications. Bourgois, Philippe (1995) In Search of Respect: Selling Crack in El Barrio . Cambridge & New York: Cambridge University Press. Bowker, Lee H., Michelle Arbitell and J. Richard McFerron (1988) ‘On the Relationship between Wife Beating and Child Abuse’, in Kersti Yll ¨o and Michele Bograd (eds) Feminist Perspectives on Wife Abuse , pp. 158–74. Beverly Hills, CA: Sage Publications. Bruckert, Chris (2002) Taking It Off, Putting It On: Women in the Strip Trade . Toronto, Ontario: Women’s Press. Callon, Michel (1986) ‘Some Elements of a Sociology of Translation: Domestication of the Scallops and the Fishermen of St Brieuc Bay’, in John Law (ed.) Power, Action and Belief: A New Sociology of Knowledge? , pp. 196–233. London: Routledge & Kegan Paul. Callon, Michel and John Law (1989) ‘On the Construction of Sociotechnical Networks’, Knowledge and Society: Studies in the Sociology of Science Past  and Present  8: 57–83. Canada (1997) Administrative Segregation: Commitment to Legal Compliance, Fair Decisions, and Effective Results. Report of the Task Force to Review Administrative Segregation. Ottawa, Ontario: Correctional Service of Canada. Carlen, Pat (1983) Women’s Imprisonment: A Study in Social Control . London & Boston, MA: Routledge & Kegan Paul. Christie, Nils (1994) Crime Control as Industry. Towards Gulags, Western Style, 2nd edn. London: Routledge. Code, Lorraine (1991) What Can She Know? Feminist Theory and the Construction of Knowledge. Ithaca, NY: Cornell University Press. Comack, Elizabeth (1993) Women Offenders’ Experiences of Physical and  Sexual Abuse: A Preliminary Report . Manitoba: Criminology Research Centre, University of Manitoba. Comack, Elizabeth (1996) Women in Trouble: Connecting Women’s Law Violation to Their History of Abuse . Halifax, Nova Scotia: Fernwood Publishers. Comack, Elizabeth (1999) ‘Producing Feminist Knowledge. Lessons from Women in Trouble’, Theoretical Criminology 3(3): 287–306. Commission of Inquiry into Certain Events at the Prison for Women in Kingston (1996) The Honourable Madam Justice Louise Arbour, Commissioner. Ottawa, Ontario: Public Works and Government Services, Canada. Crossen, Cynthia (1994) Tainted Truth: The Manipulation of Fact in America . New York: Simon & Schuster.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

