Introduction

Left Realism in Britain originated as a call to socialists to 'take crime seriously' (Lea and Young 1984, Young 1986) under conditions in which criminality and other social problems facing the working class were worsening, while many radical criminologists remained obsessed with a social constructionist view of crime as simply a reflection of media orchestrated moral panics or political diversion (Hall et al. 1976). However such an inspiration to take a problem seriously does not of itself constitute a new or an adequate theorization of the problem. The survival of Left Realism as a coherent criminology is dependent on its capacity to elaborate a body of theory which distinguishes it from other perspectives.

The development of theory in Left Realism in Britain in recent years has been largely concerned with the elaboration of a framework in which crime is analyzed in terms of the interaction between four key variables: the state, social structure, offenders and victims. This framework has enabled Left Realism to establish a relationship with existing traditions of criminological theory by seeing much of the latter as partial perspectives focusing on particular elements of the interaction rather than the process as a totality. Thus labelling theory tends to focus on the activities of the criminal justice system or society in labelling activities as criminal, classicism on the deterrent effects of the criminal justice system, much conservative theory on the efficacy of informal social control, positivism on the offender, and victimology on the characteristics of the victim. Each of these approaches may pose as a self sufficient criminology whereas each of them starts from an abstraction: one element of what is a more complex multi-faceted process. (Young 1987).

Such a framework is not presented as some theoretically neutral process of 'synthesis' of existing theory. The partiality of the latter is obviously only established from the standpoint of the more general interactionist framework of state, society, offender, victim. This framework itself is not immune from further elaboration and critique. It does not pose as a set of a priori categories defending themselves as the only logically possible coherent method for the study of crime. Rather it is a reflection of the existing division of labour in society and its legal system. The process of further theoretical development is not therefore simply a question of developing a more adequate set of categories but also of social change to which the criminologist both as theorist and policy maker is in no way external. The distinction for example between state and society is not an a priori but a question of the structure of actually existing industrial societies. It is in no way external to the realist framework that the blurring and shifting of this distinction is a feature of modern societies especially in the criminal justice area. 

Defining the terms

It is necessary to give some preliminary definition of the concepts which form the framework of Left Realist analysis. By the 'state' is meant both the criminal justice agencies in the narrow sense and the state as political system in the more general sense. The latter is an important conduit for the reception from and transmission to, other elements of the framework. In a similar way the term 'society' implies not only the micro-levels of family and community structure which may act as mechanisms of informal social control against crime but also the wider concept of 'civil society' as a set of legally, culturally and economically defined relations.

Offenders and victims are only at first sight simple concepts. The offender, it has to be remembered, is not necessarily an individual but may be a corporation a political group or other organization. Likewise victims may be chosen as individuals or as members of groups defined in various ways. They may or may not be aware of their victimization or have accurate information concerning its source. 

Action and reaction.

What relations exist between the various components of the framework? Young (1987) identifies relations of 'action' and 'reaction' in which the state and the system of social control are structures which 'react' to the 'action' of offenders and victims, redefining their activities, devoting resources to their containment etc. and thereby playing an active role in the 'production' of the final level of crime in society. This relationship is an attempt to transcend the conflict between on the one hand varieties of labelling theory which see changes in crime rates as a result of some combination of the activities of criminal justice agencies and social perceptions, and on the other hand a positivism which speaks only of 'real' changes in the numbers of offenders and victims.

This approach can made more concrete by seeing both sides as engaging in action and reaction. Society and the state do not simply respond to the problem of crime but also engage in various pre-emptive activities in the causation and definition of crime. Likewise the behaviour of both offenders and victims can be seen as not only constituting 'problematic situations' to which the state and the system of social control may react but also themselves responding and reacting to the state and the system of social control. It is of course the case that society and the state are determinant 'in the last instance' in the sense that victims and offenders do not exist except as recognized by some combination of legal and social definition whereas the state and social structure are not derived from the existence of offenders and victims [1]. 

THE STATE AND SOCIETY

A. The existence of crime: the process of criminalization.

The creation of the categories of criminality is obviously the most fundamental action of both the state and society. The criminal law can be seen as a language describing certain types of social actions and providing legitimation for the existence and activity of state agencies in their relations with individuals as offenders, victims, witnesses, holders of relevant information etc. In addition to the content of the law the rules of procedure for the establishment of culpability themselves help determine the parameters within which social problems can be dealt with through a strategy of criminalization. The fact that criminal justice systems largely deal with relations between individual agents rather than social groups per se, the rules of presentation of evidence, proof in criminal trials as well as those governing the conduct of the trial (such as rules of cross examination) all serve to determine what types of problems can be effectively dealt with by the criminal justice system.

Within a given legal code and its rules of procedure the agencies of law enforcement and justice enjoy a considerable leeway of discretion. They do not simply respond to crime but in making decisions about what types of crime to respond to and by what means, they act as a positive determinant of the pattern of criminality, There is thus a 'de facto' criminalization of activities by state agencies, which is of equal importance to the formal legal status of actions as crimes. The degree of correspondence of the two depends on the constitutional and actual accountability of the agencies, the forms of operational discretion available to them, such as the autonomy of judges and prosecutors, the operational independence of police chiefs, the day to day the relationship between criminal justice agencies and various social groups and the structure of criminal activities themselves. There is what might be called 'over criminalization'. associated with the development of particular operational policing strategies. in which whole communities are treated as if they were offenders, through for example, the use of random 'stop and search' techniques (Lea and Young op.cit.). Likewise criminal courts may tend to prejudge guilt of offenders from particular social groups. A reverse process might be called 'under criminalization' whereby some types of crime (e.g. domestic violence, neighbour disputes) are defined by police as 'rubbish crime' and not really worth dealing with, certain areas of the city may go relatively unpoliced etc. or offences by powerful corporations may go unprosecuted.

