Realist Criminology, the New Aetiological Crisis and the Crime Drop

This paper aims to provide a summary of some of the key issues outlined in my recent publication Realist Criminology (2014a). It discusses, in particular, the perceived deficiencies of both mainstream and critical criminology in relation to their explanations of the crime drop. It argues that the failure of criminologists to provide a convincing and plausible explanation of the recent decrease in recorded crime in Britain, North America and other countries has resulted in a new aetiological crisis. It is suggested that the failure to explain what is arguably the most significant development in relation to crime in living memory is no accident, but rather a function of the theoretical and methodological inadequacies that are prevalent in academic criminology.


Introduction
Writing Realist Criminology (2014a) was motivated by a number of considerations. First, it had been the intention in the 1990s to summarise the main arguments of realist criminology into one accessible text and, although Jock Young and I published two realist collections in the 1990s -Issues in Realist Criminology (1992) and Rethinking Criminology: The Realist Debate (1992)neither book provides a general introduction or overview of the realist perspective. Jock Young published a number of articles outlining some of the key features of what was known as Left Realism, which identified some of the major differences between Left realist criminology and the dominant forms of conventional criminology (see, for example, Young 1992Young , 1997. Secondly, the debate in criminology has moved on over the past two decades and it is therefore necessary to make a more up-to-date statement about the limitations and challenges facing criminology. A key factor of the development is the increased dominance of different forms of liberalism within criminology. A disturbing feature of contemporary criminology is the way in which the distinctions between critical and liberal criminology have become blurred and critical criminology has in many respects lost direction and impact. In writing Realist Criminology (2014a) the task was to identify the limits of liberalism and to develop a stance that is not antistatist or anti-punishment but rather one that works both 'in and against' the state in order to take crime, victimisation and punishment seriously.
In its original formulations in the 1980s and 1990s Left Realism provided an essentially political response to the liberal-conservative consensus in criminology. In doing so it aimed to link theory and practice in order to provide the basis for the development of a Left social democratic response. However, on reflection it did not pay enough attention to questions of theory, epistemology and methodology. Although critical of positivism and a-theoretical forms of administrative criminology on one hand, and the conservative 'broken windows' approach on the other, questions of theory and method remained largely underdeveloped (Jones, McLean and Young 1986;Kinsey, Lea and Young 1986).
Significantly, in recent years there have been developments of Critical Realism associated with the works of Roy Bhaskar (1975), Margaret Archer (1995), Andrew Sayer (2000) and others, which has provided a more sophisticated approach to social scientific investigations. Andrew Sayer's work in particular, drawing on Marxist methods, has provided a much-needed contribution to both the development of social theory and method (Sayer 2010). Thus, the integration of the insights of Critical Realism and Left Realism is the third aim of the book, which has the objective of placing the realist project on a firmer conceptual and methodological foundation.
A fourth objective of the book was to examine and interrogate the richness and subtleties of cultural criminology. Cultural criminology has provided a much needed impetus to the development of critical criminology. However, it has been accused of having idealist associations, a lack of appreciation of victimisation and little interest in policy developments (O'Brien 2005). Thus one of the central chapters of the book aims to try to integrate the contributions of both cultural and realist criminology and produce a form of critical criminology that might be termed Cultural Realism (Matthews 2014b).
The fifth reason for writing Realist Criminology (2014a) was to contribute to the development of a 'public criminology' that aims to make criminology more socially and politically relevant. In previous publications, I have developed the term 'So What?' criminology to identify those forms of criminology that are thin theoretically, adopt an inappropriate methodology, and have little or no policy relevance (Matthews 2009) I have argued that the combination of these deficiencies seriously detracts from the possibility of developing a 'public criminology' and is likely to result in a criminology that had limited social or political value. Thus a central argument in the book is the need to develop a 'joined up' criminology that can link theory, method and policy.
The book itself has two parts. One part aims to provide a framework and guide to doing realist criminology. The second part provides a critique of two of the dominant tendencies in contemporary criminology: liberal and administrative criminologies. One of the examples of the limitations of contemporary criminological approaches that is given in the book is the so-called 'crime drop' which refers to the remarkable decrease in recorded crime that has taken place in the UK, the US and at least ten other countries over the past two decades. The crime drop was not predicted and criminologists have found it very difficult to explain what is undoubtedly the most significant development in crime in living memory.
