Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Indiscipline, punishment, gender and race : examining Discipline and Punish in the context of the prison… Pemberton, Sarah 2011

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2011_spring_pemberton_sarah.pdf [ 1.75MB ]
Metadata
JSON: 24-1.0071667.json
JSON-LD: 24-1.0071667-ld.json
RDF/XML (Pretty): 24-1.0071667-rdf.xml
RDF/JSON: 24-1.0071667-rdf.json
Turtle: 24-1.0071667-turtle.txt
N-Triples: 24-1.0071667-rdf-ntriples.txt
Original Record: 24-1.0071667-source.json
Full Text
24-1.0071667-fulltext.txt
Citation
24-1.0071667.ris

Full Text

INDISCIPLINE, PUNISHMENT, GENDER AND RACE: EXAMINING DISCIPLINE AND PUNISH IN THE CONTEXT OF THE PRISON SYSTEMS OF THE UNITED STATES, AND ENGLAND AND WALES  by SARAH PEMBERTON B.A., Merton College, Oxford, 2001 M.A., The University of British Columbia, 2003  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES (Political Science)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  MARCH 2011  © Sarah Pemberton, 2011  Abstract This dissertation explores how the changing philosophy and practices of criminal punishment in the United States, England and Wales reflect broader techniques and relations of power in these societies. Two questions motivate this research: firstly, the extent to which Michel Foucault’s account of power and the prison is applicable now; and secondly, whether Foucault's later work provides an adequate conceptual framework for theorizing the aspects of power that he either overlooks or inadequately addresses in Discipline and Punish. I identify three trends in criminal justice over recent decades that challenge Foucault's account of penality: sharply rising incarceration rates, prison privatization, and the racialized and gendered nature of prison populations. I argue that although Foucault's concept of disciplinary power remains applicable, a fuller understanding of contemporary penality requires an analysis of how race and gender are constituted through biopower, and of how neoliberalism has shaped penal policy and contributed to greater socioeconomic inequality. Although I conclude that Foucault's theorization of power and the prison in Discipline and Punish is inadequate in light of the racialized and gendered nature of power relations in both historical and contemporary criminal justice systems, I draw on his later work to re-theorize power and inequality. I argue that Foucault's analysis of sex, sexuality, and race provides a valuable conceptual framework that generates important insights, particularly through the concepts of biopower and state racism. However, I critique aspects of Foucault's later work, arguing that his analysis of race is inattentive to the inter-relation of race, class and capitalism; that his analysis of sex and sexuality overlooks the question of gender; and that his account of neoliberalism is more descriptive than analytical. I therefore combine the conceptual framework provided by Foucault with insights from feminist theory, queer theory and critical race theory to show how racialized, gendered and sexed identities become constituted within institutions such as the prison. My conclusion is that criminal justice and prison systems serve to construct and reinforce racialized and gendered identities, and thereby contribute to racialized and gendered inequalities that extend far beyond the prison system.  ii  Preface Some of these research conclusions have already been published in the following article: Pemberton, Sarah. (2009) Neoliberal Prisons: Revisiting 'Discipline & Punish' in the TwentyFirst Century. In Sam Binkley & Jorge Capetillo (Eds) A Foucault for the 21st Century: Governmentality, Biopolitics and Discipline in the New Millennium (pp.255-268). New Castle upon Tyne: Cambridge Scholars Publishing.  iii  Table of Contents Abstract .................................................................................................................................... ii Preface .................................................................................................................................... iii Table of Contents .................................................................................................................... iv Acknowledgements ................................................................................................................. vi Chapter One: Introduction ........................................................................................................1 Overview ............................................................................................................ 1 Understanding Genealogy ................................................................................ 10 Analysing Discipline and Punish ..................................................................... 25 Chapter Two: Histories of the Prison - Assessing the Literature ............................................30 Overview .......................................................................................................... 30 Marxist-Inspired Analyses ................................................................................ 35 Analyses of Gender and/or Race ...................................................................... 46 Foucaultian, and Counter-Revisionist Analyses ............................................... 54 Chapter Three: Privatization, Penal Severity, and Neoliberalism .......................................... 67 Overview, and Notes on the Use of Statistics .................................................. 67 Rising Prison Populations and Incarceration Rates .......................................... 73 Prison Privatization .......................................................................................... 80 The Relationship Between Punitiveness and Privatization .............................. 85 Responding to Fraser's "From Discipline to Flexibilization" ........................... 88 Chapter Four: Neoliberalism and Foucault ............................................................................ 98 Overview .......................................................................................................... 98 Foucault on Governmentality and Neoliberalism ...........................................100 Analysing Foucault's Account of Neoliberalism ............................................ 110 1. Tensions within neoliberalism ...................................................... 111 2. Governmentality, and coercion .....................................................120 3. Economic aspects of power .......................................................... 124 Chapter Five: Theory of Sex, Gender and Race ...................................................................134 Overview ........................................................................................................ 134 Sex and Gender ...............................................................................................137 iv  Race ................................................................................................................ 153 Intersectionality .............................................................................................. 170 Chapter Six: The Constitution of Race in Prison Systems ................................................... 175 Overview ........................................................................................................ 175 Biopower and the Construction of Blackness ................................................ 178 Race, Ethnicity and Incarceration ...................................................................197 Conclusion ...................................................................................................... 214 Chapter Seven: The Constitution of Sex/Gender in Prison Systems ................................... 218 Overview ........................................................................................................ 218 Sex, Gender and Imprisonment ...................................................................... 222 1. Transgender people in the UK ...................................................... 224 2. Transgender people in the US ...................................................... 232 3. Enforcing sex/gender in prisons ................................................... 237 Intersectionality and Black Masculinity ......................................................... 248 Chapter Eight: Conclusion ................................................................................................... 258 Summary ........................................................................................................ 258 Future Directions in Penal Policy ................................................................... 267 Questions for Future Research ....................................................................... 278 Bibliography ........................................................................................................................ 282  v  Acknowledgments I am indebted to my co-supervisors Barbara Arneil and Laura Janara, and my committee member Bruce Baum, for all their advice and support. I am also grateful to the following people for their suggestions and feedback on aspects of my research: Michael Behrent, John Bretting, Christian Dean, Rita Dhamoon, Ron Haas, Susan Mezey, Lester K Spence, and Christopher Young.  vi  Chapter One: Introduction Overview This research project addresses two broad questions about Foucault's work and its relevance to the techniques and relations of power in contemporary America, and England and Wales. The first research question is most simply expressed as follows: to what extent is Foucault's account of power and the prison in Discipline and Punish persuasive today? The test of persuasiveness is threefold: firstly, whether Foucault's (1995) analysis is empirically plausible in a minimal sense; secondly, whether it provides important insights about power and penality, and; thirdly, the degree to which Foucault's analysis captures all of the important features of power relations in these prison systems. My conclusion, briefly stated, is that Foucault's (1995) account is empirically plausible and generates some important insights, but that there are crucial features of power relations around punishment that Foucault does not adequately address in Discipline and Punish. In particular, Foucault's account of penality entirely overlooks racialized and gendered over/under-representation amongst prisoners and thus fails to consider how disciplinary power relates to race and gender, and provides inadequate analysis of the impact of class and capitalism. My second research question builds on the first, by considering whether applying the insights of Foucault's (1980b, 1990, 2003, 2007 & 2008) later work generates a persuasive account of the techniques and relations of power in the United States, and England and Wales. Although Foucault (1995) ignores the subjects of race, sex, sexuality, and neoliberalism in Discipline and Punish, he explores all these issues in his subsequent books and/or his lectures at the Collège de France. I will therefore consider whether the conceptual framework provided in Foucault's later works is sufficient to develop a persuasive theorization of the aspects of power 1  that he either overlooks or inadequately addresses in Discipline and Punish. The test of persuasiveness here is twofold: firstly, whether Foucault's work on race, sex, sexuality, and neoliberalism provides valuable insights about power and penality, and; secondly, the extent to which Foucault's conceptual framework captures the important aspects of how race, gender, class and capitalism relate to penality. Here, my conclusion is that Foucault's later work provides valuable insights about both race and gender, particularly through the concept of biopower, which shows how racialized and sexed categories and identities are constructed. However, I argue that it is necessary to reject and/or modify elements of Foucault's (1980b, 1990, 2003, 2007, & 2008) conceptual framework in order to understand contemporary power relations in the United States, and England and Wales, particularly in relation to socioeconomic inequality and neoliberalism. In this dissertation I will argue that while Foucault's (1980b & 1990) account of sex and sexuality shows the constructedness of binary sex categories, it overlooks the techniques of power involved in creating gendered identities and inequalities, and the relationship between gender and sex. Similarly, Foucault's (2003) analysis of race provides the useful concepts of biopower and state racism, but fails to explore the crucial links between race, class and capitalism. I conclude that Foucault's (2007 & 2008) account of neoliberalism and neoliberal techniques of government is flawed, because it fails to identify the tensions within neoliberal theory, the coercive aspects of neoliberalism, and the normative implications of neoliberal economic policies. Although Foucault's later work certainly generates insights that help to understand the power relations around race and sex, and to a lesser extent also around neoliberalism, there are important aspects of contemporary power relations that Foucault does  2  not capture. I will therefore draw on feminist theory, queer theory, critical race theory, and Marxist theory to amend and supplement the conceptual framework provided by Foucault. My case studies for addressing these two questions are the prison systems of the United States, and England and Wales, and to a lesser extent the broader criminal justice systems of these countries, including the police and courts 1. The significance of this project lies in the fact that Foucault (1995) uses close analysis of the prison to identify and critique disciplinary power, which he argues was exemplified in the prison, but then became widespread and central to modern government. Foucault's analysis of the prison is therefore a way of generating a critical theory of power dynamics in society as a whole — the prison is a case study that enables him to develop a broader theory. In particular, Foucault regards criminal punishment as exemplifying the way that different techniques have worked upon the body, and it enables him to develop an account of how disciplinary power constitutes the subject, or as Foucault puts it: "The soul is the effect and instrument of a political anatomy; the soul is the prison of the body" (1995, 30). If Foucault's analysis of the prison is flawed, or is no longer applicable, then it calls into question the relevance and persuasiveness of his analysis of disciplinary power, docile bodies, and of the self. My research follows Foucault's approach by focusing on the prison, but using it as a case study to generate critical insights about the techniques and relations of power in contemporary societies. Like Foucault, I will be paying attention to the ways that power works through and upon the body, to how power shapes the identities of subjects, and to how penal policy and The reason I use the term 'England and Wales' instead of United Kingdom (UK), or Great Britain, is that there are three distinct legal systems within the UK, with their own criminal laws, courts, and penal institutions. Although England and Wales are two distinct nations within the UK, they are a single jurisdiction for the purposes of criminal justice and share criminal law, a court system, and a penal system. In contrast, both Scotland and Northern Ireland are distinct jurisdictions, with their own criminal law, courts, and penal systems. In this research I will be analysing the criminal justice and prison systems in England and Wales, but will not be considering Scotland or Northern Ireland. Since England and Wales is a single jurisdiction, reports and data about policing and imprisonment refer to 'England and Wales' as a unit. 1  3  philosophy relates to broader political ideologies and forms of governance. My decision to use the prison as my case study is shaped both by Foucault's choice of case study, and by other theoretical and empirical research that identifies the prison as a key institution in producing forms of identity and inequality, particularly in relation to race. There are three major reasons for selecting the United States, and England and Wales as my cases. The first reason is that both these states are important in the historical origin and development of the prison, and provide many of the examples of prison regimes discussed by Foucault in Discipline and Punish. The most important example Foucault (1995) uses to demonstrate the nature of disciplinary power is Jeremy Bentham’s design for the Panopticon, which Bentham wrote and proposed to build in England. However, Foucault (1995) also discusses the philosophy and practices of several American prisons, including the Walnut Street Prison in Philadelphia (123-6), Auburn prison in New York (238), and Cherry Hill prison (also known as Eastern State Penitentiary) in Philadelphia (239). Given that Foucault’s account of the historical development of the modern prison draws heavily on examples from the English and American prison systems, it seems appropriate to use these cases. The second reason for choosing to analyze prisons in the United States, and England and Wales, is that these cases provide a good basis for comparison, since they have both notable similarities (including the trend of penal punitiveness, and racial over/under-representation amongst prisoners) and interesting differences (e.g. the US has a far higher incarceration rate). Moreover, I believe there is particular advantage to analyzing the American prison system in a comparative manner, because many detailed studies of the US prison system tend to presume American exceptionalism and do not consider punishment in other jurisdictions. I hope that this comparative approach will help to distinguish the impact of the specific American history, 4  policies, political and socioeconomic context from broader dynamics around power and penality, such as the impact of neoliberalism. The third reason for choosing the US, and England and Wales as my case studies is that there is a substantial body of literature about the history of prisons and about the contemporary criminal justice and prison systems, including issues such as trends in sentencing, and racialized biases in policing and punishment. By analysing this preexisting empirical research it is possible to develop theoretical insights with respect to the power dynamics in the English and Welsh, and American prison systems. I begin in this chapter by exploring the nature and purpose of Foucault's genealogical method, and then briefly outlining Foucault's (1995) arguments in Discipline and Punish. In Chapter Two, I assess the persuasiveness of Foucault's (1995) historical account of power and the prison by analyzing a range of alternative historical accounts of punishment, including various critiques of Foucault's account. I argue that Foucault's (1995) argument meets the test of basic empirical plausibility, and that it generates important insights, but that there are key aspects of power relations around criminalization and punishment that he does not consider. I conclude that Foucault's (1995) genealogical analysis is partially undermined by its failure to take account of the important role of race and gender in penality, and to explore the implications of this for prevailing techniques of power. In the remaining chapters I analyze the extent to which Foucault's (1995) account of power and the prison is relevant and persuasive in regard to the contemporary prison systems of the United States, and England and Wales, and consider whether Foucault's later work adequately captures the power dynamics around race, gender, and class. In Chapter Three I identify two major trends in these prison systems over the last thirty years that challenge Foucault's (1995) account of power and the prison: the punitive trend in penal policy, which has 5  produced less rehabilitative prison regimes, rising prison populations and incarceration rates; and prison privatization. I explore the normative implications of these two trends, arguing that both trends are linked to the rise of neoliberalism, and that while Foucault's account of the disciplinary prison is less relevant than it was during the 1970s, his insights about disciplinary power remain applicable. In Chapter Four I consider the significance of neoliberalism for our understanding of contemporary techniques of power, particularly in relation to penality. Here I argue that Foucault's (2008) account of neoliberalism lacks the critical impetus and insights provided by much of his other work, and that he overlooks tensions within neoliberal theory, particularly between neoliberal policies on welfare and criminal justice. I conclude that neoliberal policies have increased socioeconomic inequality, and have changed the governance of the poor through welfare cuts and the expansion of policing and imprisonment, none of which are adequately addressed by Foucault. In Chapter Five I draw on Foucault's (1980b, 1990 & 2003) work on sex, sexuality and race, particularly the concept of biopower, and combine it with insights from feminist theory, queer theory and critical race theory in order to develop an account of how sex, gender and racial identities are constituted. This conceptual framework is then used to analyze dynamics of power around race, ethnicity and imprisonment in Chapter Six. I argue that the contemporary English and Welsh, and American criminal justice and prison systems play a major role in constructing racialized identities and inequalities, through practices such as racial profiling by police and the huge over-representation of many racialized and ethnic minorities in prison populations. Lastly, in Chapter Seven I explore the construction of sex and gender identities through prison regimes, with particular attention to the policies and practices around transgender prisoners. I conclude that binary sex and gender identities are constructed through a combination 6  of sovereignty, discipline and biopower, and that there are complex intersections between racialized and gendered power relations that mean prison regimes have a particular impact on some groups, such as young, poor, Black men. Lastly, in Chapter Eight I offer tentative observations about the future direction of penal philosophy and policy in the United States, and in England and Wales, and draw together the key points of my analysis throughout this dissertation. In assessing Foucault's (1995) account of power and the prison I have sought to distinguish the question of whether Foucault provides a persuasive historical analysis of penality (which I explore in Chapter Two) from the question of whether his analysis is persuasive and relevant now. My reason for distinguishing these two questions is that Foucault's (1995) analysis of penality was written during the 1970s, and the power dynamics around criminal punishment now may be very different. In order to understand the importance of separating these questions, it is instructive to consider Fred Alford’s (2000) article about Foucault and prison regimes. Alford sets out to analyze whether Foucault provides a compelling account of power and the prison, but proceeds to assess the persuasiveness of Foucault's account of the prison solely against contemporary American prison regimes. The conclusion Alford reaches is that Foucault’s analysis of power and the prison is wrong, because the American prisons that Alford examines do not resemble Foucault’s account of the disciplinary prison with its use of surveillance, bodily regimentation, and the timetable. There are two major methodological problems with Alford’s (2000) response to Foucault's work, and these problems largely invalidate his conclusions. Firstly, Alford seems to overlook the fact that Discipline and Punish is "a history of the present" (Foucault, 1995, 30-31) written during the early 1970s. This context is significant, because differences between what 7  Foucault argues and what Alford observes may reflect changes in prison regimes over time, and not errors in Foucault's analysis. Secondly, Alford misinterprets Foucault's argument in Discipline and Punish by treating it as a descriptive account of practices within the prison, instead of as a critical analysis of power relations. I will briefly discuss both of these methodological problems, suggest how Alford could have avoided them, and indicate how I will attempt to avoid such methodological difficulties in my own analysis. Foucault (1995) develops his account of disciplinary power through a genealogical analysis that draws out important continuities and discontinuities between the eighteenth and nineteenth centuries and Foucault’s present during the 1970s. Not only does Foucault (1995) never claim that his analysis applies beyond his own context, he also (1980a) strongly criticizes methodological approaches that assume historical continuity. Instead, the genealogical methodology that Foucault favoured and uses in Discipline and Punish is based on drawing out the continuities and discontinuities between one’s own historical context and prior periods. Since Alford (2000) does not explore the continuities and discontinuities in penality, he cannot use observations about contemporary American prisons to conclude that Foucault is 'wrong' - it may simply be that things have changed. To judge whether Foucault’s (1995) analysis of power and the prison is ‘wrong’ as opposed to outdated, one would need to assess Foucault's account in relation to the historical periods that he analyzes. This need to address the same historical context as that used by Foucault is why I begin by assessing the persuasiveness of his account of power and the prison in reference to eighteenth and nineteenth century criminal punishment in England and America. The second difficulty with Alford’s (2000) analysis is that he — like some of the counter-revisionist critics of Foucault discussed in Chapter Two — treats Foucault’s (1995) 8  account as a description of prison regimes. Alford therefore presumes that any inaccuracies in this descriptive account automatically undermine Foucault's analysis of disciplinary power. This is a significant mis-interpretation, because Foucault's (1995) account is a genealogical analysis of power, not an exhaustive description of historical prison regimes. Foucault (1995) discusses those practices within the prison that he believes are important to understanding the functioning of disciplinary power, such as the prison’s spatial organization and the regimentation of activities. Conversely, Foucault (1995) does not mention other aspects of prison regimes such as the food or bedding, presumably on the grounds that detailing these prison practices would not contribute to his analysis of power. Given that factual accuracy — which Foucault does not believe to be possible in a neutral or objective sense — is not the point of Foucault's analysis, it is not necessarily a problem if his account of historical prison regimes is incomplete or inaccurate about minor aspects of prison regimes. Whether or not one is justified in criticizing Foucault for ignoring the details of a prison regime therefore depends on whether a persuasive case is made that those details are important to understanding broader relations and techniques of power. Moreover, one of the major issues that Foucault's (1995) analysis raises is about the nature of truth, and the relationship between truth and disciplinary power, so Alford begs the question by assuming that there are neutral facts about prison regimes against which Foucault's account can be assessed. Nonetheless, it is clear that ‘factual’ accuracy is at least somewhat relevant to our assessments of whether or not Foucault's account is persuasive. Given that Foucault develops his argument about the rise of disciplinary power by tracing changes in criminal punishment, it is also important that his arguments about penality pass a minimal test of empirical plausibility. For example, Foucault’s (1995) conclusions would also be undermined if his source materials 9  for Discipline and Punish were found to be fakes and not genuine historical documents, because his analysis would therefore fail as a genealogy. However, as I will explore in more detail in the next chapter, the success of Foucault's argument does not rest on the factual accuracy or inaccuracy of his account of specific historical prison regimes, as Dreyfus and Rabinow (1983) explain succinctly: There is obviously no simple appeal to the facts involved in evaluating Foucault’s historical theses...Most of these historians have simply misunderstood his argument and hence even their minor factual corrections are simply beside the point. (1983, 126) These problems with Alford’s (2000) analysis are instructive because they caution against making hasty or misguided ‘critiques’ based on minor empirical details, and against the idea that our present should be read through Foucault’s history of his present.  Understanding Genealogy In analysing Discipline and Punish, and specifically in considering whether Foucault’s (1995) arguments regarding the prison and the nature of power are persuasive, it is necessary to have a clear understanding of the text’s subject matter and methodology. I will therefore begin by exploring the following questions: is Discipline and Punish a history, what kind of history, and a history of what? The obvious starting point for this enquiry is the book’s introduction where Foucault uses a range of different terms to explain his project, beginning with the statement that: This book is intended as a correlative history of the modern soul and of a new power to judge: a genealogy of the present scientifico-legal complex from which the power to judge derives its bases, justifications and rules, from which it extends its effects and by which it masks its exorbitant singularity. (1995, 23). However, Foucault describes his project rather differently a few pages later, stating:  10  That punishment in general and the prison in particular belong to a political technology of the body is a lesson that I have learnt not so much from history as from the present… What was at issue was not whether the prison environment was too harsh or too aseptic, too primitive or too efficient, but its very materiality as an instrument and vector of power…I would like to write a history of this prison, with all the political investments of the body that it gathers together in its closed architecture. Why? Simply because I am interested in the past? No, if one means writing a history of the past in terms of the present. Yes, if one means writing a history of the present. (1995, 30-31, emphasis added) The latter statement diverges from the former both by referring to prisons, architecture and the body instead of to souls and judgement, and by describing Foucault’s historical project as "writing the history of the present" rather than using the term genealogy. Neither the term genealogy nor this phrase "history of the present" is particularly transparent, so both require some exposition. In particular, it is important to understand whether the use of genealogy in Discipline and Punish is intended as critique of truths regarding punishment, power and the subject, or as the articulation of an alternative, critical truth claim. The distinction between these two positions is an epistemological one, but it has important implications for how we interpret and respond to Foucault's work. The Marxist ideology critique developed by theorists such as Gramsci (1971) and Althusser (2001) holds that the objective truth about capitalist society is concealed by ideology, which causes workers to have false consciousness and thereby promotes the interests of the capitalist class. Marxist theory thus suggests that if the real, objective truths about capitalism and workers' interests are revealed, then capitalist hegemony will be replaced by a true understanding. In contrast, Foucault's (1995) analysis is intended to disrupt and undermine prevailing truths about penality, but is not founded on the idea of an underlying objective truth, and may not seek to establish an alternative 'truth'. In short, Foucault's task may not be to replace one set of truths about the prison with another set, but to problematize the very idea of truth.  11  This question of how genealogy relates to knowledge is central to the objections raised against Foucault’s work by scholars such as Jurgen Habermas (1994a & 1994b) who accuses Foucault of lacking normative justification for his claims, and of contradicting himself by both rejecting the notion of truth and making truth claims. Foucault (2000g) has responded by rejecting the binary logic presupposed by Habermas’s insistence that one either is, or is not, making claims to truth and reason, stating "It’s senseless to refer to "reason" as the contrary entity to nonreason… Such a trial would trap us into playing the arbitrary and boring part of either the rationalist or the nonrationalist" (299). In "Critical Theory/Intellectual History" Foucault (1994a) expands on this theme, explaining that he is working towards "an endless, multiple bifurcation" of reason (118) and stating: I am not prepared to identify reason with the totality of rational forms which have come to dominate – at any given moment, in our own era and even very recently – in types of knowledge, forms of technique and modalities of government or domination…for me, no given form of rationality is actually reason. (125) This question of whether genealogy involves some form of truth claim as well as critique is therefore central to debates about Foucault’s work in political theory, as well as to understanding his account of power and the prison. The usual reference point for academics seeking an understanding of Foucault’s concept of genealogy is the essay "Nietzsche, Genealogy, History" which was published in 1971, implying that Foucault wrote the essay either concurrently with or in close proximity to his work on Discipline and Punish. In this essay, Foucault (1980a) pieces together Nietzsche’s views about the historical technique of genealogy in texts such as The Genealogy of Morals, and explores the nature of genealogy by contrasting it to traditional history. Foucault (1980a) makes seven lengthy observations about genealogy in this essay, beginning with the claim that genealogy is a documentary process that examines the details and specificity of events in order 12  to undermine the teleology of traditional history. Secondly, Foucault explains that genealogy is intended "to dispel the chimeras of the origin" (1980a, 144) that obfuscate the nature of history by presuming continuity rather than recognizing historical change, inconsistency and the constructed nature of truth. Both Foucault’s (1980a) third and fourth observations follow from his second, since his rejection of the notion of the "origin" requires him to provide a different basis for the study of history. Here, Foucault argues that genealogy studies "descent" (1980a, 145) in order to show that the past consisted of a series of fissures, contradictions and accidents, and that descent is inscribed in the body since the body is marked by past experience. Foucault’s fourth observation is that genealogy studies "emergence" (1980a, 148) in order to understand the interplay of forces that produce different versions of human rules regarding law, morality et cetera. Foucault directly addresses the relationship between genealogy and knowledge in his fifth observation in "Nietzsche, Genealogy, History," stating that genealogy is oppositional and aimed at destabilizing the notions of the self, body and historical events that traditional history portrays. Here, Foucault (1980a) deploys the elegant phrase that "knowledge is not made for understanding; it is made for cutting" (154), and rephrases this insight by stating that genealogy is "a curative science" (156). Foucault (1980a) then explains more precisely how genealogy inverts the principles of traditional history by becoming an "effective" (154) history that studies events at their most unique, analyzes subjects closest to it, such as the body, and affirms "knowledge as perspective" (156). In contrast, traditional history stresses the continuity of events, prefers to analyze events at a distance and tries to conceal its perspective and context (Foucault, 1980a). Foucault’s sixth observation follows from his fifth insofar as he argues that traditional history’s claim to objectivity means that history will retain "suprahistorical" (1980, 13  160) elements unless genealogy is used to negate history. Foucault’s final observation consists of the claim that genealogy has three uses: parody, which turns genealogy against reality; dissociation, which turns genealogy against the notion of continuous identity; and a sacrificial deployment of genealogy against truth. Although "Nietzsche, Genealogy, History" is an important and interesting essay, it should not be the sole reference point for understanding Foucault’s conception of genealogy. One reason to look beyond this essay is that Foucault’s focus in "Nietzsche, Genealogy, History" appears to be to be the articulation of Nietzsche’s concept of genealogy rather than a statement of his own conception, and we should not presume that the two are identical2 . Moreover, Foucault discusses genealogy at some length in other texts, and his other accounts of genealogy sit in some tension with that expressed in "Nietzsche, Genealogy, History". For instance, in the 1975-6 Collège de France Lectures, Foucault proposes a ‘provisional’ definition of genealogy as follows: We can give the name genealogy to this coupling together of scholarly erudition and local memories, which allows us to constitute a historical knowledge of struggles and to make use of that knowledge in contemporary tactics. That can, then, serve as a provisional definition of the genealogies I have been trying to trace with you over the last few years. (2003, 8) This definition is far more concise than the earlier explanation in "Nietzsche, Genealogy, History," and places more stress on the contemporary, political significance of genealogy. It is because genealogy is concerned with developing knowledge that explains and disrupts dominant  2  While Foucault does not explicitly state that he is describing Nietzsche’s concept of genealogy, he does explain genealogy with almost exclusive reference to Nietzsche’s work and ideas. This focus on Nietzsche is apparent in his analysis of Nietzsche’s use of language (Foucault, 1980, 140-142) and in his attention to Nietzsche’s inconsistent opinions regarding the use of critical history (Foucault, 1980, 164). Interestingly, the word ‘I’ is only used within quotations in "Nietzsche, Genealogy, History", suggesting the absence of the explicit authorial voice present elsewhere in Foucault’s work, for instance in Discipline and Punish (Foucault, 1995, 7, 24, 30, 31 etc.) , and regularly in his Collège de France lectures.  