Criminal Law and Philosophy

, Volume 3, Issue 2, pp 167–186

Theorizing Criminal Law Reform


    • Department of PhilosophyUniversity of British Columbia Okanagan
Original Paper

DOI: 10.1007/s11572-009-9076-7

Cite this article as:
Shiner, R.A. Criminal Law, Philosophy (2009) 3: 167. doi:10.1007/s11572-009-9076-7


How are we to understand criminal law reform? The idea seems simple—the criminal law on the books is wrong: it should be changed. But 'wrong’ how? By what norms 'wrong’? As soon as one tries to answer those questions, the issue becomes more complex. One kind of answer is that the criminal law is substantively wrong: that is, we assume valid norms of background political morality, and we argue that doctrinally the criminal law on the books does not embody those norms. Another kind of answer is that the criminal law as it stands presupposes certain empirical facts, and yet those facts do not hold. Traditionally, criminal law reform has been informed by both these answers. Analytical theorists examine doctrine for its conceptual structure, and social scientists examine the actual workings of the criminal justice system. This tidy picture is, however, challenged by social constructivist accounts of the criminal law. They challenge the stability and conceptual purity of doctrine, and they challenge the objectivity of social science. On the basis of these challenges, they undermine the ambitions of traditional criminal law reform, and argue that the only reforms to the criminal law that matter are politicized ones—that criminal law reform is pointless unless it serves the interests of the marginalized and the dispossessed. It seems undeniable that in some sense our perceptions of crime in our society are indeed moulded by social forces, and that crime does not exist independently of the social structures and processes that help to define and control it. But why should those insights have the implications for our understanding of criminal law reform that they are alleged to have? How could it follow from those insights that criminal law reform either becomes radicalized or valueless? The aim of this paper is to show that what can legitimately be taken from the emphasis on the social constructedness of crime does not require wholesale abandonment of the traditional picture of criminal law reform, even though it may require some modifications of that picture.


Criminal law reformAnalytical legal philosophyCritical criminologyCriminalizationCriminal law doctrineSocial construction of crimeCrime and power


The Traditional or Classical model of criminal law reform has two components. The first component is conceptual. The focus is on legal doctrine as a conceptual structure that must be designed aright—that is, so as to reflect some preferred background set of norms, whether these are viewed as internal or external to law. The second component is empirical. It is understood that criminal law has functions and effects. Criminology or other some social science can provide information about these effects and about the success of criminal law in performing its desired functions. Reform of doctrine can and should be appropriately adjusted on the basis of such empirical information.

Contemporary critical legal and criminological theory repudiate this traditional model. They repudiate both the ideal of the perfectibility of doctrine and the ideal of value-free social science. Law, it is urged, is no more than a social construction subject to the constraints of power, not of conceptual and empirical purity. Given this diagnosis of criminal law, reform of criminal law via reform of criminal law doctrine is pointless. Criminal law reform has no option but to become radicalized, to serve the interests of the marginalized and the dispossessed.

My focus in this paper is on this latter modality. There is an obvious kernel of truth to the linked ideas that our perceptions of crime in our society are molded by social forces, and that crime does not exist independently of the social structures and processes that help to define and control it. But how do we get from there—even, can we get from there—to the alleged irrelevance of a focus in criminal law reform on doctrinal reform?

I begin the paper by sketching the traditional model of criminal law reform and the reasons why this picture is said to be unacceptable. I then present what I think can be properly taken from the critical theorists’ focus on social construction. I then analyze the arguments for the radicalization of criminal law reform, and consider what more is needed in order to secure the soundness of those arguments from a starting point of the kernel of truth in the ‘law as social construction’ thesis. I end by considering the real extent of the damage done to the traditional model of criminal law reform by the arguments of critical legal theory and critical criminology.

The Traditional Approach to Criminal Law Reform

Criminal law reform might be viewed naively, or pre-philosophically, as amounting merely to this—the criminal law as it stands is in some way wrong or unjust: it should be changed, and in the following way. I am going to examine first what I shall call ‘The Traditional Philosopher’s Picture’ and ‘The Traditional Criminologist’s Picture’ of criminal law reform. These ‘Pictures’ are characteristically taken by their supporters to be little more than recapitulations of the simple or naïve realist view: they are, however, complex theoretical interpretations of the view. They seem very natural interpretations, and in the end we will see that they have their place in the theory of criminal law reform. However, that place can only be properly seen once the traditional pictures have undergone deconstruction.

The Traditional Philosopher’s Picture

By ‘traditional philosopher’, I mean one whose interest in criminal law reform is primarily conceptual or analytic. From the philosopher’s perspective, the ‘traditional’ or ‘classical’ law reform project focuses on legal doctrine. The criminal justice system as a social institution is in an important sense a conservative institution (Zedner 2004, p. 7). The forms and doctrines of criminal law are assumed to represent the distilled wisdom of the ages. Crime is seen as a threat to social stability. The task of criminal law is internally connected to the preservation of the social order. Criminal law doctrines may change, and have changed, but only within narrow limits. Even if it is true that gambling was once regarded as criminal and is now not, or that the rape of a woman by her husband was once not regarded as criminal but now is, the basic doctrines concerning crime in the common law have not changed a great deal over time. Anyone who has either taught or taken a class in philosophy of law, or jurisprudence, or criminal law, knows the basics (or, ‘the official basics’, if you like) as far as concerns criminal law. The focus of attention is the maxim actus non facit reum nisi mens sit rea. Criminal liability requires an actus reus, a guilty act, an act that has been defined by law as an offence, and mens rea, a guilty mind, an intention or other appropriate mental state to perform the guilty act, signifying fault. The law permits defences of various sorts, divisible into justifications and excuses. The offences typically discussed in such pedagogical contexts are the familiar ones—murder, sexual assault, theft. And of course there are always attempts and conspiracies, and the mentally gymnastic conundra they pose. The principles of guilt and innocence that can be rationally reconstructed from these sources form the General Part of the criminal law. ‘The general part consists of a range of doctrines that reflect what it is about the criminal law that is distinctively criminal’ (Simester and Shute 2002, p. 4). The special part has to do with the specific offences themselves, the definition of them and problems arising therefrom (Duff and Green 2005). The traditional philosopher’s criminal law theory is the theory of the criminal law so construed—the detailed articulation of the principles of criminality that constitute the general and special parts.

The assumption of this institutional permanence and correctness leads to a picture of law reform as a formal and conceptual analysis of the structure of criminal liability. In a truly traditional and conservative form, the exploration of doctrine is ‘scientific’, in the sense of ‘acquir[ing] knowledge on the basis of constructing logically coherent conclusions from elementary principles’ (McCrudden 2006, p. 633). However, the idea of ‘reform’ need not be, and is not purely analytical. The term ‘reform’ presupposes normative standards: there is a way of getting the analytical structure right—that is, so that the conceptual structure of the criminal law properly mirrors the relevant norms of background political morality, rather than internally adopted axioms—and many ways of getting it wrong (the criminal law does not mirror such structures). Developing the correct analytical and conceptual structure for the criminal law is an abstract task for the theoretician. The task may be one of developing a sound set of general principles for the general part, or a sound set of more specific principles for the special part.