185

186

Theoretical Criminology 8(2) Eigenberg, Helen and Agnes Baro (1992) ‘Women and the Publication Process: A Content Analysis of Criminal Justice Journals’,  Journal of Criminal Justice Education 3(2): 293–314. Eigenberg, Helen, Agnes Baro and Tammy Desnoyers (1992) ‘Women and Publication Patterns in Criminal Justice Journals: A Content Analysis’ (research note), Women and Criminal Justice 4(1): 165–70. Foucault, Michel (1975) Surveiller et Punir. La Naissance de la Prison . Paris: Gallimard. Foucault, Michel (1980) Power/Knowledge: Selected Interviews and Other Writings 1972–1977 . Brighton, Sussex: Harvester Press. Fox, Mary Frank (1995) ‘Women and Scientific Careers’, in Sheila Jasanoff, Gerald E. Makle, James C. Petersen and Trevor Pinch (eds) Handbook of  Science and Technology Studies , pp. 205–23. Thousand Oaks, CA: Sage Publications. Frigon, Sylvie (2003) L’Homicide Conjugal au F´  eminin. D’Hier a` Aujourd’hui. Montr´eal, Qu´ebec: Les  ´Editions du remue-m´enage. Fuller, Steve (1997) Science. Minneapolis, MN: University of Minnesota Press. Gaarder, Emily and Joanne Belknap (2002) ‘Tenuous Borders: Girls Transferred to Adult Court’, Criminology 40(3): 481–517. Genders, Elaine and Elaine Player (1986) ‘Women’s Imprisonment: The Effects of Youth Custody’, British Journal of Criminology 26(4): 357–71. Gibbons, Michael, Camille Limoges, Elga Nowotny, Simon Schwartzman, Peter Scott and Martin Trow (1994) The New Production of Knowledge: The Dynamics of Sciences and Research in Contemporary Societies . London: Sage Publications. Giddens, Anthony (1979) Central Problems in Social Theory: Action, Structure, and Contradiction in Social Analysis . London: Macmillan. Gieryn, Thomas F. (1983) ‘Boundary Work and the Demarcation of Science from Non-Science: Strains and Interests in Professional Ideologies of Scientists’, American Sociological Review 48(6): 781–95. Gieryn, Thomas F. (1999) Cultural Boundaries of Science. Chicago, IL: University of Chicago Press. Hackett, Robert A. and Richard Gruneau (2000) The Missing News. Filters and Blind Spots in Canada’s Press . Aurora, Ontario: Garamond Press. Hacking, Ian (1992) ‘The Self-Vindication of the Laboratory Sciences’, in Andrew Pickering (ed.) Science as Practice and Culture, pp. 29–59. Chicago, IL: Chicago University Press. Hamel, Jacques (1993) ‘Quelques Remarques sur la Singularit´e en Sciences Exactes, en Math´ematiques et en Sciences Humaines’, Diog`ene 161: 49–72. Hannah-Moffat, Kelly (2001) Punishment in Disguise: Penal Governance and  Federal Imprisonment of Women in Canada . Toronto, Ontario: University of  Toronto Press. Harding, Sandra (1991) Whose Science? Whose Knowledge? Thinking from Women’s Lives. Ithaca, NY: Cornell University Press.

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

Martel—Policing criminological knowledge  Hawkesworth, Mary (1989) ‘Knowers, Knowing, Known: Feminist Theory and Claims to Truth’, Signs 14(3): 533–77. Howe, Adrian (1994) Punish and Critique: Towards a Feminist Analysis of  Penality. London: Routledge.  Jayaratne, Toby E. and Abigail J. Stewart (1991) ‘Quantitative and Qualitative Methods in the Social Sciences: Current Feminist Issues and Practical Strategies’, in Mary Margaret Fonow and Judith A. Cooks (eds) Beyond  Methodology: Feminist Scholarship as Lived Research , pp. 85–106. Bloomington, IN: Indiana University Press. Knorr-Cetina, Karin D. (1981) The Manufacture of Knowledge: An Essay on the Constructivist and Contextual Nature of Science . New York: Pergamon Press. Latour, Bruno (1986) ‘The Power of Association’, in John Law (ed.) Power, Action and Belief: A New Sociology of Knowledge? , pp. 264–80. London: Routledge & Kegan Paul. Latour, Bruno (1987) Science in Action. Cambridge, MA: Harvard University Press. Latour, Bruno (1999) Pandora’s Hope: Essays on the Reality of Science Studies . Cambridge, MA: Harvard University Press. Law, John (1983) ‘Enr olement ˆ  et Contre-Enr olement: ˆ  Les Luttes pour la Publication d’un Article Scientifique’, Social Science Information 22(2): 237–51. Lewenstein, Bruce V. (1995) ‘From Fax to Facts: Communication in the Cold Fusion Saga’, Social Studies of Science 25(2): 403–36. Liebling, Alison (1999) ‘Doing Research in Prison: Breaking the Silence?’, Theoretical Criminology 3(2): 147–73. Lievrouw, Leah (1990) ‘Communication and the Social Representation of  Scientific Knowledge’, Critical Studies in Mass Communication 7(1): 1–10. Lynch, Michael (2000) ‘The Power of Oppression: Understanding the History of Criminology as a Science of Oppression’, Critical Criminology 9(1/2): 144–52. Maher, Lisa (1997) Sexed Work: Gender, Race, and Resistance in a Brooklyn Drug Market . New York: Clarendon Press. Martel, Joane (1999) Solitude and Cold Storage. Women’s Journeys of Endurance in Segregation . Edmonton, Alberta: ACI Communication. Martel, Joane (2001) ‘Telling the Story: A Study in the Segregation of Women Prisoners’, Social Justice 28(1): 196–215. McCormick, Chris (1995) Constructing Danger. The Mis/Representation of  Crime in the News. Halifax, Nova Scotia: Fernwood Publishing. MediaWatch (1993) Focus on Violence: Survey of Women in Canadian News papers. Toronto: MediaWatch. Messerschmidt, James (1997) Crime as Structured Action: Gender, Race, Class, and Crime in the Making . Thousand Oaks, CA: Sage Publications. Mitchell, Austin, Prem Sikka and Hugh Willmott (2000) ‘Policing Knowledge by Invoking the Law: Critical Accounting and the Politics of Dissemination’,