14  forms of knowledge, events and conceptions of the self, that Foucault refers to Discipline and Punish as a ‘history of the present’. This ‘provisional’ definition of genealogy builds upon Foucault’s other claims in the 1975-6 lectures, where he characterizes detailed, local, historical information and forms of previously disqualified perspectives (e.g. psychiatric patients) as "subjugated knowledges" (2003, 7). Lest this seem very un-Foucaultian (whatever that might mean about someone who proclaimed the death of the author) Foucault goes on to explain that genealogical claims to knowledge are quite different from other knowledge claims. Foucault’s distinction between genealogy and other knowledges is made partially on the grounds of using different techniques, but largely on the grounds of their different stances in relation to what Marxists often — and Foucault occasionally, for instance in the essay "Truth & Power" (2000b, 133) — refer to as hegemony: Genealogies are, quite specifically, antisciences. It is not that they demand the lyrical right to be ignorant, and not that they reject knowledge or invoke or celebrate some immediate experience that has yet to be captured by knowledge. That is not what they are about. They are about the insurrection of knowledges. Not so much against the contents, methods or concepts of a science; this is above all, primarily, an insurrection against the centralizing power-effects that are bound up with the institutionalization and working of any scientific discourse organized in a society such as ours (Foucault, 2003, 9). It is significant that Foucault repeatedly uses the term ‘knowledge’ to describe genealogy in the 1975-6 Collège de France lectures, which seems in tension with his earlier (1980) account of genealogy as deployed against claims to truth and useful for cutting not understanding. Foucault's (2003) later articulation of genealogy suggests that it involves knowledge claims that contest and disrupt dominant discourses, most particularly ‘scientific’ discourses, rather than a critique intended to destabilize knowledge – a subtle but significant distinction. This raises the obvious question of why Foucault would be entitled to make a claim to knowledge where others 15  are not so entitled. One possible answer is that the genealogical truth claims are justifiable because they are strategic and counter-hegemonic, disrupting existing formations of power and knowledge through the ‘insurrection of knowledges’ (Foucault, 2003, 9). This explanation is inadequate, because it does not explain how Foucault’s claims escape the power-knowledge scheme for which he criticizes the human sciences. Further, if one were to justify knowledge claims purely on the grounds that they were counter-hegemonic, this could lead to the justification of repressive, not liberatory, knowledge claims which is one of the charges Fraser (1995) levels at Foucault in "Michel Foucault: A ‘Young Conservative’?". A more plausible answer to the question of why Foucault would be entitled to make knowledge claims can be found in his methodological approach, for instance his use of archaeology. Indeed, Foucault states in the same Collège de France lecture: Archaeology is the method specific to the analysis of local discursivities, and genealogy is the tactic which, once it has described these local discursivities, brings into play the desubjugated knowledges that have been released into them. That just about sums up the overall project. (2003, 10-11) What, therefore, is archaeology, how does it relate to genealogy, and why would it justify one in making claims regarding truth? Foucault’s use of archaeology predates Discipline and Punish, since it is the main topic of The Archaeology of Knowledge, and is the technique implicitly used in his earlier studies, The History of Madness and Birth of the Clinic. There is not a single, handy definition of archaeology in The Archaeology of Knowledge; instead, as in "Nietzsche, Genealogy, History", Foucault outlines the concept in terms of what it is not. In this case, Foucault (1972) explains archaeology as being the analysis of historical discourses, which it studies in their specificity and differences, i.e. without tracing continuity or progression. Foucault (1972) also states that archaeology rejects discursive analysis in terms of oeuvre or author because one should not pre-categorize the discursive context in which something is read. 16  Instead, Foucault (1972) puts particular emphasis on the fact that archaeology does not try to recapture the origin or restore what was thought in the past: "It is not a return to the innermost secret of the origin; it is the systematic description of a discourse-object" (140). From an epistemological perspective, the significance of archaeology is that it aids an understanding of discourse by showing how meanings change over time, the discursiveformation in which a discourse existed, and the rules that shaped what could be said and found true within that discursive formation. It is sometimes claimed (Dreyfus & Rabinow, 1983, 50; Macey, 1993, 200-2) that Foucault originally intended that archaeology would provide an exterior or ‘scientific’ perspective for the analysis of discourse, and at times Foucault’s (1972) claims in The Archaeology of Knowledge imply such a view. Such a view of archaeology as scientific analysis of discourse is untenable, because one cannot categorize and study discourses without engaging with their meaning to some extent, thereby making it impossible to stand outside of discourse (Dreyfus & Rabinow, 1983, 88). However, even if archeology fails as a science, it may provide some degree of distance and critical perspective towards present hegemonic understandings, because it highlights the changes in meaning and truth over time, and therefore shows the historical contingency and constructedness of what is seen as true in any given context. Whether or not Foucault believed in the possibility of an exterior perspective in the study of discourse, he did not adhere to such a view by the time that he started using genealogy and as such the issue is not important here. For the purpose of understanding how it relates to genealogy, it is sufficient to note that archaeology is a technique for providing highly detailed, systematic and historical accounts of discourse in order to distance oneself from current discourses.  17  This account of archaeology clearly overlaps with genealogy, because the disruptive claims of genealogy rest upon a detailed analysis of the forms of historical discourse and the constructed nature of truth (Foucault, 1980a & 2003). Further, Foucault (2003) explains that genealogy enables one to make knowledge claims because it is both strategic and archaeological: archaeology demonstrates the constructed nature of truth and unearths subjugated knowledges, which enables genealogy to develop a critical account of the present. Several prominent commentators support such an interpretation of Foucault’s methodology, including Dreyfus and Rabinow (1983) who claim that archaeology has a "purifying role" (106) within genealogy by enabling Foucault to recognize the constructed and discontinuous nature of discourse. Similarly, Dean (1994) argues that archaeology is a necessary component in genealogy because it makes possible analysis that is both critically detached from and involved with the present: The roles of genealogy and archaeology appear complementary, the latter performing analyses that are a necessary condition of the former. If, from the perspective of the production of a knowledge of discursive formations, archaeology remains the indispensable methodology, from the practical polemical and strategic perspective of the use of historical analysis, genealogy holds the key. However, beyond the language of complementarity, genealogy is clearly dominant. It connects the empirical analyses revealed to concerns activated in the light of contemporary struggles. (34) Referring back to Foucault’s phrase at the start of Discipline and Punish, Dean terms the combination of genealogy and archaeology a "history of the present" (1994, 35). Nonetheless, the concept of archaeology articulated by Foucault (1972) in The Archaeology of Knowledge may not be equivalent to the detailed, historical component of genealogical analysis. This distinction may be important, because while the former approach focuses on discourse, the latter approach pays close attention to the relationship between discursive and non-discursive structures, such as the construction of the subject through 18  discipline acting on the body (Foucault, 1995). Notably, Foucault’s genealogical analyses in Discipline and Punish and The History of Sexuality both trace the ways in which subjects are produced by techniques of power that work upon the body, and therefore appear distinct from the archaeological focus on language. The importance of the construction of the subject is emphasized in Foucault’s (2000b) account of genealogy in the interview "Truth and Power," in which he states: One has to dispense with the constituent subject, to get rid of the subject itself, that’s to say, to arrive at an analysis that can account for the constitution of the subject within a historical framework. And this is what I would call genealogy, that is, a form of history that can account for the constitution of knowledges, discourses, domains of objects, and so on, without having to make reference to a subject that is either transcendental in relation to the field of events or runs in its empty sameness throughout the course of history. (118) This articulation of genealogy explains the concept as providing a means to dispense with the notion of either a fixed or historically relative subject, in favour of an account that shows how the subject is constructed by configurations of power and knowledge. This analysis of genealogy shows that there are clear continuities between the various definitions and explanations that Foucault provides, including the fact that genealogy uses detailed, historical analysis, that it both presumes and demonstrates the constructed nature of the subject, and that there is a strong critical impetus. Nonetheless, there is an apparent contradiction between the vision of genealogy as operating in opposition to truth as claimed in "Nietzsche, Genealogy, History", and of genealogy as a form of knowledge as stated in the 1975-6 lectures. Here, it is necessary to differentiate between the notion of a single, unitary ‘truth’ and a broader terrain of knowledge, which includes partial, contested and situated knowledges. Without wanting to overstate the significance of this issue, it appears that by the 1975-6 lectures Foucault’s (2003) explanation of genealogy is clearly distinct from a 19  Nietzschean approach of rejecting and undermining truth. Instead, Foucault (2003) articulates a position of rejecting scientific discourses and any centralization of truth, whilst supporting the project of forging critical understandings of oneself and of the present. On such an understanding, Foucault employs genealogy in order to make partial, situated and strategic truth claims that act as counterpoint to existing truths, whilst simultaneously showing how the current configuration of power-knowledge emerged and how alternatives were quashed. Both these accounts of genealogy retain a critical intent regarding science and dominant structures of power and knowledge, suggesting that Foucault has modified rather than abandoned Nietzsche’s thought and methodology. This is consistent with several other interpretations of Foucault, including William Connolly’s (1988) description of him as a "LeftNietzschean" (189) and Leslie Thiele’s (1990) claim that Foucault makes a "selective adaptation" (923) of Nietzsche that celebrates agonism, resistance and otherness. An explanation for this modification of Nietzsche’s view on truth is offered by Thomas Dumm (2002), who argues that Foucault is committed to a much more expansive politics than the Nietzschean position of elitism, and that this necessitates overtly political and strategic analyses: ‘What is our truth?’ is the most important question informing the structure and substance of Discipline and Punish. Moreover, in contradistinction to Nietzsche, Foucault wants that question to be asked by ordinary citizens as well as critical intellectuals. The book is intended for a wide audience by design…Foucault is more concerned to seek the widespread engagement of those who have been marginalized by the normalizing forces of the modern era. (2002, 73-4) Dumm (2002) thus suggests that by developing situated and oppositional truth claims instead of rejecting the notion of truth or knowledge, Foucault is modifying Nietzsche’s methodological approach to suit a more inclusive and democratic form of critical politics. These questions regarding Foucault’s claims to knowledge and reason are at the heart of longstanding and intractable disputes about how to interpret and categorize Foucault’s ideas, and 20  I do not claim to have resolved the epistemological questions around Foucault's work or to have presented one ‘true’ reading of Foucault. Instead, I have merely tried to clarify the conceptions of genealogy that Foucault articulated at different times and in different texts, and to identify reasons to favour one articulation over another. Insofar as it is necessary to choose between Foucault’s conceptions of genealogy in order to analyze Discipline and Punish, I believe it is preferable to adopt the version articulated in the 1975-6 Lectures rather than that outlined in "Nietzsche, Genealogy, History". Although many commentators on Foucault treat "Nietzsche, Genealogy, History" as the definitive account of genealogy, I believe that the articulation of genealogy in the 1975-6 lectures is more compatible with Foucault’s other statements about knowledge and truth, including the essay "Truth & Power" and the 1978 "Interview with Michel Foucault." Foucault's (2003) account of genealogy also seems more consistent with his account of the inter-relation of power and knowledge in the introduction to Discipline and Punish, which suggests that altering power relations inevitably involves the articulation of counter-hegemonic or subjugated knowledges: We should admit that power produces knowledge… that power and knowledge directly imply one another; that there is no power relation without the correlative constitution of a field of knowledge, nor any knowledge that does not presuppose and constitute at the same time power relations. (1995, 27) Although Foucault’s various articulations of genealogy certainly complicate the task of interpreting his work, he cautions readers not to expect consistency: "Do not ask who I am and do not ask me to remain the same: leave it to our bureaucrats and our police to see that our papers are in order" (1971, 17). This account of genealogy as a combination of historical subjugated knowledges and contemporary political strategy allows us to return to the questions posed earlier: whether 21  Discipline and Punish is a history, what kind of history, and a history of what? The genealogical nature of Discipline and Punish answers the first two questions since it shows that while Discipline and Punish is a historical analysis, it is an unusual form of history since it is attentive to discontinuities, and denaturalizes current institutions instead of providing a causal explanation of their origins. Further, the fact that genealogy is deliberately concerned with providing critical, strategic analysis of contemporary events means it functions as a ‘history of the present’. Genealogy is therefore significantly different from conventional causal histories, which, Foucault would argue, are situated in their presents but serve to buttress existing configurations of knowledge and power. However, the genealogical nature of Discipline and Punish does not answer the final question posed above: ‘a history of what?’ The answer to this last question is not simple or singular, as evidenced by the fact that Foucault (1995) provides at least two different explanations in the introduction to Discipline and Punish: "a correlative history of the modern soul and of a new power to judge" (23) and "a history of this prison" (31-2). The key to understanding this dual answer is explained in the interview "Questions of Method", where Foucault explains: "I was aiming to write a history not of the prison as an institution, but of the practice of imprisonment" (2000e, 225). In this interview, Foucault stresses that the book is a study of the rationality of the prison and thus is concerned with power formations, political technologies, the nature of the subject, and the interconnection of power and knowledge. The study of the practice of imprisonment requires one to undertake the broader project of analysing the modern soul and the power to judge: Discipline and Punish is simultaneously and inextricably a historical study of the prison, of the subject, and of power. Foucault’s arguments in Discipline and Punish will be discussed later in this chapter, but first it is useful to understand the ‘present’ in and about which Foucault was 22  writing. Although Foucault’s ‘present’ is often ignored in considering his genealogical analysis, it is important because "writing a history of the present means writing a history in the present; self-consciously writing in a field of power relations and political struggle" (Roth, 1981, 43). Discipline and Punish was published in France in 1975, and was written during the six year period since the publication of The Archaeology of Knowledge in 1969. The book seems to have been shaped both by Foucault’s intellectual endeavours and by his considerable involvement in political activism on issues of criminal justice and prisons. The most direct political influence on the text may have been Foucault’s participation in the Group d’Information sur les Prisons (GIP) during 1971 and 1972, which worked to raise the visibility of prisons as a political issue and to publicize the concerns of prisoners. Foucault’s biographers Eribon (1991) and Macey (1993) each devote a chapter to discussing Foucault’s extensive work with the GIP, and Eribon sees a lot of continuity between the concerns of the group and Foucault’s later analysis in Discipline and Punish (229). While the GIP’s activities cannot be detailed here, their activism focused on gathering and publicizing information about prisoners’ experiences in order to discredit the criminal justice institutions, as the group’s first pamphlet states: The GIP does not propose to speak in the name of prisoners in various prisons: it proposes, on the contrary, to provide them with the possibility of speaking themselves… we hope that prisoners may be able to say what it is that is intolerable for them in the system of penal repression. (Eribon, 1991, 227) Foucault was also involved in criminal justice issues outside GIP, including his arrest at a protest in December 1972 over the murder of Algerian-born Mohammed Diad by French police (Macey, 1993, 313). One can also infer that Foucault was interested in prison issues outside France, because the George Jackson case in America was discussed in the GIP’s third pamphlet (Eribon, 1991, 228) and Foucault visited Attica prison in New York during April 1972 23  (Macey, 1993, 238). Significantly, this visit took place barely six months after the infamous Attica riots in which the prisoners rebelled against terrible prison conditions, and where thirtynine people were killed when State Troopers retook control of the prison from the (predominantly Black and Hispanic) prisoners (Benjamin & Rappaport, 1974). It is also clear that Foucault’s political activism predates the GIP, since he protested the arrest, detention and torture of leftist political activists and students in Tunisia, where Foucault was appointed at the University of Tunis from 1966 to 1968. Macey (1993, 205) details some of Foucault’s activities in support of the Tunisian student protestors, including hiding a printing press used for illegal pamphlets and financing the legal defence of the students. Macey (1993) and Eribon (1991) concur that by late 1968 Foucault felt it was unsafe to remain in Tunisia and thus returned to France, where he accepted a post at a new, experimental university at Vincennes. Foucault continued to be heavily involved with leftist student activism and was arrested in support of Vincennes students during January 1969 – only days after beginning the job (Macey, 1993). Eribon (1991) concludes of Foucault’s activity during this period: "From the alembic at Vincennes emerged an enraged philosopher, one who intervened on every front, whether in action or reflection. In 1969 Foucault began to embody the very figure of the militant intellectual." (210) Although prisons were comparatively high profile issues in France during the early 1970s, not least due to the work of the GIP and media coverage of the Attica riots, it is also worth noting the broader political context in which Foucault found himself. Foucault was not in France during the student riots of 1968, but the riots had a profound impact upon the political and intellectual climate to which he returned. Following the events of 1968, assorted leftist groups were involved in French political activism, including the French Communist Party (PCF) 24  and a range of Maoist and Leninist groups. Eribon (1991) and Macey (1993) both describe Foucault as cooperating with these leftist groups as some times, whilst opposing many of their ideas such as the concept of a people’s court. The record, therefore, shows that Foucault’s political and intellectual activities positioned him in tension with the French government and with many in the political and intellectual left, despite taking up some causes in common with leftist figures or groups. Foucault's (1995) critical analysis of penality in Discipline and Punish is therefore underpinned both by his earlier academic work, and by his various political activities, including his activism on criminal justice issues in Tunisia and in France.  Analysing Discipline and Punish In Discipline and Punish, Foucault (1995) traces the development of techniques and institutions of discipline during the period from the late eighteenth to mid-nineteenth century 3. While the chronology of the text is clearly defined, the geographical scope is more uncertain: Foucault is attentive to developments in France, but also draws on texts, ideas and institutions elsewhere, particularly England and America. Foucault pays detailed attention to the development of the prison, but his real focus is on the changing nature and techniques of power, and on identifying the roots of the new disciplinary power in Enlightenment conceptions of knowledge and the self. Three eras of punishment are identified by Foucault, or more precisely two eras in which distinct forms of punishment were employed and a period of indeterminacy in between. Foucault begins with an account of criminal justice during the ancien regime in France, where legal power and judgement were derived from the sovereign, and the ‘truth’ of criminality was revealed on the accused’s body by means of torture. In this period both  3  Foucault (1995) judges that the transition from the ancien regime to discipline in prisons was complete when the Mettray penitentiary was constructed in 1840 (293).  25  investigation and punishment were aimed directly at the body of the accused, and Foucault uses the execution of Damiens to illustrate the publicity and physical brutality of punishment (1995, 3-5). Nonetheless, law and punishment were highly inconsistent since the sovereign power over punishment included powers of appointment or removal for judges, to suspend or alter court decisions, and the right to grant clemency (Foucault, 1995, 80). The second era outlined by Foucault (1995) is a period during the late eighteenth century, during which public execution was criticized for being excessive and irregular, and various alternative punishments were proposed. Simultaneously, Foucault identifies socioeconomic changes due to the decline of a feudal class system and the development of a propertied middle class that sought to regularize the poor and to control property crime effectively (1995, 83-8). Foucault believes there was a complex relationship between changing penality and these socioeconomic factors, stating: The shift from a criminality of blood to a criminality of fraud forms part of a whole complex mechanism, embracing the development of production, the increase of wealth, a higher juridical and moral value placed on property relations, stricter methods of surveillance, a tighter partitioning of the population, more efficient techniques of locating and obtaining information: the shift in illegal practices is correlative with an extension and a refinement of punitive practices. (77) Foucault notes that the alternative punishments proposed by reformers such as Beccaria were certain, rational, efficient and had calculable consequences, arguing that their intent was "not to punish less, but to punish better" (1995, 82). However, given the wide range of suggested penalties, Foucault regards the prison’s prompt emergence as the dominant punishment as "surprising" (1995, 117). This claim that the prison’s rise as the major form of punishment was uncertain and contingent is important to Foucault’s genealogy, since the identification of historical alternatives serves to de-naturalize the prison. Foucault then describes early prisons in  26  England and America to show how they attempted to control and transform the behaviour of prisoners, and thus began the "development of knowledge of individuals" (1995, 125). The third era identified in Discipline and Punish is that of disciplinary power, which produces ‘docile bodies’ and thus shapes the minds of subjects through the performance and assessment of minutely detailed activities. Foucault (1995) devotes considerable attention to identifying and describing disciplinary techniques, which include control of people’s spatial organization, activity, time, and their separation from or combination with others (149-169). Discipline also involves the production of knowledge about the subject through assessment and normalization, which lead to the internalization of these norms and assessment by the disciplinary subject. Foucault therefore argues that the individualized modern subject comes into being through disciplinary power. Later in the text, Foucault (1995) describes discipline more simply as consisting of two central principles: "hierarchical observation and normalizing judgement" (192) which are combined in the form of the examination. Here, the Panopticon is introduced as the exemplar of disciplinary power, because it combines the enclosure of prisoners in individual cells with constant visibility and uncertainty regarding whether one is currently being watched. The Panopticon, Foucault concludes, is not merely an architectural plan but "the diagram of a mechanism of power reduced to its ideal form…a figure of political technology that may and must be detached from any specific use" (1995, 205). This constant threat of surveillance is important, because it is the reason inmates internalize disciplinary norms and therefore become self-disciplined, i.e. discipline becomes self-imposed as the result of surveillance. Foucault (1995) states that three specific practices came to dominate disciplinary prison regimes: isolation (either through silence or solitary confinement), work, and adjustment of the 27  penalty to the individual i.e. individuation (237-256). While Foucault argues that the intent of the prison was corrective, he claims that prisons necessarily fail to correct and that this failure is built into their ideology since it justifies the intensification and expansion of discipline: For a century and a half the prison had always been offered as its own remedy: the reactivation of the penitentiary techniques as the only means of overcoming their perpetual failure; the realization of the corrective project as the only method of overcoming the impossibility of implementing it. (1995, 268) Foucault concludes that prisons both transform their occupants into ‘delinquents’ through the necessary failure of normalization, and produce delinquency in the wider society by marking out ex-prisoners and their families, and by making prisoners’ families destitute, thereby creating a sub-culture of poverty, criminality and recidivism (1995, 264-272). As such, Foucault rejects the notion that the purpose of the prison is to control crime, claiming that its real benefit is to subject the population and dominate sections of it: "the differential administration of illegalities through the mediation of penality forms part of those mechanisms of domination" (1995, 272). Finally, Foucault describes the extension of discipline throughout society, to form a network of disciplinary institutions (including schools, hospitals, mental asylums etc.), surveillance and normalization which he terms "the carceral archipelago" (1995, 301). The conclusion of Foucault's (1995) analysis is that imprisonment has the same effect regardless of whether the regime is purportedly punitive or rehabilitative: the prison always and inevitably fails to reform inmates, or to reduce crime (268). Since there is no neutral or objective definition of what constitutes crime, or of what constitutes deviant behaviour, the meanings of crime and deviance are constructed through the criminal justice system and disciplinary institutions. The prison differentiates inmates from the rest of the population by marking them out as delinquents, and prisoners are released to face increased surveillance, restrictions on their movement, and difficulty in finding work, all of which make them likely to be re-convicted and 28  returned to prison. The necessary failure of the prison thus serves to continually justify the exercise, intensification, and extension of disciplinary power. Penality is thus revealed to be a way of exercising power and producing truth by differentiating between legality and illegality, acceptable and unacceptable behaviours, normal and abnormal subjects. Towards the end of Discipline and Punish Foucault makes explicit its continuity with his earlier work on medicine, stating "the social enemy was transformed into a deviant, who brought with him the multiple danger of disorder, crime and madness" (1995, 299-300). Foucault's (1995) analysis in Discipline and Punish therefore combines an account of changes in penality during the late eighteenth and early nineteenth century with analysis of changing techniques of power, and of how disciplinary power operates in modern society. In the next chapter I will evaluate Foucault's (1995) claims about power and the prison by considering alternative historical accounts of late eighteenth and early nineteenth century punishment in England and America, including criticisms of Foucault's argument. I will therefore be surveying the historical literature about criminal punishment to see whether other historical accounts of the history of punishment support or challenge Foucault's (1995) account of power and the prison.  29  Chapter Two: Histories of the Prison - Assessing the Literature Overview In this chapter I will provide a broad literature review that surveys and analyzes the scholarly literature about the history of punishment. In order to assess the persuasiveness of Foucault's historical account of power and the prison, I compare his analysis with alternative accounts of the history of criminal punishment, and particularly prisons, in America, and England4. I evaluate whether these historical accounts of penality are consistent with Foucault's arguments in Discipline and Punish, and where they are inconsistent with Foucault's (1995) account I assess which is more persuasive. The intention of this analysis is to assess the theoretical contributions made by alternative approaches to the history of punishment, and other normative accounts of the power relations involved in criminal justice. The conclusions I reach in this chapter will therefore inform my own analysis of contemporary power relations around criminal punishment in the later chapters of this dissertation, including most notably, the importance of race and gender in shaping forms of criminal punishment. The literature discussed here examines the period that Foucault (1995) identifies as the birth of the prison, from the late eighteenth century to the mid nineteenth century, and focuses on the criminal justice systems in England and the United States. However, I do not pretend to discuss all of the literature which examines the history of punishment in these countries and period, which is extensive and beyond the scope of this research project. Instead I have chosen a selection of the works published on this subject, which present a variety of perspectives and  As stated in Chapter One, the case studies for this thesis are the criminal justice and prison systems in the United States, and England and Wales. In the present day England and Wales is a single jurisdiction for the purposes of criminal justice, and therefore all reports and data about crime rates, policing, imprisonment etc. refer to 'England and Wales' as a unit. In contrast, the historical literature surveyed in this chapter discusses the history of criminal punishment in England, but tends to make no mention of Wales. My literature survey in this chapter therefore considers accounts of the history of criminal punishment in England (i.e. excluding Wales), and the United States. 4  30  yield insights about the possible flaws and strengths in Foucault's account of power and the prison. The literature considered represents a range of theoretical approaches and empirical cases (England and/or the United States, and sometimes also other European states), so for clarity I have divided this literature into three thematic categories. I examine the literature in each of these categories in turn, but will refer across the categories when necessary to identify a similarity or disagreement between these works. The first category is Marxist-inspired analyses, which focus on how forms of criminal punishment relate to the labour market, control of the working class, and underlying socioeconomic structures. In this category I consider the analyses provided by Rusche and Kirchheimer (1967), Ignatieff (1978), Melossi and Pavarini (1981) and Hirsch (1992). There is considerable variation in the works I have grouped as Marxist-inspired, from the fairly narrow economic approach of Rusche and Kirchheimer, to Ignatieff's broader analysis of social, political and economic forces, to Melossi and Pavarini's attempts to combine Foucaultian and Marxist conceptual approaches. The label of 'Marxist-inspired' is therefore meant only to indicate a broad overlap between these works in their account of the influences on and purpose of criminal punishment, and not to obscure or minimize the differences between them. There are also two Marxist-inspired historical analyses that I do not explore in this category: the works of Colvin (1997) and Davis (1998; 2003), both of whom analyze class alongside race and gender and are therefore discussed in the following category. I will argue that there is significant overlap between Foucault's account and Melossi and Pavarini's (1981) work, but that many of their claims are already present in his analysis. However, Foucault's genealogical approach is in tension with the causal histories provided by most of these scholars, and his account of power is  31  incompatible with that of the more economically focused analyses, such as the work of Rusche and Kirchheimer (1967). The second thematic category is scholarship that analyzes race and/or gender in relation to criminal punishment, by exploring how race and gender relate to the rates at which one is criminalized and punished, the forms of that punishment, and the ends that this punishment serves. In this category I consider the analyses by Freedman (1981), Rafter (1990), Zedner (1991), Colvin (1997) and Davis (1998; 2003). Again there is considerable variation between these analyses and the accounts of power in relation to criminal justice that are presented, but all of them argue that either gender, race, or both were important determinants of the forms and purpose of criminal punishment. While Freedman (1981) and Zedner (1991) only consider how gender related to punishment, analyses of the roles of both race and gender are provided by Rafter (1990), Colvin (1997) and Davis (1998; 2003). Here I will argue that the analyses of gender and punishment extend Foucault's analysis, but are largely consistent with his claims about the constitution of the subject through disciplinary power. The scholarship around race and punishment poses more of a challenge to Foucault's argument, because of the interconnection of race, class, and labour exploitation, particularly in penal slavery. The third and most diffuse category comprises counter-revisionist critiques of Foucault (1995) that challenge his argument on the grounds of factual accuracy, Ignatieff's (1981) rebuttal of counter-revisionist arguments, and analyses using a broadly Foucaultian conceptual approach. I consider two Foucaultian-inspired accounts of punishment: Dumm (1987) and Meranze (1996), both of whom analyze criminal punishment in America. I also discuss three counterrevisionist analyses: Brown (2003) and Forsythe (1987) who argue that Foucault misrepresents the nature of prison regimes in England, and Semple (1993) who provides a more specific 32  analysis that challenges Foucault's account of the Panopticon. Lastly I discuss Ignatieff's (1981) rebuttal of such counter-revisionist critiques, and Foucault's own response to such criticisms. There are three inter-related grounds on which these historical analyses implicitly or explicitly challenge Foucault's analysis of power and the prison. Firstly, some of these analyses challenge Foucault's (1995) account on the grounds of factual accuracy but do not present an alternative account of power and the prison, which is the usual approach taken by counterrevisionist critics. Secondly, some of these historical analyses challenge Foucault's (1995) arguments on the grounds of completeness, for instance by arguing that Foucault ignores the important racialized and gendered dimensions of power in relation to criminal punishment. Thirdly, a number of these analyses propose an alternative account of power relations around criminal punishment that they argue better fits the empirical facts and/or captures more about the nature of power than Foucault's (1995) analysis. The Marxist-inspired analyses of the history of criminal punishment fall into this final category, as do some of the analyses in terms of race and/or gender. Challenges to Foucault's (1995) argument on the grounds of factual accuracy largely fail, because they do not directly address Foucault's account of power and therefore tend to criticize him on points of detail that are tangential to his main argument. As a result, most of the counterrevisionist critiques of Foucault (1995) are not persuasive. Challenges to Foucault (1995) on the grounds that his account of penality is incomplete are more successful, and are most persuasive when their argument has important implications for techniques and relations of power. In particular, there are several compelling analyses that show how the English and/or American criminal justice systems both worked through and played a significant role in creating race and gender identities and inequalities, and these analyses identify racialized and gendered power 33  dynamics in criminal punishment that Foucault (1995) did not discuss. Interestingly, these analyses of race, gender and punishment also link to Marxist-inspired analyses, by identifying class-specific gender roles, such as middle class norms of femininity, and the exploitation of Black labour in penal slavery and the Convict Lease system. Although some of the arguments about gender and punishment are partially consistent with Foucault's (1995) account, the analyses of race and labour exploitation fundamentally challenge his account of the transition, forms and use of criminal punishment, and its broader significance for power relations. Many of the challenges to Foucault (1995) proposing an alternative account of power are Marxist-inspired, and the success of these analyses is mixed. Some, such as Rusche and Kirchheimer (1967) present arguments that are economically determinist and therefore unconvincing, while others such as Melossi and Pavarini (1981) and Hirsch (1992) bring out important aspects of the ideological links between capitalism and the prison. There is some overlap between Foucault's (1995) arguments and Marxist-inspired accounts of punishment, which broadly agree about the timing of changes in criminal penalties, and identify strict disciplinary rules and work as part of prison regimes. However, these Marxist-inspired accounts clash with Foucault (1995) over the relationship between capitalism, prison regimes and disciplinary power, and make persuasive arguments about the influence of the profit motive on some forms of punishment, from the Convict Lease system to Bentham's plan for the Panopticon. In particular, Melossi and Pavarini (1981), Hirsch (1992) and Semple (1993) make a good case that disciplinary techniques were shaped by the rise of industrial capitalism and the need to constitute productive workers.  