Then there is a second, more concrete but still largely intellectual task—a process which takes the analytic structure just mentioned and checks it against the actual content of some body of black-letter criminal law, the Canadian Criminal Code for example. Either the general assumptions of principle that may be extracted from the Code, or the structure of some specific Code offence, may not conform to the best normative standards. Criminal law reform then amounts to making recommendations as to the changes needed in criminal law doctrine, in order to make doctrine conform to principle.1

The Traditional Criminologist’s Picture

One implication of the traditional philosopher’s approach to criminal law reform is that what matters most for getting in the real world the desired normative results from law reform is legal decision-making. Legal decision-making is the application of legal rules or norms to fact situations. Get the legal rules right, and, once the facts are in hand, the desired result will follow. This optimistic assumption has been long challenged by Legal Realism. Legal Realists pointed out that much more determined the outcome of litigation than an abstract mapping of rule on to fact situation. As long ago as 1910 Roscoe Pound urged attention to the distinction between ‘law in action’ and ‘law in books’ (Pound 1910). Even if (regrettably) there is no hard evidence that Jerome Frank really made the famous claim always attributed to him that judicial decisions may be caused by what the judge had for breakfast, still it is clear that realists all urged greater attention to how social forces might affect legal decisions, both in the taking thereof and in how such decisions actually played out in the real world.

A certain Traditional Criminologist’s Picture then emerges of the contribution social science, and especially criminology, can make to law reform. The pattern of thought is this. The law is much more than legal decision-making, as opposed to what the traditional philosopher and traditional doctrine assume. The actual effects of a legal rule are often quite different from the intended effects. The actual realities of the legal process are often quite different than we imagine them to be. Social science needs to collect data in order to test our assumptions about laws and legal procedure. Sound legal doctrine needs to be based on sound social-scientific research.

There is a certain complementarity between these two traditional pictures. Social science in such a role is subversive of traditional legal doctrine to the extent that the latter has empirical exposure and makes implicit empirical commitments that it cannot discharge. But social science is not subversive of legal doctrine überhaupt. The thought is simply that legal doctrine needs to be empirically sound, rather than empirically weak. The thought is not that there is something disreputable about the very idea of legal doctrine as such. Criminology in this mode supports, rather than undermines, and is therefore implicated in, what might be called the Official Version of the law, that it is an impartial, neutral, and objective system for resolving social conflict (Comack 1999, p. 21). Social science on the traditional picture is just as neutral and objective as law according to its official version. If law and legal doctrine fail to be neutral and impartial, then that is because the social science is not right, not because it can be shown, for example, that impartiality and neutrality are inherently impossible for law as an object of social science.

The Traditional Model of Criminal Law Reform

A certain model of the relationship between criminal law doctrine and criminology—call it the traditional model—emerges from these traditional conceptions. The task of social science is to do the empirical work that will confirm or disconfirm the instrumentalist implications of any proposed reform to criminal law doctrine presented by the doctrinal analysts. There are two aspects of this model. One I call the ‘research assistant’ or RA aspect: analysis provides hypotheses and social science confirms or disconfirms. The second I call the ‘division of labour’ or DL aspect: analysis has one job, social science another, and the greater good will be realized only if neither tries to do the other’s job. The theoretical question I am interested in is this. In many ways, this ‘research assistant/division of labour’ model of criminal law reform has been strenuously attacked and vilified in much recent legal and criminological theory, and it has been assumed that the traditional model in terms of the strictly instrumentalist role of criminology has to go as well. But is that true? Must we think of the place of doctrine and social science in criminal law reform as researcher and assistant, or as dividing the labour? Do we in fact have to revise fundamentally our conception of criminal law reform if we repudiate the traditional philosopher’s and criminologist’s pictures?

The Critiques of Traditionalism

Both dimensions of the traditional model have come under severe attack in twentieth-century legal theory. With respect to legal doctrine, critical legal theory in the form of the Critical Legal Studies movement strenuously argued that analytical jurisprudence greatly overestimated the tidiness of legal doctrine. Two classic ‘Crit’ pieces are illustrative. Duncan Kennedy in 1976 argued that private law doctrine is inherently contradictory (Kennedy 1976). Private law revolves around two conceptual axes—that of Form and Substance, and that of Individualism and Altruism. Individualism tends to be associated with form and with rules; altruism tends to be associated with substance and with standards, norms that are vaguer and more open-ended than rules. A superficial ideal for the law is to find the right balance in any given area of doctrine of each polarity or axis. This ideal is spurious, as the axes are fundamentally contradictory. Private law is a site for the conflict between individualism and altruism, not for a resolution of the conflict. Mark Kelman 5 years later wrote about criminal law that at the core of the criminal law in practice are a variety of what he called ‘interpretive constructs’—powerful concepts or images that structured courts’ readings of the facts, rather than having their applicability revealed by the facts (Kelman 1981). Thus, the alleged ‘givenness’ behind the idea of an accused person being ‘guilty on the facts’ is spurious.

Note that these critical arguments if sound are not threats to the idea promoted by legal positivism that the existence and content of laws is to be established by a test of social fact, rather than a normative test. In fact the arguments represent in a sense a somewhat quirky version of the social fact thesis. The critical arguments if sound repudiate what I have called the official version of the law—the idea that the law is an impartial, neutral, and objective system for resolving social conflict. If private law has the inherently contradictory character that Kennedy says it has, for example, then any legal decision applying private law will not be impartial, but will represent a value-laden choice among the incompatible versions available. Any verdict of guilty under the criminal law, if Kelman’s argument is sound, will impose a value-laden interpretation on proffered facts, and will not be a neutral and impartial application of a rule to those facts.

Sociological criminology analogously was revealed to have ‘positivist’ background assumptions, in the sense of scientific, not legal, positivism. If legal doctrine was to be validly reformed, arguments for reform had to be based on facts about human behaviour and human nature. Theories and hypotheses had to be formulated with empirical import, and the validity of that empirical import tested. Such a methodology was classically exemplified in what has been called the ‘penal-welfarist’ phase of penal policy-making. The presupposition of placing emphasis on the rehabilitation of prisoners was that there existed a body of psychological and sociological knowledge of human behaviour that could be brought to bear on the understanding and modification of prisoners’ attitudes and behaviour.

Positivist social science of this behaviouristic form came under attack from different directions. Some of these attacks were general in scope applying to any science—for example, the argument that positivist methodology presupposed a rigorous distinction between theory and observation, whereas no such distinction could be reliably made. Other criticisms were specific to the social, or human, sciences. It was urged that the special character of the human mind and of human societal interactions and arrangements ruled out purely objective scientific study. Understanding human behaviour, rather, is more a matter of sympathetic interpretation than neutral observation.