Some years ago, Jock Young (1997) made reference to the 'aetiological crisis' and the inadequate explanations that criminologists provided for what appeared to be the inexorable increase in crime. Now that crime has decreased dramatically we have arguably a new aetiological crisis as criminologists from various persuasions have been unable, to date, to develop a convincing explanation of this dramatic change. This paper aims to provide an overview of the ways in which different criminologists have tried in recent years to address the question of the crime drop on the presumption that an indicator of the health of a subject area lies in its ability to explain the most significant developments in its field of enquiry.

The aetiological crisis
In his influential depiction of the 'aetiological crisis,' Jock Young (1997) pointed out that, on both sides of the Atlantic between the 1960s to the end of the 1980s, crime continued to increase despite periods of full employment and decreasing levels of poverty and deprivation. This would seem to undermine liberal arguments about the causes of crime. At the same time the observation that recorded crime continued to increase in both Britain and the US despite the significant growth of incarceration served to undermine the conservative claims that getting tough on crime would serve to reduce it.
In Young's view, the various theoretical explanations for the apparently inevitable rise in crime were found wanting, and although some explanations proved to be more plausible than others, none of the major theories were found to be able to provide a comprehensive or consistent explanation. What concerned Young most was that this lack of explanatory power was not restricted to the conventional forms of liberal and conservative criminology but was also evident amongst critical and radical criminologists. It was Young's central task to develop a viable critical criminology that could address the major issues of the day and provide a credible alternative to mainstream criminology (see Young 2013). If critical criminology was to compete successfully with conventional criminology, the implication is that it needed to provide a credible explanation for the apparently relentless increase in crime. Although such an explanation never fully materialised, it was clear that the dominant criminological approaches including positivism, subcultural theory, strain theory, control theory, and labelling theory were all deficient in this respect. The bankruptcy of these theoretical approaches signalled the limitations of criminological theory and paved the way (unfortunately) for the rise in forms of administrative criminology.

The crime drop
However, since the early 1990s in Britain, the US and a number of other countries, recorded crime has decreased significantly year on year. This systematic decrease was not only unpredicted but for a number of years it remained unacknowledged amongst the criminological community.  (2000), for example, in his examination of the role of incarceration concludes by saying that imprisonment 'was a contributing factor' to the crime drop and then goes on to say that most of the responsibility for this decrease rests with improvements in the economy, changes in the age structure, and other unexamined social factors.
Focussing on the role of policing in relation to violent crime, John Eck and Edward Maguire (2000) provide a similar set of equivocal conclusions. While the authors credit the police with having some impact on violent crime, the degree of impact is unspecified. They did add, however, that: 'the most plausible explanation is that police actions interacted with other criminal justice policies'. The 'other policies,' however, remain unspecified, as does the nature of the proposed interaction.
Other chapters in the book are similarly inconclusive and we remain no clearer at the end of the book about the causes of the crime drop than we were at the beginning. What is also evident is that the authors and editors of the book also seem impervious to the fact that similar developments have been taking place in the UK and other countries which have not experienced mass incarceration, US-style policing practices, or the problems associated with gun use. An international perspective on the crime drop may have injected some refinement into the analysis.
In the same year Andrew Karman (2000) produced a more thoughtful and well researched account of the crime drop in New York. The book set out to review the role of imprisonment, drug control, changing demographics and policing on the homicide rate in New York. He questioned the claims that zero tolerance policing, and other changes that took place in the era of the former Mayor Giuliani were responsible for the decrease in homicide in particular, and the crime drop in general. Karman pointed out that crime dropped in a number of other North American cities that had adopted very different policing strategies. Despite providing a perceptive critique of many of the dominant explanations for the crime drop in New York, he concludes with the statement that: The best way to describe the city's situation in the 1990s was that a 'fortuitous confluence' of underlying factors materialised. Luckily for city residents -and for those in a position of authority and trust -every one of the causal factors known to affect crime rates moved in the same direction. (Karman 2000: 259) If it was the case that all of the known factors were moving in the same direction, it might be reasonably assumed that there is some underlying generic causal mechanism that was driving this change. However, it is not the case that all the possible causes were moving in the same direction in all the countries experiencing a decrease in recorded crime. Furthermore, in North America the high rate of recidivism, the fragmentation of communities, economic fluctuations and other 'factors' would all seem to have been pushing in another direction.