34  Marxist-Inspired Analyses Rusche and Kirchheimer’s book Punishment & Social Structure provides a Marxist historical narrative of criminal punishment, from medieval Europe to the 1930s in Europe and America. Rusche and Kirchheimer’s (1967) theoretical approach is based on Marxist analysis and the critical theory of the Frankfurt School, and their overarching argument is that the form and function of criminal punishment is determined by socioeconomic structures, particularly economic factors such as the labour market. Rusche and Kirchheimer's work is regularly claimed (Garland, 1990; Howe, 1994) to be the work that initiated the historical study of the practices, determinants and significance of criminal punishment, instead of focusing solely on contemporary forms of punishment. Although the book was originally published in 1938 it received little attention until its republication in 1967, and this edition is mentioned by Foucault (1995) who credits Rusche and Kirchheimer with the important insight that punishment is not merely repressive but has "a whole series of positive and useful effects" (24). Foucault's acknowledgment of the influence of Rusche and Kirchheimer's work on his own thinking makes it particularly interesting to explore this work in relation to Discipline and Punish. Punishment and Social Structure has two distinct parts, because chapters two through eight were written by Georg Rusche in Germany, and the remainder was written later by Otto Kirchheimer in the United States. Although Rusche and Kirchheimer analyze a range of countries and historical periods, the common thread is their argument that criminal punishment is largely socioeconomically determined and often serves the purpose of class control. Like Foucault (1995), Rusche (1967) identifies a trend towards the use of imprisonment during the late eighteenth century, but Rusche attributes this change to the profit motive: both in the narrow sense of making prisons profitable through forced labour, and in the wider sense of contributing 35  to a mercantilist society. Although Rusche (1967) notes that Enlightenment ideals influenced penal philosophy, he claims these efforts only succeeded during the late eighteenth century because "humanitarian principles coincided with the economic necessities of the time" (84). Rusche's account of prison regimes during the late eighteenth and nineteenth centuries also differs substantially from Foucault's analysis. Rusche (1967) argues that prison conditions were determined by the principle of ‘less eligibility’, which stated that prisoners should have a living standard below that of the poorest free men (94). The combination of less eligibility and punitive work, such as the treadwheel, was intended to deter crime by ensuring that prison conditions were extremely unpleasant (122). However, Rusche notes that the bare subsistence conditions of the working class undermined less eligibility and therefore the goal of deterrence, because it was impossible for prisoners to have a lower standard of living than the working class without dying as a result. Rusche therefore views the prison in the context of the industrial revolution and working class poverty, arguing both that imprisonment was part of a broader effort to control the working class, and that prison conditions were determined by working class living conditions. Rusche then analyzes punishment in America during the nineteenth century, and draws on economic motives to explain differences between prevailing European and American prison regimes. Rusche (1967) argues that in Europe there was no shortage of labour and hence prison regimes centred on solitary confinement and the effort to deter crime (133). In contrast, the labour shortage in the United States provided an incentive for prison regimes that would enable profitable industries, making the Auburn system of silent, congregate work more popular in America than prison regimes based on solitary confinement (Rusche, 1967). Lastly, Kirchheimer (1967) provides a thematic overview, arguing that while rising living conditions for the working 36  class during the late nineteenth century produced a more rehabilitative approach to punishment prisons nonetheless remained ineffective at reforming criminals. Rusche and Kirchheimer's (1967) account of the history of punishment differs from Foucault's (1995) arguments in several key respects. Firstly, there is an obvious methodological difference between the causal explanations provided by Rusche and Kirchheimer and Foucault's genealogical approach that emphasizes discontinuity and historical contingency in order to denaturalize the prison. Secondly, whereas Foucault argues that punishments were focused on the body during the eighteenth century, through torture and execution, Rusche argues that punishments such as transportation were more common than capital penalties. Thirdly, Rusche and Kirchheimer's accounts of nineteenth century prison regimes in both Europe and America diverge from Foucault's account of the disciplinary prison. Rusche's account of English prisons contradicts Foucault's account by identifying the punitive and deterrent nature of prison regimes, and his account of American prisons contradicts Foucault by identifying prison regimes based on congregate work to extract profit. These disagreements about prison regimes are significant because they relate to the function of the prison and the forms of power exercised in relation to criminal justice. Rusche and Kirchheimer's analysis therefore challenges Foucault's account of sovereign power during the eighteenth century, his account of the disciplinary prison during the mid nineteenth century, and his account of power more broadly. Although Rusche and Kirchheimer (1967) provide an interesting critical analysis of the prison, their argument is weakened by its over-reliance on economic explanations. Rusche’s explanation of the rise of imprisonment as driven by the profit motive is unconvincing given the considerable cost of imprisonment and the epidemic disease risks posed by late eighteenth century prisons. Similarly, Rusche's account of punishment in America relies on the assumption 37  that prison industries were profitable, but this claim is refuted by Rothman (1971) who argues that few prisons made a profit from their industries. By privileging economic explanations for changes and forms of punishment over other factors such as prevailing ideas and social norms, Rusche and Kirchheimer (1967) provide an account of the history of criminal punishment that appears partial and over-simplistic. A number of the themes of Rusche and Kirchheimer’s (1967) analysis are taken up in Michael Ignatieff’s (1978) book A Just Measure of Pain, which analyzes the development of the penitentiary in England from 1770 to 1850. Although Ignatieff follows Rusche and Kirchheimer in arguing that criminal punishment served the purpose of providing social control over the working classes, he is attentive to the ideological origins and significance of the prison. Similarly, Ignatieff agrees with Rusche and Kirchheimer that transportation was the most common punishment during the late eighteenth century, with capital punishment being reserved for the most serious crimes. Although Ignatieff identifies an increase in the use of imprisonment in the early nineteenth century, as Foucault (1995) describes, he attributes this to the ending of transportation to America after the War of Independence and the resultant need for a new criminal penalty (82). However, Ignatieff (1978) argues that the new penal theory was based on the ideas of John Howard, not Bentham, and on broader concerns about social disorder in an increasingly urban, industrial society: The penitentiary, in other words, was more than a functional response to a specific institutional crisis. It exerted a hold over men’s imaginations because it represented in a microcosm the hierarchical, obedient and godly social order, which they felt was coming apart around them (85) Ignatieff (1978) describes Bentham’s plan for a prison based on continual surveillance, but he observes that Bentham planned to run the Panopticon for profit as a private contractor, which led Parliament to reject the proposal (110-113). Ignatieff's account of of Bentham's plans 38  for the Panopticon therefore emphasizes very different things to Foucault's (1995) analysis: that the Panopticon was intended to be privately run and to generate profit, and that as a result it was never built. Bentham's intention to profit from the Panopticon challenges Foucault's account of the disciplinary prison as not motivated by profit, seems to support Marxist accounts of the prison and provides a precedent for present-day prison privatization. Ignatieff also echoes Rusche and Kirchheimer’s (1967) account of the punitive conditions in English prisons that were intended to deter crime, including a bad diet and use of the treadwheel. This harsh prison discipline is explained with reference to the labour market, as Ignatieff argues that low wages and rising crime led to a harsh prison regime in an attempt to maintain social hierarchy and stability (179). Although Ignatieff acknowledges that a strict new prison regime was introduced with the intent of reforming prisoners, he notes that these regimes were highly unpopular with prisoners and were criticized as cruel by some contemporaries. Ignatieff (1978) therefore concludes that although people became convinced that prisons should rehabilitate, prisons nonetheless failed to reform criminals and served a punitive and deterrent purpose (209-210). A third important Marxist-influenced analysis is Hirsch’s (1992) book The Rise of the Penitentiary, which considers prisons in the context of forced labour and traces the roots of American prison regime to the workhouse. Hirsch argues that efforts to reform criminals drew on earlier theories that idleness could be cured by forcing people to work and thereby to develop good habits (14-15), and that this philosophy underlay the Auburn system of silent, congregate work. Hirsch thus follows Rusche and Kircheimer (1967) by linking the development of the prison to forced labour, but rejects their explanation in terms of the profit motive in favour of the view that the institutions shared similar ideas about the roots of poverty in idleness.  39  Hirsch (1992) then compares the penitentiary with slavery in America, arguing that there were many points of similarity because both involved isolation from the general population, confinement, forced labour, and subordination to others (71). Hirsch notes that both slaves and penitentiary inmates were viewed as "prone to criminality" (1992, 73) and that slavery and the prison were combined in the Convict Lease System. However, whereas prisons aimed merely to recoup their costs, Hirsch notes that slavery and the Southern Convict Lease system were highly profitable and therefore more heavily driven by economic motives (95-96). Moreover, Hirsch (1992) notes that the penitentiary involved only a temporary loss of citizen’s rights and provided some education, unlike slavery and the Convict Lease system. By considering criminal punishment in the Southern US, Hirsch (1992) presents a broader analysis than many other Marxist-inspired theories of punishment and begins to explore the links between criminalization, race, and labour exploitation. The final Marxist-influenced analysis of the prison is Dario Melossi and Massimo Pavarini’s (1981) book The Prison and the Factory, which combines attention to class and the labour market with a Foucaultian account of disciplinary power. Melossi and Pavarini (1981) argue that capitalism led to the development of prisons, which in turn used disciplinary techniques to turn criminals into productive workers. The book is divided into two parts: the first two chapters, written by Melossi, focus on penal policy in Europe; while Pavarini’s three chapters discuss penal policy in the US. Like Rusche and Kirchheimer (1967) and Ignatieff (1978), Melossi argues that European penal policy was shaped by the transition to an industrial economy, and by labour market trends. However, like Hirsch (1992), Melossi (1981) pays particular attention to the workhouse, which he describes as "disciplinary training for capitalist production" (21) and views as the origin of the modern prison. Melossi (1981) shows the 40  Foucaultian influence on his analysis by diverging from Marxist explanations of the prison in terms of exploitation of labour or deterrence of crime, instead arguing that prison labour was intended to constitute labour-power via bodily regulation. Pavarini (1981) argues that penal policy in nineteenth century America was shaped by the labour market, because the poor were viewed as lazy and/or deviant, and hence in need of forced labour. The popularity of the Auburn system of silent, collective work is therefore explained in terms of the need to create efficient and disciplined workers, and Pavarini (1981) describes prisons as factories "producing proletarians, not commodities." (114-5). This claim is then expanded to argue that the prison is an institution of class hegemony, and that the identification of criminals with poverty and idleness meant that propertylessness came to be regarded as virtually equivalent to criminality (148-149). Pavarini (1981) therefore provides an account of punishment in terms of ideas about work, poverty and human nature, instead of relying on narrowly economic explanations. Melossi and Pavarini (1981) explore the possible overlap between Marxist analysis and Foucault's account of disciplinary power and the constitution of the subject, producing an interesting analysis that avoids the economic reductionism of Rusche and Kirchheimer (1967). The authors tease out the implications of Foucault’s rather opaque statements regarding class, capitalism and the ‘dangerous classes’, and identify some overlap between Foucault's account of the prison and Marxist claims, but it is doubtful how much Melossi and Pavarini (1981) add to Foucault’s analysis. For instance, Foucault (1995) argues that the self-evidence of prison as a penalty derived partly from the existence of wage-labour (232), and that the prison might serve to produce disciplined workers for an industrial economy (242). It therefore seems that many of Melossi and Pavarini's (1981) claims are already present, implicitly or explicitly, in Foucault’s 41  work. Moreover, there are aspects of Melossi and Pavarini’s (1981) analysis that Foucault would clearly reject, including the claim that the development of industrial capitalism determined the form of the prison, and their inattention to the link between disciplinary power and the human sciences. While Rusche and Kirchheimer's (1967) accounts of English and American prison regimes are unconvincingly reductionist, it is significant that forced labour was a component of most prison regimes. It is also noteworthy that the potential to profit from convicts’ labour contributed to the preference in Northern US states for prison regimes based on silent, congregate work instead of solitary confinement (Hirsch, 1992; Colvin, 1997). Melossi and Pavarini (1981) attempt to combine this attention to work and the influence of capitalism on prison regimes with a Foucaultian account of disciplinary power, and argue that work within the prison was designed to produce disciplined workers. Marxist-influenced analyses therefore produce a range of accounts of how capitalism and class relate to penality, and there is greater conflict between Foucault's account and economically focused analyses such as Rusche and Kirchheimer (1967) than with the accounts that pay more attention to ideas, such as Melossi and Pavarini (1981). Foucault (1995) briefly explores the relationship between capitalism and disciplinary power in Discipline and Punish, arguing that discipline enabled capitalism because the "body becomes a useful force only if it is both a productive body and a subjected body" (25-6). Although Foucault acknowledges that labour was a part of prison regimes, he attributes this to the disciplinary effects of work on the subject, and not to the profit motive. However, by tracing the rise of the prison to Enlightenment notions of science and the human subject, not to economic factors or capitalist ideology, Foucault rejects the central theme of Marxist-inspired 42  analyses of criminal punishment. Foucault's analysis also has a different methodological basis than these Marxist-inspired accounts, because Foucault provides a genealogy and not a causal history. Foucault (2000f) expands on his understanding of the relationship between capitalism and disciplinary power in the lectures "Truth and Juridical Forms", in which he argues that industrial society required disciplinary techniques for extracting time and for "converting people’s bodies into labour power" (2000f, 82). Foucault (1995) therefore seems to invert Rusche and Kirchheimer's (1967) argument by claiming that the disciplinary power exercised in institutions such as prisons enabled the development of industrial capitalism, instead of being determined by it. Foucault provides further detail about the distinctions between his account and Marxistinspired analysis of the prison in the 1983 interview "What is called punishing". Here, Foucault (2000d) argues that one can distinguish the aim of something from its result, and the use to which it is put, and that where the use of an institution diverges from the initial aims there may be a reconfiguration to take account of this new use: They are results that are adapted to different uses, and these uses are rationalized – organized, in any case – in terms of new ends…This game is quite capable of solidifying an institution and I think that the prison has been solidified, in spite of all the criticism that was made, because several strategies belonging to different groups have converged on that particular site. (386) This reasoning disconnects the eventual strategic function of a form of criminal punishment from the reasons why that penalty was introduced, and the purpose it was intended to serve. This interview clarifies how Foucault's (1995) genealogical account of the prison co-exists with his argument that prisons construct delinquency and enable the intensification of discipline over poor and marginalized populations, making it clear that Foucault is not arguing that the prison was originally intended to serve this purpose. 43  The overlap between Foucault's (1995) account of prison discipline and Marxist-inspired arguments about the development of a productive workforce may stem from earlier Marxist historical analysis about industrial productivity and discipline. EP Thompson’s (1967) article "Time, Work-Discipline and Industrial Capitalism" pre-dates Discipline and Punish by several years, and argues that industrial capitalism required synchronization and a ‘time orientation’ amongst workers, which were instilled through education. Thompson (1967) claims that workers internalized this discipline and the new conception of time, and gradually came to use this time orientation for resistance against their employers: The first generation of factory workers were taught by their masters the importance of time; the second generation formed their short-time committees in the ten-hour movement; the third generation struck for overtime or time-and-a-half. They had accepted the categories of their employers and learned to fight back within them. They had learned their lesson, that time is money, only too well. (86) It is not surprising that Marxist historians should observe changes in conceptions of time and behaviour due to wage labour and industrialization, but it is interesting that Foucault (1995) echoes Thompson's language of ‘discipline’, regularity and synchronization. Although Thompson’s (1967) argument may be consistent with Foucault’s emphasis on the roots of discipline in the human sciences, Thompson suggests that the desire for profit shaped disciplinary practices – a claim that Foucault would likely reject. Foucault (1995) would probably also reject the idea that one can exercise resistance by internalizing disciplinary techniques and turning them against one's employer, instead of by challenging those techniques. There is therefore considerable agreement between Foucault and Thompson about the techniques of micro-power used to create disciplined subjects and about the approximate era of the changes, if not about the underlying cause and broader significance of discipline for power relations. 44  Although there is significant disagreement between these Marxist-influenced analyses and Foucault's (1995) account of power and the prison, the question of deciding which theoretical approach is more persuasive is not an easy task. This task is complicated by the fact that some of the most persuasive Marxist-influenced accounts of the prison also involve analysis of race and/or gender, and therefore speak to the incompleteness of Foucault’s account in persuasive ways. In particular, Foucault's (1995) account ignores penal slavery and the Convict Lease system in America (Colvin, 1997; Davis, 1998 & 2003) both of which suggest that forms of punishment were driven by the profit motive and by ideas about race. The Marxist account is also persuasive to the extent that Foucault (1995) ignores the normative implications of Bentham's plan to run the Panopticon for profit and the basis of its design on an earlier design for a factory, as Semple (1993) points out. Foucault (1995) thus seems to understate the significance of the profit motive in determining forms of punishment, at least in some jurisdictions and during some periods, and may be underplaying the extent to which industrial capitalism contributed to the development of disciplinary techniques. Although Foucault's (1995) certainly generates important insights, particularly about the internalization of discipline and the connections between power and knowledge, there are important power dynamics around capitalism, labour exploitation and penality that are not captured in Foucault's account. I will return to this question of the persuasiveness of Marxist-inspired accounts of the history of punishment and the implications of such arguments for understanding power relations at the end of this chapter, after assessing the analyses of race and/or gender, and counterrevisionist analyses.  45  Analyses of Gender and/or Race Freedman’s book Their Sisters’ Keepers is a study of American reformatories between 1820 and 1920, which were women's penal institutions intended to provide moral and sexual reform. Although the reformatories were intended to ‘help’ women and not to punish or coerce, Freedman (1981) claims that in practice reformatories were punitive. Freedman notes that these institutions were designed for convicts considered capable of reform, and that the criminals sent to reformatories consequently tended to be young, White women who had committed minor offences (79). These early-nineteenth century reformatory regimes were strongly gendered and modelled on a domestic, feminine environment, including the use of communal ‘cottages’ instead of individual cells (Freedman, 1981, 76). There is some overlap between this analysis and that of Foucault, since Freedman describes institutions that used the control of activity and space with the intent of reforming inmates. However, Freedman focuses on the good intentions of prison reformers whilst failing to analyze the effects of their work or the power relations involved. There is also little attention to class or race in Freedman’s analysis of the prison reformers, prisoners or institutional regimes, resulting in a rather over-simplified portrayal of the relationship between gender and punishment. Further, Freedman’s work suffers from an unacknowledged bias in case selection, since she draws almost exclusively on historical material from Indiana, Massachusetts, and New York where she admits that reformatory movements were strongest. As a result, Freedman only considers women’s imprisonment in those states where reformatories were least penal and most attentive to ideals such as education, producing a partial and rather skewed account of women’s punishment in nineteenth century America.  46  Nicole Rafter’s (1990) book Partial Justice fills many of the gaps in Freedman’s analysis, since she analyzes a wider range of American states and penal institutions, including reformatories, women’s prisons, labor camps, and penal slavery. Crucially, Rafter (1990) employs a critical approach to women’s incarceration, explaining that the treatment of women in the criminal justice system reflected battles over gender definition, which was strongly shaped by class and race, and that reformatories were used to impose a particular version of femininity: The reformatories institutionalised bourgeois standards for female propriety, making it possible to "correct" women for moral offences for which adult men were not sent to state penal institutions… For these reasons, the women’s reformatory served special, female specific functions with regard to social class and social control. (158) Rafter (1990) argues that women’s reformatories instilled middle class norms of femininity, and believes that they were partially successful in imposing these norms, despite the restrictions that this placed on the work that women could pursue. The deeply gendered nature of penal regimes and conceptions of appropriate behaviour leads Rafter (1990) to conclude that women were subject to more severe punishment and greater social control than men, including discipline of their moral and sexual conduct (36). Rafter (1990) also examines how class and race shaped women’s treatment within the criminal justice system, noting that in some Southern states racially targeted institutions such as the Convict Lease system were used to exploit prisoners’ labour (150). Even outside the South, Rafter (1990) notes that race impacted on women's treatment, because official judgements about a convict’s capability to reform depended upon that person’s race, age, and previous offences, meaning that: Black women were overrepresented in the populations of custodial institutions and "impure" reformatories like that of Ohio owing to three factors: overt racism, covert discrimination, and higher rates of offending by black than white women for some serious crimes. (134) 47  This analysis therefore demonstrates how ideas about gender, race and class determined the nature of penal regimes, which institutions a convict was sent to, and whether someone was seen as capable of reform. Rafter's (1990) analysis also links the trends in women's criminal punishment to broader gender politics, by showing how women's reformatories sought to teach inmates middle class feminine norms, from sexual propriety to domesticity (158). While Rafter's analysis is broadly consistent with Foucault's account of how detailed disciplinary norms constitute the subject, and about the attempts to normalize criminals, she extends this approach to show the gendered nature of discipline and how it constitutes normatively gendered subjects. Lucia Zedner's (1991) analysis of punishment in England during the mid nineteenth century confirms Rafter's (1990) account of how middle class ideals of femininity shaped views about women, crime and punishment. However, while Rafter (1990) suggests that American women's reformatories sought to rehabilitate prisoners, Zedner (1991) argues that Victorian ideals about women's moral superiority meant that female prisoners were stigmatized more than men (30). Despite feminist efforts to characterize female criminals as victims who were capable of reform, Zedner (1991) states that in England "criminal women were portrayed as sunk beyond redemption and certainly beyond the supposedly reformatory powers of the prison" (43). Zedner (1991) also contests Foucault's account of the disciplinary prison, arguing that the vast majority of women imprisoned in the nineteenth century served short sentences in local prisons that were never intended to be reformatory (132). Nonetheless, Zedner (1991) notes that regimes in convict prisons varied by gender, because the work given to female prisoners was based around an ideal of domestic femininity, and that both schooling and religion were more important components of prison discipline for women than for men (197).  48  This account of female prisoners as beyond reformation and as being largely placed in non-rehabilitative local prisons challenges Foucault's (1995) account of penality, but Zedner (1991) also advances an explicit critique of Foucault: In Foucault’s analysis the prison becomes a conceptual apparatus for the imposition of a discipline which extends throughout the whole social body. Endowed with total power, it knows no exceptions, failings or inconsistencies; it has no need for change. Once reformed, the prison is, he implied, immutable. In Foucault’s work the reality of continuing administrative chaos and human error is lost. His cavalier treatment of evidence, the gap between his idealized account and the reality of nineteenth century prisons, and, above all, his failure to account for change over time combine to restrict the value of his work for historians. (95) These criticisms of Foucault position Zedner as one of the counter-revisionist historians of criminal punishment, and such arguments will be considered in more detail later in the chapter. However, even if Zedner's claims about the theory and practice of punishment are essentially correct, it is not clear that this undermines Foucault’s account of penality or disciplinary power. Despite arguing that criminal women were constructed as beyond reformation and often sent to local jails, Zedner (1991) acknowledges that convict prisons for women were rehabilitative, albeit in different ways than men's convict prisons (183). Although Zedner’s account of the gendering of prison regimes is valuable, she lacks Rafter’s (1990) critical scope and fails to recognize that specific gendered identities were being constructed – not merely represented – in penal discourse and institutions. Instead of refuting Foucault's (1995) arguments about the constitution of the subject through discipline, Zedner largely ignores his theoretical claims. A more powerful critical analysis of gender and criminal punishment is provided by Angela Davis (1998 & 2003), who combines Marxism, feminism and attention to race in her analysis of punishment in America. Davis (1998) argues that Foucault's account of power is undermined by the fact that he ignores the impact of race and slavery upon the history of punishment. Davis (1998) states that American prison regimes were deeply gendered, producing 49  substantial differences between the institutional practices applied to men and to women. Moreover, Davis (1998) argues that punishments oriented at the ‘soul’ were not applied to Blacks, and therefore criticizes Foucault’s account of criminal punishment in terms of disciplinary power: If, as Foucault insists, the locus of the new European mode of punishment shifted from the body to the soul, black slaves in the US were largely perceived as lacking the soul that might be shaped and transformed by punishment…As white men acquired the right to be punished in ways that acknowledged their equality and the racialised universality of liberty, the punishment of black slaves was corporal, concrete and particular. (1998, 99) Davis (1998) also insists that the exploitation of convicts' labour was an important determinant of criminal justice in America, particularly the criminalization of free Blacks during the Postbellum era in order to facilitate penal slavery and the Convict Lease system. Finally, Davis (1998) refutes Foucault’s argument about the cessation of cruelty and torture as punishment, identifying the continuation of cruelty and bodily punishment towards Black convicts in the South. These arguments are further developed in Davis's book Are Prisons Obsolete? where she argues that race has historically been important to conceptions of criminality and cites Frederick Douglass’s observation of the tendency to "impute crime to color" (Davis, 2003, 30). Like Rusche and Kirchheimer (1967), Davis (2003) believes that the labour supply shapes the forms of criminal punishment, and she argues that this drove the introduction of Black Codes (which criminalized behaviour for Blacks that was legal for Whites) and the Convict Lease System in the American South (28-29). Davis also draws parallels between the Convict Lease System and prison privatization in the twenty-first century, since both are sexist, racist, and exploit prisoners for profit. Davis (1998; 2003) thus challenges Foucault's account of the transition from penality based on sovereign power to one based on disciplinary power, his account of the disciplinary 50  prison, and his broader arguments about disciplinary power in society. Davis (1998; 2003) makes a convincing case that Foucault's analysis of criminal punishment is incomplete, both because the prison is only part of the power dynamics around criminalization and punishment in America, and because criminal justice cannot be fully understood without analysing race and gender. Although Davis's (1998; 2003) claims about gendered prison regimes are probably consistent with Foucault's (1995) analysis, her arguments about the role of capitalism and labour markets in determining criminal justice institutions are not. The last analysis of race, gender and punishment is Mark Colvin’s (1997) book Penitentiaries, Reformatories, and Chain Gangs, which examines the insights provided by a range of theoretical approaches to the history of punishment. Although Colvin argues that Foucault is correct that human sciences and the desire for internal control shaped the penitentiary, he argues that the civilizing and humanitarian intent of reformers was more important than the desire to discipline prisoners. Nonetheless, Colvin (1997) claims that prisons "served as a mechanism of class control" (80) and that a Marxist perspective explains the popularity of the Auburn system of silent collective work over regimes based on solitary confinement, because the former facilitated prison industries. On the subject of women’s punishment in the Northern US, Colvin (1997) observes that the transformation in the punishment of women accompanied changing accounts of ideal femininity. Colvin argues that the development of women’s reformatories is best understood through a Foucaultian framework and that Foucault’s account of discipline is a more accurate account of women’s prisons than of men’s: The early reformatories for women can be seen as a series of experiments in which new technologies of power were first tested and developed…The point of the reformatory was to create self-control so that inmates would ‘do right without compulsion,’ the goal emphasized by Foucault in his discussion of penal discipline. (1997, 191) 51  Colvin (1997) also considers punishment in the South, arguing that the Convict Lease System provides the best case for a Marxist analysis of punishment, but that it was facilitated by racism including "‘scientific’ racial doctrines that promoted linkages between heredity, race and crime" (249). Although Colvin provides an interesting comparative analysis of different explanations of punishment, and makes the plausible claim that these differing explanations tend to fit different examples of penal institutions, his account lacks overall theoretical coherence. Colvin makes no effort to resolve the theoretical clashes between the Foucaultian, Marxist and humanitarian ‘good intentions’ accounts of nineteenth century punishment, which is problematic given the clear tensions between these accounts. As a result, Colvin's work functions more as a literature review than as an analysis of power in relation to criminal punishment. These analyses about gender and/or race in regard to criminal punishment demonstrate that Foucault's analysis of power and the prison is incomplete because he does not consider how ideas about criminality, appropriate punishment, and therefore penal regimes are shaped by race and gender. Given that Foucault (1995) does not discuss gender in Discipline and Punish it is difficult to say whether or not these analyses of the gendering of punishment are consistent with his analysis, but there is no obvious inconsistency and Foucault did not argue that discipline and penality were not gendered. Some of these analyses seem consistent with Foucault’s account of disciplinary power, for instance Colvin’s (1997) claim that women’s reformatories exemplified the disciplinary regime, while others such as Davis's (1998; 2003) analysis undermine Foucault's arguments about the nature of power in relation to criminal punishment. Moreover, Colvin (1997), Rafter (1990), and Zedner (1991) all argue that penal institutions were used as a means to discipline women into particular, middle class ideals of domestic femininity, which suggests that gendered identity formed part of the constitution of the subject that Foucault argues 52  occurred through discipline. These analyses of the gendered nature of discipline in prison regimes therefore extend and deepen Foucault's (1995) account of how subjects are constituted, and of the uses of imprisonment. The question of whether these arguments about the gendered nature of disciplinary power and the constitution of gender identities are consistent with Foucault's (1980b & 1990) analysis in Herculine Barbin and The History of Sexuality will be considered in Chapter 5. The issue of race in regard to punishment is potentially more disruptive to Foucault’s analysis, because several of the historians discussed argue that interconnections of race, class and criminality were central to power relations around punishment in America. The accounts of penal slavery and the Convict Lease system in Southern states provided by Davis (1998 & 2003), Colvin (1997) and Hirsch (1992) make it clear that one cannot separate the issues of race and exploitation of labour in nineteenth century punishment. Moreover, Rafter (1991) argues that even in Northern states non-disciplinary forms of punishment were targeted at those racialized as non-White, because they were viewed as less capable of reformation. While there is little attention to race in the historical literature about English prisons, it is clear that ideas about race and colonialism have shaped conceptions of the subject and of social order, including discourses and practices about gender and sexuality. Foucault's (1995) inattention to race in Discipline and Punish is even more problematic given that his work is a history of the present, and race was clearly an important factor in criminal justice and punishment in France during the 1970s – as demonstrated by Foucault’s own political activism. Foucault (1974) also acknowledged in the interview "Michel Foucault on Attica" that race is salient to American prisons, and observed the substantial over-representation of Black men during his 1972 visit to Attica prison (157). The lack of reference to race in 53  Discipline and Punish is odd given Foucault's awareness of the racialized dimension of criminal justice, and undermines the text as a critical diagnosis of power relations in his present. However, Foucault (2003) analyses race in the 1975-6 Collège de France lectures, where he develops the concept of biopower, and I explore the insights provided by this analysis in Chapters 5 and 6. The question of whether attention to race requires the modification or rejection of Foucault’s (1995) account of power and the prison will be explored in these chapters in more detail.  