In various ways, and for various reasons, then, theorizing the criminal law and criminal law reform a decade or two ago lost its analytical and positivistic innocence. Reform of the criminal law and of the penal system could no longer be plausibly represented as the revision and emendation of doctrine based on sound conceptual analysis and sound social science.

The story as I have presented it so far, however, has been told simply in technical or methodological terms. Little or nothing of what I have said constitutes a political critique of the law and of law reform. ‘Little’ perhaps is better than ‘none’, since the claim that the criminal law fails to live up to law’s official version of itself is a normative, and so if you like political, critique—but just so far an open-ended one, leaving many options for its substantive completion, including the ‘conservative’ option that law should be changed so that it did conform to the official version. It is important to make this point, not so much because I rely on it explicitly in what follows, but because the core point is of a piece with a central theme of the remainder of this paper. The political critique of the criminal law that figures so prominently in critical legal theory and critical criminology may have emerged from prior views about the methodology of criminal law reform, but it is not entailed by those views. Positivism in social science and analytical jurisprudence both may be misguided in the way that they implicitly approach criminal law reform, but that in and of itself does not imply that critical legal theory and critical criminology are correct in their politicized accounts of the criminal law.

The Social Construction of Law

We move closer to political critique of traditional law reform when we ask the question, if the analytic and positivist accounts are not correct, what abstract or theoretical account of criminal law is correct? The mistake, it is claimed, of analytic and positivist approaches to theorizing the criminal law, and so to criminal law reform, is to think that there is a truth about the relevant normative concepts that doctrine should reflect, and a truth about human behaviour that criminology as a social science should reflect. Neither of these theoretical assumptions, the argument goes, can be sustained. The criminal law is simply a set of social constructions.

In a banal sense, the criminal law is an artifact, because law of any kind is artifactual. Acts or omissions in real life are not in themselves or as they occur ‘crimes’. They may arguably be in themselves long or short, harmful or harmless, foolish or sensible, occur at night or on Tuesday or in the High Street, and the like. But they are not in themselves ‘crimes’. An act/omission—a taking of someone else’s property with the intention of taking it, for example—is a ‘crime’ because an official organ of the state (usually, the legislature, but occasionally a court) has defined a criminal offence that makes that act/omission into an official or legal crime.

Everyone commits a theft who fraudulently and without colour of right takes, or fraudulently and without colour of right converts to his use or to the use of another person, anything whether animate or inanimate, with intent to deprive, temporarily or absolutely, the owner of it, or a person who has a special property or interest in it, of the thing or his property or interest in it.

That is the formal definition of (one form of) theft in s.322 (1)(a) of the Canadian Criminal Code. That definition is what makes Anita’s making off with Bernard’s bicycle, intending not to bring it back, into a crime.

Here we have a first and basic instance of the artifactuality of crime, which is simply a reminder as to what law is. Law itself is a human artifact. Laws construed as rules are made by human beings—enacted by the legislature, or formulated by courts. That includes the criminal law. Intentional killing may be morally abhorrent in and of itself. But it is not a crime until the law defines it as a crime. So murder as a crime is a human artifact, a legal construction, if you like.2

Those who insist on the criminal law being seen as a social construction have in mind, however, something more than this banal point. Consider: ‘Crime is not an inherent property of individuals but a status conferred on the individual by those who make and enforce rules. Crime and deviance are social creations.’ (Comack 1999, p. 32) ‘Crime is not the object but the product of criminal policy. Criminalization is one of the many ways to construct social reality.’ (Hulsman 1986, p. 71) ‘Crime and criminal justice are conceptualized and implied as though they are objective realities having real consequences, consequences we attribute to their claim, but that they do not possess in any intrinsic sense. Institutions of law are the organized acting out of discursively produced ‘control thoughts’, whose very action reflects on the reality of that which they are organized to defend.’ (Henry and Milovanovic 1994, p. 118).

The thought behind these comments can be explicated this way. Unreflective public opinion would likely converge on the following as paradigm crimes—murder, manslaughter, sexual assault, robbery, arson, assault, kidnapping, burglary, theft. These are the familiar kinds of offences against personal and property rights. So, if one saw Charlie raining punches and kicks down on Dennis in the street, describing what one saw as an ‘assault’ seems as natural as calling that object over there with four legs, a seat and a back a ‘chair’. Already, whether the object over there is a ‘chair’ is not so straightforwardly a fact. It might look very like what in our culture would be the paradigm example of a chair, but it might not be. It might be an avant-garde work of art. It might be a religious object from some quite different culture. When we turn to Charlie raining punches on Dennis and our immediately deciding we are observing a crime taking place, matters are different again from the chair example. It is not just that ‘crime’ is a legal category and we have no way yet of knowing whether Charlie is legally sane, or has a valid legal justification or excuse. We might not be watching a crime at all, but a scene from a movie being shot, for example, or a piece of performance art.

The point, however, is not simply that, in calling what we see a ‘crime’, we are stating a conclusion which might not be true, and for which we really ought to have more evidence before we express an opinion. The point is why we would immediately leap to the conclusion that we are watching a crime taking place. We identify what we see as a crime taking place, because we are conditioned or socialized to see such a happening as a crime. Crime is not an inherent property of individual actions or states of affairs, but a status conferred on those individuals and states of affairs by those who make and enforce rules, especially legal rules. Our current conceptions of what is criminal and what is not are historically contingent: that is, they are artifacts not only at a particular point in history, but of the particular configuration of social forces operating at that point of history. ‘Those deemed to threaten order are less likely to be city brokers engaged in insider dealing or tax evasion, but street gangs, drug users, and discontented youth…. it is the delinquent youth not the besuited broker who is seen primarily to threaten law and order.’ (Zedner 2004, p. 7) ‘Rowdy behaviour by local youths is thought about quite differently from the “high spirits” of undergraduate celebrations even where, objectively, the behaviour in each case is all but identical’ (Zedner 2004, p. 40).

Nor do we need simply to be aware that our notions of crime change and evolve. We need to be aware how they change and evolve. Our notions of crime are developed through our experiences, as well as through our interactions with our family, friends and other people whom we encounter in our daily lives. We also derive our notions of crime from sources such as newspapers, television, radio, books, and films. We receive messages about crime and unwanted conduct through these various sources, which in turn help shape our perceptions about what is and is not a crime. ‘Power, control, justice, and communication are reflexive. One aim of social control is to define the time, place, and manner of appropriate behaviour. Social control is manifested through images and claims about order and disorder, stigma and identity, enforcement, and moral denunciation…. Gonzo justice has emerged as a new cultural form to address the mass mediated public perception of unsuccessful social control. Popular culture provides a way to participate or play with horror, banditry, crime, and justice, as we are presented with a range of scenarios and enactments through which we can interactively arrive at meaningful interpretations.’ (Altheide 1992, n.p.).