Alongside the state agencies and the language of criminal law, society also has its language which define particular behaviours as crime. The extent to which the language of ordinary social interaction overlaps with that of the criminal law is complex. Some abolitionists (Hulsman 1986) like to imagine that the criminal law is quite redundant and that society has its own language of 'problematic situations' which is autonomous from and superior to that of the criminal law. It does not seem likely that the language of interpersonal disputes can replace that of the criminal law when ordinary citizens may not be able to agree on spontaneous definitions of situations (e.g. when does physical violence constitute assault) or be otherwise confused as to the nature and extent of their victimization.

In democratic societies the criminal law might be expected to simply reproduce and echo social sanctions. But where society is divided by structures of class and power the criminal law can be both a progressive and a reactionary force; It may import the language of criminality into some areas where ordinary language, reflecting established power relations of class or gender, is blind to the existence of conflict and harm while in other areas it may turn a blind eye to situations criminalized by ordinary discourse.

Indeed rapid social change may produce conditions in which the same activity is both criminalized and to some extent still sanctioned by social values. For example the morality associated with traditional family structures may still tolerate a certain amount of violence by husbands against wives even while the advance of democratic rights, and the political influence of feminist movements combine to sensitise society to the rights of married women as citizens to defence against violence from their husbands. These conflicts are reflected in the working of the criminal justice system with the result that a formal acknowledgement of a problem is often combined with a practical reluctance of agencies to act effectively despite a formal requirement to do so. 

Reaction: the response to changing social problems

The process of creation of the categories of criminality by the interaction of state and society involves, in modern industrial societies, a process of constant reaction and adaption to social changes. While there may be considerable stability in general ideas of crime or harm, there are substantial changes in how, and to which activities, such categories are to be applied. Social and political movements may result in simultaneous pressure on the state to expel the concepts of criminality from some areas (e.g. soft drugs) and bring it more strongly into others (e.g. marital rape) or, on the level of the day to day activities of criminal justice agencies, devote more resources to the containment of some crimes and less to others. Such dynamics, unfortunately for abolitionists, do not seem to result in any general reduction in the role of the criminal justice system the containment of social problems. 

B. The existence of offenders and victims.

Having established the categories of criminal behaviour both the state and society respond and attempt to contain criminal activity itself. But at the same time the creation of such activity, for all but the most hardened biological positivist, is a product of society. The fact is that modern industrial societies put as much effort, albeit unconsciously, into the creation of crime as they do to its containment. 

Action: the production of potential offenders and victims

There are a number of respects in which criminal activity has to be seen as a product of the normal working of state and social processes rather than their 'breakdown'. Even the commission of crimes by the criminal justice system itself should not be thought of as solely a characteristic of authoritarian states or 'states of emergency' in which normal legality is suspended. Violation of criminal law by state security agencies, albeit concealed, appears to be normal enough activity even in liberal democratic states. (Wright 1987, Chambliss 1989).

More important for the study of 'ordinary' crime however are the unintended consequences of the normal functioning of criminal justice institutions of which the most obvious example is that of the prison as the producer of crime through its latent function as a channel for the socialization of the inmate into established criminal subcultures. Other examples are to be found in the unintended consequences of the activities of law enforcement agencies. Young described the 'deviancy amplification' effect whereby tough policing strategies against soft drug users drove the latter further into criminal activity (Young 1972). More generally, one of the effects of the 'overcriminalization' mentioned above may be that of legitimizing offences which are seen by local communities as resistance to heavy and indiscriminate policing.

However it is predominantly at the level of social structure rather than state institutions that the major causes of criminal activity are to be found. It is important to note that Left Realism has not hitherto attempted to elaborate any new theory of the causes of crime; rather the strategy has been that of elaborating and adapting existing bodies of theory into the 'action and reaction' model. A major influence has been the theory of anomie or relative deprivation derived from the classic contributions of Merton (1938) and Cloward and Ohlin (1960) combined with a theory of social and economic marginality. Relative deprivation, a conflict between socially diffused goals and needs and the restricted availability of the legitimate means for their achievement would likely be resolved by a turn to criminality among those groups additionally marginalized from participation in the political processes of modern industrial societies. (Lea and Young op.cit.).

Serious critics of Left realist criminology have pointed to the restricted applicability of the theory to particular areas of crime. Street crime such as household burglary, shop lifting, and street robbery are those which come most readily to mind when considering relative deprivation as 'illegal means to socially sanctioned goals'. However relative deprivation theory as developed by Cloward and Ohlin (op.cit.) and others saw the causes of delinquency not so much in an instrumental response to deprivation as might be evidenced by engaging in burglary as an income supplement but in the development of a subculture in which alternative values develop precisely as a way of coping with the frustrations of exclusion from legitimate routes to success. In this sense the role of interpersonal 'expressive' violence, or ritualised forms of conspicuous consumption as ways of establishing status in the absence of conventional means and symbols can be understood. Also 'crimes of passion' such as homicide and interpersonal violence such as sexual assault and rape have a concentration among the poor and deprived and can be seen as arising from dynamics of relative deprivation. Box describes the dynamics of rape by men from poor and deprived backgrounds:

  "When men from this latter group rape they rely primarily on physical violence because this is the resource they command. Being relatively unable to 'wine and dine' females or place them in a position of social debt, and being less able to induce in women a sense of physical and emotional over-comeness these 'socially' powerless men are left with a sense of resentment and bitterness which is fanned and inflamed by cultural sex-role stereotypes of 'successful' men being sexually potent."(Box 1983 p.152) [2].

However there are important areas of criminality to which theories of relative deprivation and marginality might be thought inapplicable. "What does anomie contribute to the analysis of organized crime or of so-called white collar crime?" asks Tamar Pitch (Pitch 1987). In fact quite a lot, at least as regards organized crime, though it is true that Left realists have as yet to do substantial work in this area. It is worth remembering that, as Arlacchi has recently noted (Arlacchi 1987), one of the impulses to the work of relative deprivation theory, especially at the time Merton was developing his account in America, was the attempt to combat the widespread 'alien culture' view of organized criminality as associated with Italian immigrants rather than the internal dynamics of American society. Relative deprivation theory presupposed socialization into the values of mainstream american culture. The adaption of relative deprivation theory by Lea and Young (op. cit.) was for a very similar purpose; that of rebutting the media image of the 'alien culture' of young blacks as a cause of street crime in British cities.