Foucaultian, and Counter-Revisionist Analyses In this section I discuss two broadly Foucaultian accounts of the nature and development of disciplinary power in the prison, three counter-revisionist accounts that involve critiques of Foucault, and Ignatieff's (1981) rejection of such counter-revisionist critiques. The first Foucaultian argument is by Thomas Dumm (1987) who assesses the critiques of Foucault’s account of the prison and argues that counter-revisionist objections on the grounds of empirical detail miss the point, because Foucault’s (1995) focus is on disciplinary power rather than prisons per se. Dumm (1987) then outlines the disagreement between Foucault and Marxist accounts of power, claiming that disciplinary power is consistent with the exploitation of labour identified by Marxists: Foucault in Discipline and Punish thus decentred Marx, rearranging the order of the relationship between Marx’s most fundamental categories of value and production. This rearrangement was designed to show the importance of production in regard to power and knowledge even as it asserted that Marx was mistaken to make the struggle between labor and capital so central…What thus emerges, at least in this reading of Discipline and Punish, is a remarkably specific analysis of the operations of power at a level below that recognized by Marxist politics. These operations are compatible with but different from those of Marxist exploitation. (53)  54  This passage is carefully phrased and does not state that Foucault’s analysis of power is compatible with Marxist analysis, but Dumm does suggest that disciplinary power regarding the constitution of the subject could co-exist with exploitation of labour. Dumm then presents his own historical analysis, arguing that the American Revolution and introduction of democracy drove the rise of the prison, because disciplinary techniques were required to produce responsible, self-governing citizens. Since Tocqueville visited both the Auburn and Pennsylvania prisons, Dumm turns to him for insight on the relationship between democracy and penal discipline. While the Pennsylvania system consisted of solitary confinement and aimed at deep moral reform through contemplation, the Auburn system involved silent congregate work under constant observation and intended merely to teach good habits of behaviour (Dumm, 1987, 117-118). Dumm therefore regards the debate about prison regimes as a fundamental dispute over what sort of citizen should be produced, and concludes that the popularity of the Auburn system privileges obedience over autonomy. Dumm (1987) draws a parallel between this trend in prison regimes and Tocqueville’s arguments about the triumph of democratic despotism over liberty, concluding: The petty rules, the intricate procedure, the schedules, and finally the silence were born of an expressed desire on the part of responsible citizens to make responsible citizens out of others. Thus democratic despotism is born…Tocqueville came to the United States to study the prison, and left to write Democracy in America. No irony need be made of that coincidence, nor should anyone be surprised. After all, the penitentiary was the ideal liberal democratic institution. (140) By arguing that disciplinary power is necessary for democracy, Dumm (1987) links the prison to Enlightenment notions of the rational individual in a manner that is similar to Foucault’s account of the link between liberal political philosophy and discipline (139). The other Foucaultian analysis is Meranze’s (1996) book Laboratories of Virtue, which analyzes Philadelphia from 1760 to 1835, using local detail to demonstrate the development and 55  extension of disciplinary power. Meranze (1996) begins by observing that the initial replacement for public, bodily punishment after American independence (at least in Philadelphia) was not the prison, but a system of public, penal labor that begun in 1786 (55). This use of labor as a means of punishment was intended to combine deterrence with reform through public shame and the inculcation of habits of work; a rationale similar to that behind the penitentiary (Meranze, 1996, 80). However, Meranze (1996) notes that the system of public labor led to "disarray and violence" (87), and was quickly abandoned in favour of imprisonment. Eastern State Penitentiary, the exemplar of the regime of solitary confinement in the US, was in Philadelphia and so Meranze devotes considerable attention to this prison regime and its broader impact. Meranze (1996) argues that Eastern State employed a strategy of spatial control to counter the problem of inmates corrupting one another, but also used labour as a key component of penal theory and practice (187). Meranze also observes that similar disciplinary techniques were later introduced across the city, from town planning, to Refuges for children, to Madgalens for prostitutes, all in an effort to regulate working class life (289). Meranze (1996) argues that these forms of discipline were "deeply gendered" (282), and argues that the extension of discipline beyond the prison demonstrates the intent to enforce gender and class control. This account of the development of discipline in the penitentiary and its extension to other social institutions and city spaces supports Foucault’s (1995) claims about the spread and intensification of disciplinary power. Meranze's analysis also backs up Foucault’s (1995) claim that imprisonment was not the initial or inevitable successor to corporal punishments, and that liberal political philosophy both masked and required new, more insidious, forms of power. However, Meranze combines these Foucaultian insights with  56  consideration of class and gender to show that subjects were constituted according to middle class norms of masculinity and femininity, and therefore extends Foucault's analysis. There have been a number of counter-revisionist 5 accounts of the history of the prison, but given the repetitive nature of this literature I will discuss only three: the work of William Forsythe (1987) and Alyson Brown (2003), both of whom provide broad analyses, and Janet Semple's (1993) analysis of Bentham’s plans for the Panopticon. Although I have categorized these thinkers as distinctly ‘counter-revisionist’, there are several points of overlap between these accounts and some of the analyses considered earlier, especially between Brown (2003), Forsythe (1987) and Zedner (1991) all of whom focus on the patchy nature of the reforms that Foucault describes. There is also a common theme between Semple (1993) and Ignatieff (1978), both of whom observe that Bentham intended the Panopticon to be run by a private contractor and that this was a major reason behind Parliament’s rejection of his plan. Further, Ignatieff (1978), Forsythe (1987) and Brown (2003) agree that the prison was used for explicit punishment and not only for rehabilitation, indicating that the divide between revisionist and counter-revisionist histories of the prison is a matter of degree and not a clear bifurcation. Brown (2003) analyzes prisons in England from 1850 to 1920, and challenges Foucault's analysis on three central grounds. First, Brown (2003) claims that Foucault (1995) leaves no room for resistance amongst prisoners, whereas records of prison disturbances and riots show the presence of resistance and the only partial success of discipline. Secondly, Brown (2003) argues that Foucault’s claims apply primarily to convict prisons and are much less convincing in  Discipline and Punish is one of a number of histories of criminal punishment that historians label as 'revisionist', because it critically re-examines and challenges the earlier orthodoxy about the history of punishment (e.g. liberal narratives of progress). Counter-revisionist analyses of the history of punishment are those that have challenged and refuted central claims made by the revisionist historians. 5  57  regard to local prisons 6, where disciplinary regimes were undermined by incompetent staff, overcrowding and disciplinary problems amongst prisoners (70). Brown (2003) therefore follows Zedner (1991) in arguing that during the mid nineteenth century local prisons were varied and less rehabilitative in intent than convict prisons. Thirdly, Brown identifies a trend away from reforming prisoners and back to severe, physical punishment in convict prisons from the mid-1860s to the mid-1890s, as a result of prison riots and a loss of public confidence in rehabilitation. Brown (2003) concludes that these new, punitive regimes constituted physical cruelty, because they involved a poor diet, hard labour, and severe punishment for prisoners who misbehaved (92), including the use of handcuffs, leg irons, and sometimes electric shocks or burns. Brown's (2003) first objection to Foucault's (1995) account seems to rest on a misunderstanding of Discipline and Punish, which as a genealogy was primarily intended to critically analyze the techniques and rationality of power, rather than to describe penal theory or practices. However, Brown does raise the point that – for all Foucault’s (2003) talk about subjugated knowledges – his analysis of the prison is based on the records of prison designers and administrators, and does not draw on the experiences of prisoners who were subjected to discipline. Brown's second claim repeats Zedner's (1991) argument that local prisons in England during the mid nineteenth century did not aim to rehabilitate prisoners, but this does not really address Foucault's (1995) claims about the nature of power. The third claim advanced by Brown, that after 1860 prisons became punitive and involved physical cruelty, is perhaps the most significant because it challenges Foucault's (1995) account of the replacement of violent, sovereign power by disciplinary power. Nonetheless, Brown identifies a return of rehabilitative Until the end of the nineteenth century there were two different levels of government running prisons in England. Convict prisons were run by the national government and held prisoners on longer sentences. Local prisons were run by the local government, and held offenders who were imprisoned before trial and those on short sentences. 6  58  approaches to punishment during the late 1890s, and does not advance her own account of power and penality, so her criticisms of Foucault (1995) largely miss their target. Brown's (2003) second and third critiques of Foucault (1995) are also advanced by Forsythe (1987), who accepts Foucault's broad account of changing forms of punishment, but adds caveats about the slow and imperfect process of transformation in prison regimes. Forsythe does not address questions about the broader social function and significance of the prison as an institution, nor about gender or class biases, and concludes by counselling against oversimplification of the history of the prison. Both Brown (2003) and Forsythe (1987) therefore criticize Foucault's analysis of the prison based on an interpretation of Discipline and Punish and an approach towards history that Foucault does not share, and as a result their criticisms do not respond to the key points of Foucault's analysis. The final counter-revisionist analysis is provided by Janet Semple (1993) who challenges Foucault’s (1995) account of Jeremy Bentham's plans for the Panopticon. Foucault (1995) describes the Panopticon as based on constant surveillance of inmates who were kept in solitary confinement and refers to it as "the diagram of a mechanism of power reduced to its ideal form" (205). In contrast, Semple (1993) argues that the Panopticon was designed to be a contractor-run and profitable institution, stating that "in the panopticon writing, at least, Jeremy Bentham is the forerunner of the application of modern free-market ideology" (34). Semple (1993) explains that the design drew heavily on Samuel Bentham’s (brother of Jeremy) design for a Russian factory based on continual surveillance (100), and that the architectural principle of constant surveillance therefore begun as a plan to increase the productivity of workers. Semple (1993) also states that Bentham did not advocate longterm silent or solitary confinement, since he believed prisoners’ work would produce both individual reform and profit 59  for the governor. Although Bentham planned multi-layered observation (Semple, 1993, 140), allowing the public to observe the Panopticon and thereby preventing cruelty, Parliament rejected Bentham’s plan due to fears about the risk of prisoner abuse (Semple, 1993, 274). Semple concludes, in an apparent attack on Foucault: The panopticon is not a paradigm of government or society but a harbinger of humane competence, a light leading through the darkness of error and superstition to a new heaven and a new earth. It would also have made Jeremy Bentham a very rich man (1993, 232). Semple’s (1993) account suggests that Foucault misrepresents the techniques and relations of power involved in the Panopticon in several respects. Firstly, it seems that the rationality of the market and exploitation of workers was built into the design of the Panopticon, which leads Semple to regard Bentham as the forerunner to neoliberal ideas about privatization. Secondly, Semple explains that the public would have had access to the Panopticon and hence surveillance would have been used over the guards and governor rather than simply over prisoners as Foucault (1995) suggests. This multi-layered observation implies a different, and perhaps more democratic, model of power than Foucault's account of normalization through surveillance and assessment by unobserved authority figures. Finally, Semple joins Ignatieff (1978) in noting that Bentham's plan for the Panopticon was rejected by Parliament, and therefore the prison was never built, which raises questions about whether it is appropriate for Foucault to use the Panopticon as the exemplar of the disciplinary prison. It is rather unclear what conclusions one should draw about the nature of power from Semple's arguments about Bentham's plans for the Panopticon, because Semple does not offer an alternative account of power and the prison. On the one hand, Semple's account of Bentham's aim to profit from running the Panopticon bolsters Marxist-inspired arguments about the significance of the profit motive in shaping penality. On the other hand, Parliament rejected 60  Bentham's plan because they were concerned about a private contractor building and running a prison for profit. The rejection of Bentham's plan on this grounds suggests that although the profit motive may have shaped penal theories, it did not directly shape penal practices in late eighteenth and early nineteenth century England. However, combining Semple's (1993) analysis with Thompson's (1967) arguments about the use of discipline to develop productive workers suggests that capitalism had a major influence on the development of disciplinary techniques. Arguably, the origin of Jeremy Bentham's ideas about constant surveillance suggests that discipline did not develop in the prison, as Foucault (1995) suggests, but instead in the factory. The analyses by Semple, Dumm (1987) and Meranze (1996), like that of Melossi and Pavarini (1981), hint at a degree of overlap between Foucault's (1995) account of power, and Marxist-inspired analyses that highlight the significance of class and labour exploitation in shaping penality. Dumm (1987) argues that disciplinary power can coexist with the exploitation of labour, and if there is no necessary inconsistency between these aims and techniques then some prison regimes might have used work both as a disciplinary technique and to generate income. Dumm's (1987) analysis therefore offers a way to combine Foucault's conception of disciplinary power with Semple's (1993) argument that Jeremy Bentham planned to profit from the panopticon, and with both Semple's and Thompson's (1967) claims that disciplinary techniques were used in factories. If discipline and the capitalist exploitation of labour can coexist then many disputes about the origin, causes, and uses of penal regimes become a matter of differing emphasis, instead of requiring one to reject one explanation in favour of another. What this suggests is that Bentham's design for the panopticon can be understood both as a mechanism of disciplinary power, as Foucault (1995) argues, and as a means of generating profit, as Ignatieff (1978) and Semple (1993) emphasize. 61  Moreover, Meranze (1996) both follows Foucault (1995) in regarding disciplinary power as being based on the human sciences and liberal political philosophy, and argues that disciplinary norms were strongly shaped by class and were intended to control the working class. This provides a point of overlap between Ignatieff's (1987) Marxist-inspired arguments about the prison serving the function of class control, and Foucault's (1995) emphasis on normalization, by implying that the disciplinary normalization described by Foucault may have aligned with capitalist ideology. Nonetheless, there are clearly cases where the goals of class control, the exploitation of labour, and disciplinary power did not co-exist, and here it is important to take into account the intersections between gender and race, which shaped beliefs about who was capable of being disciplined. The analyses by Hirsch (1992), Davis (1993 & 2003), Colvin (1997) stress that the the convict lease system in the Southern US was based on the exploitation of convicts's labour and maintenance of racialized inequality, not upon any ideal of disciplinary normalization. Further, Rafter (1990) argues that even in Northern states Black women were regarded as less likely to be capable of reform than White women, and were therefore sent to penal institutions with less rehabilitative regimes. The most productive analytical approach therefore seems to be to draw insights from Foucault's (1995) account of penality in terms of disciplinary power, from Marxist-inspired analyses that are attentive to the direct and indirect impact of capitalism in shaping penal regimes, and from analyses attentive to race and gender. Ignatieff (1981) responds to early counter-revisionist critiques in an article published several years after his own account of penal reform in England. Ignatieff (1981) notes that he (1978) and Foucault (1995) agree on the overall trends in punishment: a reduction in punishment involving public infliction of pain, the use of imprisonment as the pre-eminent 62  penalty for serious offences, and the imposition of strict rules in the new prisons. Then Ignatieff rejects the counter-revisionist accusation that he and Foucault exaggerated the transition from traditional law to rationality and discipline, arguing that their claims about the changes in punishment are essentially correct. Ignatieff (1981) also refutes the criticism that his and Foucault's accounts describe changing penal theory rather than actual practice, on the grounds that even changes in ideology rather than practices require analysis and explanation (164). These disagreements about the history of the prison reveal a fundamental dispute over the nature and meaning of history. Whereas Dumm (1987) and Meranze (1996) present critical historical accounts that support Foucault's analysis of disciplinary power, Brown (2003) and Forsythe (1987) focus on the details of prison regimes and challenge the empirical accuracy of Foucault's (1995) account. There are two major difficulties with this counter-revisionist approach taken by Forsythe (1987), Brown (2003), Zedner (1991) and to a lesser extent Semple (1993). Firstly, these counter-revisionist critics judge Foucault's (1995) historical analysis according to criteria of whether or not he provides an accurate description of changes in prison regimes, which is not the primary purpose of Discipline and Punish. Foucault (1995) states clearly that his work is a genealogy, and hence it provides a historical account that critically analyzes forms of power and denaturalizes them by showing how history could have been different. Foucault (1995) therefore does not share a methodology focused on causal explanation, and indeed explicitly criticizes such accounts of history on the grounds that they portray existing institutions and forms of power as historically inevitable. The second difficulty with these counter-revisionist analyses is that they tend not to provide an alternative account of the broad trends in criminal punishment or the power relations 63  involved, and therefore the criticisms come across as nitpicking and not as the basis for a fundamentally different account of penality. This point is made by Ignatieff (1981) and eloquently expressed by Phillips (1983) who observes that: Rather than offer alternative theoretical contributions to those of the ‘revisionists’, most of the contributors to that volume [Bailey ‘81] simply take refuge in a detailed but sterile empiricism, as if this somehow refuted larger theoretical overviews. They quote details, gleaned from the sources, to show that events and personalities were more complex than can be easily accommodated within the large ‘revisionist’ patterns. But this is to abdicate the historian’s true function…To explain events as simply ‘one damned thing after another’ is unsatisfactory history at any time. (68) Given the substantial overlap between Foucault's (1995) account of the broad changes in criminal punishment and multiple other historical accounts, including the counter-revisionist histories produced by Brown (2003) and Forsythe (1987), it appears that Foucault’s arguments pass the test of empirical plausibility. Nonetheless, those who favour an approach to history based on ‘detailed but sterile empiricism’ will no doubt continue to find Foucault’s account of power and the prison unsatisfactory. Foucault responded to critiques made by unnamed but presumably counter-revisionist historians in the 1978 interview ‘Questions of Method’, explaining that he provides a fragmentary analysis and not claims about the totality of punishment (2000e, 205). Foucault (2000e) defends his genealogical approach and states that his intent is to disrupt the selfevidence of the prison and to analyze its rationality: If I had wanted to describe ‘real life’ in the prisons, I indeed wouldn’t have gone to Bentham. But the fact that this real life isn’t the same as theoreticians’ schemes doesn’t entail that these schemes are therefore utopian, imaginary, and so on…These programs induce a whole range of effects in the real (which isn’t of course the same as saying that they take the place of the real): they crystallize into institutions, they inform individual behaviour, they act as grids for the perception and evaluation of things…They are fragments of reality that induce such particular effects in the real. (233)  64  Foucault (2000e) thus acknowledges that his account of changes in punishment does not correspond exactly to actual practice, but argues that his intent was to identify and critique the rationality of the prison that influenced and motivated these practices. Foucault (2000e) then turns the criticisms of counter-revisionists against them, claiming that the disillusionment over the failure of the prison reveals a disciplinary rationality: "If the prisons were seen to have failed, if criminals were perceived as incorrigible…it’s precisely because this type of programming didn’t remain a utopia" (233). This claim is interesting, since it both supports Foucault’s (1995) claims regarding the significance of the disciplinary prison, and acknowledges public loss of faith in the institution’s efficacy – a point that may sit in tension with his earlier claim that the prison "had always been offered as its own remedy" (1995, 268). Temporary crises of confidence in the possibility of reforming prisoners may help to explain the persistence of severe prison regimes and deterrence that Brown (2003) and Forsythe (1987) identify. I will consider the rise of less disciplinary and more severe penal theories and practices in the late twentieth and early twenty-first centuries in the following chapter. The remaining chapters of my thesis draw on the alternative theoretical accounts of power and punishment discussed in this chapter in order to explore contemporary power relations around criminal punishment in the United States, and England and Wales. In Chapter Three, I analyze the recent trends towards more severe penal regimes and prison privatization, which picks up on the themes of Semple's (1993) account of Bentham's plan to run the Panopticon for profit as a private contractor. My analysis in Chapter Three also discusses the use of prisoners' labour to produce good or services for corporations, which connects to the Marxistinspired analyses of the exploitation of prisoners' labour discussed in this chapter. In Chapters Five and Six I analyze the constitution of racialized identities and inequalities through the 65  criminal justice and prison systems, drawing on Colvin's (1997) and Davis's (2003) insights about the way criminal punishment is racialized, and its contribution of penality to racialized inequality in America. Finally, in Chapters Five and Seven I discuss the constitution of gender identities, and draw on Rafter's (1991) insights about the gendered nature of discipline and the attempt to instil normative ideals of femininity amongst women prisoners. Many of the alternative historical accounts in this chapter therefore include theoretical analyses of penality that generate insights about the techniques and dynamics of power in the contemporary prison systems of the United States, and England and Wales.  66  Chapter Three: Privatization, Penal Severity and Neoliberalism Overview, and Notes on the Use of Statistics In Chapter Two I surveyed the historical literature on criminal punishment in England and America, and identified three broad conceptual approaches that challenge Foucault’s account of power and the prison: Marxist-inspired analyses, analyses of race and/or gender, and counter-revisionist analyses. In this chapter I test the questions these analyses raised about Foucault’s account of punishment against the empirical realities of contemporary penal policy and prisons in America, and England and Wales. I identify two major empirical trends in criminal punishment in the thirty years since Discipline and Punish was published, and consider the implications these trends have for our understanding of power in relation to criminal punishment: firstly, the rise in both prison populations and incarceration rates; and secondly, the introduction and expansion of privately-operated prisons. I use literature and data about the prison systems in America, and England and Wales to explore these two empirical trends, and assess whether they are consistent with Foucault's account of power and penality, or whether they lend credence to alternative accounts such as those described in the previous chapter. My analysis of the normative significance of these trends includes a critical examination of Nancy Fraser's (2003) argument about the decline of disciplinary power, the changing role of the prison, and the replacement of disciplinary normalization of criminals by repression. Although I share Fraser's view that prison privatization and increased penal severity are significant, I argue that her account of the change in techniques of power is over-simplistic and mischaracterizes Foucault's (1995) account of disciplinary power.  67  These two trends of prison privatization and rising incarceration rates mean that the power dynamics around criminal punishment have altered significantly since the publication of Discipline and Punish during the mid 1970s. The recent practice of prison privatization means that there is a ‘new’ profit motive involved in incarceration in both England and Wales, and the United States, and the existence of such a profit motive lends support to Marxist accounts of penality as involving the exploitation of those being punished, and of economic influences on criminal justice policy. Whereas Foucault (1995) argued that prison regimes sought to produce disciplined subjects and not profit, I argue that the economic exploitation of prisoners cannot be so easily dismissed in the era of prison privatization. Further, there has been a punitive shift in penal philosophy and policies in both England and Wales, and the US, including longer prison sentences and more restrictive use of parole, which has driven the rising prison populations and incarceration rates. This recent punitive trend lends some credence to counter-revisionist analyses that have questioned whether the institution of the prison is characterized by disciplinary normalization. I conclude that contemporary penal philosophies and regimes are less rehabilitative than they were during the 1970s, but argue that discipline continues to play an important role in the English and American criminal justice and prison systems. Before tackling these subjects it is necessary to address a methodological and epistemological issue about the use of empirical data and literature. One of the central themes of Foucault's work is the interconnection of power and knowledge, and thus the impossibility of 'neutral' or 'objective' perspectives. On the one hand, Foucault's analyses of the prison, madness and sexuality seem to suggest that one should reject and counter claims of official or scientific truth. On the other hand, Foucault (2003) seems to favour the articulation of strategic, oppositional truth claims, and this critical stance underlies his genealogical method. In some 68  cases there is no tension between Foucault's support for strategic critiques and his opposition to scientific and hegemonic forms of knowledge, but it is unclear whether these positions are always consistent. For example, the use of official statistics about prison populations might facilitate an analysis of racial bias in the criminal justice system, or a critique that too many people are imprisoned. If there are situations in which one might mobilize scientific or official truths in order to better articulate a counter-hegemonic perspective, then should it be done? To put the question differently, does Foucault's support for strategic, critical truth claims outweigh his opposition to forms of official or scientific knowledge? The best indication of how Foucault might answer this question is found in the interview "Truth and Power" where he advances a number of 'propositions' about truth and the attitude intellectuals should take towards it: 'Truth' is to be understood as a system of ordered procedures for the production, regulation, distribution, circulation, and operation of statements. 'Truth' is linked in a circular relation with systems of power that produce and sustain it, and to effects of power which is induces and which extend it – a regime of truth…The essential political question for the intellectual is not to criticize the ideological contents supposedly linked to science, or to ensure that his own scientific practice is accompanied by a correct ideology, but that of ascertaining the possibility of constituting a new politics of truth. The problem is not changing people's consciousnesses – or what's in their heads – but the political, economic, institutional regime of the production of truth. (Foucault, 2000b, 132-3) My reading of this passage and of "Truth and Power" as a whole is that Foucault favoured a change in the conditions whereby a system of truth is produced, and not that he had an overriding objection to particular claims or forms of 'evidence'. In some cases scientific truth claims can and have been mobilized in a counter-hegemonic manner, for instance Anne FaustoSterling's (2000a) use of biological knowledge regarding organs, genes and hormones to argue that binary sexual categories are not natural. Given that political, social and economic disputes  69  are frequently reflected within scientific discourse (i.e. 'science' itself is often deeply divided), there is no necessary connection between hegemony and any given scientific claim or statement. However, there is a further problem with the data I will be using in this chapter, because much of our knowledge regarding criminal justice – and especially about prison populations – relies on official reports and statistics. Analysing trends in prisoner numbers, arrest rates, and levels of crime reported to the police means relying on data that is produced, collated, analyzed and published by agencies of the state. It would be naive to believe that such official figures are not influenced by the prevailing ideological framework, institutions and even the political aims of past or present state authorities. The questions posed and the categories used in compiling statistics create biases that cannot be removed by later interpreting the figures in the light of a critical perspective. For example, if the statistical categories presume binary sexes and the resultant data categorizes everyone as either 'male' or 'female', then later reinterpreting the flawed data will not remove the bias introduced by the erasure of people who identify outside these binary sex categories. The feminist sociologist Dorothy Smith (1990) therefore warns that official statistics constitute a self-reinforcing "ideology circle" (94) that may be more resilient and less visible than the ideological nature of other forms of knowledge. As such, there is cause for scepticism about whether analysis using such official statistics can do anything other than reinforce existing power relationships, particularly since such figures deploy the notions of hierarchical observation and objectivity that Foucault views as inherently elitist and by implication hegemonic. To complicate matters even more, Foucault (1990; 2003) shows that statistical knowledge and its use in the management of populations are characteristic techniques of  70  biopower, so using this statistical information may make one complicit in this form of governance. Despite all these concerns, Foucault's (2000b) comments about the need to advance counter-hegemonic truth claims and political judgement provide the best argument that the use of official statistics should depend on strategic decisions about how best to challenge prevailing relations of power and knowledge. In this case, I think that a strategic judgement indicates that one should use official data regarding prison populations, arrest rates etc. because official figures are often the only source of information about important criminal justice indicators. Gathering independent information about prisoners is particularly difficult because security procedures in prisons create restrictions on the access provided to researchers, and because the authorities may not give permission for such access. There are also important ethical concerns about conducting research in prisons, because prisoners are a vulnerable population with impaired autonomy and privacy, high incidences of mental health problems, and high rates of learning disabilities. Foucault's involvement during the 1970s with the Group d’Information sur les Prisons (Eribon, 1991; Macey, 1993) shows that he was aware of such difficulties in collecting information about imprisonment, and of the political role and consequences of this exclusion of observers. Although valuable sociological research about prisons and prisoners has been conducted despite these limitations, large-scale research such as analyses of trends in national prison populations inevitably relies on official data. Nonetheless, researchers can and have used this official data to criticize prevailing policies and practices, for instance by arguing that punishment is too severe, that rising incarceration rates are an costly and ineffective way to reduce crime, and to point to broader social problems such as racialized biases and socioeconomic inequality. 