The proper strategy of law reform therefore, it is said, must be to investigate and interrogate, not the doctrines of criminal law, but society’s practices of criminalization. The term ‘criminalization’ has two different but associated meanings. To criminalize a certain act is to pass a law making performance of that act a crime. Such a passing, though, is also to criminalize persons, to make certain people into criminals. Crime is not a behaviour or a set of behaviours identified in isolation, but is rather a process. ‘The criminal law must be understood as something that society does,’ (Bilionis 1998, p. 1302) and it is a complex ‘something’. ‘What we are dealing with is a complex and dynamic process of categorization and discrimination rather than a static practice of deductive reasoning from premises set by a legal definition.’ (Lacey 1995, p. 8) It may seem as though a study of criminalization and criminality is a study of actors and actions the outer limits of which are set by legal doctrine, by legal definition. But that is not so. The focus must instead be on the complex social practices whereby not only through the formal mechanism of legislators passing, courts applying and police enforcing laws, but also through the informal mechanisms of media both journalistic and entertainment, schools, churches, families, clubs, and societies, corporations, water fountains, pubs, common rooms, and coffee shops, our ideas of what is a crime and who is a criminal develop and harden. It would be a mistake to think of criminalization as a one-time act by the state in enacting a piece of legislation. The state itself is composed of diverse agencies, diverse in ‘structure, political orientation, responsibilities, professional training, and culture’, and ‘a wide number of groups which are not clearly “state” or “state-sponsored” agencies have an important influence on decisions about, and practices of, criminalization’ (Lacey 1995, p. 5).

I happen to think that some form of social-constructivist account of the criminal law is correct, and that criminal law reform must take that result on board. To urge that point, however, is not the main goal of this paper. Rather, I want to pursue two questions that will turn out to have some connection. The first is this. The implications of adopting a social constructionist account of criminal law and criminal law reform are widely held to be inevitably political, and broadly-speaking left-political. But is this true? I do not mean merely, Is it true that, when you adopt the social constructionist perspective, you will see that the criminal justice system is in fact no ‘justice’ system at all, but the creator and sustainer of a whole variety of social injustices? That’s as may be. I mean rather, is it true that acknowledgment of the socially constructed character of law entails or necessitates adopting a left-critical normative perspective on the criminal law? Once criminal law reform has been theorized in social constructionist terms, then only left-leaning reform recommendations are possible. The second question is this. I introduced earlier what I called the RA and DL aspects of the traditional model for the relationship between criminal law doctrine and criminology. What happens to this model and its two aspects if we take on board a social constructionist account of criminal law reform? Does the model have to be abandoned, or merely revised? I will deal with these issues in that order.

Why Focus on Practices of Criminalization?

Consider first the thought that criminal law reform, and serious study of the criminal justice system in general, should focus, not on doctrine narrowly conceived, but on ‘practices of criminalization’. As said, I am willing to accept this thought as methodologically sound. Note though that such soundness is neither an implication nor a proof of the thesis that criminal law is socially constructed, at least not in a sense of that thesis that makes it stronger than simply the view that criminal law should be seen as a social institution or social phenomenon, not merely a technically legal institution or phenomenon. Criminology as a social science need not be constructivist, and it could still be important to study practices of criminalization even if criminal law doctrine reflected transcendental truths. So from where can the focus on practices of criminalization derive?

The best immediate argument would seem to be empirical: ‘Look and see’. Such a move is a repetition of the lessons of legal realism. There are two stages in the argument. The first is to extend the focus within the legal system narrowly conceived from doctrine alone to all aspects of criminal justice. There are many, many sub-practices within criminal justice—‘the criminal law can and should be understood as part of an integrated process of criminalization.’ (Lacey 2001, 258) Among these practices are sub-practices as to how cases are selected for trial and processed after conviction, as to the operation of rules of evidence and procedure, as to the institutional structure of criminal courts, and sub-practices incorporating all stages from the articulation of offences through investigation, diversion into processes other than formal criminal law, prosecution, trial, sentencing, and the operation of incarceration and parole. But these practices are all interlinked: what happens at one stage affects what happens at another; what values rule one stage typically rule others as well.

The second stage of the argument is to draw out the implications of the fact that the legal system, and a fortiori the criminal justice system, is a social institution. It is embedded in wider society. Its officials are human beings, men and women with all the strengths and weaknesses of humanity (‘neither devils nor angels’, as H.L.A. Hart famously observed: Hart 1961, 1994). Insofar as laws, and a fortiori criminal laws, emanate from political institutions, the creation of laws will be subject to political forces. Laws, and a fortiori criminal laws, are interpreted, applied and enforced by human beings in social roles, who will then be as subject to social forces as any human being doing anything anywhere. A judge, say, can treat a defendant, or a peace officer treat a suspect, as simply a legal object and not an individual person—just as a surgeon can treat a patient as simply the next hip replacement on the OR schedule and not as a person. Whether from the point of view of professional ethics these are defensible stances are complicated issues: perhaps such a mode of address can be justified on grounds of efficiency as long as the system overall takes individual needs seriously, and perhaps not. It is patently obvious, though, that it would be a mistake to think that there was nothing to the criminal justice system except the interaction of judge and defendant, or peace officer and suspect, just as much as it would be a mistake to think that there is nothing more to the health care system than the interaction of surgeon, physician, nurse, and patient.

The recommendation to focus on practices of criminalization is at bottom no more (and no less of course, in view of its importance) a recommendation to look holistically at criminal justice, to see various discrete moments in the operation of the criminal justice system as at best conceptually discrete, not functionally or socially or politically or psychologically discrete, and maybe not even conceptually discrete. As such, the recommendation is simply methodological: in and of itself it is neutral as to substantive norms for criminal law reform. It is compatible with a wide range of normative principles that might be brought to bear in criminal law reform on the changing of such practices.

Practices of Criminalization as Practices of Power

It may be argued, though, that once a person adopts the holistic gestalt of seeing the criminal justice system as a set of practices of criminalization, then it becomes just as immediately obvious that practices of criminalization are practices of power. One can see this thought implicit in the broker or high-spirited undergraduate versus disaffected or delinquent youths examples of a few pages above.

The meaning of ‘practices of power’ needs a brief explanation. Terms like ‘power’, ‘influence’, ‘cause’, ‘effect’ have a tendency to be interpreted in terms of intentional agency. If Andrew influences Brian, it must be because Andrew intends to influence Brian. If Andrew intends to influence Brian, then Andrew can be personally criticized or blamed for influencing Brian, if influencing Brian turns out to be the wrong thing to do. Terms like ‘power’, ‘influence’ and so on, however, need not imply agency of this blame-attracting kind. Consider an engineer explaining how an internal combustion engine works. She may say, ‘The power of the exploding mixture of air and gasoline drives the piston down’. There’s no agency there, or agency only in a metaphorical sense. The engineer is simply identifying the components of a system, the forces operating on those components, and the way that those forces operating on those components in that configuration produces certain results. In emphasizing the way in which social power and social forces affect how a criminal justice system operates, one need intend thereby no attribution of malign or blameworthy agency, but intend simply in the manner of the hypothetical engineer to be explaining how the system works, to be explaining what might be called the ‘system hydraulics’.