John Conklin (2003) also questions the viability of a mono causal approach and concludes, like Karman, that a range of 'factors' probably came together but claims that the role of imprisonment was the most important reason why crime rates fell in the 1990s. Using a form of regression analysis, Conklin suggests that the problem is that many of the variables 'defy easy measurement' and that 'the simplification of independent variables can obscure their impact on the dependent variable' (Conklin 2003: 192) The real problem, however, is to know a priori what counts as an 'independent variable' and the possibility that the relevant independent variables are not included in this analysis. Although there is a widespread assumption that regression equations are virtually synonymous with causal analysis, this is not the case. Regularities are not sufficient conditions for the identifications of causes. In fact, they are not even necessary conditions (Sayer 2010). The perennial problem is distinguishing causal from contingent relations. Unsurprisingly, after listing a number of variables that might have contributed to the decrease in recorded crime, Conklin (2003: 201) writes: 'My hope is that this book contributes to an understanding of why crime rates declined in the 1990s and that it will generate thoughtful discussion of why crime rates fell and how they can be kept down or further reduced.' Unfortunately the book adds little to our understanding of this issue and is unlikely to stimulate much constructive discussion about how to reduce crime further.
As a result of the limited explanatory value of most of the main 'factors' which have been presented in these publications, other authors have developed more imaginative explanations of the crime drop such as the legalisation of abortion, which Steven Levitt (2004) argues reduced the 'at risk' population of potential criminals. Franklin Zimring (2007) has taken issue with Levitt's thesis. Zimring argues that the legalisation of abortion thesis is limited by the lack of fit between the introduction of abortion legislation and the beginning of the reduction of crime. Also, other methods of birth control were available and probably widely used before the new legislation, which would have reduced birth rates amongst the poor. Most importantly, Zimring points to the methodological limitations of this thesis and in particular the question of causality and the problem of distinguishing generic from contingent causes.
Zimring (2007) himself, however, fails to offer a convincing alternative explanation. Rather, he claims that there is not a single cause of the crime drop and that it was, he suggests, a 'classic example of multiple causation, with some of the many conflicting causes playing a dominant role' (Zimring 2007: 195). However, he notes that crime declined in Canada over the same period as in the US although economic, social and criminal justice developments were very different in Canada during this period. Although this looks like an argument against parochialism, the decline in Canada would appear to undermine his general conclusion about the multi-factor nature of the crime drop. He also fails to resolve the methodological problems that he identifies in his multi-factor approach since he does not explain how these factors might be causally connected.
In contrast to the multi factor approach, some criminologists have recognised that the steady decrease in crime is almost certainly linked to wider developments. Even James Q Wilson, who is probably best known for his arguments on the need to focus on minor crimes, incivilities and anti-social behaviour (see Kelling and Wilson 1982), has recently suggested in an article in the Wall Street Journal that: At the deepest level, many of these shifts, taken together suggest that crime in the United States is falling -even through the greatest economic downturn since the Great Depression -because of a big improvement in the culture. The culture argument may strike some as vague, but writers have relied on it in the past to explain both the Great Depression fall in crime and the explosion of crime during the Sixties. (Wilson 2011: 4) Although Wilson notes that measuring and examining 'culture' raises methodological difficulties, he is no doubt correct that explanations based on broad socio-cultural terms offer the possibility of developing a more plausible and credible explanation of the crime drop than many of those accounts that have been offered to date. Again, taking a broader cultural perspective, Robert Sampson (2008) has examined the relation between immigration and the crime drop. He has argued in line with other researchers that first generation immigrants are held to be generally hard working and law abiding and that the recent influx of immigrants in North America which are mostly Hispanic, may have had some impact in diluting the criminogenic culture of some US cities. Research on immigration and crime in Europe, however, presents a more mixed picture, but there is some indication that the involvement in crime by some immigrant groups is relatively low (Bell and Machin 2011).