71  In some cases, the availability or unavailability of statistics about a particular subject depends on the outcome of political battles to make the criminal justice system more accountable and/or to render biases or inequities more visible. One example is the collection and publication of data about the race or ethnicity of people that the police stop and search in the UK, and in the past this data has demonstrated the existence of 'racial profiling' for drug and terrorism offences (UK Ministry of Justice, 2007, 31-34). In contrast, there is no systematic collection of data about the race or ethnicity of those searched by police in the US, and as a result it is more difficult to substantiate accusations of racial profiling by police. However, analysis of longterm trends in the incarceration rate in America is enabled by the publication of annual reports that provide both the numbers of people imprisoned and the incarceration rate (i.e. the number of prisoners per 100,000 population). In the UK, such analysis of trends in the incarceration rate is more difficult, because the government figures usually specify the prison population but not the incarceration rate. The information that is provided or withheld about aspects of the criminal justice system therefore has political causes and consequences, enabling or hampering particular truth claims. Despite my concerns about the biases and occlusions within official criminal justice statistics, I believe that the advantages of using this data outweigh the disadvantages. Using official statistics about prison populations enables many critiques of existing power relations, and allows one to critically analyze how hierarchical knowledge links to criminalization and imprisonment. While Foucault (2003) observes that the use of statistics is part of the exercise of biopower and contributes to the production of racialized categories, the history of racialized struggles in the US and UK suggest that statistics about racialized or ethnic biases can also be valuable to challenge racialized discrimination and inequity. Similarly, scientific analysis has 72  provided valuable critiques of contemporary penal policy, particularly in areas such as sentencing and drug addiction (Caulkins et al, 1997; Tonry, 2004; Ribeaud, 2004; Muir, 2008). While Foucault's work demonstrates that we should be careful what we wish for in terms of defining what counts as evidence, expertise or truth, the opposition of many 'experts' to the recent punitive trend suggests that scientific knowledge is not necessarily hegemonic in regard to penal policy. My task in the remaining chapters is to understand the nature of contemporary penality, its techniques of power and accompanying regime of truth.  Rising Prison Populations and Incarceration Rates The first trend in penal policy considered here is the rise in prison populations and incarceration rates (i.e. the number of prisoners per 100,000 population) in the United States, and England and Wales. Between 1980 and 2009, the total US prison population increased almost fourfold7, while the federal prison population increased over sevenfold8 , greatly exceeding both population growth and any rises in crime (Bureau of Justice Statistics, 1981, 2; 2010b, 1). By 2009, America had a prison incarceration rate of 502 per 100,000 residents (Bureau of Justice Statistics, 2010b, 1) and an incarceration rate for prison and jail inmates of 748 per 100,000 (Bureau of Justice Statistics, 2010c, 2). The rising prison population began later in England and Wales, but the population in custody increased 85 percent between 1990 and December 20099 (UK Home Office, 1991, 2; HM Prison Service, 2009, 6). According to the World Prison Population List the United States has the highest incarceration rate in the world at 756 prison and jail inmates per 100,000 residents, and the incarceration rate in England and  7  The increase was 390%, from a total prison population of 329,122 in 1980 to 1,613,740, in 2009.  8  The increase was 754%, from a federal prison population of 24,363 in 1980 to 208,118 in 2009.  9  From 45,600 in 1990 (UK Home Office, 1991, 2) to 84,231 in December 2009 (HM Prison Service, 2009, 6)  73  Wales of 153 prison and jail inmates per 100,000 residents is far above that of comparable European states such as France or Germany (Walmsley, 2009, 5). Given that there have not been corresponding increases in crime (Blumstein & Beck, 1995; Caplow & Simon, 1999; Millie et al, 2003; Mauer, 2006; Tonry, 2010), analysts regard the rising incarceration rates in England and America as evidence of greater severity in criminal punishment. This trend towards harsher sentences over the past thirty years has been described using terms such as "the new punitiveness" (Pratt et al, 2005, 128) and the "new culture of crime control" (Garland, 2001, 175). Baker and Roberts (2005) explain that the ‘new punitiveness’ consists of "a moral stance that criminals should suffer through enduring ‘hard treatment’ and being afforded lesser rights than their victims, with a ‘common sense’ conviction that harsh punishment deters crime" (Baker & Roberts, 2005, 128). The recent trend of longer sentences (as shall be discussed in detail shortly) and resultant rise in incarceration rates therefore reflects a change in both penal policy and philosophy during the past three decades. Moreover, the rise in prisoner numbers as a result of more punitive sentencing has often led to overcrowding in prisons, thereby causing prison conditions to deteriorate and often reducing prisoners' access to rehabilitative programs such as education and drug or alcohol treatment programs. Changes in English and American penal regimes therefore reflect both deliberate changes in policy, and unintended consequences such as the impact of widespread prison overcrowding. Incarceration rates in America have been rising since the 1970s, when there were major changes in the theory and practice of punishment. In the early 1970s penal policy in the US was characterized by indeterminate sentencing based on a rehabilitative penal philosophy, which allowed for individualized punishment and gave a lot of discretion to judges and parole boards (Tonry, 2005). This approach of individual assessment and penal institutions with the putative 74  goal of normalization is reflected in Foucault's (1995) account of disciplinary power, and specifically the disciplinary prison. However, by the mid 1970s this penal policy was widely criticized, including concerns about arbitrariness, racialized sentencing disparities, and a loss of confidence in the efficacy and normative justification of efforts to rehabilitate criminals (Tonry, 2005). Between 1975 and 1980 there was an overhaul of sentencing policy across the US, in which indeterminate sentencing was replaced by determinate sentences that often included mandatory minimum terms (Tonry, 2005). The political momentum for this change came from an unlikely coalition of critics of the previous system, including civil liberties and prisoners' rights activists, opponents of the medicalized rehabilitation model, and Republicans seeking tougher sentences (Tonry, 2005). Garland (2001) also argues that Foucault's (1995) critical analysis of the prison contributed — probably inadvertently — to the shift away from rehabilitation, and towards a more punitive approach. This shift towards more severe sentences continued in many US jurisdictions during the 1980s and 1990s, and included 'three strikes' laws such as Proposition 184 passed in California in 1994. One of the major reasons for the rise in prisoner numbers in the US since the 1970s is more intensive policing and harsher sentences for drug offences, often referred to as the 'War on Drugs'. Between 1980 and 1996 the incarceration rate for drug offences in US state prisons rose from under 15 inmates per 100,000 adults to 148 inmates per 100,000 adults – nearly a tenfold increase (Blumstein and Beck, 1999). To put this figure into perspective, the proportion of the American adult population incarcerated in state prisons during 1996 for drug offences was significantly higher than the total incarceration rates (i.e. for all offences) recorded in England and Wales, or in Canada, and was almost 65 per cent higher than the total incarceration rate in  75  France10 (Walmsley, 1999). Blumstein and Beck (1999) calculate that between 1980 and 1996 in state prisons "the contribution of drug offences to total incarceration growth is 33 per cent" (22), and that in federal US prisons the drug incarceration rate increased more than elevenfold (45). Drug offences are measured in terms of arrest rates, so Blumstein and Beck (1999) cannot tell whether there was a rise in drug offences as well as more intensive policing and severe punishment. However, Mauer (2006) cites evidence from the Household Survey on Drug Abuse indicating that drug use declined significantly from 1979 to 2000, and thus argues that arrests and incarceration for drugs soared despite falling drug use. Sentences for other crimes in the United States also became significantly more severe from the late 1970s onwards. The time served in prison for offences such as murder, burglary, robbery and sexual assault in the US increased due to the imposition of longer sentences (including mandatory minimum terms), restricted use of parole, and fewer pardons (Blumstein and Beck, 1999, 35). Although some forms of crime rose during the 1980s, such as motor vehicle theft (Bureau of Justice Statistics, 1998), Blumstein and Beck (1999) attribute the growth in American incarceration rates for non-drug offences to more severe sanctions and not rising crime: For offences other than drugs, approximately 40 percent of the growth in incarceration in state prisons is accounted for by increases in decisions to incarcerate and approximately 60 percent of the growth by increases in the time served by those sent to prison. (1999, 54) Similarly, Blumstein and Beck (1999) conclude that the increase in incarceration rates for federal prisons reflects a combination of "increases in both commitment rates and, more  10 According  to Walmsley (1999) England and Wales had 125 prisoners per 100,000 adult population, Canada had 115 prisoners per 100,000 adult population, and France had 90 prisoners per 100,000 population. Walmsley explains that these figures rely on estimates of the adult population in each state and have been rounded to the nearest five "in order to avoid a spurious appearance of precision" (1999, 1). These recorded incarceration rates may not be for the year of 1996, because Walmsley (1999) states that his figures refer to dates between 1994 and 1998.  76  strongly, in time served" (52). In short, Blumstein and Beck's (1999) conclusion is that the rise in US prison populations and incarceration rates reflects a punitive trend in criminal justice policy, not an increase in crime rates. The imposition of more severe criminal penalties and the resultant rise in the prison population in England and Wales began during the 1990s, under the Conservative Home Secretary Michael Howard, who adopted the slogan 'prison works' (Coyle, 2003). The penal philosophy introduced by Howard involved the belief that there should be greater use of prison sentences, and that prisons conditions should be punitive, i.e. "that the experience of imprisonment should be unpleasant" (Coyle, 2003, 14). The New Labour governments from 1997 to 2010 continued this punitive trend by adopting a populist 'tough on crime' stance. Between 1991 and 2001 the adult custody rate11 rose by 65 percent, and the custody rates for some offences more than doubled (Millie et al, 2003, 372). Over the same period, there was a 139 percent increase in the prison receptions for sentences under a year, and a 62 percent increase in sentences from 4 years to life (Millie et al, 2003, 373). Millie at al (2003) therefore conclude that: The main reasons for the rise in the prison population are that sentencers are s e n d i n g a higher proportion of offenders to prison and that when they use custody they are passing longer sentences. (375). This punitive trend also involved the use of life sentences for a wider range of crimes, with the result that between 1990 and 2002 the number of life-sentenced prisoners in England and Wales increased by 66 per cent 12 (Coyle, 2003, 74).  11  The adult custody rate is the proportion of those aged 21 or older who were sent from court to immediate custody. The adult custody rate for all courts in England and wales increased from 17% to 28% between 1991 and 2001 (Millie at al, 2003, 372). 12  The increase was from 3095 life-sentenced prisoners in 1990 to 5150 in 2002 (Coyle, 2003, 74)  77  There has been less attention to the contribution of drug offences towards rising incarceration rates in England and Wales than in the United States, but Millie et al (2003) identify "a large increase in the number of convictions for drug offences" (371) between 1991 and 2001, despite a general pattern of falling crime. They note that legislative changes have introduced harsher penalties for some drug offences, and that there has been a more punitive attitude amongst sentencers and the general public, but do not rule out the possibility that drug crime also rose over this period (Millie at al, 376-380). More severe punishment for drug offences thus explains much of the increase in prisoner numbers in the United States from 1980 to 1996, and some of the increase in prisoner numbers in England and Wales between 1991 and 2001. The New Labour government in the UK also expanded the range of criminalized behaviour through their support for tactics such as zero tolerance policing, and through the introduction of the Anti-Social Behaviour Order (ASBO). ASBOs were introduced in 1998 as a new penalty for behaviour such as public drunkenness, graffiti, begging, loitering, and verbal harassment (Fitzgibbon, 2007). Although an ASBO is not itself a criminal penalty and is similar to a civil injunction, breaching an ASBO is a criminal offence and can result in up to five years imprisonment (Tonry, 2010). The use of ASBOs has been criticized as ineffective, due to the high rates of breach, and as part of a strategy of "pre-emptive criminalisation" (Fitzgibbon, 2007, 130) of those deemed most likely to commit crimes, particularly disadvantaged young people. In his review of criminal justice policy under New Labour, Tonry (2010) argues that ASBOs are illiberal, have "increased public discontent and fostered intolerance" (399), and are almost certainly ineffective because "there is no credible evidence that their use increased public confidence or decreased the prevalence of antisocial behaviour or public anxiety about it" (390). 78  Given this punitive trend in criminal justice policy in both the United States and in England and Wales, is Foucault's account of penality in terms of disciplinary power still relevant? On the one hand, the shift from rehabilitation to penal philosophies of deterrence, retribution, and/or incapacitation means that Foucault's (1995) account of the disciplinary prison seems more similar to early 1970s penal philosophy than to contemporary penal theory or practices. Whereas rehabilitation targets those who are currently imprisoned, deterrence is aimed both at preventing recidivism among current prisoners, and more importantly at deterring others from committing crimes. While Foucault stresses that disciplinary power targets individual prisoners, and indeed constructs the individual, penal policy based on deterrence targets the broader population. Moreover, while rehabilitation requires close surveillance and assessment of prisoners, which Foucault describes as "hierarchical observation, normalizing judgement, and their combination...the examination" (1995, 170), prison regimes based on deterrence, incapacitation and/or retribution do not require close monitoring of prisoners' behaviour. To the extent that the punitive trend in penal philosophy and sentencing policy have translated into changes in prison regimes, one would expect disciplinary power to play a far less significant role in contemporary prisons than in prisons during the 1970s. Garland (2001) reaches a similar conclusion, arguing that contemporary prisons are more explicitly about "exclusion and control" (177) than was the case in the 1970s. On the other hand, rehabilitative institutions such as probation, parole, and educational provision within prisons continue to exist in both the United States and in England and Wales (Garland, 2001). There have also been innovations that take a more rehabilitative approach to punishment, including the introduction of state Drug Courts in the US to provide treatment for non-violent drug offenders. Although there is less emphasis on rehabilitation now than during 79  the 1970s, there are still efforts to normalize criminals through education, drug treatment, counselling etc. (Coyle, 2005). Moreover, the changes in prison regimes may be far less significant than the change in penal philosophy implies, and may be more the product of overcrowding, cost-cutting and/or prison privatization than of the decline of the rehabilitative ideal. As I will discuss in later chapters, the racialized and gendered nature of prison populations is also important, because the punitive trend has had a disproportionate impact on those groups who are more likely to be imprisoned. To understand how the rise in prison populations and incarceration rates have changed power relations around criminal punishment, one needs to take into account the direct and indirect consequences of the carceral boom, and the broader socioeconomic and political context, including neoliberalism. Later in this chapter, and in Chapter Four, I will explore how neoliberalism relates to the change in penal philosophy and policies, and to techniques and relations of power – key to this dimension of penal philosophy is the trend towards the privatization of prisons.  Prison Privatization The second trend explored in this chapter is prison privatization. Privately run prisons operated for profit were introduced in the US during the late 1980s, and in England and Wales during the mid 1990s, although, as noted in the previous chapter, Jeremy Bentham proposed the idea in the early nineteenth century. In December 2009, 129,336 prisoners were held in privately operated federal or state prisons in the US and constituted 8 percent of the overall prison population (Bureau of Justice Statistics, 2010b, 9). There are privately operated prisons in the federal prison system and in 32 of the American state prison systems, while in each of New Mexico, Montana, Alaska, Vermont, Hawaii, and Idaho over a quarter of prisoners were held in 80  private facilities in 2009 (Bureau of Justice Statistics, 2009, 34). In England and Wales during May 2010 there were eleven contractor-run prisons that held a total of around 9,618 prisoners or 11.3 percent of the prison population — the highest proportion of prisoners in private facilities of any state in Europe (Prison Reform Trust, 2010, 56). Although the proportion of prisoners held in privately run prisons in the US, and in England and Wales, is relatively small, this trend is more significant than those figures would imply. As I discussed in the last chapter, the presence and role of the profit motive in prisons is conceptually significant because it is central to the disagreements between Foucault and Marxist theorists about the power relations involved in criminal punishment. Moreover, as I will explain later in this chapter, the introduction of contractor-run prisons has altered the ethos and practices in publicly-run prisons, including through the process of market testing whereby private contractors and the public sector compete on grounds of cost to determine who will run a prison. The first contract for a privately run prison in the United States was awarded by the state of Texas in 1988, and the practice was quickly adopted by other jurisdictions (Harding, 2001). The growing prison population (described in the first part of this chapter) led to overcrowding and deteriorating prison conditions, which prompted legal challenges, and as a result by 1988 the prison systems of thirty-nine states were under court supervisory orders or consent decrees (Harding, 2001). The practical advantage of introducing contractor-run prisons was that it enabled states to expand their prison capacity whilst spreading the cost over a longer timeframe, through a contract with the private provider who would design, build, manage and run the facility (Harding, 2001; Hallett, 2006). Moreover, prison privatization fit with neoliberal ideology since it promised to reduce the size of the state, introduce competition in the provision of services, and therefore potentially reduce the cost per prisoner. Prison privatization thus 81  appealed to American politicians for both practical and ideological reasons, and contributed to the carceral boom by enabling prison populations to rise without necessitating short-term government spending on new infrastructure. The introduction of prison privatization in England and Wales was driven by a similar combination of practical and ideological reasons. Prison privatization was consistent with the neoliberal policies of privatization pursued by Conservative governments since the early 1980s, and it was promoted by the influential free market think-tank the Adam Smith Foundation (Cavadino & Dignan, 2003). As was the case in the US, privately run prisons enabled the provision of extra places in an overcrowded prison system while deferring much of the cost, but Harding (2001) argues that the UK government also hoped to weaken the prison staff unions and thus reduce wage levels. Contractor-run prisons also promised to distance politicians from problems in the prison system, such as the high profile breakout of three prisoners from Parkhurst maximum-security prison in 1995 (Coyle, 2003). Prison privatization may also have gained momentum as a result of close links between Conservative politicians, former prison service employees, senior civil servants, and companies interested in the prison sector (Cavadino & Dignan, 2006). The New Labour governments from 1997 to 2010 expanded the use of privately operated prisons in order to minimize the short-term cost to the government of the rising prisoner numbers discussed earlier (Coyle, 2003). However, the introduction of privately run prisons has also led to changes in the ethos and practices of publicly run prisons in the UK, and has therefore had an impact on the prison system as a whole. The introduction of market thinking through prison privatization means that "value for money" has become one of the central principles of the public prison service in England and Wales (HM Prison Service, 2008, 7). 82  Decisions about who will run an existing prison or receive the contract for a new prison are made through a process of competitive bidding (also known as market testing), whereby the proposals of private contractors are weighed against those from the public sector (UK Home Office, 2006). The effects of competition are supposed to reduce costs and improve prison regimes, but Cavadino and Dignan (2006) express doubt that there have been cost savings, and note the public sector has sometimes been excluded from bidding for a prison contract. It is unclear whether there has been an improvement in the cost-effectiveness of prisons in England and Wales, and, therefore, whether neoliberal claims about efficiency through competition have born out in practice. Moreover, evidence provided in a Parliamentary written answer in 2007 suggests that for most categories of prison in the UK the cost per place in a privately run prison was significantly higher than that in a public prison (Prison Reform Trust, 2010). Many concerns have been raised about privately run prisons, particularly the risk that profit maximization leads to unacceptably low expenditure on staff, healthcare etc., resulting in sub-standard or unsafe conditions for prisoners (Sinden, 2003). Some evidence suggests that in America, and England and Wales privately operated prisons have lower staff to prisoner ratios than public prisons, and that the relatively low wages produce high staff turnover, inexperience and difficulty in maintaining order (Sinden, 2003; Prison Reform Trust, 2010). There is particular concern that cost-cutting has produced sub-standard prison conditions in some American states, because many contracts for privately run prisons specify that the cost must be significantly lower than for public prisons, and prison conditions are not rigorously monitored (Harding, 2003). At a state level privately run prisons in America also provide fewer educational programs than public prisons, because a smaller percentage of privately run prisons offer educational programs, and the range of programs available is narrower in privately run 83  facilities 13 (Bureau of Justice Statistics, 2003, 4). This inferior provision of educational programs to prisoners in privately run facilities is a significant failing, particularly in light of the low educational achievement of prisoners and their consequently restricted opportunities for legal employment. Prison privatization has therefore had an impact on prison regimes, potentially making prisons less safe and reducing prisoners' access to rehabilitative programs. While the criticisms of prison privatization often relate to the incentive to maximize profits by providing inadequate services, there is a deeper concern that the introduction of a profit motive in criminal punishment may distort policymaking. Both Davis (2003) and Hallett (2006) argue that the private prison industry is a clear beneficiary of increased incarceration and therefore has a direct interest in promoting the continuation and expansion of imprisonment. This financial interest in carceral expansion remains even if regulation and monitoring ensures that private prisons provide a service that is equivalent to, or better than, that of state facilities. Although there are standards of probity during the formal procurement process for prison contracts, there are often concerns about undue influence during the periods before or after this formal process (Harding, 2001). Prison contractors have sought to influence politicians through lobbying and political donations, and in some cases prison corporations have been directly linked to Republican politicians (Hallett, 2006). There have also been cases in the UK and US where public officials involved in awarding a contract have subsequently gone to work for the bidder, introducing a conflict of interest (Harding, 2001).  13 According  to the Bureau of Justice Statistics (2003), all federal prisons provided education programs in the period from 1995 to 2000, but there were differences in the provision of educational programs in state prisons: in 1995 educational programs were provided by 88% of public prisons and 72% of private prisons, and in in 2000 by 91% of public prisons and 88% of private prisons (4). Moreover, public prisons generally offer a wider range of educational programs than is provided in privately run facilities: in 2000 basic adult education was provided in 80.4% of state public prisons as compared with 61.6% of private prisons; secondary education was provided in 83.6% of state public prisons versus 70.7% of private ones; and special education was available in 39.6% of state public prisons versus 21.9% of private ones (Bureau of Justice Statistics, 2003, 4).  84  In addition to the obvious profit motive involved in privately run prisons there are other financial incentives in incarceration, including companies who market goods or services to prisons. Davis (2003) argues that in America "public prisons have become so thoroughly saturated with the profit-producing products and services of private corporations that the distinction is not as meaningful as one might think" (100). Another profit motive exists for companies who use prisoner labour, because prisoners are paid exceptionally low wages, receive no benefits and are legally prohibited from unionizing, making them a cheap captive labour force that offers the opportunity for huge profits (Davis, 2003; Price, 2003). It is also important to recognize that the people employed in the public prison system and privately run prisons have a financial interest in criminal punishment and therefore seek to influence policymaking (Christie, 2000). The existence of multiple profit motives towards imprisonment leads Price (2003) to identify a circular relationship of prison expansion: An increase in the number of convicts in convict labor programs leads to more income to prisons, which equals more profit from prison labor, which results in a trend to privatize prisons, which leads back to the beginning of the cycle: more convicts. (153)  The Relationship Between Punitiveness and Privatization There are good reasons to view rising incarceration rates and prison privatization as inter-connected, and in both England and Wales, and the US the shift towards privately run prisons accompanied the introduction of more punitive sentencing and rising prisoner numbers. Nonetheless, I am doubtful about Price's (2003) argument that the profit motive of private corporations involved in the prison industry has driven rising prisoner numbers. In the US the beginning of a punitive trend with attendant rising incarceration rates pre-dated prison privatization by a decade, so prison privatization clearly did not begin the trend towards greater penal severity. On the contrary, in both England and America rising prisoner numbers and 85  serious overcrowding in existing prisons were major reasons behind the introduction of privately run prisons (as discussed earlier). However, prison privatization has contributed to carceral expansion insofar as the use of private prison contractors offered the ability to add extra prisoner spaces faster and at less shortterm economic cost than would have been the case for new public prisons. Current incarceration rates might well be lower if American politicians in the 1980s and British politicians in the 1990s had been forced to choose between rising incarceration rates and high short-term expenditure on building new prisons, instead of deferring much of the cost by introducing private contractors. Prison privatization has therefore enabled the continuation of an existing trend of rising prisoner numbers and incarceration rates, which were caused by greater penal severity. My analysis therefore suggests that the punitive trend contributed to the introduction of prison privatization, but that private prison contractors have both benefited from and enabled the punitive trend. It also seems likely that private prison corporations would lobby against policies that would reduce or remove the need for their services, and therefore pose an obstacle to the reduction of incarceration rates (Davis, 2003; Hallett, 2006). In the course of this chapter I have argued that both greater penal severity and prison privatization raise questions about the contemporary relevance of Foucault's (1995) account of power and the prison. Rising prison populations and incarceration rates are the product of a punitive shift in penal philosophy and policy, which means that both prison regimes and the broader penal system are less rehabilitative than in the past. Whereas Foucault characterized the prison as an institution aimed at disciplinary normalization, some commentators have suggested that contemporary prison regimes focus on simply containing the ever-increasing prison population and that this undermines Foucault’s thesis (Garland, 2001). Similarly, the 86  introduction of privately operated prisons challenges Foucault's account of the prison, including prison work, as concerned with disciplining subjects and not with profit. In both cases, however, these trends are not necessarily directly opposed to Foucault’s theory of discipline as some critics suppose. Foucault's argument that prisons play an important role in constituting the subject is largely correct in my view and I explore it in great detail in later chapters with regard to race and gender, but there is no reason to believe that such disciplinary power is inconsistent with the profit motive in criminal punishment. Similarly, while I agree with Davis's (2003) argument that the contemporary US prison system allows the unjust exploitation of prisoners for profit, I believe that the profit motive is one amongst many reasons for prison privatization, and that its influence on carceral expansion is probably slight and indirect. While these two trends of prison privatization and carceral expansion suggest that Foucault's (1995) account of penality is less relevant now than it was thirty years ago, they do not mean that his analysis is irrelevant. Although Foucault's analysis in Discipline and Punish is about criminal punishment, he uses prisons as a case study to generate broader insights about power in modern society. Despite the changes in penality, it is possible that Foucault's analysis of disciplinary power remains as relevant as ever to other social institutions and processes, such as the education system, or social work. Moreover, Foucault provides insights about the interconnection of power and knowledge, the operation of subtle techniques of power upon the body, and the constitution of the subject, all of which have applicability for political theory and many other fields of study. Even if Foucault's analysis of the prison was wholly outdated — and, as I have explained above, it is not — then it would still be possible to use elements of Foucault's theoretical framework to analyze power dynamics that he does not discuss in Discipline and Punish, as I do in my 87  analysis of race and gender in Chapters Five, Six, and Seven. In the remaining chapters of this thesis I will argue that Foucault's account of disciplinary power continues to generate valuable insights both about contemporary criminal justice systems, and about contemporary techniques of power more broadly.  Responding to Fraser's "From Discipline to Flexibilization?" One of the most prominent critiques of Foucault's work in recent years is provided by Nancy Fraser (2003), and her arguments merit attention here. There are two reasons why it is important to consider Fraser's analysis: firstly, because her topic is similar to my own, insofar as she assesses the extent to which Foucault's account of disciplinary power is still relevant given the socioeconomic changes over the past thirty years. The second reason for discussing Fraser's argument is that she explicitly (albeit briefly) discusses the normative significance of changes in criminal justice policy and prisons. Fraser argues that Foucault's account of disciplinary power is outdated because it presumes three empirical facts that are no longer true: the national organization of social regulation; the non-marketized nature of social regulation through the Keynesian welfare state; and the subjectifying and individualizing logic of disciplinary self-regulation. In particular, Fraser contends that the rehabilitative approach to punishment has been replaced by explicit repression through more intensive policing and imprisonment, particularly of poor, racialized communities (2003, 166). Fraser therefore interprets the punitive trend in criminal justice as evidence that repression has replaced the disciplinary prison described by Foucault, and as an indication of the broader changes in techniques of power as a result of neoliberalism. I share Fraser's view that neoliberalism had an impact on penal policy and broader techniques and 88  relations of power (a topic I discuss in depth in Chapter Four), but she is wrong about the end of the disciplinary power and the rise of repression. Under close scrutiny both Fraser's empirical and theoretical analysis fails to hold up, because she oversimplifies the empirical trends over the past three decades, and fails to provide sufficient justification for her theoretical conclusions. The first problem with Fraser's (2003) analysis is the slippage between disciplinary power and the concept of "fordist regulation" (160). Fraser develops the concept of fordist regulation in the attempt to combine Foucault's account of disciplinary power with a Marxist materialist analysis, and her claim that disciplinary power is outdated rests on her claims regarding the death of fordist regulation. However Fraser does not demonstrate that Foucault's concept of discipline and her concept of fordist regulation are consistent with one another, let alone equivalent. Fraser's failure to show that disciplinary power equates to fordist regulation means that she sets up a straw-man argument in place of Foucault's (1995) analysis, and then attacks the straw man by arguing that its account of society and modern techniques of power are outdated. Given Fraser's (2003) claim that fordist regulation requires that social regulation is nationally organized, non-marketized, and involves a Keynesian welfare state, it seems implausible that fordist regulation is equivalent to disciplinary power. As Thomas Lemke (2003) points out, Foucault traced disciplinary power to the early nineteenth century, and therefore it long predates the welfare state. Further, the local and capillary nature of disciplinary power (Foucault, 1995, 27) means that social regulation does not have to be organized at the national level. In fact, the historical literature analyzed in the previous chapter shows that nineteenth century prison regimes were shaped by changing configurations of national, state and local governments (Forsythe, 1987; Meranze, 1996; Colvin, 1997; Brown, 2003). Foucault (2007) 89  himself notes these complex relations between city governance and other levels of political authority in Security, Territory, Population, where he makes it clear that he does not presume power is organized around a national government. Lastly, Foucault (2007) developed the concept of governmentality precisely in order to theorize the decentralized exercise of power through political economy, and he argued that the techniques of governmentality and discipline co-existed and reinforced one another (107-8). All of this suggests that Fraser (2003) is wrong in her belief that a shift away from the national organization of social regulation or the welfare state necessarily involves a decline or cessation of disciplinary power. However, if one makes the charitable assumption that Fraser is capable of demonstrating satisfactorily that discipline is equivalent to fordist regulation, then we are left with the task of assessing her claims about the replacement of discipline by new forms of power. I will briefly assess whether Fraser's (2003) first two empirical claims are applicable to contemporary criminal justice: that social regulation is no longer "socially concentrated within a national frame" (163), and that social regulation is increasingly marketized. I will then analyze Fraser's account of the change in penal policy, and her claim that the punitive trend means that disciplinary techniques have been replaced by repressive modes of power in criminal punishment. Criminal justice is an area that remains under national jurisdiction and policymaking is primarily determined by domestic political and administrative pressures. Criminal justice policy therefore varies considerably between nation states, including substantial differences in what activities are criminalized, the penalties that are imposed for a given offence, the numbers and proportion of people incarcerated, and in prison regimes. Some indication of the scope of this variation in criminal justice policy is provided by the World Prison Population List, which 90  shows that in 2008 Norwegian incarceration rates were only one eleventh of those in America, and that Icelandic incarceration rates were only one eighteenth of American levels14 (Walmsley, 2009, 3-5). It is also instructive to contrast the American figure of 756 prison and jail inmates per 100,000 population with the Canadian figure of 116 prison and jail inmates per 100,000 population (Walmsley, 2009, 3) despite the cultural similarities, and economic integration under NAFTA. While incarceration rates are undoubtedly a blunt instrument for indicating variations in criminal justice or penal policy, such variations provide cause to doubt that Fraser's claim about the globalization of social regulation is accurate in regard to criminal justice policy. One area in which it could be argued that penal policy is becoming more globalized is through international agreements about the treatment of prisoners. Although several such international agreements exist, they have limited efficacy because they may be disregarded by national courts and officials; for example, American officials and courts have failed to apply the UN Standard Minimum Rules for the Treatment of Prisoners within the US prison system (King, 1999). International institutions have had more impact on penal policy within Europe, since cases brought under the European Convention on Human Rights have led to changes in prison policies and conditions in England and Wales (Coyle, 2003). Although there is some evidence that supra-national institutions within Europe have influenced criminal justice policy, the transfer of powers is certainly not sufficient to warrant Fraser's claim about the loss of national responsibility for social regulation. Fraser’s (2003) second empirical claim is much more persuasive in regard to penal policy, because privatization and market thinking have been introduced in the English and The World Prison Population List compares the numbers and incarceration rates of prison and jail inmates (i.e. those detained either before or after sentencing) internationally, instead of solely counting prisoners (i.e. those detained after sentencing). The Norwegian incarceration rate was 66 prison and jail inmates per 100,000 population, the Icelandic incarceration rate was 40 prison and jail inmates per 100,000 population, and the American incarceration rate was 738 prison and jail inmates per 100,000 population (Walmsley, 2009, 3-5). 