However it is true that a too close association between relative deprivation and socio-political marginality leads to a restricted focus on disorganized street crime. Organized crime may originate in and recruit from social groups marginalized from legitimate opportunity structures but once established is able to secure a measure of stability in its environment through protection, bribery, threats of violence etc and a degree of penetration of legitimate financial and political institutions (Catanzaro 1988, Santino 1989). To deploy the term 'marginality' at this stage is purely rhetorical and means little more than that the activity is illegal[3]. Indeed successful organized crime as a structure of illegal opportunities for the accumulation of wealth and status is better understood in this context as a form of demarginalization.

This analysis seems to be in accordance with Cloward and Ohlin's classic treatment in which a distinction between the criminal career and the criminality of the disorganized -i.e. marginalized slum is stressed. "Among the environmental supports of a criminal style of life are integration of offenders at various age levels and close integration of the carriers of conventional and illegitimate values" while "Just as the unintegrated slum cannot mobilize legitimate resources for the young neither can it provide them with access to stable criminal careers for illegitimate learning and opportunity structures do not develop" (op.cit. pps 162,173).

White collar and business crime paradoxically present less problem for the theory of anomie and relative deprivation than does street crime. As far as the latter is concerned one of the main criticisms of the Mertonian approach was that of the plurality of values in the cultures of modern industrial societies and the consequent implausibility of attributing criminality to innovating attempts to achieve the goal of economic success (Lemert 1967). This criticism has suffered from declining plausibility with the expansion of mass media and education disseminating standardized expectations. But as Box remarked the modern capitalist corporation never suffered from the problem of ambiguity concerning its goal of profit maximization. "This defining characteristic -it is a goal seeking entity - makes a corporation inherently criminogenic for it necessarily operates in an uncertain and unpredictable environment such that its purely legitimate opportunities for goal attainment are sometimes limited and constrained" (Box 1983 p 35, Passas 1990). At this level marginality becomes, in Stan Cohen's phrase. a 'mickey mouse concept' (Cohen 1985) Worse, it can fulfil an ideological function in denying the criminogenic nature of the capitalist system itself by implying that crime is always only a product of breakdown or absence of 'normal' social or political structures whereas in fact for powerful corporations violation of the criminal law or pressure on governments to legalise activities technically 'criminal' are strategies often pursued in combination (Pearce 1976)

The context of white collar and corporate crime places emphasises that relative deprivation is not to be confused with material poverty. The frustration of inability to meet expected goals and norms of 'success' is diffused throughout all social classes in a capitalist society. The young middle class business executive who takes out his failings in the competitive arena by violence towards his wife or children, or in bouts of public drinking is no less a victim of relative deprivation than the inner city mugger. The emphasis here is on 'relative'

However this brings us to the second weak spot identified by critics: the question of the precise mechanisms whereby high levels of relative deprivation and, in a subset of cases, marginality are linked to the commission of criminal acts (Pitch 1987, de Leo 1987). An account of the production of potential offenders has to be supplemented by an account of the production of criminal opportunities. The theory of relative deprivation precisely because of the wide range of its application cannot explain the particular types of crime that occurs unless supplemented by a theory of criminal opportunities into which those suffering from such deprivation may move.

The study of crime as a structure of opportunities has been one of the preoccupations of rational choice theory. While some criminologists -economists in particular- have attempted to portray rational choice theory as a self sufficient general account of criminal behaviour such an approach usually find it necessary to make a prior distinction between those who would or would not take advantage of -or seek out- criminal opportunities. Thus Becker's classic analysis of rational choice theory applied to criminology (Becker 1968) having claimed that "...a useful theory of criminal behaviour can dispense with special theories of anomie ...and simply extend the economist's usual analysis of choice" proceeded to add 'willingness to commit an illegal act' as a variable in the formula for the calculus of rationality of criminal action. The best approach seems therefore to regard accounts of the production of potential offenders and accounts of the production of criminal opportunities as essentially complementary and interdependent aspects of the process whereby society produces crime.

Two key components of the criminal opportunity structure can be distinguished. Firstly the existence of a demand on the part of both criminals and the non criminal public for illegal commodities or services (drugs, vice, arms etc) or for legal commodities illegally acquired: that is, the existence of the criminal economy. Secondly, a source of legitimate - from the standpoint of offenders - victims who are 'fair game' as targets for crime or situations in which criminal violence is seen as justified. In this respect it is a mistake to see the 'criminal subculture' or the 'subculture of violence' as entirely isolated from dominant values. The dominant culture provides a set of labels which, while not directly legitimizing crime, enable the offender to elaborate his own legitimation. Thus the murder of a prostitute is considered less socially obnoxious than that of a 'respectable ' woman [4]. Social groups defined as 'outsiders' may become particular targets for crime (e.g. racial attacks). At the level of corporate crime particular state agencies such as tax authorities or other national states, particularly in the third world may be seen as legitimate victims for corruption, low safety standards, dumping of hazardous products etc. (Braithwaite 1984, Jones 1988).

There is often a close relationship between the legitimacy of victims and their vulnerability. The former may be a cause of the latter as where racial prejudice legitimises violence against ethnic minorities, maximizes vulnerability by assuring them that the victim is likely to get little support -e.g. from neighbours in cases of racial harassment on housing estates and at the same time relegates ethnic minorities to the most criminogenic areas of the city. The two may run parallel as where the weak law enforcement capacities of some third world states parallels their legitimacy as victims in the eyes of multinational corporations. The two may also conflict as with elderly people in large cities living lonely isolated lives with a high vulnerability to crimes such as burglary and mugging or young women having to walk home during unsocial hours with a high vulnerability to rape though in both cases such crimes are regarded as particularly vicious by society at large.