This conception of power is associated with the work of Michel Foucault, and has been named the ‘practice conception’ of power (Lacey 1995, p. 13). According to this conception, power is not owned or wielded by any one agent or institution. Rather, power subsists, less tangibly but no less influentially, in a wide array of discourses and practices spreading throughout the social body. Power has productive effects via the operation of practices and discourses. These effects are, in Foucault’s image, ‘capillary’: ‘[power] does not emanate from some central source, but circulates throughout the entire social body down even to the tiniest and apparently most trivial extremities.’ (Fraser 1981, p. 278).

‘Power’ just so far, however, is a normatively neutral notion. It abandons neutrality insofar as the discourses and practices that embody power are, in Foucault’s term, disciplinary or hierarchical—that is, insofar as the discourses and practices create distinctions between the normal and the deviant, the acceptable and the unacceptable. Power can then become oppressive insofar as individuals internalize these distinctions and modify their behaviour accordingly. The discourse of crime and criminality of course is a paradigm example of a disciplining discourse: the distinction between the normal and the deviant is integral to the discourse.

Rose deploying a different metaphor has put the matter this way:

The control of conduct [is] now immanent to all the places where deviation could occur, inscribed into the dynamics of the practices into which human beings are connected…. Control society is one of never-ending modulation where the modulation occurs within the flows and transactions between the forces and capacities of the human subject and the practices in which he or she participates. Control is not centralized but dispersed; it flows through a network of open circuits that are rhizomatic and not hierarchical. (Rose 1999, p. 234)

Rose’s thought here has two parts. First, even if we restrict reference to the formal apparatus of the state, the criminalizing power of the state is not confined to the formal authority of Parliament and courts. Rather, it is diffused ‘rhizomatically’ (think here of spreading quack grass) throughout the criminal justice system. Prison guards, parole officers, peace officers, by-law officers, regulatory agencies and their employees, prosecutors, clerks, private security companies and guards to whom tasks have been outsourced, all are sources of state criminalizing power: all have the power to criminalize, to declare an individual a criminal. Moreover, and second, these formal agencies exist within social structures and discourses about crime of informal and multifarious but influential kinds—conventions and patterns of crime reporting by news media, for example, or conventions and patterns of the depicting of crime by entertainment media, educational professionals, and so forth. These social structures exercise criminalizing power as well: all can influence perceptions of what is a crime and who is a criminal. Practices of criminalization seen as practices of power constitute a criminal justice system of extraordinary complexity.

Note, however, that in moving from individual acts of criminalizing by courts or officials to practices of criminalization to practices of power, we still seem to be following simply a trajectory of ever more sophisticated methodologies for the comprehension of criminal law seen as a social institution. The combination of the two Traditional Pictures is too simplistic in its focus on doctrine and empirical research. We move then from the empirical to the interpretive and finally to the social theory-based approaches to the study of crime and criminal law. The latter leads us to focus not on doctrine, but instead on practices of criminalization. Now, there is a fourth move. The conception of law as a social construction and the focus on practices of criminalization are too simplistic in their omission of a criminology of power. But how do we get from here to a supposed necessity for a left-critical political perspective on criminal law and criminal law reform?

It may be true that critical criminology, especially Foucaultian critical criminology, has brought to light facts relevant to criminal law reform that were left undisturbed by mainstream criminology and criminal law scholarship. Moreover, if a critical normative perspective is already presumed, then there is an easy complementarity between that perspective and the Foucaultian conception of power. Critical criminology may even think that it would not have been able to reach its insights if it had not made the assumption that the criminal law is a social construction and other kindred assumptions, but the argument for that is at the most an empirical one about peoples’ thought-processes. At a time when Foucault’s work on power was just beginning to achieve recognition in the English-speaking world, Nancy Fraser argued that ‘Foucault’s valuable accomplishment consists of a rich empirical account of the early stages in the emergence of some distinctively modern modalities of power’ (Fraser 1981, p. 272). Foucault brackets, she says, normative frameworks and concentrates instead upon the actual ways in which power operates (Fraser 1981, p. 273). There is no reason to abandon this view of Foucault’s account of power as empirically rich, rather than transcendentally inevitable.

The perspective that the notion of practices of power must be at the centre of criminal law reform is theoretical in the sense that it is an assertion at a high level of abstraction and generality, one that rests on broadly expressed principles accompanied by at most the detailed study of a few examples that the reader or researcher or student is expected to appreciate to be paradigmatic and universalizable. Methodologically, that amounts again to a ‘Look and see’ kind of argument. One could try to argue the thesis more discursively by urging that it follows from practices of criminalization being essentially political practices that they are practices of power, or from their being essentially social practices that they are practices of power. But the same kind of holistic gestalt is needed to see not only the alleged fundamentally political character or fundamentally social character of the relevant practices, but also that being an essentially political or essentially social practice implies being a practice of power. So it does not seem as though the argument to a left-critical normative perspective is advanced linearly by these moves.

Taking Power Seriously

Such a modest conclusion, however, arguably cannot be sustained. The step from seeing mere practices of criminalization to seeing practices of criminalization as practices of power is seen by its supporters as crucial to left-critical conceptions of criminal law reform3 since it makes available an impressive argument against the traditional model of criminal law reform. The argument if sound would seem to take us closer to the inevitability of radicalized criminal law reform. The argument goes as follows.

The traditional conception of the source of power for the legal system obscures the complexity of modern power. Traditionally, the source of state power is seen as unitary, the sovereign—not necessarily a single monarch, but in a parliamentary democracy Parliament, or perhaps citizenry who elect Parliament. Parliament, and the courts operating under the final authority of Parliament, are the sources of the formal rules and doctrines of the criminal law. Just as a highly complex electronic system such as a computer only works once it is plugged into a source of electric power, so the criminal justice system only operates because of the authority of the sovereign. On the ‘classical’ model of parliamentary power over the criminal law as unitary and sovereign, the choice to criminalize or not appears as a power-producing choice.

This model though is just another version of the traditional or ‘classical’ emphasis on legal doctrine. Parliament, and the courts operating under the final authority of Parliament, are the sources, not of criminalization overall, but merely the formal rules and doctrines of the criminal law. In order to understand power compatibly with the emphasis on the social structures and processes that define and control criminality, power has to be reinterpreted. Parliament is merely one part, and not necessarily the most significant part, of the criminal justice system construed as a system of diffused power. The appearance of a free parliamentary choice to criminalize, of a choice external to the criminal justice system and affecting it but not being affected by it—this appearance is exposed as an illusion. The parliamentary choice is the effect of power. Parliament’s decisions will have an effect, of course, but those decisions are also affected by the antecedent distribution of power within the system, which in the end is the political and social system as a whole.