14  91  Welsh, and American prison systems (as I discussed earlier). However, Fraser does not adequately explain or justify her claim that the spread of privatization indicates the decline of disciplinary power. Fraser's belief that marketization is incompatible with disciplinary power may stem from her assumption that discipline occurs through a centralized state, but as I argued earlier this assumption is contradicted by Foucault's (1995) account of discipline as capillary. Alternatively, Fraser's claim that marketization indicates a decline in disciplinary power may be based on the belief that post-fordist social regulation is less focused on the individual, and thus involves less disciplinary self-regulation. If Fraser's view is that contemporary social regulation does not target the individual, then her position clashes with other analyses of neoliberal techniques of government. Notably, both Rose (1996; 1999) and Cruikshank (1999) show how neoliberal techniques of government have increased personal responsibility and self-discipline, and hence the use and internalization of disciplinary norms has continued despite neoliberal reforms. This question of neoliberal techniques of power is explored in more detail in Chapter Four. Fraser's third claim is that the use of disciplinary techniques to reform criminals has declined and has been at least partially replaced by straightforward repression. It is somewhat difficult to evaluate Fraser's (2003) claim about "the return of repression" (166) because she does not explain what she means by 'repression', how it works, why it has occurred, or what techniques it involves. It is unclear whether Fraser's use of the term 'repression' refers to the use of violence and coercion, to the economic aspects of the contemporary prison regimes such as the for-profit incarceration of those from poor communities, to a combination of these things, or to something else altogether. The best indication of what Fraser means by repression is in the  92  contrasts she draws between the American prison system and Foucault's account of power and the prison: In the US, accordingly, some observers posit the transformation of the social state into a "prison industrial complex," where incarceration of male minority youth becomes the favoured policy on unemployment. The prisons in question, moreover, have little in common with the humanist panopticons described by Foucault. Their management often subcontracted to for-profit corporations, they are less laboratories of self-reflection than hotbeds of racialized and and sexualized violence – of rape, exploitation, corruption, untreated HIV, murderous gangs, and murderous guards. If such prisons epitomize one aspect of postfordism, it is one that no longer works through individual self-governance. Here, rather, we encounter the return of repression, if not the return of the repressed. (2003, 166) In this passage, Fraser discusses the three empirical trends in prison regimes that I have identified in America, and England and Wales: increased punitiveness; the introduction of privately operated for-profit prisons; and the gendered and racialized nature of power and penality. I agree with Fraser that these trends raise questions about the extent to which Foucault's account of disciplinary power adequately captures power relations in the US criminal justice system. Moreover, I share Fraser's (2003) concerns about prison privatization and her criticisms of the "racialised and sexualised violence" (166) within contemporary prisons. Nevertheless, these deeply troubling features of imprisonment in America do not justify the over-simplistic conclusion that individual self-discipline and normalization has been replaced by 'repression'. Fraser (2003) is appropriately critical of the prevalence of sexual assault in the American prison system, but I think she is wrong to interpret sexual violence within prisons as simply a generalized form of repression. As I discuss in Chapter Seven, sexual assault is often targeted at prisoners who are gender-nonconforming or perceived as effeminate, and is therefore connected to disciplinary power through the constitution of sexed and gendered identities. The practice of sexual assault in American prisons is therefore linked to the internalization and enforcement of 93  gendered disciplinary norms, particularly around hegemonic masculinity. Once again, this suggests that Foucault’s (1995) theory is not inconsistent with the trends observed by Fraser, and that the extension of Foucault's work can generate new insights about contemporary forms of power. Similarly, I argue in Chapter Six that the over-representation of Black men in American prisons indicates that the US criminal justice and prison systems are playing key roles in constructing racialized identities and inequalities through biopower. These power dynamics around the constitution of gendered and racialized subjects are overlooked in Fraser's (2003) characterization of prisons as simply repressive. Fraser (2003) is right to suggest that the end of efforts to rehabilitate criminals would challenge Foucault's account of penality in which the prison is institutionally defined by the intent to reform subjects: One thing is clear: the prison was not at first a deprivation of liberty to which a technical function of correction was later added; it was from the outset a form of ‘legal detention’ entrusted with an additional corrective task…Penal imprisonment from the beginning of the nineteenth century covered both the deprivation of liberty and the technical transformation of individuals. (Foucault, 1995, 233) However, she is wrong to suggest that disciplinary normalization has been completely rejected in favour of punishing and containing inmates. As I argued earlier in this chapter, there are still rehabilitative institutions and practices within the English and American criminal justice systems, including parole, probation, and educational and vocational programs within prisons. Although prison regimes vary according to factors such as the security classification of the inmates, the definitive disciplinary techniques of "hierarchical observation, normalizing judgement and their combination" (Foucault, 1995, 170) persist in many prisons in America, and England and Wales. Prisons usually involve extensive surveillance of prisoners, and most inmates are required to follow detailed rules of conduct, a strict timetable, and prescribed 94  patterns of spatial distribution (Morris, 1998) — all of which are disciplinary techniques identified by Foucault (1995). Penal policy and philosophy have therefore become more punitive over the past thirty years, but they are not wholly punitive, meaning that disciplinary techniques now coexist with other techniques of power. Fraser's (2003) account of American prisons as characterized by repression is more than an argument about changing prison regimes — it is a direct challenge to Foucault's crucial insight about the productive nature of power. Attention to the productive and not merely repressive aspects of power involved in criminal punishment is what Foucault (1995) describes as the first rule of his analysis in Discipline and Punish: 1. Do not concentrate the study of punitive mechanisms on their 'repressive' effects alone, on their 'punishment' aspects alone, but situate them in a whole series of their possible positive effects, even if these seem marginal at first. As a consequence, regard punishment as a complex social function. (23) Fraser's focus on the repressive aspects of power is therefore a challenge both to Foucault's analysis of the prison, and to his broader account of power as constitutive of the subject. As I shall argue in later chapters, the English and American criminal justice and prison systems play an important role in constituting sexed, gendered, and racialized identities, both through discipline and through the exercise of biopower. By ignoring Foucault's account of power as productive, Fraser therefore overlooks many of the techniques and dynamics of power in contemporary criminal justice systems. Finally, Fraser's analysis seems to make an erroneous assumption about the historical succession of different modes of power, for instance the replacement of the disciplinary prison by the repressive prison. Although Foucault (1995) suggests in Discipline and Punish that sovereign power was succeeded by disciplinary power, he amends this argument in the 1977-8 Collège de France lectures: 95  The idea of government as a government of population makes the problem of the foundation of sovereignty even more acute…and it makes the need to develop the disciplines even more acute…So we should not see things as the replacement of a society of sovereignty by a society of discipline, and then of a society of discipline by a society, say, of government. In fact we have a triangle: sovereignty, discipline, and governmental management. (2007, 107) Foucault's (2007) account of the co-existence and inter-action of multiple techniques of power suggests that there are strategic combinations of different techniques and apparatuses of power, not simply the replacement of one dominant form of power by another. I draw on Foucault's (2007) model of a triangle between the three forms of power to analyze contemporary power relations in regard to criminal punishment. Shifting the framework of analysis to the 'triangle' of sovereign power, disciplinary power, and governmentality makes it possible to critically analyze the deployment of multiple techniques of power, instead of having a rather sterile academic debate about whether or not contemporary prisons are disciplinary. Moreover, by analysing the combination of multiple techniques of power one is better able to explore dynamics around race and gender, both in prison regimes and more broadly in English and American societies. Although Fraser observes the racial and gendered disproportionality of prison populations she does not explicitly relate these phenomena to her critiques of Foucault, nor to her identification of new forms of power under neoliberalism. In the following chapters I will analyze power relations around race, gender and criminal justice, drawing on Foucault's work around sex, sexuality, and race, and on queer theory, feminist theory, and critical race analysis. However, before considering the issues of race and gender it is first necessary to provide a closer analysis of neoliberalism in order to understand how changing penal policy relates to the wider ideological and political context over the past thirty years. This requires a definition of neoliberalism, an analysis of neoliberal approaches to crime, and a brief account of the social, 96  political and economic consequences of neoliberal reforms. In the next chapter, I use Foucault's analysis of neoliberalism in the 1979 Collège de France lectures, The Birth of Biopolitics, as a starting point for exploring neoliberal theory. I argue that Foucault (2008) fails to identify the tensions in neoliberal theory, particularly around criminal justice policy, and that he does not provide a critical perspective on neoliberalism. I then consider the analyses of neoliberal techniques of power provided by other scholars working in a Foucaultian conceptual framework, and the more economically focused analysis of David Harvey. This analysis of neoliberalism will explore the ideological basis for the punitive trend and prison privatization, and lay the foundation for the analysis of race and criminal justice in later chapters.  97  Chapter Four: Neoliberalism and Foucault Overview In the previous chapter I analyzed two trends in penal policy in the United States, and England and Wales that have altered techniques and relations of power: prison privatization, and the rise in prison populations and incarceration rates. I argued that although disciplinary power persists in English and American prisons, these trends indicate that contemporary criminal punishment is less focused on disciplinary normalization than Foucault (1995) argues, and that profit incentives and market thinking play a greater role than Foucault's analysis suggests. However, neoliberal ideas contributed to the introduction of prison privatization and greater punitiveness, so a full understanding of the normative significance of these trends requires closer attention to neoliberalism. In this chapter I analyze neoliberal theory, the techniques of power used in neoliberal government, and particularly neoliberal philosophies and policies that relate — directly or indirectly — to criminal punishment. I begin by briefly outlining Foucault's (2008; 2009) understanding of neoliberalism, and then critically assess both Foucault's account of neoliberal theory and the concepts he develops to analyze it, notably that of governmentality. I also consider how Foucault's (2009) account of neoliberal government compares to the analyses provided by a number of subsequent theorists working within a Foucaultian conceptual framework, along with a Marxist analysis of neoliberalism provided by David Harvey (2005). In this chapter, I hope to show that while Foucault's (2007; 2008) analysis provides a useful description of a large body of neoliberal theory, and generates some insights about neoliberal techniques of power, there are four major problems with his account of neoliberalism. Firstly, Foucault (2008) fails to identify the tensions within neoliberal theory, for instance between cost-effectiveness and the neoliberal account of the state. I argue that the logic of 98  neoliberal theory demonstrates that social and economic policies such as welfare cuts will undermine neoliberal policies on crime, and that neoliberal criminal justice policies may be more influenced by neoliberal conceptions of the state and the subject than by concerns about efficacy or cost-effectiveness. I conclude that Foucault's (2008) inattention to the conflicting impulses within neoliberal thought leads him to mischaracterize neoliberalism, and particularly neoliberal approaches to criminal justice. The second major problem with Foucault's (2007; 2008) account of neoliberalism is that he overlooks the coercive aspects of neoliberal government, while overemphasizing the freedom of the subject. Despite Foucault's (2007) account of the co-existence and inter-relation of governmentality, disciplinary power, and sovereign power, Foucault's (2008) analysis of neoliberalism does not explore the strategic combination of these multiple forms of power. Moreover, while Foucault's (2007) concept of governmentality is useful for showing how power works through subjects' freedom, I suggest that it lacks the critical impetus and insights that would challenge and denaturalize neoliberal forms of power. I argue that Foucault (2008) provides a largely uncritical account of neoliberalism that repeats neoliberal claims about freedom and choice, whilst ignoring the coercive implications of neoliberal theory, including the expansion of policing and imprisonment. The third problem with Foucault's (2007; 2008) analysis is that he ignores the economic dimensions of neoliberalism and their normative implications. The neoliberal theory described by Foucault (2008) outlines a policy programme that includes minimal welfare provision, the privatization of risks such as illness and unemployment, and a tax system that does not promote redistribution in favour of the poor. This neoliberal social and economic policy can be expected to cause increased socioeconomic inequality and poverty, but Foucault (2008) ignores these 99  issues. I argue that understanding neoliberal techniques of power requires a synthesis of poststructuralist analysis, including Foucault's insights about the combination of sovereign power, discipline and governmentality, with Marxist analysis of the economic aspects of neoliberalism, such as the work of David Harvey. Lastly, Foucault's analysis of neoliberalism — like his account of disciplinary power — does not draw out power relations around race or gender. I argue that a fuller understanding of neoliberal techniques and relations of power, particularly in relation to criminal justice, requires attention to race and gender. These issues of race and gender are explored in Chapters Five, Six and Seven, where I draw on Foucault's (2003) account of biopower, and his (2007) account of the multiple mutually-reinforcing techniques of power.  Foucault on Governmentality and Neoliberalism Since this dissertation examines Foucault's analysis of modern techniques of power, it is logical to begin a discussion of neoliberalism with Foucault's own comments about it. Foucault devotes the 1978-9 Collège de France lectures to the subject of neoliberalism, and in the 1977-8 lectures he introduces the concept of governmentality, which is a form of power that works through the market and economic knowledge. Both of these lectures were unpublished until recently 15, with the exception of the fourth of the 1977-8 Collège de France lectures 16 which was published in English translation under the titles "On governmentality" (Foucault, 1979) and "Governmentality" (Foucault, 1991). This single lecture about governmentality and Gordon's (1991) essay on the subject have shaped much of the subsequent interest in this concept amongst  15  The 1977-8 Collège de France lectures were published in French in 2004 and in English translation in 2007, while the 1978-9 Collège de France lectures were published in French in 2004 and in English translation in 2008. 16  This was delivered by Foucault on February 1st 1979 (2007, 87)  100  Anglophone academics.17 A body of secondary literature has developed around Foucault's concept of governmentality, much of which uses governmentality as a framework for understanding contemporary techniques of neoliberal government such as privatization, reduced welfare provision and a focus on managing risk (Burchell, 1996; O'Malley, 1996; Rose, 1996; Rose, 1999). Now that English translations of Foucault's 1977-8 and 1978-9 Collège de France lectures are available, it is possible to compare Foucault's own analysis of neoliberalism with the work of these later scholars who use a Foucaultian conceptual vocabulary. Foucault's discussion of neoliberal approaches to government begins in the 1977-8 Collège de France lectures published under the title Security, Territory, Population. In these lectures Foucault (2007) provides an account of the historical emergence and functioning of governmentality, which he describes as targeting the population, working through apparatuses of security, and using political economy as its form of knowledge. Foucault argues that in France government through security began in the eighteenth century, and that it changed both the techniques and the objectives of government. Government through security targets the population as a whole, as opposed to individuals, works through subjects' freedom, and through supposedly natural processes such as the free market. Security was applied to criminal justice through the use of statistical and probabilistic assessment of risks, costs, and efficacy, in the attempt to reach "socially and economically acceptable" levels of criminality (Foucault, 2007, 6). This security approach to criminality operates through voluntarism and self-interest, and is therefore distinct from sovereign power, and from the disciplinary approach of "surveillance and correction" (Foucault, 2007, 6). Government through security involves a minimum of active  17  For example, Nikolas Rose (1999) cites the published lecture 'Governmentality' as "the best introduction" (3) to the concept and lists Colin Gordon's (1991) essay as one of the four other recommended texts.  101  governmental intervention, but like disciplinary power it still aims to create normative subjects — albeit through different techniques. Foucault then presents an account of historical transitions between different forms of power, modifying his earlier (1995) claim about the replacement of sovereign power by disciplinary power in criminal punishment: There is not a series of successive elements, the appearance of the new causing the earlier ones to disappear…In reality you have a series of complex edifices in which, of course, the techniques themselves change and are perfected, or anyway become more complicated, but in which what above all changes is the dominant characteristic, or more exactly, the system of correlation between juridico-legal mechanisms, disciplinary mechanisms, and mechanisms of security. (2007, 8) This account of multiple techniques of power, as opposed to the wholesale replacement of one form of power by another, could be seen as a concession to those who argue that the historical changes in punishment were more complex and incomplete than Foucault suggests in Discipline and Punish (Forsythe, 1987; Brown, 2003). However, this statement about the transition between different techniques of power does not undermine Foucault's critical diagnosis of the nature of disciplinary power, it merely suggests that discipline co-existed with other techniques. Whether or not Foucault (2007) is making a concession to his critics, this account of historical change enables him to identify the emergence and growing use of governmental management without implying that discipline has ceased. Foucault (2007) argues that disciplinary power continued even after governmental management through security became important, and claims that discipline "was never more important or more valued than when the attempt was made to manage the population" (107). It is shortly after this claim (made in the fourth lecture of the 1977-8 series, previously published as 'Governmentality') that Foucault makes the important and much-cited argument about the inter-connection of different forms of power, stating: 102  We should not see things as the replacement of a society of sovereignty by a society of discipline, and then of a society of discipline by a society, say, of government. In fact we have a triangle: sovereignty, discipline, and governmental management, which has the population as its main target and apparatuses of security as its essential mechanism. (2007, 107-8) Most of the remaining 1977-8 lectures focus on the historical forms of government that preceded security, but in the final lecture Foucault returns to the subject of governmentality and emphasizes that it operates through "natural phenomena of economic processes or processes intrinsic to the population" (2007, 353). Foucault's observations about laissez faire economics and governing through the freedom of subjects provide the conclusion of the 1977-8 lectures and flow almost seamlessly into the subject of the following year's lectures: neoliberalism. In the 1978-9 Collège de France lectures published under the title The Birth of Biopolitics, Foucault (2008) begins by outlining liberal forms of government, which he links both to political economy and to a principle of internal restrictions or "how not to govern too much" (13). Foucault (2008) argues that during the 1970s there was a "crisis of liberalism" (69) because the Keynesian welfare institutions introduced to prevent a reduction in freedom actually curtailed freedom. Here, Foucault reiterates his (1995) argument that liberalism is intimately connected to the disciplines, and provides the example of Bentham's belief that the panopticon would both reform prisoners and ensure the productivity of workers, thereby generating profit (2008, 67). The interconnection of liberalism and discipline suggests that effective resistance against disciplinary techniques may disrupt liberalism, which Foucault confirms by describing the revolt against disciplinary power as part of the crisis of liberalism in the late 1970s. Foucault (2008) goes on to discuss the rise of neoliberalism, which he divides into two main forms: the German Ordoliberals, and the American neoliberals led by the Chicago School. Foucault traces German neoliberal thought to the period after the Second World War and the 103  Nazi regime, arguing that the Ordoliberals associated Keynesianism and state economic planning with state expansion and a corresponding loss of freedom in a manner similar to — though far less extensive than — Nazism. In the course of this analysis Foucault (2008) observes that there is no "autonomous socialist governmentality" (92) thus implying that socialism relies on existing liberal techniques of government, including discipline, and cannot resolve the current crisis of liberalism. The Ordoliberal reaction against the Keynesian state prompted the development of a new logic for the relationship between the state, the economy and the market (Foucault, 2008). While there are similarities between the Ordoliberal framework and earlier liberal economic theory propounded by the likes of Adam Smith, Ordoliberalism calls for a greater role for the market, creating "a state under the supervision of a market rather than a market supervised by the state" (Foucault, 2008, 116). The Ordoliberals also rejected the classic laissez-faire approach in favour of a government that actively facilitates the conditions for market competition, leading Foucault (2008) to conclude that "government intervention is no less dense, frequent, active and continuous than in any other system" (145). This is an interesting observation, but unfortunately Foucault does not expand upon it by explaining how neoliberal government intervention works, or the degree to which is works directly upon individuals as opposed to indirectly, for example through policies applied to businesses. According to Foucault (2008), German neoliberal thought thus favours government intervention to create appropriate economic policies for business and to encourage the development of enterprising individuals. This principle of the government enabling instead of restraining or counter-balancing the market requires a new form of social policy that does not attempt to equalize incomes, provide full employment or guarantee individuals against risks (Foucault, 2008). 104  Foucault pauses in this analysis to observe that the critics of the Ordoliberals have largely mis-identified the nature of neoliberalism and the techniques of power that it involves. Here, Foucault (2008) repeats his claim that neoliberalism is not the re-activation of classical liberal economic theory, and further observes that it is not merely market relations or intensified administrative intervention in the guise of market freedom. These barbed remarks about the tendency of critics to oversimplify help to clarify Foucault's belief that neoliberalism is a distinct rationality and assemblage of power that requires dedicated analysis. Foucault (2008) also identifies the influence of German neoliberal thought amongst the contemporary French political elite, noting the influence of neoliberal ideas on social policy around illness, unemployment et cetera. This explicit acknowledgement of the salience of neoliberalism in France at the time of the lecture strongly suggests that Foucault intended his analysis to be of political — and not merely academic or conceptual — interest and significance. Foucault (2008) then discusses the American neoliberalism of the Chicago School. Like neoliberal thought in Germany, Foucault sees American neoliberalism as essentially reactive and argues that it developed as a critical response to the New Deal, state planning of the economy during the Second World War, and government programs around poverty reduction and education. Foucault identifies two aspects of American neoliberalism that are distinct from German approaches: the attention to ‘human capital’ or the productive power of the individual; and the tendency to generalize economic logic throughout the social body. This latter characteristic of American neoliberalism leads to the application of economic thinking to criminality and law enforcement through assessments of the costs of delinquency, courts and punishment (Foucault, 2008). I will return to this issue of neoliberal approaches to criminal justice shortly. 105  Although Foucault regards German and American neoliberalism as historically and theoretically distinct, both approaches oppose Keynesianism, advocate government intervention to promote a competitive market, and treat ‘economic man’ as a premise and a goal of policy. This concept of the rational, self-interested and entrepreneurial economic man, or what Foucault (2008) refers to as "homo oeconomicus" (270), is important because neoliberal techniques of government are shaped by their conception of the subject 18. According to neoliberal theory, economic man is both the subject of freedom and "eminently governable" through the manipulation of his/her environment and incentives (Foucault, 2008, 270). Foucault argues that while neoliberals share Adam Smith's belief that knowledge of the whole economy is beyond the capacity of the state, they depart from Smith by emphasizing the rationality of individual actors within this process. Neoliberal thought thus views individuals as knowable, but identifies "an essential incompatibility between the non-totalizable multiplicity of economic subjects of interest and the totalizing unity of the juridical sovereign" (Foucault, 2008, 282). This gap between knowable individuals and the un-knowable complexity of the whole economy is the reason for using governmentality to harness the freedom of individuals. Neoliberal policies therefore attempt "the strategic reprogramming of individuals’ activity" (Foucault, 2008, 223) by harnessing the freedom of subject, presumably through practices such as the individual empowerment programs identified by Barbara Cruikshank (1999). Foucault (2008) depicts neoliberal social policy as being centred on the privatization and individualization of risk via a system of insurance that shifts the burden of provision for illness or old age onto individuals, instead of society or the state. The combination of this privatization  In economics, the rationality of an agent is defined in a very limited and specific way. Put simply, a rational agent in economics is one who is self-interested, has clear preferences, and acts based on calculations about how to maximize his/her utility or wellbeing. This thin definition of rationality and of the agent in economics is distinct from broader debates about reason, madness and individual behaviour in fields such as psychology and sociology. 18  106  of risk and the opposition to a planned or state controlled economy means that neoliberal thought favours minimal efforts to alleviate poverty, and holds that welfare support should be focused on those unable to support themselves, such as the seriously disabled (Foucault, 2008, 143). Foucault then outlines the neoliberal account of criminality, drawing on Gary Becker’s (1968) economic analysis of crime as an activity based on rational self-interest, and contrasting this with his own (1995) account of criminals as delinquents: The criminal is nothing other than absolutely anyone whomsoever. The criminal, any person, is treated only as anyone who invests in an action, expects a profit from it, and who accepts the risk of a loss. (Foucault, 2008, 253) If crime is viewed as the product of a cost-benefit calculus by rational economic agents, then it follows logically that one should be able to engineer a socially acceptable level of crime by increasing the cost and/or reducing the benefits of criminal behaviour. Neoliberal thinking thus suggests that crime rates would fall if law enforcement were expanded and intensified, because this would increase the severity of the penalty and/or the likelihood of being punished, thus altering the cost-benefit analysis of criminals (Foucault, 2008). This increased law enforcement could take various forms including an increase in the quality and quantity of policing, more rapid prosecutions, and/or more severe sentences. Within this economic conceptualization of crime, the efficacy of deterrence would depend on the certainty of punishment, and thus be affected by changes in the extent to which an individual's punishment could be modified by officials such as prison officers (Foucault, 2008, 254-5). It then follows that reductions in the discretion of criminal justice officials, such as the use of determinate sentences and mandatory minimum terms, should make punishment more certain and thus make deterrence more effective. This logic of deterrence suggests that the 107  individualized discipline described by Foucault (1995) in Discipline and Punish would be somewhat ineffective at reducing crime, because the length and severity of imprisonment is dependent on officials' evaluation of the offender and is therefore uncertain. Foucault (2008) then turns to a more obscure topic: the elasticity of crime. In economic parlance, elasticity is a measure of how much one variable changes in response to a change in another variable, for instance how an increase in the price of a newspaper alters its sales. Elasticity relates to criminal justice insofar as shifts in the likelihood and/or the severity of punishment change the supply of crime by altering the cost-benefit calculus of potential criminals. If crime (or at an individual level, the propensity of someone to commit criminal acts) is relatively inelastic then more certain and severe punishment will only produce small reductions in the crime rate. Alternatively, if the supply of crime is highly elastic then an increase in the level of policing or the severity of punishment will greatly reduce crime rates. Foucault (2008) suggests that the economic logic behind neoliberalism requires an understanding of the varying elasticities for different types of crime and between different offenders because "the supply of crime is not indefinitely or uniformly elastic" (255). Foucault also observes that drug use amongst addicts is highly inelastic because "the addict will want to find his commodity and will be prepared to pay any price for it" (2008, 257) even if this involves committing other crimes in order to fund drug purchases. The relative inelasticity of consumption amongst drug addicts means that efforts to prohibit drug sales or increase drug costs will not significantly reduce consumption among existing users and hence severe penalties for addictive drugs are not an effective means of crime reduction (Foucault, 2008, 257). Foucault describes neoliberal theory in terms of governmentality, as a form of government that works through economic knowledge, through the natural process of the market, 108  and through using incentives to shape the behaviour of individuals. Foucault (2008) repeatedly stresses that the neoliberal rationality of government is one of "self-limitation" (17), which means that neoliberalism is defined as much by how and what neoliberalism does not govern (for instance the lack of efforts to reduce socioeconomic inequality), as by what and how it does govern. One of the few areas where Foucault provides a detailed account of the neoliberal approach to a policy area is criminal justice, and he seems to apply the framework of governmentality to penal policy, by explaining how deterrence works through the freedom of subjects and by stressing cost-efficiency. Perhaps the most interesting part of Foucault's analysis is his observations about the neoliberal conception of the subject as economic man, which operates as both the premise and the goal of neoliberal policy. Economic man is a premise insofar as neoliberal theory assumes that people behave as rational and self-interested individuals, but it is a goal because this behaviour is what makes people intelligible and manipulable, as Foucault (2008) explains: Economic behaviour is the grid of intelligibility one will adopt on the behaviour of a new individual. It also means that the individual becomes governmentalizable, that power gets a hold on him to the extent, and only to the extent, that he is a homo oeconomicus. (252) This statement is important, because it shows that the neoliberal techniques of government only work if the neoliberal conception of the subject is accurate. If people do not behave like rational economic agents then the incentives and disincentives targeted at individual behaviour will fail to yield their desired results, and governing through the freedom of subjects is likely to be ineffective. The self-limitation at the heart of neoliberal government therefore relies on having the right sort of subjects — people whose behaviour, if not their thinking, approximates that of homo oeconomicus.  