A final important component of legitimation of victims concerns the attribution of responsibility for the criminal act itself. Victim blaming and 'victim precipitation' can be regarded as the polar case of the legitimate victim in which the latter rather than the offender comes to be regarded as the cause of the crime. Such an ideological construct needs to be carefully distinguished from the fact that many offences may arise out of prior interaction in which both victim and offender participate. 

Reaction: deterrence, control and support for victims

Both the state and society react to crime, attempting to control and contain that which they have produced. An orthodox distinction would see the state as the sphere of law and legitimate force aimed largely at detection and deterrence with society as the sphere of informal social control and prevention. This distinction is in fact blurred in a number of ways. The operation of state agencies even at the most 'technical' level rests on assumptions concerning social relations. The state itself may attempt to initiate, and organize informal crime prevention measures by citizens ranging from surveillance, dispute resolution, to 'community courts' (Matthews 1987). The motives of particular governments in taking such initiatives may include those of attempting solutions to what are considered growing legitimation or fiscal problems faced by criminal justice agencies. Finally the boundaries between state and private institutions become increasingly blurred through the development of intermediate structures of surveillance and control (Cohen 1985)

The deterrent effect of the criminal law cannot be considered independent of its relationship to collective moral sentiment. Where the criminal code reflect moral sentiments widespread in society then the simple fact of illegality itself has a deterrent effect, although this may be mediated by factors such as the general social status of the offender and the absence of feelings of guilt or shame (Braithwaite 1989) with a consequent reluctance on the part of the criminal justice system to act forcefully in such areas. (Johnson 1986). While the severity of the penalty may have some deterrent effect attempts in research to specify this have been inconclusive. (Cook 1980). What is reasonably established however is that it is less the penalties for particular offences prescribed by the law than the certainty of detection which constitutes the bulk of the deterrent effect. The certainty of detection itself has both technical and social components. The social component, concerns the general willingness of individuals to give information and report crime to the police, appear as witnesses in court etc. This depends on the relations between police and local communities, on the relations of power between victims and offenders and on the socio-economic structure of localities determining such matters as the daily movement of population and the likelihood of crimes being witnessed. (Cohen and Felson 1979).

An important social and political component of crime detection concerns the nature of terrain on which they are committed, The political power and surveillance capacities of a state are not necessarily evenly spread throughout its terrain. In third world countries the de facto power of the criminal justice system may be concentrated in urban centres. In the large cities of modern industrial societies economic changes over the last fifty years, have produced an 'internal periphery' in the older urban areas in which high unemployment, political and social marginalization, alienation from both criminal justice and welfare state agencies, high rates of crime and other social problems are concentrated. Police operations in both types of areas may frequently take the form of episodic military expedition rather than regular patrolling.

Analogous problems arise when the terrain over which the crime takes place exceeds the geographical boundaries of the state. Modern computer assisted financial crime usually involves sophisticated movement of funds on a world scale and tracing the activities of criminals e.g. the 'laundering' of profits from drug traffic may involve cooperation between the criminal justice systems of several nation states. The differences between states both in terms of the content of law and in the efficiency of criminal justice agencies may enable certain types of offenders to select the terrain on which they operate or focus the illegal aspects of their international activities (Braithwaite 1984).

The technical questions that concern detection are numerous. Money is difficult to trace especially if it is nothing other than electronically stored information. Unique antique art objects are more easily traceable. Between these polar cases stands the vast bulk of 'mass private property' of motor vehicles, televisions etc., which in their identity may present problems of traceability akin to that of money. Secondly there is the process whereby the crime is committed. In modern business fraud the specialist skills required to detect and assemble the evidence of the crime are considerable and pose a problem for police forces considered as generic multi-task bureaucracies. A different type of complexity can be seen in crimes committed within the family such as child sexual abuse where the evidence -may be ambiguous and depend on medical judgement. In both areas policing has become increasingly 'multi-agency' involving experts and agencies from outside the criminal justice system.

A crucial question both for criminological theory and policy making is the extent to which the average 'solvability' of crime is decreasing. Technical questions of solvability may in fact be social factors in disguise. How much of the alleged -by police- decrease in solvability of household burglary (Newman 1985) is due to the untraceability of mass private property and how much to the fact that the state of police community relations reduces the flow of information from public to police is difficult to assess. Nevertheless the hypothesis of an increase in the amount of crime coupled with a general decline in solvability would certainly account for the recent popularity of the preventative approach to crime control. Other 'technical' problems are ultimately questions of social relations. The complexity of computer bank frauds is in the last instance a question of the secrecy and discretion dictated by the regime of competition in industrial and financial capitalism. The problem of detecting stolen 'mass commodities' like TV sets or video recorders is a question of the development of an economy of illegal consumer goods into which such items once inserted, become just commodities like any other, and the 'technical' problem of diagnosing child sexual abuse is at root one of the relations of power and collusion in family relations.

Like crime detection, prevention has both social and technical dimensions. What all preventative strategies have in common is the attempt to prevent the activity of criminal offenders by means of measures which apply to all individuals irrespective of intention. Traditionally prevention strategies have been conceived of as technically based and concerned with 'target hardening' which consists in the identification of potential targets (usually buildings, vehicles, safes, computer programs etc) and making them generally difficult to penetrate without the cooperation of the appropriate authority. Timelocks, entry phones, computer passwords are all of this nature. A large part of the skills of the professional criminal are concerned with the ability to circumnavigate such obstacles. Screening focuses on the people entering or leaving a particular area or area or activity which is vulnerable to crime. Employees leaving a workplace may be subject to random searches as a measure against pilferage, passengers entering commercial aircraft will have their baggage searched as an anti-terrorist measure. Many technical 'target hardening' devices provide an automatic opportunity for screening: the owner of an answerphone system can refuse entry to those who arouse suspicion. Again, the skills of the criminal are largely dedicated to outwitting such measures. Burglars may penetrate answerphone systems by posing as public officials. Terrorist bombs made out of non-detectable explosive carefully disguised as cassette recorders may penetrate baggage control systems. General surveillance and visibility may be increased by the use of video cameras e.g. in shopping precincts, railway stations etc, by improving street lighting o increase visibility, by architectural changes to remove some of the spatial opportunities for crime mentioned above, or by attempts to mobilise residents in a local neighbourhood into 'Neighbourhood Watch' Schemes to keep an eye open for suspicious circumstances and increase the flow of information to the police.