I call this argument ‘impressive’ for the following reason. The contrast between traditional doctrine-based criminal law theory and critical criminal law theory is often put this way. Traditional criminal law scholars, even if they acknowledge that moral ideas to do with fairness and wrongdoing lie in the background of criminal law doctrine, take the idea of ‘crimes’ as given by acts of law-creation. Critical criminal lawyers assume that the power and meaning of criminal laws depend on a more complex set of processes and underlying factors than the mere positing of prohibitory norms to be enforced according to a particular procedure. Traditional scholars leave out of the legal and reform agenda both political and criminological issues: critical scholars foreground them.

As a characterization of what has happened in criminal law scholarship in the last few decades, that is pretty much accurate. But one feels inclined to respond, So what? This difference between traditional and critical criminal law scholarship can be easily explained. The critical scholar’s work is driven by a political commitment and the traditional scholar’s work by an analytic one. Both are valid choices. It is as proper to be a-political about criminal law doctrine and reform as it is to be political. Some economists are more politically driven than others, some parish priests are, some bus drivers are. It is all very well to sound clever and say that the choice to be a-political is a political choice. But why is it?

The argument just given tries to give the argument. The traditional or ‘classical’ emphasis on legal doctrine in criminal law reform makes a political statement in that it makes two connected assumptions. The first is that only the power of the state to enact and apply laws affects who and what is criminalized. The second is that the law does not affect the exercise of state power: the law is impartial and neutral, as per the official view. The continued adherence to the project of perfecting analytically or conceptually the normative content of the law is, then, an inherently political project because it is an inherently conservative project. It takes the form of claiming that what we have always had we should continue to have. It can only be justified by continual arguments that show the changes in social conditions that have undoubtedly occurred and will keep occurring do not necessitate any changes in the basic doctrinal scheme for the criminal law. That is a political issue, an issue of political morality, and, at least from the perspective of the left, a sadly mistaken political view. It seems then that the interpretation of practices of criminalization as practices of power is not a matter merely of methodology.

Moral Soundness and Political Legitimacy

Undoubtedly, there are traditional doctrinal criminal law scholars who do look upon the proper design of criminal law doctrine as merely a technical and conceptual problem. But it is unfair and a mistake to say that the traditional doctrinal approach to criminal law reform must be normatively insensitive. Many contemporary analytic scholars with an ultimately doctrinal focus are deeply concerned about the moral standing of the criminal law. Much of this concern revolves around the idea of ‘the moral limits of the criminal law’, to echo the title of Joel Feinberg’s famous treatise (Feinberg 1984–1988). The ‘moral limits’ can be examined in very broad terms—the morally proper limits to criminalization as such, for example, or the moral principles around which a truly just system of criminal punishment should be designed (Husak 2002; Husak 2004; Husak 2008; Ashworth 2000). Or the ‘moral limits’ can be examined in very close-grained terms—the morally proper standards for the criminalization of attempts, for example (Duff 1996) or endangerment (Duff 2005), or for the interpretation of the traditional common law defences taken severally (Horder 2004). Such scholars focus on the moral soundness of the criminal law, rather than its political legitimacy. Or rather, they view the problem of the legitimation of the criminal law as being exactly the problem of the moral soundness of criminal law procedure and doctrine—an analytic and conceptual problem insofar as it is a problem of the isomorphism or lack of between two conceptual schemes, the legal and the moral. Criminal law reform needs no credentials other than the generation by the intellect of moral soundness.

For the rival approach, the political legitimacy approach, moral soundness is not enough. Criminal law reform needs democratic credentials as well as intellectual ones. Political legitimacy functions as an independent demand on criminal justice systems. The coercive character of criminalization implies accountability on the part of the state to those on whom such power is exercised. In this context, the thesis that practices of criminalization are practices of power is therefore crucial. Power in the sense used here is blind to issues of legitimacy: the power relations in a society are just how it is in that society, that society’s system hydraulics. If, then, a society is to enjoy political legitimacy the exercise of power in that society must be judged against standards of political legitimacy.

Critical legal theory and critical criminology then urge that the practices of criminalization in contemporary neo-liberal democracies face a crisis of legitimacy. The criminal law represents paradigmatically the coercive power of the state. Liberalism standardly regards the exercise of the criminal law power against citizens as justified because (in part: the whole story is a lot longer) the act that attracted the coercive restriction of liberty was voluntary: hence the alleged centrality of doctrines of mens rea, justification and excuse and the like to criminal law doctrine. The social conditions faced by those most commonly on the receiving end of the criminal law’s coerciveness—the poor, the young and the racialized; the atrocious conditions in which most of Canada’s aboriginal peoples live is a prime example—give rise to a legitimation crisis because it is not at all clear any longer how accurate it is to insist that the actions of theirs that attracted the attention of the criminal law were fully voluntary and thus their convictions justified. Or this argument might be made.

Suppose the defendant belongs to a group that has suffered persistent, severe disadvantages…. If we have collectively failed to include them in the community of citizens, how can we now claim the standing to call them to account for their alleged wrongdoings? Serious social injustice does threaten the legitimacy of criminal punishment, and indeed of the whole criminal process, by undermining the polity’s standing to call the victims of such injustice to answer for their crimes. (Duff 2003, pp. 709–710)

Here we see again the force of the assertion that our practices of criminalization are practices of power. A truly legitimate criminal justice system will be one that mitigates the effects of power. Criminal law reform must acknowledge that the socially constructed character of criminal law makes the criminal law in place at any given point in the life of a society simply a resolution of the moments or forces of power at work in that society at that time. Power distorts legitimacy because it distorts patterns of consent, of the consent that creates legitimacy. Counteracting the effects of power requires taking seriously the position—the needs, interests and opinions—of those oppressed by power: the marginalized, the racialized and the dispossessed.

If the analysis of sections “Taking Power Seriously” and “Moral Soundness and Political Legitimacy” is correct, then it becomes clear why the interpretation of societal practices of criminalization as practices of power finds favour with many left-critical conceptions of criminal law and criminal law reform. This interpretation leads directly to two assertions of major differentiation between traditional and critical approaches to the theory of criminal law reform. The first demonstrates that the focus on legal doctrine, not practices of criminalization, in the context of power is a political act. The second demonstrates that the focus on moral soundness, not political legitimacy, in the context of power is a political act.

The ‘practices of power’ interpretation only throws into relief, however, how the step to it from the methodological recommendation to focus on practices of criminalization, not doctrine, is again simply empirical or intuitive. Or rather, the proponent of the argument faces a dilemma. Suppose the introduction of practices of power is founded on empirical insight. Then the move from practices of criminalization to practices of power cannot ground the necessity of a critical perspective on criminal law reform, but only propose its observational adequacy. Suppose on the other hand the fact of the influence of power is assumed in itself to raise issues of legitimacy. Then a fundamentally liberal perspective on criminal law reform is being adopted, and the debate is simply about whether assumed preconditions for legitimate coercion are in actual fact satisfied. Still no inference lies to the necessity of a critical perspective.