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012

187

188

Theoretical Criminology 8(2) paper presented at the Sixth Interdisciplinary Perspectives on Accounting Conference, University of Manchester, England, July. Motiuk, Larry and Kelly Blanchette (1997) Case Characteristics of Segregated  Offenders in Federal Corrections. Report no. R-57. Ottawa, Ontario: Correctional Service Canada. Moyer, Imogene (1986) ‘Gatekeepers for Academic Criminology: An Exploratory Study of the Status of Women’, paper presented at the Twelfth International Conference on Improving University Teaching, Heidelberg, Germany. O’Neill, G. Patrick and Paul N. Sachis (1994) ‘The Importance of Refereed Publications in Tenure and Promotion Decision: A Canadian Study’, Higher Education 28(4): 327–435. Pir`  es, Alvaro P. (1993) ‘Recentrer l’Analyse Causale? Visages de la Causalit ´e en Sciences Sociales et Recherche Qualitative’, Sociologie et Soci´  et´  e 25(2): 191–209. Pir`  es, Alvaro P. (1997) ‘ ´Echantillonnage et Recherche Qualitative: Essai Th´eorique et M´ethodologique’, in Jean Poupart, Jean-Pierre Deslauriers, Lionel-Henri Groulx, Anne Laperri`  ere, Robert Mayer and Alvaro P. Pir` es ´  emologiques et M´  (eds) La Recherche Qualitative. Enjeux Epist´  ethodolo giques, pp. 113–69. Montr´eal, Qu´ebec: Ga¨etan Morin. Rosenblum, Gerald and Barbara Rubin Rosenblum (1990) ‘Segmented Labour Markets in Institutions of Higher Learning’, Sociology of Education 63(3): 151–64. Said, Edward W. (1994) Representations of the Intellectual: The 1993 Reith Lectures. New York: Pantheon Books. Sanch`  ez-Jankowski, Martin (1991) Islands in the Street: Gangs and American Urban Society. Berkeley, CA: University of California Press. Shaylor, Cassandra (1998) ‘It’s Like Living in a Black Hole’: Women of Color and Solitary Confinement in the Prison Industrial Complex’, New England   Journal on Criminal and Civil Confinement  24(2): 385–416. Solicitor General (1998) Report on the Provisions and Operations of the Corrections and Conditional Release Act . CCRA five-year review, consolidated report. Ottawa, Ontario: Ministry of the Solicitor General. Spender, Dale (1981) ‘The Gatekeepers’, in Helen Roberts (ed.) Doing Feminist  Research, pp. 186–220. Boston, MA: Routledge & Kegan Paul. Stebbins, Robert A. (2001) ‘The Aid to Scholarly Publication Programme: Contributions to Canadian Sociology’, Canadian Journal of Sociology 26(3): 453–68. Strauss, Anselm L. (1987) Qualitative Analysis for Social Scientists . Cambridge: Cambridge University Press. Task Force on Federally Sentenced Women (1990) Report Creating Choices. Ottawa, Ontario: Correctional Service Canada. Van den Hoonaard, Will (2001) ‘Is Research-Ethics Review a Moral Panic?’, Canadian Review of Sociology and Anthropology 38(1): 19–36. Voumvakis, Sophia E. and Richard Ericson (1984) News Accounts of Attacks

Downloaded from tcr.sagepub.com at University of Warwick on October 26, 2012