The operation of technical crime prevention measures always has a crucial social element. The numbers of people and the nature of their use of public space determines the way in which geographical and architectural factors may act as a deterrent to crime (Jacobs 1962, Newman 1972, Coleman 1986). A newly installed system of entry phones and door locks may, by impeding movement, increase the isolation of elderly people and so escalate their fear of and their vulnerability to crime. A sophisticated factory or airport security system may easily be undermined if the quality or working conditions of the labour force does not match the sophistication of the equipment. Neighbourhood watch schemes designed to increase surveillance will fail if the majority of people are absent from the area during the times when crimes are most likely to be committed (Cohen and Felson op.cit.), or simply do not look out of their windows. Increasing the intensity of street lighting enables rapists and muggers to see their victims more clearly every bit as much as it enables ordinary citizens to see where they are going. A purely 'technicist' attitude to crime prevention is incapable of understanding why the same technical innovation works on some occasions and not others because it fails to grasp how technical 'effects' are always mediated through the social relations between people. (Lea 1986)

Social crime prevention so called is a more recent innovation. Its basic aim is to construct or reconstruct as a deliberate aim of social policy the type of social relations, usually at a micro-level within a neighbourhood of a city, that are thought to maximise social pressure against crime. Neighbourhood Watch schemes are seen to have a social cohesion effect as much as an opportunity reduction effect by increasing the occasions for neighbourly cooperation in the production of newsletters, and other means of disseminating information. Such informal bonding increases peoples' confidence in their locality they are more likely to be out on the streets and take an interest in what is going on and hence increase the informal controls against crime. Even the state agencies such as the police by their very presence (irrespective of their crime detection activities) can be seen to contribute to a sense of public security and so help create a self reinforcing process of greater public sense of security and diminishing crime (Wilson 1975). Some governments have attempted to infuse an element of legal coercion is the creation of such informal control. Obviously the 'neighbourhood' or 'community cannot be an object (yet) of legal regulation but the family can. Proposals to increase the responsibility of parents for the conduct of their children will, it is thought, give some legal backing to the 'reconstruction' of informal social control. A final dimension of social prevention is the tendency in recent years to identify a crime prevention aspect to almost any activity that occupies people harmlessly, as with youth clubs, summer camps etc.

At the other end of the criminal spectrum preventative strategies have been deployed as strategies against corporate crime. The granting of licensing powers to business trade or financial investment associations with the associated activities of surveillance and monitoring is regarded as an effective, if not preferred method of crime control to that of criminal investigation (Clarke 1983 Levi 1987). It is here that we should expect to find some of the problems of strategies of social prevention clearly illustrated. In the sphere of corporate crime those entrusted with the role of prevention are not clearly differentiated from those likely to engage in criminal activities. The question is not so much that of positive collusion as the failure of a normative system of control to impede criminality because it implicitly sanctions it (Pearce 1976). This is illustrated in the duplicity of the corporate criminal offender's role the broader social structure in which he also appears as an exemplar of the good law abiding citizen (Solivetti 1987).

In the prevention of 'street' crime analogous problems prevail. Neighbourhood watch schemes are biased in the direction of household burglary and vehicle theft with an inbuilt bias to view the offender as an outsider to the neighbourhood (LRET 1986, Rosenbaum 1987) Such schemes ignore crimes such as domestic violence because the latter are shielded by conventions of non-interference in the private sphere. Crime prevention discourse also frequently reinforces the separation increasingly arbitrary, between street crime and organized crime, by imagining a world consisting only of the former. Against organized drug dealing networks neighbourhood watch may be rather ineffective.

In other negative ways crime prevention strategies may reproduce the patterns of social inequality and victimization which they seek to combat. Those areas of highest rates of street crime are the hardest areas in which to establish preventative initiatives precisely because such areas also suffer heavily from other social problems like unemployment, poverty, poor housing etc. Physical crime prevention may therefore fall victim to general decay of housing stock and streets while the lack of a sense of community, the prevalence of ethnic conflict etc may impede either participation or make it highly unrepresentative. The long run effect may be simply the displacement of crime into the poorest and most socially disorganized areas of cities where crime prevention is ineffective. Meanwhile in the area of corporate crime regulatory bodies may end up simply displacing crime (e.g. dumping unsafe products, violation of safety standards etc, to those areas of the third world where all such 'surveillance' is powerless to act.

The state reacts not only to offenders but also to victims. It reacts most effectively to those considered as illegitimate victims though in this respect state agencies many not necessarily duplicate definitions of the victim or prioritisation of forms of victimization widespread in society. This may be both due to the particular class and group interests or cultural values that predominate in the state and to forms of rationality which determine the distribution of resources, Where the state does not recognize those forms of victimization defined as such by significant social groups then alternative forms of victim support are likely to develop e.g battered women's refuges, rape crisis centres. These may then in turn become the basis for the organization of social movements around victimization to change the priorities of state activity. 

OFFENDERS AND VICTIMS

It is one dimensional to see crime and victimization as simply the product, or end result of a particular set of social or other determinants. A great deal of criminology has been preoccupied with the aetiology of crime. This is only one side of the 'square of crime' The sociology of crime cannot rest at the sociology of the production of crime but must recognize the relative autonomy of crime both as a subcultural phenomenon and as a form of economy with its own dynamics of reproduction and change and capacity to effect not only its immediate victims but society and the state in general. Just as the state and society can react and respond to crime, so can crime react and respond to the state and society. 