The Traditional Model Revisited

According to what I have termed the traditional model (see section “The Traditional Criminologist’s Picture” above), the task of social science is to do the empirical work that will confirm or disconfirm the instrumentalist implications of any proposed reform to criminal law doctrine. This model is instantiated via a strictly doctrinal approach to criminal law reform coupled with a positivistic approach to social science. On the social constructionist view of crime and criminal law, however, an instrumentalist role for social science conceived in such terms is ruled out. Can the traditional model survive suitably reformulated? Does it even need reformulation?

On the face of it, the holistic study of social practices of criminalization, given its difference in methodology from empirical social science based on a positivist conception of scientific method, challenges the traditional model. It still appears able to preserve the ‘division of labour’ and ‘research assistant’ aspects of the model. Holistic hypotheses and theories tend as a result to be at a high level of generality that defies immediate and straightforward empirical support. Practices of criminalization have been said to exist alongside and to relate in an intimate albeit complex way to a variety of other—political, economic, moral, psychiatric, religious, educational, familial—normative, labelling and sanctioning practices (Lacey et al. 2003, Chap. 1). But need such grand holism threaten the twin ideas of social science as the ‘research assistant’ to legal doctrine, and of a ‘division of labour’ in criminal law reform between doctrinal analysis and social science research?

Consider the following example. In late 2005 a new Section 162 of the Canadian Criminal Code came into force, which created a new offence of sexual voyeurism. Briefly, the section is aimed at criminalizing those who make or distribute surreptitious visual recordings of sexual nudity or activity in contexts where the person(s) filmed have a reasonable expectation of privacy. As a candidate or actual offence, sexual voyeurism fits the profile of being relevantly situated within a wide range of ‘normative, labelling and sanctioning practices’. Political: The section incorporates a ‘public good’ defence (s.162(6)), but its parameters are unclear. The imagined victims are clearly women and children and the imagined perpetrators men; thus all the familiar issues of power and gender are raised. Economic: the background parliamentary material to the creation of offence focuses a lot on the dangers and attractions of the commercial distribution of such visual recordings: there is a clear consciousness of the profitability of such distribution. Moral: there is a general undercurrent of moral prudery to the creation of the offence. Psychiatric: the possible status of voyeurism as a psychiatric disorder and thus not amenable to control by punitive means is clearly relevant. Religious: the clear influence of religious groups on the creation and form of the bill to introduce the section. Educational: not clear whether bona fide but unannounced research would be an offence under the section. Familial: Material knowingly created within an intimate context of the family or similar relationship, but later published by one member of such a context without the consent of the other or others, would not be an offence as the section stands. Thus, questions of power relationships within families become highly pertinent.

In a review of the proposed section when it was still under parliamentary debate, Eric Ward raised a number of questions about the section (Ward 2005): I have drawn generally on his discussion here. One concern is that we should have much more information on who the likely potential victims of the offence would be, to ensure that indeed there was a cognizable harm being committed. Another concern was that we would just as much need information on the likely potential perpetrators. Ward’s fear was that a large number of potential offenders would be teen-age males, and that there are serious questions about the wisdom of criminalizing such persons. A third concern was that, in comparison to similar offences in other jurisdictions, the Canadian version lacked any reference to consent, and thus became both over- and under-inclusive. A person could willingly circulate sexually explicit photographs surreptitiously taken of them by a partner and be guilty of an offence. A person could circulate without that partner’s consent sexually explicit photographs overtly taken of a partner and not be guilty of an offence. Ward recommends that attention be given to incorporating consent where appropriate.

The section was enacted with little or no change from the draft version that Ward was criticizing. So his proposals if brought forward again would have the status of proposals for criminal law reform. Notwithstanding the undoubted embeddedness of the offence within a wide range of ‘normative, labelling and sanctioning practices’, it seems clear that the research Ward’s questions implicitly demand is both relevant and conventional. Ward suggests, for example, that much of the activity of operators of CCTV cameras would fall within the definition of the offence, insofar as they might spend undue time focusing on areas where sexual activity might be expected. Yet government and corporate power being what it is, it seems unlikely that police would seriously trawl CCTV operating companies for potential offenders. Relatively conventional empirical academic research would reveal just whom the police are targeting as potential offenders. A Foucaultian sense of the modes of power might lead one to see the potential problem: but whether the potential problem was a real one would seem to be a matter of fact, not theory. Even a social theory holistically conceived in principle generates factual hypotheses whose accuracy can be checked.

Lacey has posed a different kind of challenge to the traditional model. On the traditional model, it is clear that criminal law doctrine takes priority over social science research—hence the idea of a research assistant. The reformers of doctrine take the lead in the formulation of hypotheses. As Lacey sees it, the methodological shift to a focus on practices of criminalization reverses this priority: ‘legal philosophers are indeed intellectually dependent on sociologists of law’ (Lacey 2000, p. 39). One might call this the TA model (Teaching Assistant), rather than the RA model. However, it is not clear what force this change really has. It has force against the RA component as the traditional model of the law interprets it—against the idea that the familiar doctrinal structure and content of the criminal law is fundamental and at its core unchanging. But if we abandon that idea in favour of an acknowledgment of the centrality of practices of criminalization and the fluid character of such practices, the insistence on the priority of the social seems now to have nothing to be insisting against. There is only value in insisting on the priority of the social over the legal or doctrinal if ‘social’ and ‘legal or doctrinal’ are clearly defined antecedent categories. The effect of decentering and deconstructing doctrine, however, coupled with the insistence on the ubiquity of theorizing, is to undermine the very idea that the ‘social’ and the ‘legal or doctrinal’ are in fact clearly defined categories.

The claim that legal philosophy is dependent on sociology of law is extreme. Even if it cannot be sustained, a more moderate claim, on the other hand, does threaten the traditional model of criminal law reform. Lacey herself in fact has more recently described the relation between analytic and sociological jurisprudence as one of mutual dependency (Lacey 2006, p. 948). Zedner has spoken of criminal justice as a ‘rendezvous subject, enjoying little coherence of approach and method but providing a meeting place where scholars of different disciplinary backgrounds, but common substantive interests, converge and converse’ (Zedner 2004, p. 34). These idioms and images imply more that the role of legal doctrine strictly conceived and of criminology or other social science is to be research collaborators than for either one to be the research assistant to the other. What are the implications of such a view?

These ‘mixed’ or ‘ecumenical’ models of the relation between legal doctrine and criminology force apart the two components of the traditional model of criminal law reform. They rule out any idea of a ‘division of labour’ in the conventional sense. That is, there would be nothing corresponding to the baker sticking to baking and the cobbler sticking to cobbling. Social insight and progress seems to come only from a blending of approaches and ideas, an openness to the theoretical Other. The roles of research director and research assistant, if you like, would be roles that any given discipline slips into and out of as needed. The roles could be abstractly separated and identified, but they could not be essentially associated with any given discipline.