The main contribution to the debate about the decrease in recorded crime in the UK centres round the presumed effects of the increasing use of security and crime prevention methods over the past two decades. Graham Farrell and his colleagues (2011) have presented what they refer to as the 'security hypothesis'. Drawing on routine activities and opportunity theory, they claim that the implementation of various forms of crime prevention measures such as car locks, and other target hardening techniques has deterred prospective offenders. Although it is no doubt the case that the billions of pounds and dollars that have been spent on security measures over the last two or three decades have had an impact on crime, the security hypothesis suffers from two major limitations. First, the timing of the introduction of the various security measures that the authors refer to do not match up well with the timeline of the decrease in different forms of crime. Second, the security hypothesis does not go very far in explaining the simultaneous decrease in violent crime. Therefore, as an attempt to provide a comprehensive explanation of the crime drop, it is less than convincing. Even the UK's Home Office researchers remain unconvinced about the explanatory capacity of this hypothesis. They state that: The hypothesis is largely silent on why violence has fallen alongside theft. And for acquisitive crime, the case that better security caused the drop rests on the largely untested assumption that car immobilisers also prevented or deterred thieves from committing other types of theft. Data suggests that the opposite is equally likely -that as one thing becomes harder to steal, thieves switch to something else. So, because all types of theft fell markedly at the same time in the mid 1990s it seems likely that a change in offender propensity for crime is more likely to provide the main explanation. (Home Office 2015: 1) Graham Farrell (2013) has, however, in another publication identified five tests that he suggests any credible explanation of the crime drop must meet. These include, first, whether there are reasonable empirical grounds to consider the hypothesis. Second, that the hypothesis stands up to cross-national comparisons. Third, that the hypothesis is compatible with the fact that crime increased over previous decades. Fourth, that it explains variations in trends amongst different types of crime. Lastly, that the hypothesis is compatible with variations in the timings of crime falls between countries and crime types. Farrell claims that the only hypothesis that meets these criteria is the security hypothesis. Nonetheless, as suggested above, the security hypothesis is not only weak in relation to timing of changes but also fails to explain that, given the variation in the introduction of different crime prevention measures, all forms of crime appear to have declined simultaneously. Thus, although the security hypothesis may go some way to explaining the decrease in car crime, it does not explain variations in violence or drug-related crime well. Moreover, it employs a mono-causal explanation, which, as various authors have pointed out, is unlikely to grasp the complex and wider structural changes that are associated with the crime drop.
In my own research on armed robbery, for example, I found that there was a significant decrease in commercial robbery in the UK in the 1990s (Matthews 2001). However, this was not associated with the introduction of any new security measures in this period. In fact, it was during the 1980s that a range of measures including cameras, alarms and screens were introduced in the UK. These made little impact on the number of commercial robberies during this period. The decrease in commercial robberies in the 1990s was a function of a cultural shift involving the demise of old style professional robbers by more amateur and spontaneous offenders and this also involved a shift towards more accessible options.
The other major response in the UK to the decrease in recorded crime involves a denial of its occurrence and a claim that the apparent decrease is a function of the manipulation of the crime figures by the key agencies or that crime has not so much decreased as been displaced and that 'street crime' and 'normal' crime are being replaced by cybercrime. In relation to the manipulation of the crime figures, there is clear evidence that the police have changed recording practices in recent years and have admitted 'massaging' the crime figures. In December 2014, the gold standard 'national statistics' status was withdrawn from the police in England and Wales due to discovery that they had been 'fiddling' the figures (Travis 2014). However, the continued and widespread decrease in a range of crimes over time, particularly those where the police are able to exercise limited discretion, or where there is not much room for manipulation such as car crime, robbery and serious violence, suggests that a significant decrease has taken place, at least amongst certain crime types. The international nature of the crime drop also suggests that, despite dubious police recording practices in the UK, crime is decreasing in different countries where recording practices have remained relatively stable. In addition, the simultaneous decrease in victimisation recorded by the Crime Survey for England and Wales tends to reinforce the conception that crime is decreasing.
In relation to the increase in new forms of cybercrime there can be little doubt that this is taking place, but it is extremely questionable that traditional forms of 'street crime' are being displaced in this way. Cybercrime would in all probability increase in the current period irrespective of changes in other forms of crime. However, it is important to note that cybercrime involves a distinctly different group of offenders than traditional forms of street crime as well as having a different set of victims (McGuire 2016;Wall 2007).

The crime drop and criminological theory
In her classic but often overlooked text Women, Crime and Society (1982), Eileen Leonard argued convincingly that traditional criminological theory is incapable of explaining the nature or patterns of female crime. A similar critique arises in relation to the attempts by different criminologists -drawing on these traditional theories -to explain the crime drop.
Standard liberal theories that attempt to explain the crime drop in terms of poverty, deprivation, unemployment or economic fluctuations have proved to be inadequate. One of the significant features of the crime drop is that it appears to transcend economic changes including changes in the level of unemployment. Standard versions of strain theory appear less than helpful in this respect and, despite the increasing gap between the rich and poor in many advanced countries, recorded crime has decreased.
Socio-biological and standard positivist accounts appear largely irrelevant and all those theories that present individualistic accounts, or claims that criminality is a product of innate propensities or personal attributes, are difficult to take seriously. Similarly, notions of anomie, although having the benefit of considering wider social and cultural influences, fail to capture the nature of shifting moral values in late modernity. As Zygmunt Bauman (1991) and others have pointed out, the era of 'liquid modernity' involves new forms of inclusion and exclusion. Increasing social divisions occur alongside the search for community and identity. It is a world of moral ambivalence characterised by greater tolerance on one hand and indifference on the other. This is a world, according to Jock Young (2003), of changing agency and structure and, as he suggests, the art of criminological analysis is to capture these two dimensions simultaneously.