Relations between offenders and their victims

Entirely random victimization is rare: the crazed gunman who walks down a street shooting at random is not the typical criminal offender. Most offenders have some social or economic relationship to their victims. They may live with them as in domestic violence or spousal murder, they may live in the same neighbourhood as with most opportunist street crime or street violence, they may identify them as an economically suitable target in terms of the aims of criminal enterprise as with bank robbery or commercial fraud, or they may be in some economic market relationship such as that of customer or competitor. An understanding of the relations between victims and offenders is crucial to explaining both the perceived impact of crime and many of the problems associated with its detection. A useful way of classifying such relations is in terms of whether the offender is an individual or an institution and whether the victimization takes place as a result of direct intention or is indirect; that is to say it occurs as the unintended result of the pursuit of other goals. The most obvious case of victimization where the offender is an individual and the victimization direct is murder or physical assault. Indeed 'crime' as such is frequently thought of as predominantly actions of this type. Domestic burglary involves a less direct relationship between victim and offender -they may not meet, but the offender is an individual and the offence is intended and the target is subject to a degree of selection.

By contrast much corporate crime so-called involving the offender as an organization involves indirect victimization[5]. The ignoring of safety regulations is undertaken from the normal corporate goal of profit maximization and the assumption is generally that the organization can successfully function without them. That is not to say that corporate decision makers are unaware of the likely victimization consequences of their actions. They may quite cynically calculate that 'if' people die as a result of corporate actions few consequences will fall on the corporation because of the dependence of the government e.g. in a Third World country, on the presence of that corporation. Alternatively a calculation may be simply made of the chances of 'accidents' occurring and risks taken. Here corporate crime begins to move towards direct victimization where a company quite deliberately markets a product it knows to be unsafe as with the notorious Ford Pinto case (Cullen et.al. 1987), or engages in the bribery of governments, assassination of trade union organizers etc. Individual victimization can of course be indirect as in drunken driving manslaughter or other forms of victimization resulting from individual failure to anticipate the consequences of actions.

Victims, like offenders may be individuals or groups. The individual victim of rape or murder stands out in the community and whether he or she is shunned or supported by the immediate community depends on many factors noted already. The victim may be seen as chosen only because of membership of a group -as in racial harassment. This is likely to maximise community support. In other cases where a group is being victimized, e.g. the consumers of contaminated products the spread of victimization may be a factor in hiding, at least for a period, some of its impact since individuals are unaware of differences in their own situations compared to those of other similar people. 

Action: the impact of crime

The impact of crime is perceived not as some simple observable body of facts but through the interaction between society, the state and victims themselves. State agencies seek objective measurable criteria such as average likelihood of victimization, average value to property lost in burglary or criminal damage etc. The question of resource distribution between alternative areas of social policy creates a pressure towards a form of measurement in terms of which the impact of a vast array of social problems can be compared. Translation of the impact of crime into monetary terms enables governments to make comparable estimates of the 'seriousness' of different crime problems and of crime generally in relation to other problems competing for public resources. However there are good and bad ways of making such calculations. Statistical averages over a whole country can be misleading because different types of crime are unevenly distributed and have different types of impact. Such calculations as 'the average citizen stands a chance of being a victim of crime one every four hundred years' and 'the average property lost by burglary and theft is £50' can become part of a political strategy to minimise the impact of crime and society under conditions in which it is the poor and marginalized who are the major victims of crime and the same value of property stolen from a poor person obviously has a greater impact than if stolen from a wealthy person (Kinsey et.al. 1986).

However the opposite strategy of allowing the victims to define impact of crime through simple responses to social survey questions has drawbacks and may in fact embody a similar relationship between victim and state - one of marginality and lack of political power - as that embodies in statistical averages of property values lost in burglary. Social surveys assume that victims are conscious of their victimization. In terms of the classification of victim - offender relations outlined above it is obviously the direct and individual forms of offending that victims will be most aware of. This can result in ignoring many forms of corporate crime of which victims may be unaware in the short run or which may appear as simply 'accidents' to anyone not familiar with legal regulations or the processes of corporate decision making and where the extent of victimization - as for example with illegal environmental pollution - may only become clear over a number of years.

The impact of crime on victims is closely allied to the question of the fear of crime. State researchers making statistical calculations of 'risk rates' in the manner noted above often come to the conclusion that citizens have an 'exaggerated' fear of crime in terms of their 'real exposure'. Quite apart from the problems of using statistical averages already mentioned a crucial issue concerns how far the victim is seen as idiosyncratic or as a typical member of a social group. Where the victim is regarded as 'legitimate' -in the sense previously discussed- or as having provoked the offence him or herself then the impact of the crime on others in terms increasing the fear that similar victimization will occur to them also will be minimized. Fear will be maximized not only where the victim is seen as representative or typical but where the offence is seen as linked to other offences. Women as a group are more frightened of burglary than men, not because they are more irrational at calculating risk rates than men but because for a woman, encountering a burglar may involve also the risk of rape or sexual assault whereas more men may feel they could scare off the offender. Women thus tend to have more similar rates of fear across the spectrum of different types of crime than do men (Warr 1985) and they are quite rational so to do. This is not to say that fear of crime. or lack of it. is always rational. Where the impact of crime is not immediately grasped as such it will not be feared. The irrationality of victims fear is more likely to err on the side of underestimating crime by ignoring the impact of corporate crime than it is to exaggerate the importance of street crime.

Fear of crime has its own impact the most important of which is a de facto curfew on movements around the city. especially after dark coupled with feelings of insecurity even within the home. The unintended consequences of fewer people on the streets may well be to contribute further to the social decay of the area and to increase the likelihood of street crime (Wilson 1975).

In conclusion neither impact of crime nor the rationality of fear can be adequately measured by the most frequently used techniques of average risk rates or victim surveys. Both may underestimate the real importance of crime in people's lives unless they are seen as one, albeit important, input to the democratic process where such exists, that is most capable of enabling ordinary people and their elected representatives to become aware of the extent of crime and its real impact on society. In all societies in which, even with the formal trappings of democracy the media of communication are monopolized and distorted and even fear of crime becomes itself a means to other political ends, this will not be achieved. 