109  What Foucault (2008) does not analyze in these lectures, and indeed explicitly states does not interest him, is the actual processes of neoliberal government: I have not studied and do not want to study the development of real governmental practice by determining the particular situations it deals with, the problems raised, the tactics chosen, the instruments employed, forged, or remodeled, and so forth. (2) Although both Foucault's (1995) account of penality and his (2008) analysis of neoliberalism focus on theory and not actual practices, it is notable that his approach to governmentality is far less specific and critical. Whereas Discipline and Punish reveals the operation of micropower, there is no equivalent focus on the fine details of neoliberal forms of power. Similarly, where Foucault (1995) distinguishes between the logic of the disciplinary prison and its actual use in terms of constructing delinquency, he (2008) makes no distinction between the logic of neoliberalism and its effects or use. Foucault (2008) therefore describes and summarizes neoliberal theory in some detail, but without the critical analysis that typifies most of his other work.  Analysing Foucault's Account of Neoliberalism The largely uncritical portrayal of neoliberalism in the 1978-9 lectures has been noted by several Foucault scholars, including Colin Gordon (1991) who observed that Foucault’s discussion of neoliberalism suggests "a sense of (albeit value neutral) intellectual attraction and esteem" (6). Similarly, Mitchell Dean (2007) concludes that "it is apparent that he [Foucault] approaches the discussion of liberalism and neoliberalism with openness and generosity..." (86). Hamann (2009) goes further, suggesting that Foucault’s (2008) sympathetic account of neoliberal subjectivity in terms of autonomy in the 1979 lectures is connected to his later work on practices of the self, and concludes that in this regard Foucault is a "somewhat naive 110  advocate of neoliberalism" (48). The historian Michael Behrent (2009) shares the view that Foucault (2008) seems to be in favour of neoliberalism, and justifies this position by putting Foucault's remarks into historical and political context. According to Behrent, Foucault's sympathy for neoliberalism in the 1978-9 lectures expresses both an intellectual attraction to a set of new ideas, and a timely and intentionally political intervention on a subject that was current amongst the French elite. Behrent argues that on an intellectual level Foucault was drawn to neoliberal theory because it shared his critique of state authoritarianism, and a non-humanist conception of the subject. On a more political level, Behrent (2009) suggests that Foucault’s comments on neoliberalism should be read in the context of French politics, including Foucault’s longstanding criticisms of academic Marxism, his opposition to the traditional policies of the Socialist Party, and his support for the 'Second Left' (553). Behrent's analysis therefore suggests that Foucault’s sympathy for neoliberalism is certainly not ‘value neutral’ as Gordon (1991) suggests — instead, Foucault's uncritical account of neoliberal theory in the 1978-9 Collège de France lectures was motivated both by political strategy, and philosophical affinity. It therefore seems possible that Foucault was blinded to the flaws and inconsistencies within neoliberal thought by the points of overlap between his philosophical approach and neoliberalism, and by his longstanding scepticism towards socialism and class-based analyses. 1. Tensions within neoliberalism Regardless of Foucault's motivation in selecting the topic for his lectures, there are several problems with his (2008) discussion of neoliberalism, the first of which is that he ignores the tensions within neoliberal theory. Scholars such as David Harvey (2005) have identified many tensions within neoliberal thought, and exploring all of these is beyond the 111  scope of my project. Instead, I will focus on one obvious and important tension that Foucault (2008) overlooks, which results from the connection between neoliberal policies of reduced welfare provision, and neoliberal approaches to crime. The neoliberal conception of the subject as a rational economic agent means that criminality and welfare provision are interconnected, because people choose between the available options. If potential criminals behave in a rational, self-interested manner using cost-benefit analyses, then changes in welfare provision, employment levels, wage rates, tax regimes et cetera will alter crime rates. If the motive for committing a crime is financial19 then, ceteris parabus, the poor will find lucrative crimes more appealing as a result of reduced welfare provision and the cessation of redistribution to promote income equality. If welfare cuts are accompanied by rising unemployment and/or a less progressive system of taxation for employment 20 income then crime will be even more appealing in comparison to the alternatives, and logically the crime rate should rise. I do not believe that the neoliberal account of ‘economic man’ adequately captures the complexities of individual or social behaviour, and as such I am not arguing that this posited relationship between crime and welfare provision is necessarily accurate. However, the interconnection of crime with welfare provision, employment rates and other aspects of economic policy is a logical consequence of the neoliberal thinking that Foucault (2008) describes. This link between neoliberal approaches to crime and welfare means that neoliberal policies can be expected to have two contradictory effects on the crime rate. On the one hand, neoliberal criminal justice policies will take measures to increase the costs of crime that ceteris parabus would reduce crime rates. On the other hand, neoliberal socioeconomic policies such as welfare 19  The motivation for many, though certainly not all, forms of crime is at least partially financial, including thefts, prostitution, drug dealing and thereby much of the violence and gang activity associated with drug sales. 20 A progressive  tax system is one that taxes those earning high incomes at a greater rate than those earning low incomes; in contrast, a regressive tax system is one where the tax rates are higher for those on low incomes than for high-earners.  112  cuts are likely to make lucrative crimes more appealing — or even necessary for survival — for those individuals, families and communities affected. The introduction of neoliberal social and economic policies therefore breaches the ceteris parabus assumptions of neoliberal crime policy, i.e. the changes in welfare, wages, taxes et cetera mean that all other things are not being held constant. If both sets of neoliberal policies are introduced then it is plausible – even likely – that much of the government expenditure 'saved' through welfare cuts and privatization schemes will be spent on law enforcement. Moreover, the logic of ‘economic man’ and rational criminality suggests that some of this increased spending to deter crime will merely be offsetting the increased relative appeal of crime caused by the welfare cuts. It is even possible that the increased spending on law enforcement might exceed the money 'saved' through welfare cuts, in which case neoliberal policies on crime and welfare could lead to a rise in the overall level of government spending. Although Foucault (2008) provides a lengthy explanation of the neoliberal approach to punishment, he completely overlooks this problem and its normative implications. The neoliberal philosophy described by Foucault (2008) thus proposes to shift government spending on poor communities away from welfare support and towards policing, prosecution and incarceration, but it is not clear that either state spending or crime will fall as a result. Foucault's (2008) failure to mention the connections between poverty, employment, welfare provision and crime may reflect neoliberal thinkers' reluctance to acknowledge the logical implications of their own theories, or it might reflect some other motivation on Foucault’s part. However, the omission of this obvious and crucial point suggests that Foucault is both insufficiently rigorous in analysing the internal logic of neoliberalism, and that he fails to consider the likely effects of neoliberal policies. This is a significant omission, because this 113  punitive approach towards governing the poor challenges the neoliberal rationality of government, which claims to be based on a principle of self-restriction, individual autonomy, cost-efficiency et cetera. These tensions within neoliberal theory suggest that neoliberal approaches to criminal justice policy are deeply flawed — and perhaps doomed to failure — whether or not the broader neoliberal social and economic policies succeed in constituting people as rational, self-interested and entrepreneurial. If neoliberal social policies do not succeed then people will not behave like rational, self-interested agents, and the attempt to deter crime through harsher penalties will therefore fail. However, if neoliberal social and economic policies do succeed in shaping people's behaviour along the lines of homo oeconomicus, then the welfare cuts and the cessation of redistribution will make crime relatively more appealing to these rational, self-interested agents. Either way, it seems likely that the neoliberal approach to social and criminal justice policy that Foucault (2008) describes is essentially self-defeating. Foucault's (2008) inattention to the links between property ownership, inequality and criminal justice in the 1978-9 Collège de France lectures contrasts to his earlier (1995) observations about the interconnection of these issues. In Discipline and Punish, Foucault (1995) notes that criminal punishment relates to socioeconomic inequality, and identifies the bourgeoisie's concern for protection of property as a key part of the historical context around changes in punishment during the eighteenth century. Foucault (1995) also observes economic aspects within 1970s penality, stating that "Delinquency, controlled illegality is an agent for the illegality of dominant groups…the existence of a legal prohibition creates around it a field of illegal practices, which one manages to supervise, while extracting from it an illicit profit" (281-2). Given Foucault's (1995) observations about punishment, property, dominant 114  groups and profit, it is odd that he ignores the same issues in the 1978-9 lectures, which were delivered only four years after the publication of Discipline and Punish. The difficulties with Foucault’s account of neoliberalism are particularly clear if one considers empirical research about the deterrence-based criminal justice policies introduced over the last thirty years. If, as Foucault (2008) suggests, neoliberal criminal justice policy were based on evidence about cost-effectiveness and the varying elasticities of crime then it would target crimes where deterrence is effective, whilst tolerating relatively inelastic offences like drug use. In fact, American criminal justice policy over the past thirty years has done precisely the opposite: massively expanding policing and imprisonment for drugs, despite the high cost of these policies and their inefficacy for reducing crime (Western, 2006; Mauer, 2006). Moreover the combination of high incarceration rates and reduced welfare provision has exacerbated poverty and familial breakdown amongst disadvantaged urban communities, thus creating the conditions that encourage criminality whilst removing other sources of income (Western, 2006). The deterrence-based approach to criminal justice in recent decades thus fails the short-term economic or utility calculus, whilst contributing to social conditions that tend to increase crime in the longer term. The inefficacy of deterrence for drug crime that Foucault (2008) notes has been confirmed by a Federal Judicial Center (1994) study on the effects of lengthy mandatory minimum terms for drug offences. The Federal Judicial Center found that neither crime nor drug availability fell as a result of the introduction of mandatory minimum terms for drug offences, and explained this according to the logic of rational agents: To be deterred, offenders must stop to weigh the costs and benefits, be aware of the penalties, find those penalties intolerable, and have other attractive options. Even if some potential offenders are deterred, drug trafficking will not be curtailed is there are other 115  persons willing to take the place of convicted offenders. This appears to be true in the profitable drug business. (1994, 11) Research also suggests that drug treatment programs are more effective and far more costeffective than increased policing and incarceration (Federal Judicial Center, 1994; Caulkins et al, 1997). Moreover, medically supervised provision of heroin to addicts may be even more effective than regular drug treatment at reducing illegal drug dependence and associated crime (Ribeaud, 2004). Given Foucault’s (2008) claim that neoliberal policymaking is based on costeffectiveness, why did the Reagan, Bush, and Clinton administrations in the US, and the Major, Blair, and Brown governments in the UK all use ineffective deterrence-based policies towards drug crime? One explanation for the divergence between Foucault's (2008) account of neoliberal approaches to criminal justice and recent policies towards drug crime is that these policies do not reflect neoliberal penal philosophy. Although Foucault explains that neoliberal penal philosophy favours a deterrence-based approach with severe and certain punishment, which is consistent with the punitive trend discussed in the previous chapter, there may have been additional influences on penal policy. Garland (2001) argues that the introduction of longer sentences and mandatory minimum terms is the result of the combination of neoliberal and neoconservative approaches, which overlapped to produce an ethos of "economic freedom and social control" (100). While neoliberalism suggests the need for rationalization, costeffectiveness, a small state and a view of offenders as rational actors, neoconservatives favour greater social control via increased policing, imprisonment and the stigmatization of offenders as morally evil (Garland, 2001). Although there are tensions between neoliberalism and neoconservatism over the size and role of the state, Garland argues that in practice they combined politically to produce a consensus in favour of greater incarceration of the poor. 116  The implication of Garland's (2001) argument is that neoliberal concerns about the high cost and low cost-effectiveness of a punitive approach to drug crime may have been set aside as a result of neoconservative enthusiasm for greater social control. If Garland is right, then one would expect the tensions between neoliberal and neoconservative approaches to criminal justice to become more visible over time as the continuing rise of prison populations makes punitiveness increasingly expensive. One might also expect that political tensions over the high levels of expenditure on law enforcement would become more pronounced if high prison populations coincided with a serious economic recession or budgetary crisis. For example, a 2008 article in The Economist argues that the budgetary crises in New York and California are partially due to the very high levels of criminal justice expenditure, and are therefore the product of poor judgement by politicians 21. There is some evidence that American policymakers and criminal justice officials have indeed adopted a less punitive approach in response to the recent recession, because the total number of prisoners held in state prisons fell during 2009 — the first reduction in this population for 27 years (Bureau of Justice Statistics, 2010b, 2). The reductions in state prison populations were particularly pronounced in Michigan, California and New York22 , each of which had over 1500 fewer people imprisoned during 2009 than 2008 (Bureau of Justice Statistics, 2010b, 2). Moreover, the total (i.e. federal and state) imprisonment rate in the US fell slightly in both 2008 and 2009, and now stands at 502 prisoners per 100,000 population as  21  'Local Zeroes', Nov 13th, 2008 issue of The Economist. Retrieved from http://www.economist.com/world/ unitedstates/displaystory.cfm?story_id=12608223. The total number of prisoners held in US state prisons fell by 0.2% or 2857 prisoners from December 2008 to December 2009, but an increase in the number of federal prisoners meant that the total population in federal and state prisons in the US increased by 0.2% — the smallest rise in the past decade (Bureau of Justice Statistics, 2010b, 1-2). 22  117  compared with 506 prisoners per 100,000 population in 200723 (Bureau of Justice Statistics, 2010b, 1). According to the Pew Center (2010), the desire to reduce spending has been one of the major driving forces for this change in policies. The temporary alliance between neoliberals and neoconservatives over greater penal severity that Garland (2001) identifies may therefore be breaking up in the face of the inhospitable economic climate following the 2008 recession, and the resultant state and federal budget deficits. However, it is important to recognize that the tensions between punitiveness and costefficiency arise within neoliberal theory, and not only between neoliberal and neoconservative approaches to criminal justice. Although deterrence-based approaches to drug crime are not effective or cost-effective, they do have the benefit of being consistent with the neoliberal conception of the subject, and with neoliberal ideas about the role of the state. Neoliberal accounts of the state argue that it should provide only minimal welfare support for the poor, but regard law enforcement as part of the necessary and legitimate function of the state. Law enforcement is one of the few functions that virtually all social and political theorists believe the state should pursue, even libertarians such as Robert Nozick who argued for a minimal or "night watchman state" (Nozick, 1974, 24). Conversely, the neoliberal preference for a small state and privatization of risk is difficult to reconcile with the universal provision of drug treatment and mental health care that would provide a more cost-effective way of dealing with drug addiction and associated crime. In short, the neoliberal account of the appropriate role and scope of the state privileges law enforcement over healthcare and welfare provision as a means for responding to problems of drug crime.  In 2008 and 2009 the total number of prisoners rose, but the incarceration fell (Bureau of Justice Statistics, 2010b, 1-2). This means that in both those years the American population grew at a faster rate than the prison population. 23  118  Similarly, a deterrence-based approach to drug policy fits with the neoliberal conception of the subject. Understanding why deterrence is not effective for drug crimes would require a more nuanced understanding of subjectivity as embodied, and rooted in a socioeconomic context. In contrast, neoliberal theory regards the rational, autonomous and self-interested subject, which Foucault (2009) terms "homo oeconomicus" (252) as a both a premise and a goal of social policy, which means it privileges policies that centre on this economic conception of the subject. Wendy Brown (2005) captures this prescriptiveness in the neoliberal account of the subject, stating that "neoliberalism involves a normative rather than ontological claim about the pervasiveness of economic rationality and it advocates the institution building, policies and discourse development appropriate to such a claim" (40). If proponents of neoliberalism claim not merely that people behave rationally but that they ought to behave rationally, not that deterrence works but that it should work, then the failure of policies based on the assumption of rational individualism can be used to justify a re-intensification of those efforts. Just as Foucault (1995) observes that the prison's failure to normalize inmates leads to the continual "reactivation of the penitentiary techniques" (268), the failure of deterrence-based approaches to crime may lead to the introduction of increasingly severe penalties. Neoliberalism therefore produces a punitive, deterrence-based approach to crime, including drug crime, both because of the economic logic of deterrence, and because nonpunitive responses rooted in an understanding of the socioeconomic origins of crime are unthinkable within a neoliberal conceptual framework. The policies that follow from neoliberal conceptions of the state and the subject are not necessarily rational according to other principles that neoliberalism purports to uphold, such as cost-effectiveness, but the tensions within neoliberal theory make it necessary to prioritize some principles over others. This contradiction 119  within neoliberalism is not confined to criminal justice policy, because Burchell (1996) identifies similar tensions in neoliberal policies on education, which he describes as: A paradoxical situation where the conduct of government is rationalized and justified in terms of liberal principles of economic government, but where it is quite possible to argue that it is failing completely and causing poor economic performance at high sociopolitical cost (26). By failing to critically analyze the neoliberal theory he describes, Foucault (2008) overlooks the tensions within neoliberal thought, and the consequences of these tensions for neoliberal government and techniques of power. 2. Governmentality, and coercion The second difficulty with Foucault's (2008) account of neoliberalism is that he overlooks the coercive aspects of neoliberalism, while presenting an analysis in terms of governmentality that does not challenge or denaturalize neoliberal techniques of power. In the 1977-8 lectures, Foucault (2007) emphasizes that governmentality works through freedom: "the insertion of freedom within governmentality...as an element that has become indispensable to governmentality itself. Henceforth, a condition of governing well is that freedom, or certain forms of freedom, are really respected" (353). In the following year's lectures, Foucault (2008) shows how neoliberalism uses governmentality through the expansion of market rationality, and the use of incentives to shape the behaviour of subjects by harnessing their freedom. However, while Foucault (2007) argues that governmentality, discipline and sovereign power co-exist and reinforce one another, his analysis of neoliberalism in the 1978-9 lectures does not explore this inter-relationship. Instead, Foucault's (2008) account of neoliberalism focuses on governmentality to the exclusion of other techniques of power, and in the process overlooks the coercive implications of the neoliberal philosophies and policies he describes.  120  The concept of governmentality provides valuable insights into neoliberalism by showing how power works through subjects' freedom, particularly in relation to neoliberal reforms around choice and individual responsibility. However, Foucault's (2008) analysis of neoliberalism using the concept of governmentality largely has the effect of describing neoliberalism in its own terms, through a language of freedom, choice, limited government, and efficiency. As a result, Foucault's (2008) account of neoliberal techniques of power does not challenge or denaturalize neoliberalism, but instead seems to repeat and reinforce the ideological claims made by neoliberal theory. Whereas Foucault's (1995) analysis of disciplinary power makes subtle techniques of power visible and therefore open to resistance, no equivalent unmasking of neoliberal forms of power is apparent in the 1978-9 Collège de France lectures. Similarly, while Foucault (1995) differentiates between the logic of disciplinary power and its actual consequences, in which the failure of disciplinary normalization justifies the reintensification of discipline, he makes no distinction between the logic of neoliberalism and its effects. Foucault (2008) therefore repeats neoliberal claims about freedom, autonomy, and minimal government, but does not pose the question of whether these claims are bourn out, or whether they serve to justify the further intensification of power. As discussed above, criminal justice policy is an example of Foucault's (2008) failure to critically analyze the logic of neoliberalism, and to distinguish between the claims of neoliberal theory versus the consequences of neoliberal government. The use of coercion is clearly implied in Foucault's (2008) account of neoliberal crime policies, which involve the imposition of severe and certain punishment to deter people from committing offences. This neoliberal penal philosophy translates into policies of more intensive policing and longer, mandatory prison sentences, but Foucault (2008) does not analyze the normative implications of these policies. 121  Foucault's (2008) account of neoliberalism therefore emphasizes freedom — both of the market and of the subject — whilst skirting over the fact that coercion has been intensified through more severe criminal punishment. Foucault's (2008) failure to consider the coercive aspects of neoliberalism is particularly problematic given the interconnection of crime and welfare policies explained above, whereby the logical consequence of welfare cuts is to undermine the deterrent effect of neoliberal crime policies. The combination of welfare cuts and more punitive criminal justice policy thus means that state spending is diverted from welfare provision to policing and prisons, without necessarily reducing crime. The overall effect of neoliberal crime and welfare policies is therefore a more coercive approach towards poor and marginalized people, who receive less welfare support but are subject to more intensive policing and harsher penalties for crime. This coercive aspect of neoliberalism has been bourn out the punitive trend in criminal justice policy in both the UK and US, as I explored in Chapter Three. If as Foucault (2007, 2008) suggests neoliberalism has involved a shift towards governmentality, then it has been accompanied by the increased use of coercion towards segments of the population. This coercive element in neoliberalism has been noted by several scholars working within a Foucaultian conceptual framework. Mitchell Dean (1999) argues that neoliberal governmentality should be seen as only one of several different techniques of power used in contemporary societies, including the use of disciplinary power, and of coercive sovereign power towards welfare dependants and the criminalized. Similarly, Thomas Dumm (2000) argues that high incarceration rates are evidence of a new understanding of freedom as security that "allows for there to be a strategic use of the rhetoric of freedom to intensify control over  122  populations at large" (132). The use of coercion in neoliberal government is also noted by Nikolas Rose (1999), who states: The undoubted persistence and salience of coercive tactics — in the policing of the inner cities and the urban poor, in the surveillance and control of political dissidence, and of course in the various international adventures of advanced liberal nations — must also, today, be justified as the price necessary for the maintenance of freedom. (10) Although both Dean (1999) and Dumm (2000) seek to challenge and denaturalize neoliberal techniques of power, including the coercion they identify, Rose's (1999) analysis does not seem to share this critical impetus. In a later work, Dean (2007) comments that "Rose seems to endorse the relatively benign character of liberalism...There is a narrowing of the relationship between the normativity of liberal conception of government and the analysis offered by governmentality studies." (86) The lack of critique of neoliberalism in much of the governmentality literature has also been noted by Jacques Donzelot (2008) who argues that analyses focused on the concept of governmentality have tended to produce descriptions of social technologies, but not political engagement or critique. Although Donzelot (2008) does not implicate Foucault in this criticism, he suggests that the subsequent governmentality literature has been inadequate, asking, "does this political ambivalence in the notion of governmentality not condemn it to serving an ideological function, determined by political circumstance?" (56). Given that Foucault's (2007, 2008) discussion of governmentality does not seem to generate critical insights that denaturalize or disrupt neoliberalism, and that the same problem afflicts much of the later governmentality literature, it seems possible that the problem lies with the concept of governmentality. Whereas Foucault's concept of genealogy is inherently critical and involves "the insurrection of knowledges" (2000b, 133), as discussed in Chapter One, the concept of governmentality is not defined in opposition to prevailing forms of power and knowledge. As a 123  result, analyses based on governmentality may fail to provide critical purchase or insights about prevailing techniques of power. For example, analyses based on governmentality tend not to capture issues of inequality or marginalization, and more specifically do not tend to address the issues of sex, sexuality and race that Foucault raises in his discussions of biopower (1990, 2003). This is not to say that analyses using the concept of governmentality can never provide critical insights, but it seems possible — indeed likely — that the concept of governmentality is not designed to be critical, and is therefore far less conducive to critique than genealogy is. In short, governmentality, unlike genealogy, is not "made for cutting" (Foucault, 1980a, 154). As I will discuss in the next section, there are also questions over whether the concept of governmentality can illuminate the economic aspects of neoliberal techniques of power. 3. Economic aspects of power Not only does Foucault (2008) overlook the tensions between neoliberal policies on welfare and crime, and the coercive aspects of neoliberal crime policies, he also fails to provide a critical analysis of economic power under neoliberalism. Neoliberal policies such as welfare cuts, the privatization of risk, and the cessation of redistribution towards the poor can be expected to increase socioeconomic inequality and poverty. Moreover, such policies may also alter the demographic composition of the poor, for instance contributing to the overrepresentation of children, the elderly, the disabled, or female-headed households amongst the poor. Foucault's (2008) silence about the connections between neoliberalism, poverty and socioeconomic inequality, and his seemingly uncritical description of neoliberal economic theory around efficiency, cost-effectiveness et cetera seem to be in tension with the critical approach and interest in marginalized groups that is characteristic of his genealogical analyses.  124  Foucault's inattention to economic forms of power is noted by Tellman (2009) who suggests that Foucault's reluctance to engage with issues of economy is a product of his longstanding struggle with Marxism, but argues that it is a betrayal of the critical impulses usually found within Foucault’s thought: A more typical Foucauldian approach would commence to undo the invisibility of the economy and the market as an invisible "telescope" and "information-machine". This would mean rendering visible the market’s own "machine of seeing", rather than seeing like the unseen market itself (22). Tellman argues that Foucault's inattention to economic power extends to his concept of governmentality, which is conceptualized in a way that challenges the traditional concept of the state and makes visible the broader operation of governmental reason, but renders the economy invisible. As a result, Tellman (2009) believes that Foucault re-articulates the liberal view of the economy that serves to justify the limited state because "within this discursive construction, the market becomes the sole site legitimately producing this knowledge of the whole. The invisibility of the market and the construction of its epistemological authority go hand in hand" (22). While the concept of governmentality clearly has limitations, I am unconvinced by Tellman's (2009) claim that it renders economic forms of power invisible. One benefit of governmentality is that it is well suited to analysing the techniques of power involved in the privatization of functions and agencies that were previously the domain of the public sector. The perspective of governmentality suggests that such privatization is not a reduction in government, but instead evidences governance at greater remove by introducing markets and market thinking in order to promote an ideology of autonomy and choice. Analyses in terms of governmentality also show that autonomous and marketized behaviour is created amongst subjects and organizations, rather than being a reversion to a ‘natural’ state of competitive market activity 125  (Burchell, 1996). This creation of marketized behaviour is clearly visible in prison privatization, where the free market arguments for efficiency through free competition are ill suited because two or three companies dominate the market and the ‘consumers’ are not freely purchasing a product. Instead, prisoners are incarcerated against their will and as a result of the actions of multiple state agencies, which creates a problematic three-way relationship between prison contractors, prisoners and the state: At the very least there may well be a conflict of interest between the needs and desires of the parties to this triadic relationship, which underlies the importance of not accepting at face value the free-market rhetoric that is usually used in support of prison privatization. (Cadavino & Dignan, 2006, 67). These incentives lead to a situation whereby the state seeks to promote competition and cut its own costs, and the contractors seek market domination and profit maximization, but both may lose focus on the quality of the service and its value-for-money. Although  the above example demonstrates that it is possible to use the concept of  governmentality to generate insights about economic power under neoliberalism, the economic aspects of neoliberalism have been neglected by many of the scholars using a Foucaultian conceptual framework (Gordon, 1991; Rose, 1999; Dean, 1999; Dumm, 2000). Instead, analyses of the economic impact of neoliberalism have tended to take place amongst economists, and amongst political and social theorists who draw on a Marxist conceptual framework, such as the work of Angela Davis (2003) and David Harvey (2005). One of the major findings of scholars interested in the economic implications of neoliberal policies has been a significant rise in income inequality in the UK and US over the past thirty years, which is attributable in large part to neoliberal policies on welfare, taxation, privatization et cetera. The GINI co-efficient is a simple and widely used measure of income inequality and applying this measure to US census figures suggests that from 1978 to 2005 there was a 126  significant and continuous increase in income inequality amongst American households, from a GINI co-efficient of 0.402 to 0.46924. The richest households have benefited disproportionately from this shift in income distribution, since the top 10% of households received one third of the income during the 1970s, but nearly one half of the income by 2006 (Piketty & Saez, 2006). Piketty and Saez (2006) show that much of the increase in income share amongst the top decile is accounted for by changes in the top 1% of households, whose income in the US ranged from "about 8 percent during the 1960s and 1970s, and back to almost 17 percent by 2000" (201). Breaking down these figures even further shows that the income share of the top 0.1% of households in the US has tripled since the 1970s, and they received over 7% of the income share by 199825 (Piketty & Saez, 2006, 202-3). The UK has experienced a similar trend towards greater income inequality and large gains for the richest, with a rise in the GINI co-efficient from 0.25 in 1979 to 0.34 by the early 1990s (Brewer, Sibieta & Wren-Lewis, 2008, 2-3). Since the early 1990s the UK's GINI coefficient has fluctuated between 0.33 and 0.35, and there has been a small but statistically significant rise in inequality despite the poverty reduction programs of New Labour governments since 1997 (Brewer, Sibieta & Wren-Lewis, 2008)26. The increase in inequality in the UK is largely explained by the sharp increase in the earnings of the top 10% of individuals during the 1980s and in the years since27 . Brewer, Sibieta and Wren-Lewis (2008) therefore  24  Retrieved from http://www.census.gov/hhes/www/income/histinc/p60no231_tablea3.pdf .  25  The income share of the top 0.1% of households in the US rose from around 2% in 1978 to over 7% in 1998, while the income share of the top 0.1% of households in the UK rose from around 1% in 1978 to over 3% in 1998. (Figure 3, Piketty & Saez, 2006, 203). 26  The increase from 0.33 in 1997 to 0.35 in 2008 reflects a small, but apparently significant, increase in income inequality under New Labour (Brewer, Sibieta & Wren-Lewis, 2008, 3) 27  From 1996-7 to 2004-5 the average annual real income growth for the top 10% of individuals was over 2% each year, while the top 1% of individuals gained on average over 3% a year, and the top 0.1% gained on average 4% a year (Brewer, Sibieta & Wren-Lewis, 2008, 20).  127  conclude that the historically high income inequality in the UK has occurred because "those on high incomes (about the top 10%) have seen still faster income growth and the very poor (about the poorest 15%) have seen weak income growth" (30). Since the UK figures are for individuals rather than households, it is possible to see that in 2004-5 men comprised over 70% of the top decile of earners and over 90% of the top 0.1% of earners (Brewer, Sibieta & Wren-Lewis, 2008, 14). Given that only slightly over 50% of taxpayers were male during the same period, this suggests that the disproportionate gains of top earners may have contributed to gender discrepancies in earnings. A comparative perspective on income inequality shows that there has been "a sudden rise of top wages in Anglo-Saxon countries since the 1970s" (Atkinson and Piketty, 2007, 11) whilst top income shares in mainland Europe have remained fairly stable. These changes in the UK and US can be clearly traced to neoliberal policies such as reduced tax rates for high earners in the UK and US during the 1980s (Atkinson, 2007; Piketty & Saez, 2007), and in America the reduction in tax rates for dividend income in 2003 (Piketty & Saez, 2007). Although income inequality and wealth concentration have risen in both the UK and US, the changes over the past three decades are especially pronounced in America. Piketty and Saez (2007) calculate that if one excludes all transfers such as pensions and welfare then the incomes of the bottom 99% of US households stagnated from 1973 to 2003, whist the real incomes of the top 1% more than doubled (174). Given the clear links that analysts draw between growing income inequality, wealth concentration and the neoliberal economic policies pursued in the UK and US, Foucault’s failure to consider socioeconomic inequality means that he neglected an important aspect of neoliberalism.  