Does that matter? Let us begin for a moment at the other end, as it were. A committed caricature postmodernist approach to social theory4 might respond in the following way to what I said about Ward and the crime of sexual voyeurism. The story told contains such key expressions as ‘commercial distribution’, ‘profitability’, ‘disorder’, ‘bona fide research’, ‘consent’, ‘family’, and the like. The story speaks as though these terms had determinate meanings, and indeed the terms must have determinate meanings if the compatibility of the holistic approach with the traditional model is to stand. But these terms do not have determinate meanings. They are themselves sites of contests of power. The struggle to define those terms is a political struggle of a piece with the struggle to establish the offence itself. There is no hard ground here exempt from political contestation. Or so some might say.

Is the ‘ecumenical’ approach focusing on roles rather than occupants of roles necessarily an instance of such ‘postmodernism’? Arguably, it is not. The fact that the role of proposer of hypothesis is not associated solely with the analyst of legal doctrine and that of investigator solely with the student of law as a social institution does not imply that there are no facts to be discovered, or no concepts to be analyzed. Suppose criminal law reform to be approached through such a collaborative mixing of disciplines—what I called ‘an openness to the theoretical Other’. Important gains are then made in matching criminal law reforms to ideals of criminal justice. The emphasis on practices of criminalization, whether or not such practices are interpreted strictly as practices of power, allows for a greater range of data to sustain and enrich investigation than could occur when the determination of data requirements came only from existing legal doctrine. The emphasis on roles as available to any discipline allows the output from the holistic approach to criminal law reform to feed back into the analysis of legal doctrine.

There is a major substantive difference as a result of the introduction of a focus on practices of criminalization, whether or not they are also seen as practices of power. The developments in social science that have led to the extended role played by theory in social science have broadened hugely the range of data, and the range of images, concepts and ideas, that may be brought to bear on criminal law reform. Some of these indeed do open up the way to appreciating the place of politics and power in the determination of who and what is criminalized and why. In terms of substantive content, criminal law reform should have a very different focus and meet a different range of interests than perhaps it traditionally has. Law—especially criminal law—continues to be what it has for some time already been: both a social, not a technical, institution, and one of society’s major tools of social ordering. How law orders society is of as much interest as how society orders law. It is in the end law that has to be seen as a social institution, that has to be seen holistically. Our practices of criminalization inextricably involve the law—legal doctrine and legal process—however broad the context that best illuminates them. To involve the law, however, is inescapably at some point to involve legal doctrine.


I began this paper with the gulf between, on the one hand, a traditional model of criminal law reform based on a focus on doctrine and on criminology as instrumentally related to reform, and, on the other hand, contemporary left-critical views of law reform as properly motivated only by a concern for the marginalized, the powerless and the dispossessed. It is not hard to discern an incompatibility between these views, in the sense that one does and one does not place doctrine in a central position. My project in this paper, though, was to go beyond the apparent incompatibility and investigate how far the repudiation of the traditional models rested on solid ground, and how far-reaching were the implications of any such ground if discovered. I focused the investigation on two questions. First, does acknowledgment of the socially constructed character of law entail or necessitate adopting a left-critical normative perspective on the criminal law? Second, what happens to the traditional model in terms of ‘division of labour’ and ‘research assistance’ if we take on board a social constructionist account of criminal law reform? The back-story to these questions is that I accepted the contention of critics of the traditional model of the law that the focus of criminal law reform should be social practices of criminalization, rather than doctrines of criminal law.

With respect to the first question, I argued that a left-critical perspective on criminal law and criminal law reform could be validly derived from a perception of practices of criminalization as practices of power, but not from the recommendation to focus on practices of criminalization in itself. On the other hand, the acknowledgment of the role of power in the sense understood here seemed left as observationally based, rather than derived from discursive argument.

With respect to the second question, the social constructivist approach to criminal law displaces legal doctrine from a position of centrality to criminal law reform. Therefore versions of the traditional model that depend on foregrounding doctrine must be abandoned. The most appealing picture of the relation between the analysis of doctrine and the study of social practices of criminalization, insofar as that relation is central to criminal law reform, is one of mutual dependence and collaboration. The idea of a ‘division of labour’ between doctrinal analysis and the social study of law cannot stand. The idea of ‘research assistance’ survives, but only in an abstracted form. ‘Research assistance’ names a role in criminal law reform that may be occupied by doctrinal analysis as much as by social study. The analytic study of legal doctrine—especially via what I termed the ‘moral soundness’ approach—has its place in criminal law reform, as long as it is responsive to, and if appropriate constrained by, critical criminology’s emphasis on social practices of criminalization. Critical criminology’s emphasis on practices of criminalization and practices of power also has its place in criminal law reform, as long as it can accept legal doctrine as one means, though not the end, of criminal law reform.5


Within the traditional philosopher’s approach a further distinction may be made between those of a formalist, or natural law-ish, or internalist, bent who believe that the relevant normative principles are internal to the concepts being analyzed, and those of a more externalist, or positivistic, bent, who take background political morality in some form to be the source of the relevant normative principles. This distinction, however, is not material to the present paper, however much it might matter in other contexts.


I am thinking here of Nicola Lacey’s distinction between legal and social constructions (Lacey 2002, p. 264).


As a referee for this journal pointed out, there are many varieties of critical criminology that would place other concepts such as force, desire, violence, symbolic identification at the core of critical criminological analysis. So it would not do for this paper to identify a focus on power with critical criminology as such. Nonetheless, the focus on power is characteristic of influential and appealing versions of critical criminology. With the above qualification noted, I shall continue to foreground it in this paper.


I leave it to the reader to decide how accurate the caricature may be.


Earlier versions of this paper were presented to the Political Theory seminar at the University of York, the Jurisprudence Discussion Group at the University of Oxford, the Department of Philosophy at the University of East Anglia, and the Canadian chapter of the IVR at the University of Saskatchewan, May 2007. I am grateful to all these audiences for criticism and discussion, especially to Nathan Brett’s commentary on the latter occasion. I also thank an anonymous reviewer for this journal for valuable criticism. My debt to Nicola Lacey’s work is obvious, and although she has encouraged this paper she should not be assumed to agree with it. The paper grew out of work for the Law Commission of Canada in 2005–06 in relation to their research project ‘What Is A Crime?’. I would like to thank Steven Bittle, Senior Research Officer for the Commission in charge of this project, for many conversations in person, by phone and by e-mail on the topic of the project: I learned an enormous amount from these conversations, although he too must not be supposed to agree with what is said here. The research was supported by the Social Sciences and Research Council of Canada, grant no. 845-2005-003: I thank the Council for their support. The paper was written during tenure of a H.L.A. Hart Visiting Fellowship at the Centre for Ethics and Philosophy of Law, University of Oxford. I am grateful to the Master and Fellows of University College, especially John Gardner, for their hospitality, and for providing a unique environment for philosophical reflection.


Copyright information

© Springer Science+Business Media B.V. 2009