Reaction: victim mobilisation

A cardinal principle of social stability in modern industrial societies embodied in the criminal law is that it is the state rather than the victim that responds to crime. The victim in most criminal legal codes exists as witness for the state. Crime prevention strategies can be pursued by anyone because they are aimed at citizens in general as long as they are not of a punitive nature and do not infringe civil liberties such as the right of way in public spaces. However the legitimacy of state agencies may be compromised in particular sections of society due to their partisanship or inefficiency and give rise in extreme cases to the growth of vigilante or other forms of self policing with the support of victims and those who fear victimization. In revolutionary situations such organs may approach the status of embryonic alternative criminal justice systems (Santos 1980) directly challenging the existing state. Under less dramatic conditions such alternatives may become easily co-opted or out- manoeuvred by adaptions from existing criminal justice agencies. In a similar way social groups perceiving that the state does not take their victimization seriously enough may develop forms of alternative victim support as illustrated by the growth of rape crisis centres and voluntary refuges for women victims of domestic violence. These may in turn become the basis for political campaigns around victimization and may in the long run accept funding from or be incorporated into a quasi-official relationship with the state. 

Offenders, victims society and the state

Alongside the impact of crime on its victims is its impact on society and the state. These two are frequently not distinguished, victims after all are members of society, the fear of crime and its consequences extends beyond those immediately victimized and the criminal justice system and various state financed victim support services divert social resources to deal with the consequences of crime. However as noted above, it is important sociologically to understand criminality as a form of social behaviour, as a subsystem of society, in fact as a 'normal' part of social activity which acts to reproduce itself and defend the conditions of its own existence. It is otherwise difficult to understand the continuity of various types of criminality in modern societies. 

Action: reproducing the social structure of offenders and victims

Crimes of passion do not generally exist as a particular form of social activity with their own economy, subculture and division of labour. Spousal murder or domestic violence might be thought of as arising spontaneously out of the dynamics of the individuals involved in destructive marital relationships. This simply means that the social structure and culture which sustains of such criminal activity, namely the modern family and its associated patriarchal values and stresses and strains on personal relations, is generally diffused throughout the social system, a point that feminists have frequently emphasized. A similar point could be made about much commercial crime which is produced by the entirely normal values of business organization and methods.

At the other end of the spectrum are the elaborate organizations of international organized crime and its leading edge: the illegal drugs and armaments trade. While not part of normal -i.e. legal - capitalist enterprise, it is precisely the similarity of the manufacture and trade in dangerous and illegal commodities to legitimate enterprise in terms of the extended chain of production, marketing and financing which requires measures to be taken to secure the continuity and predictability of such enterprise in the face of determined - although hitherto unsuccessful - attempts by law enforcement agencies to interdict. In the centre of this continuum are various types of criminal semi-economies with varying degrees of organization such localized trade in stolen consumer goods such as video recorders or credit cards. The more organized and continuous over time the criminal activity the more 'security conscious' it is forced to become and the more steps it must take to secure its conditions of existence. Within these criminal economies all the usual dynamics of competition and amalgamation pertain. Such criminal economies, in order to secure their continuity have to both secure the compliance of local populations and neutralize the criminal justice system.

Securing the compliance of local populations depends in the first instance on the nature of the criminal activity. If the latter is one that members of the local community can benefit from compliance may not even a problem: few people in any locality are going to complain at the presence of cheap tape cassettes or video-recorders for sale. If the activity is one that the local population is in principle opposed to then other steps will have to be taken. The narco-capitalists of Medellin and other Latin American drug production centres invest a considerable number of dollars in local schools social and health care centres and other forms of 'Robin Hood' activities which, combined with violence and terror, aim at the maintenance of a quiescent local population. In the older traditional Mafia regions of Sicily and Calabria mafiosi traditionally maintained their status in the community as 'men of honour' by helping local citizens sort out legitimate grievances with the authorities or with powerful landowners. (Arlacchi 1985 Hess 1973). In a similar way powerful criminal organizations may finance political parties or particular candidates sympathetic to their interests both at a national and regional level.

Alongside the activities of powerful criminal organizations the criminal economy may exercise its own form of social control by bringing employment and income to marginalized groups in society who would be otherwise unemployed. Mike Davis's graphic description of a large city in the United States in which as the dynamics of the modern capitalist economy moved legitimate employment out of the older urban areas young male black teenagers have become the main family bread winners though being able to avail themselves of the only form of well paid economic activity in the area street dealing in cocaine and its derivatives. (Davis 1988) 

Reaction: containing and galvanising the state

Keeping the criminal justice system at arms length or otherwise rendering ineffective its activities is important for the survival of organized crime. A mixture of bribery and threat of violence against police or judicial personal may be attempted depending on the relative balance of power between the two institutions. At the lower levels of small time organized crime police officers may be periodically found who are prepared to accept bribes and even participate in the planning of criminal actions. In Britain in the early 80's the 'Operation Countryman' investigations of showed something of the extent of police involvement in such relations even in a country with a relative lack of a tradition of state corruption and clientalism. (Doig 1984) On a more organized level powerful criminal groups may attempt to mix bribery with the threat of violence or assassination. The process is by no means unknown in Western Europe in which police officers and prosecutors -and on occasion even journalists pursuing organized crime operate under conditions of threats and assassinations.