128  An analysis of neoliberalism that focuses on the economic aspects of power is provided by David Harvey (2005), who proceeds from a broadly Marxist perspective, or what a recent reviewer termed "Marxian metatheory" (Johnston, 2008, 946). However, it is necessary to provide two important caveats before comparing Harvey's (2005) analysis of neoliberalism to Foucault's (2008) account: firstly, they are writing in very different contexts, and secondly, they are analysing slightly different things. In the 1978-9 lectures Foucault (2008) provides an account of a political and economic theory that was rising in significance in France (and beyond), but was not yet dominant. In contrast, Harvey’s (2005) account of neoliberalism was written during a period in which neoliberalism was one of the leading ideologies and forms of government in the US and internationally, if not the hegemonic ideology. This difference in context is important both for its implications for critique based on disrupting dominant power relations, and because the two theorists have different examples and experiences of neoliberalism to draw on. Whereas Harvey analyzes the ‘history’ thusfar of neoliberal theory and practice, Foucault's discussion is confined to neoliberal theory because most of the practice of neoliberal government post-dated his lectures. When compared to Foucault’s bold effort at conceptualizing an emerging theory of government, Harvey has the benefit of hindsight and evidence about the actual historical impact of neoliberal policies. Harvey's overall argument is that neoliberal ideology has used claims about freedom to construct popular support for policies that have reconfigured class relations to increase inequality and benefit the rich, whilst sacrificing broader ideals of freedom. Significantly, Harvey (2005) both identifies tensions within neoliberal theory, and argues that in practice neoliberal theory has tended to be disregarded when it contradicts the best interests of capital: We can, therefore, interpret neoliberalization as either a utopian project to revitalize a theoretical design for the reorganization of international capitalism or as a political 129  project to re-establish the conditions for capital accumulation and to restore the power of economic elites. In what follows I shall argue that the second of these objectives has in practice dominated… The theoretical utopianism of neoliberal argument has, I conclude, primarily worked as a system of justification and legitimation for whatever needed to be done to achieve this goal. (2005, 19) Whereas Foucault (2008) seems to presume that neoliberal theory is internally consistent, the strength of Harvey's analysis is his reluctance to oversimplify what neoliberalism is. Instead, Harvey shows that the task of defining neoliberalism is complicated by deep-seated disagreements amongst economists about subjects such as the consequences of unregulated markets and of state intervention to restrict the market 28. Harvey (2005) identifies tensions within neoliberal thought on subjects such as monopoly and market failure, and argues that neoliberal rhetoric about efficiency and 'free' markets is misleading to the point of being "either innocently utopian or a deliberate obfuscation" (68). These theoretical disagreements between economists are the reason for Harvey's uncertainty about whether neoliberal economic arguments are misleading by accident or design: neoliberal thinkers may genuinely believe punitive sentencing will deter criminals and reduce crime, or more sinister motives may be at work. However, Harvey is less interested in the intentions of neoliberal policymakers than in determining what the effects of neoliberal governance have been, just as Foucault (1995) distinguishes between the intentions and results of the disciplinary prison. Regardless of what the intentions of neoliberal theorists and policymakers were, Harvey argues that the overall consequence of their policies has been a redistribution of wealth and power towards the rich, and a weakening of democracy and noneconomic forms of freedom. 28  For key examples of this disagreement one can contrast Adam Smith's arguments about the invisible hand and need for the state to refrain from market intervention, with Karl Marx's account of the development of monopolies over time and the resultant poverty and economic crises. The disagreement between neoclassical (or in other words neoliberal) and Keynesian economists in the present day continues to reflect some of the differences between Smith's and Marx's accounts of the relationship between the state and the economy.  130  Despite the different contexts in which Foucault (2008) and Harvey (2005) were writing and their methodological differences, I believe that it is valuable to draw on aspects of both analyses. While Foucault overlooks the tensions within neoliberal theory and the economic aspects of power, Harvey's analysis focuses on these issues. Moreover, whereas Foucault’s account seems to biased in favour of neoliberal ideas, Harvey is strongly critical of neoliberalism as a political and economic project. However, the flaws I have identified in Foucault's (2007, 2008) analysis of neoliberalism do not invalidate his account of how power works upon and through the subject, including through the combination of sovereign power, discipline and governmentality. In order to understand of contemporary forms of power in relation to criminal justice, I believe it is necessary to draw on aspects of both Foucault's and Harvey's accounts of neoliberalism and its associated techniques of power. In the chapters that follow I will be drawing on Foucault's work to analyze how the UK and US prison systems constitute people as particular types of subjects, and drawing on Harvey's insights about the heightening of economic inequality as a result of neoliberal policies. Both Foucault's (2008) and Harvey's (2005) analyses of neoliberalism help to show the causes and significance of the trends discussed in Chapter Three. The punitive trend in penal policy, which has produced large rises in the English and American prison populations and incarceration rates, should therefore be understood in the context of the neoliberal conception of the subject as rational, self-interested, economic man. However, changes in penal policy also need to be understood in the broader context of neoliberal social and economic policies, including welfare cuts, and a tax system more favourable to the rich. In this context, it becomes clear not only that the prison has become less disciplinary over the last thirty years, as I argued in the previous chapter, but that government strategies and spending on poor and marginalized 131  populations have shifted from welfare provision towards increased policing and imprisonment. The neoliberal philosophy of choice, autonomy, and minimal government described by Foucault (2008) has therefore produced rising socioeconomic inequality and incarceration rates. Similarly, this broader analysis of neoliberalism provides useful context for understanding the trend of prison privatization, and its relationship to rising prison populations. I argued in Chapter Three that prison privatization did not begin the punitive trend, but that it enabled this trend by reducing the infrastructure costs associated with rising incarceration rates. Foucault's (2008) account of neoliberalism and neoliberal governmentality helps to show that prison privatization is part of a broader political strategy of governing indirectly and through the market, in the hope of achieving greater efficiency through competition. Moreover, Harvey's (2005) analysis exposes the way that such privatization initiatives redistribute wealth towards the rich by boosting shareholder revenue and transferring public assets into private ownership. The introduction of contractor-run prisons thus enabled private prison providers to profit from prisoners, but is only one part of a wider political and economic strategy that has contributed to increased socioeconomic inequality, and in many cases to reduced democratic accountability over state services and institutions. Neoliberalism has had a significant impact on the English and Welsh, and American prison systems through the punitive trend and prison privatization, but the sheer scale of incarceration in recent years also means that these prison systems have major implications for neoliberal governance. Although Harvey’s (2005) analysis of neoliberalism corrects some of the failings of Foucault's (2008) account, there are issues that both theorists neglect. In particular, the issues of gender, race and ethnicity are ignored by Harvey and Foucault (both in the 1978-9 lectures and in much of his other work including Discipline and Punish) and I will now turn to these 132  subjects. In the next chapter I explore the theoretical literature around sex, gender, race, and ethnicity, drawing on Foucault’s analysis of these subjects, and more recent work in feminist theory, queer theory, and critical race theory. In Chapter Six I apply this conceptual framework around race and ethnicity to the English and Welsh, and American criminal justice systems, and argue that practices of policing and imprisonment play a major role in constructing racialized identities and in contributing to racialized inequality. In Chapter Seven I analyze issues of sex and gender in the prison systems of the US and England and Wales, arguing that these prison systems contribute to the construction of normative sex and gender identities, to the marginalization of gender-nonconforming people, and to the development of aggressive forms of masculinity.  133  Chapter Five: Theory of Sex, Gender and Race Overview Having argued in Chapter Two that the literature on the history of criminal punishment in England and America shows the importance of race and gender in shaping penal theory and practices, the next three chapters of my dissertation will focus on contemporary power dynamics around race and gender. In this chapter I develop a theoretical account of how sex, gender, and race are constituted, and in Chapters Six and Seven I will use this theory to explore the prison systems of America, and England and Wales. Several themes from the earlier chapters of the dissertation will be continued and further developed in the course of my analysis of race and gender. In Chapter Three I argued that the trends of increased penal severity and prison privatization suggest that disciplinary normalization plays a less significant role in the current English and Welsh, and American prison systems than was the case during the 1970s, but I suggested that discipline contributes to the construction of sexed and gendered identities. In this chapter I discuss how disciplinary power links to gender identities, and in Chapter Seven I will show how prison regimes constitute masculinity and femininity through a combination of disciplinary power, sovereign power, and biopower. Further, in Chapter Four I argued that neoliberalism has lead to an increase in socioeconomic inequality, and to a shift in the governance of the poor, in which there is reduced welfare provision and increased use of coercion through policing and imprisonment. In this chapter I explore the constitution of race through biopower and briefly address the interconnection between race and class. Then in Chapter Six I analyze the disproportionately high incarceration rates for racialized and ethnic minorities in the United States, and England and Wales, arguing that the racialized biases in these criminal justice systems plays a major role 134  in constructing racialized identities and inequalities. I conclude that changes to criminal justice policy as a result of neoliberalism have had an impact not only on the governance of crime and poverty, but on the governance of race. The purpose of this chapter is to briefly outline, analyze, and synthesize the theoretical literature about race and gender in order to develop a theoretical framework for analysing the racialized and gendered power dynamics in the prison systems of the United States, and England and Wales. In the first part of the chapter I draw on feminist theory, queer theory and Foucault's account of the construction of sex and sexuality, to develop an account of the constitution of sex, gender and sexuality. I also use the concept of somatechnics, which highlights the ways in which subjects use technology to shape their body in accordance with their self-image, including by using practices such as diet, exercise, and surgery to alter their body in accordance with norms of masculinity or femininity. In the second part of the chapter I draw on Foucault's account of the construction of race through biopower, and on contemporary theories about race including Omi and Winant's (1994) account of racialization. I outline the theoretical debates about the use of racialized language and categories, and argue that in America there is evidence of a double standard whereby explicit racialized criteria are used in policy areas that increase racialized inequality (e.g. policing), but not in policy areas that reduce racialized inequality (e.g. employment and education). After analysing the theoretical literature about both gender and race, I combine these accounts to argue that neither sex, nor race, are natural or pre-existing ways of categorizing bodies. Given that there is no unmediated or wholly objective access to the body, I argue that ways of categorizing human bodies — whether they consist of genetic variation that is interpreted in racialized terms, or the interpretation of sex in terms of two, three, or five 135  categories — are always shaped by discourse and power. I therefore criticize attempts to compare the bodily 'realness' or 'constructedness' of sex and race, arguing that this establishes a false dichotomy that serves to erase and marginalize transgender and intersex people of colour. Instead, I call attention to how multiple relations of power interact, including in relation to bodily difference. I use Foucault's (1990; 2003) analysis of sex, and of race, as starting points for my analysis and draw on his accounts of biopower and state racism. However, my analysis suggests that Foucault's theoretical frameworks around both sex and race are inadequate and I therefore amend, supplement and reject parts of Foucault's analysis. I critique Foucault for his inattention in Discipline and Punish to how penality is shaped by sex, gender and race, which leads him to largely overlook the gendered — and to a lesser extent racialized — nature of disciplinary power. I also criticize Foucault’s interpretation of Herculine Barbin as a narrative about the advantages of non-identity, and argue that he is too quick to equate resistance to hegemonic forms of sex and sexuality with the refusal of labels. Although Foucault’s account of race provides insights about biopower and state racism, his focus on violence leads him to neglect the way that capitalism, exploitation and class have historically often been — and to a lesser extent still are — entwined with race. I then suggest that Foucault (2003) uses the term 'state racism' in two different senses, and argue that the broader sense (which encompasses civic exclusion and an increased risk of death) is more useful than the narrow sense that focuses on violence (Foucault, 2003). Finally, I argue that an adequate theoretical understanding of race and gender requires one to take an intersectional approach by exploring the overlap and inter-relation of race, class, sex, gender and sexuality.  136  Part of the challenge in analysing sex, gender and race lies in the choice of terminology, which inevitably has political implications. I discuss the question of what terminology is most appropriate for analysing race in the second part of this chapter, and argue that it is preferable to refer to racialized instead of racial identities. However, there is no satisfactory answer to the question of what terminology should be used to discuss sex and gender. In this chapter I will make a distinction between sex and gender insofar as sex conventionally refers to bodily difference in organs, chromosomes, hormones etc. whereas gender conventionally refers to differences in social behaviour and roles. I am not suggesting that there is a clear distinction between sex and gender — on the contrary, I argue that our interpretation of bodies as falling into binary sexes is shaped by beliefs about binary gender categories. However, I will use the terms sex and gender to explore Foucault's work around sex and sexuality, and the responses to Foucault from gender theorists, because it is difficult to understand and analyze this literature without beginning from the conventional distinction between these concepts. At points in this chapter and the following two chapters when it is not necessary to make a distinction between the bodily organs, hormones etc. and social norms I will refer to 'sex/gender' in order to indicate that these concepts are not distinct.  Sex and Gender Male pronouns are used throughout the English translation of Discipline and Punish29 , which suggests that disciplinary subjects and those responsible for conducting surveillance and The question of gendered language is more complex in the original French Surveiller et Punir, because several of the key terms used by Foucault are feminine, including 'la discipline' (discipline), 'la prison' (prison) and 'la délinquance' (crime). Analysing the gendered implications of the French language in general, or specifically in Surveiller et Punir is beyond the scope of this dissertation. However, the terms homme and hommes (man and men) are used far more frequently than femme and femmes (woman and/or wife, women and/or wives), which suggests that Foucault's subjects of analysis are presumptively male. A search of Surveiller et Punir (1975) using Googlebooks reveals that the word 'homme' is used on 41 pages in the book and 'hommes' on 61 pages, whereas 'femme' is used on 6 pages, and 'femmes' on only 5 pages. 29  137  assessment are presumptively male. Nonetheless, Foucault (1995) makes no comment about the sex or gender of prisoners, and does not consider how penality or discipline relate to masculinity. Later, Foucault engages at length with issues of sex and sexuality in the first volume of The History of Sexuality, and his commentary to the memoirs of Herculine Barbin, published under the self-explanatory title Herculine Barbin: Being the Recently Discovered Memoirs of a Nineteenth-Century French Hermaphrodite. Although Foucault's later work analyzed the constitution of sex and sexuality, and he even argued that disciplinary power contributed to the construction of sexuality (Foucault, 1990), he did not explore how the criminal justice system and the prison discipline or construct certain kinds of sexuality and sexed bodies. Foucault's (1995) inattention to sex and gender in Discipline and Punish is problematic for two reasons: firstly, because nineteenth century penal philosophies and practices were extremely gendered and indicated the gendered nature of power in these institutions, as I discussed in Chapter Two. According to many historians, gender was extremely important in understanding criminal punishment, because ideas about criminality, deviant behaviour and appropriate punishment were shaped by normative gender ideals (Freedman, 1981; Rafter, 1990; Zedner, 1991; Meranze, 1996; Colvin, 1997). Secondly, and more importantly given the genealogical nature of Foucault's analysis, by ignoring gender in Discipline and Punish he overlooks the possibility that disciplinary power might be deeply, even irreducibly, gendered. Foucault's (1995) inattention to the sex or gender of prisoners and to how discipline links to gender might therefore lead him to mis-identify the techniques and relations of power. A number of feminist theorists have indeed criticized Foucault for ignoring women and for neglecting  138  gender in his work (Hartsock, 1996; McNay, 1992), although others defend Foucault's analysis and draw heavily on his later accounts of the constitution of sex and sexuality (McLaren, 2002). Given that Foucault encountered Herculine Barbin’s memoirs while researching The History of Sexuality it is unsurprising that there is considerable overlap between these texts. Herculine Barbin’s narrative begins with her/his30 education in a girls’ boarding school, and later experiences of teaching in a similar school where s/he developed a romantic relationship with Sara, the daughter of the school’s owner (Foucault, 1980b). After people grew suspicious of Herculine’s relationship with Sara, Herculine confessed the relationship and her/his concerns about her/his body to a prominent priest, leading to a medical examination. The doctor who performed this examination concluded that Herculine was more male than female, advised her/ him to pursue a legal change of sex, and informed Herculine’s employer who fired her/him. After this revelation of her/his "true sex" (Foucault, 1980b, 122), Herculine was prevented from contacting Sara, adopted a male identity and moved to Paris, where s/he was unhappy, unable to find work and eventually committed suicide. Throughout the memoirs Herculine avoids reference to her/himself in sexed or gendered language and Herculine’s self-identified sex and/or gender identities are therefore unclear. The closest indications of how Herculine may have interpreted herself/himself are the reports that Sara referred to her/him using masculine qualifiers (Foucault, 1980b, 58), but we are not told whether or not Herculine assented to this practice. Despite the fact that Herculine does not 30  It is very difficult to know what pronoun to use in reference to a gender-variant person whose self-identified gender and pronoun preferences are unknown. Since Herculine Barbin was what we now call 'intersex' and Herculine's narrative does not contain a self-identified sex or gender, neither male nor female pronoun seem appropriate. I considered using a gender neutral pronoun such as 'ze' in reference to Herculine, but such pronouns often carry a connotation of a sex or gender identity outside of the male-female binary, and risk retroactively imposing an interpretation of Herculine's gender and/or sex. Given the lack of a self-identification by Herculine and the fact that Herculine was legally classified first as female and later as male, I have chosen to refer to Herculine by 's/he' and 'her/his' since this captures the legal change of sex and gender that is central to Herculine's narrative. While this is an imperfect solution, particularly given that 's/he' and 'her/him' are often considered offensive terms when applied to transgender people, I think it is the 'least worst' way of referring to Herculine given the available pronouns in English.  139  provide a self-identification, Foucault projects an identification — or more accurately a lack of identification — on to the narrative: What she evokes in her past is the happy limbo of a non-identity, which was paradoxically protected by the life of those closed, narrow, and intimate societies where one has the strange happiness, which is at the same time obligatory and forbidden, of being acquainted with only one sex. (Foucault, 1980b, xiii) This interpretation by Foucault (1980b) is highly problematic, because Herculine never states that s/he considered herself not to have an identity, nor that s/he was 'happy' about a lack of identity, and Foucault provides no evidence to support his reading. Foucault (1980b) imposes his own meaning onto Herculine's body and narrative, and it is not clear what makes Foucault's exercise of power in interpreting Herculine any more legitimate than that of the doctors and priests. Foucault’s introduction to Herculine Barbin also discusses the relationship between sex and truth that is the central theme of The History of Sexuality, commenting that "at the bottom of sex, there is truth" (Foucault, 1980b, xi). Notably, the institutions that produce sex, sexuality and truth in Herculine Barbin’s story are the same as those identified by Foucault in volume 1 of The History of Sexuality, namely the medical profession and the Catholic Church. Foucault’s central argument in The History of Sexuality is that what we understand as sexuality and as sex are not natural, but rather are categories that come into being through discourse, i.e. through power and truth. This insight about the constructed nature of sexuality leads Foucault (1990) to reject the idea that there are natural, biological categories of sex, i.e. of male and female. Instead, Foucault (1990) argues that the concept of sex is a "fictitious unity" (154) of anatomy and sensations that is the product of discourses of sexuality. Since sexuality emerged during the nineteenth century as a privileged site for truth about the self, particularly in the Catholic practice of confession, Foucault (1990) rejects the idea that power operates to repress sexuality. Instead, he argues that 140  sexuality is produced both through medical discourses and the requirement that people should talk about it, and connects sexuality to broader strategies of power through the concepts of disciplinary power and of "bio-power" (Foucault, 1990, 140). Biopower involves knowledge of and government over the population as an entity, through means such as the measurement and regulation of birth and death rates, life expectancies, disease etc. (Foucault, 1990). According to Foucault (1990), sexuality is uniquely important because it is located at the intersection of biopolitical governance of the population, and disciplinary power, which he describes as "an anatamo-politics of the human body" (139). The importance of sexuality as a target for biopower stems from its role in the reproduction and health of the population, while it is a target for discipline because sexuality relates to the anatomical regulation and normalization of the body, and thereby of the subject. Foucault’s (1990) claims about the overlap of discipline and biopower in constituting sex and sexuality thus reiterate his (2007) argument in the 1977-8 Collège de France lectures that different forms of power co-exist and support one another. Therefore while Foucault (1995) does not consider the link between sex and discipline in Discipline and Punish, he states in The History of Sexuality (1990) that disciplinary power is part of the constitution of sex. If one puts Foucault's (1980b, 1990 & 1995) analyses in Discipline and Punish, The History of Sexuality and Herculine Barbin together, the common thread is that our understandings of bodies are not natural or outside power, but rather are shaped by discourses and techniques of power. Disciplinary techniques work on the body through the spatial distribution of subjects, the control of activity through means such as timetables, the imposition of minutely detailed exercises, and the threat of continuous surveillance and assessment (Foucault, 1995). In addition, biopower operates to collect information about and to regulate the 141  life of the population as a whole31 . Biopower and discipline overlap to construct normative heterosexuality and the accompanying ideas of sexual intercourse, and of binary biological sexes that are defined in terms of genitals. The use of discipline and biopower to construct and naturalize binary categories of biological sex therefore erases the existence of ‘hermaphrodites’ (now termed intersex people) such as Herculine. What Foucault does not explicitly discuss is how sex and sexuality relate to the prison, and how sex relates to gender, i.e. how conceptions of binary, biological sexes of male and female relate to masculinity and femininity, and the accompanying categories of men and women. Given Foucault’s (1995) arguments about the minutely detailed norms involved in the exercise of disciplinary power in institutions such as prisons and schools, and his claim that discipline is linked to the constitution of sex and sexuality, is seems to follow that prisons and schools construct binary sexes and normative heterosexuality. One way that disciplinary institutions might construct sex and sexuality is by spatially distributing their subjects according to sexed identities, e.g. by segregating male and female subjects, or by pairing them with the opposite sex. Another way discipline might construct sex and sexuality is by assessing individuals against sexed and heterosexualized norms of appearance and behaviour until subjects internalize these norms and become self-disciplining. At this point another problem becomes apparent with Foucault’s (1980b) interpretation of Herculine’s life as a student and then teacher in an all-female environment as an experience of "non-identity" (xiii), because this contradicts the idea that discipline imposes sexed and heterosexualized norms. If Herculine was both subject to and later imposed disciplinary rules within an all-girls school, an institution that by definition has employed the disciplinary technique of spatially organizing subjects by sex, Foucault's (1990) account of biopower in The History of Sexuality is noticeably similar (although not identical) to his account of how governmentality operates through statistical knowledge and supposedly natural processes such as births, the food supply, and infectious diseases (Foucault, 2007). 31  142  then surely it follows that her life was shaped by sexed and/or gendered disciplinary norms? However Herculine chose to identify it would therefore have been in the context of sexsegregation, her/his experiences of feminine disciplinary norms within the school, and later of masculine disciplinary norms once Herculine was classified as male. The 'missing link' in Foucault's work about the connection between the detailed techniques of disciplinary power described in Discipline and Punish and the constitution of sex and sexuality is explored by Sandra Lee Bartky (1988). Whereas Foucault focuses on the constitution of sex by way of ideas about sexuality and genitalia, Bartky avoids the subject of genitalia and focuses on how gender identities are constructed through bodily practices. Bartky's (1988) central argument is that femininity is constructed precisely through "those disciplinary practices that produce a body which in gesture and appearance is recognizably feminine" (64). Bartky provides numerous examples of sex and/or gender-specific disciplinary practices, from diet and exercise, to the differences in how men and women walk and sit, to practices of depilation and skincare, to cosmetics and hair styling. The multiplicity of rules about women’s appearance and the media circulation of images of the ‘perfect’ woman’s body make the ‘norms’ of femininity largely unachievable, or, as Bartky (1988) states bluntly: The disciplinary project of femininity is a "setup": it requires such radical and extensive measures of bodily transformation that virtually every woman who gives herself to it is destined in some degree to fail. Thus a measure of shame is added to a woman’s sense that the body she inhabits is deficient: she ought to take better care of herself. (71) Bartky's conclusion is that the excessively detailed rules of femininity, and the sanctions for breaking them, mean that women are subject to greater bodily discipline than men — in short, women’s bodies are rendered more docile. Bartky (1988) explores what she sees as the gendered disciplinary norms of the contemporary US, and therefore these norms are not applicable to Herculine's context in 143  nineteenth century France. However, one could imagine a historical analysis that would outline the gendered and heterosexualized disciplinary norms that applied in Herculine's context, from the girls boarding schools in which s/he studied and taught to her/his later experiences seeking work in Paris. By reading Herculine's narrative with attention to the gendered and heterosexualized disciplinary norms of her/his historical context, one would develop a greater understanding of the norms that Herculine and her/his lover Sara were conforming to and/or resisting, and of what a masculine or feminine identification would have meant. Such an analysis would help us to understand the implications of Herculine’s concern that her/his hips are insufficiently curvy, and her/his body is too hairy, in relation to the gendered norms of Herculine's context. If Herculine is indeed self-disciplining her/himself according to sexed and gendered norms, then Foucault's (1980b) interpretation of Herculine's narrative in terms of nonidentity misses the point by failing to recognize these exercises of power. Any attempt to interpret Herculine's identity — or lack thereof — is problematic because of the exercise of power involved, as I argued earlier, but it is important that our theoretical account of power acknowledges that sexed, gendered and heterosexualized norms are a background presence in Herculine's narrative. Whether or not one accepts Bartky's (1988) claim that women are disciplined more than men, as opposed to simply being disciplined differently to men, her analysis helps to show how minutely detailed disciplinary norms relate to the construction of sex, gender and sexuality. Part of what is interesting about Bartky's analysis is that there is no clear distinction between sex and gender in many of the examples she provides about the normalization and disciplining of the body. For example, the presence or absence of hair in particular parts of the body is related both to conceptions of binary biological sex (via hormones that promote the development of 144  secondary sexual characteristics such as beard growth) and to gender, i.e. socially constructed norms about the ways that one inhabits and modifies a given body. Although Bartky (1988) does not explore this issue about the relationship between sex and gender, or between what is innate versus what is modifiable about the body, her analysis hints at it by stating that the norms of femininity are unachievable — one's 'natural' body is incompatible with the ideal. Whereas Bartky (1988) focuses on the construction of femininity, Bob Connell (1987) analyzes the constitution of masculinity. Connell uses the Gramscian concept of hegemony, or ideological dominance backed up by force, to identify hegemonic masculinity, i.e. a dominant masculinity that is maintained by prevailing beliefs, and by the threat of violence. Connell (1987) argues that hegemonic masculinity is constructed both in opposition to femininity — of which there is no hegemonic version, because femininity is subordinated — and to subordinated masculinities, and that it is produced by the state in institutions such as the military, police and prisons. Since the characteristics of soldiers and policemen are very similar to those of the criminals, and the dynamics of coercion, hierarchy and repression overlap, "the state both institutionalizes hegemonic masculinity and expends great energy in controlling it" (Connell, 1987, 128). Connell (1987) does not provide the kind of detailed analysis of how masculinity operates upon the body that Bartky (1988) provides in relation to femininity, although arguably such disciplinary practices of hegemonic masculinity are precisely what Foucault (1995) describes in Discipline and Punish. However, Connell hints at these disciplinary practices by explaining that gendered power relations and a masculine self-identity are involved in how the body is used in work and sexual relations, including the size, shape, posture and ways of moving: The social definition of men as holders of power is translated not only into mental bodyimages and fantasies, but into muscle tensions, posture, the feel and texture of the body. 145  This is one of the main ways in which the power of men becomes ‘naturalized’, i.e. seen as part of the order of nature. (1987, 85) Although Bartky (1988) and Connell (1987) advance Foucault's argument by exploring the constitution of femininity and masculinity respectively, there is an unresolved question about the relationship between sex and gender. If, as Foucault (1990) argues, sex and sexuality are constructs and the binary sex categories of male and female are not natural, then where does gender come in? Foucault's work invalidates theories that treat sex as the natural, pre-discursive basis of socially constructed gender identities, but gives us no alternative account of gender or how it relates to sex. Here, it is useful to draw on the work of contemporary queer theorists such as Anne Fausto-Sterling and Judith Butler, both of whom build on Foucault's account of the constructed nature of sex and sexuality. Fausto-Sterling (1993) suggests that we replace the current two categories of biological sex with five categories and creates three new categories to describe people who are neither straightforwardly male nor straightforwardly female: ferms, herms, and merms. Fausto-Sterling's (1993 & 2000b) argument is that the two sex model misrepresents the variation in bodies and erases the existence of intersex people at the level of discourse, leading doctors to perform non-consensual genital surgery on intersex children in order to normalize their sex — often without even consulting their parents. The proposal that we re-classify sex demonstrates the biologically arbitrary nature of the two sex schema, and subverts the idea of definitive sex categories or objective biological knowledge, thus opening the way for people to accept a greater range of bodies and gender expression. Butler (1993 & 1999) also argues that the existence of intersex people shows that binary categories of biological sex are not a natural or pre-discursive, but she develops this claim by arguing that binary sex categories are instead are a product of binary categories of gender:  146  If the immutable character of sex is contested, perhaps this construct called ‘sex’ is as culturally constructed as gender; indeed, perhaps it was always already gender, with the consequence that the distinction between sex and gender turns out to be no distinction at all. (Butler, 1999, 10-11) Having rejected the distinction between sex and gender, Butler (1999) develops an account of gender as performative, i.e. not as a property or identity that one possesses, but as an iterative practice that brings into existence what it purports to represent. This performative account of gender therefore suggests that gender is something you do and not something you are: masculinity and femininity are practices that only give the impression of being a pre-existing identity. Butler then turns to Foucault’s (1990) critique of the repressive hypothesis or the ‘incest taboo’ and its role in gender formation according to psychoanalyti