Even if the criminal justice system cannot be terrorized into inactivity it may be possible for criminal organizations to enter into various sorts of collusion. On an individual level an offender, once caught, may be granted immunity from prosecution in return for information leading to the successful prosecution of others. Such is the case with much organized crime and political terrorism. Periodic 'confessions' by mafia notables, by no means all of it producing reliable evidence, are a constant feature of the 'war on organized crime' in both the United states and Italy. However it is the possibility of collusion between the criminal organization and elements of the criminal justice system with the aim of securing the continuity of criminal activity which is of importance here. On occasion governments may deliberately pursue a policy of collusion with criminal groups. The latter may for example be able to provide a supplement to state agencies in the maintenance of a particular political regime of 'law and order'. Some Latin American regimes in recent years have openly tolerated the existence of right wing paramilitary murder squads as a form of control of left wing political movements. States may turn a blind eye to the use by employers of criminal gangs as strike breakers etc. The possibility of such an alliance and its degree of openness depends on the balance of political and social forces. Governments may miscalculate the political situation and find their collusion with criminal groups an embarrassment. It is currently a matter of some embarrassment to the United States government that, as the 'Irangate' hearings have revealed, US clandestine operations in the early 1980's a med at supplying the Nicaraguan contras for armed struggle against the government of that country involved a close relationship with drug traffickers in which US financed aircraft taking illegal arms to the rebels were used for drug shipments on the return flight. Older examples of such collusion can be found in European history as in the years immediately following Italian unification in which the central government whose power in the Sicily and the South was still weak allowed the Mafia to act as an unofficial agency of law and order (Arlacchi 1985)

The final avenue open to offenders is to secure the decriminalization of their activity. Offender groups generally do not have the collective political power to influence their decriminalization except as part of a wider social movement. Decriminalization of homosexual intercourse between consenting adults required a significant liberalisation of public opinion in general for its achievement. Drug users are a particularly powerless group. However socially powerful groups such as corporate capital may well be consulted by governments in the process of legislation and thereby be able to influence the formulation of law to their advantage, even being able to engage in 'anticipatory decriminalization' Organized crime has a much more ambiguous attitude to decriminalization since much of the vast profits to be made by trade in illegal commodities depends on the very situation of illegality. Consequently a more sophisticated form of 'decriminalization' is that in which the profits of illegal enterprise can be ploughed into safer forms of legitimate investment through sophisticated 'laundering' procedures. (Taylor 1989 Santino 1989)

The reaction of victims to the state - as opposed to direct action by victims against offenders - is generally to the perceived inactivity of the latter. An exception is of course where the state itself is seen as the offender and where the relatives of those directly victimized by the state attempt attract political support from wider social groups, from abroad or from subsequent governments to prosecute members of predecessor regimes. The activity of victim groups pressing increased activity and resources from the criminal justice system is often closely allied with movements to widen the boundaries of criminalization itself. The key variable is the extent to which wider groups in society will identify with the form of victimization in question. This is governed by the factors concerning the visibility and directness of victimization, the status of the victims in society etc discussed above. Victim groups are at their most effective when they can become seen as the exemplary case of a wider social issue. Thus rape victims have in recent years benefited from a generalized support from the women's movement - which was instrumental in setting up rape crisis centres and mounting campaigns for changes in police and judicial treatment of rape - which saw the treatment of such victims as epitomizing the generalized treatment of women in modern society. They may be contrasted with the poor and elderly victims of street mugging who during much of the same period found that the seriousness of their victimization was being loudly contested by radical groups and being denounced as media orchestrated 'moral panic', or the powerless victims of corporate crime killings such as Bhopal who still any years after the event still await adequate compensation. 

Conclusion.

Most of the topics touched on in this essay could be, and are the subjects of lengthy studies in themselves. The aim of this article has been deliberately eclectic: to show that the development of criminology has to proceed concretely and synthetically by bringing a wide range a different factors together in the analysis of particular situations rather than developing one sided generalized theories of 'the causes of crime' or 'the causes of criminalization'. The framework of the 'square of crime' developed by Left Realism forces us to approach seemingly simple questions like 'why has crime risen over the last decade' as the process of a complex structure of interactions. The task now is to demonstrate the utility of this framework in the concrete analysis of different types of criminality and policies aimed at controlling it. 

NOTES

[1] except rhetorically that is. "A philosopher produces ideas, a poet verses, a parson sermons, a professor text books etc. A criminal produces crime. The criminal produces not only crime but also the criminal law: he produces the professor who delivers lectures on this criminal law, and even the inevitable text books in which the professor presents his lectures as a commodity for sale in the market....Further the criminal produces the whole apparatus of the police and criminal justice, detectives, judges, executioners..." (Marx, cited in Bottomore & Rubel 1963)

[2] Radical feminists who wish to assert with Dworkin that "The annihilation of a woman's personality, individuality, will, character is prerequisite to male sexuality" (Dworkin 1988) and that any man in any situation is as likely as any other to rape have to make recourse to ever widening and blurring of boundaries in the notion of sexual violence. Box himself was guilty of this. (see 1983 esp. pps 121-9) and Liz kelly (Kelly 1988).

[3] The concept of marginality can end up as identical with relative deprivation in the sense than any group with blocked access to legitimate means of goal attainment is by definition marginalized. The only coherent use of the term is therefore as a reference to blocked access to those legitimate political processes such as trade unions, pressure groups, political parties etc. which provide collective means to open up new channels for the achievement of legitimate goals or secure welfare compensation for those who have no access to existing means.

[4] For example some feminist commentators on the recent Yorkshire Ripper murders in Britain have argued that it was only when the victims were no longer prostitutes but began to include 'respectable women' that the level of public outrage and the activities of the police were really galvanized into action. (Jouve 1987, Cameron and Frazer 1988)

[5] This is a separate question from the legal arguments concerning negligence and matters in which the offender is constrained to take responsibility for negligence. Indeed a good part of the legal prosecution of this type of crime consists precisely in the attribution of criminal responsibility for 'unintended' actions as when as large corporation is successfully penalized for environmental damage or ignoring safety procedures -in the hope that no deaths would in fact occur- which have in fact resulted in death. At the time of writing British legal history is being made by the decision to criminally prosecute the owners of the Ferry company whose alleged negligence resulted in the sinking of a vessel with the deaths of over a hundred passengers at Zeebrugge in 1987

REFERENCES