Labelling theory, realists have argued, has always suffered from the deficiency of being onesided by focussing almost exclusively on the social reaction to deviance (Young 1992). Realists instead have presented the 'square of crime,' pointing out that crime is not only an act and a reaction but also has a number of formal and informal dimensions that combine to create the phenomenon of 'crime' (see Lea 1992). However, apart from this limitation, labelling theory has always been much better at accounting for the application of labels rather than their non- application. Even in periods in which attempts have been made to 'define deviance down' (Moynihan 1993) through forms of diversion, the use of cautions and lesser penalties, these strategies appear to have minimal effect on the overall crime or custody rate (Garland 1996). Consequently there is little to suggest that the crime drop has been a formation of fundamental changes in the application of labels or simply an increase in the level of tolerance.
Social disorganisation theory also appears to be deficient. Increased mobility and immigration in a period of globalisation should, according to this perspective, result in an increased crime rate as a result of the breakdown of both formal and informal controls. The demise of the community together with the 'death' of the modern family alongside forms of economic displacement should, according to social disorganisation theory, lead to increases in crime.
Finally, rational choice, routine activities and opportunity theories have historically provided a limited, if not distorted, form of explanation. Conceptions of the offender as a rational actor weighing the cost benefit of criminal involvement, has always been an exaggeration. Cultural criminologists have emphasised the emotional and risk-taking nature of much criminal involvement and rejected forms of explanation based on notions of instrumental rationality (Hayward 2007). Routine activities theory suffers similar limitations. It is not so much a theory but a set of propositions that claims that crime is likely to occur when there is a motivated offender, attractive target and a lack of appropriate guardians. However, this is little more than a set of tautologies. We only know that the defender was motivated, that there was an attractive target and that there was a lack of suitable guardians after the crime has taken place. Similarly, opportunity-based theories do not account for the planning of crimes or the differential predispositions to commit crime, or that many individuals deny given opportunities (Ekblom and Tilley 2000).
Other traditional criminological theories and perspectives could be cited but arguably all would be found wanting in terms of providing a plausible explanation of the crime drop. This means that there is little to draw on from the available body of criminological theory and that, if we are to begin to try to account for the crime drop, we need to develop a different type of account.

Conclusion
It is evident from this survey of the contributions to explanations of the crime drop that we are facing a new aetiological crisis. The criminological approaches that have been cited are deficient in many respects. In fact, the body of literature on the crime drop makes a significant contribution to 'So What?' criminology providing at best a partial explanation of the decrease in recorded crime. These approaches display both theoretical and methodological deficiencies. The end result is that these accounts have little or no policy relevance.
The major conceptual limitation stems from the narrow conception of possible determinants. Both conservatives and liberals are reluctant to engage in an analysis of the wider socioeconomic and cultural processes that almost certainly underpin the crime drop. Conservatives are reluctant to engage in an analysis of 'deep structures' and are more comfortable focussing on the immediate and directly observable phenomenon. Liberals, on the other hand, tend to view socio-economic analysis as a form of economic reductionism and prefer to operate with some form of political reductionism.
A critical realist explanation of this international development would begin from an examination of the shift from Fordism to Post-Fordism and in particular the decline of industrialisation and the growth of a service economy accompanied by increased national and international mobility. This involves patterns of mass immigration and associated cultural shifts. These developments I would suggest have occurred alongside major changes in gender relations in general and notions of masculinity in particular. As we enter the 'have a nice day' IJCJ&SD 10 Online version via www.crimejusticejournal.com © 2016 5(3) economy and cappuccino society, it is no longer desirable for men to be physically tough. Instead the modern man is cool, flexible and smart. In addition we are witnessing significant changes in the nature of social control and modes of governance.
It is from this conceptual vantage point that we might reasonably begin to explain the crime drop and move beyond narrow mono-causal accounts and multi-factor approaches. It is only by providing more comprehensive forms of causal analysis that can identify the mechanisms involved that we are able to move beyond 'So What?' criminology and develop a form of explanation that is both critical and useful.
Towards the end of the book (Matthews 2014a) I discuss the decreasing social and political significance of crime in western societies and raise the question of whether we are becoming a post-disciplinary society in Foucauldian terms. In contrast to Jonathan Simon's (2007) 'Governing Through Crime' thesis, I suggest, following Giles Deleuze (1995), that the emerging forms of governance are shifting away from a focus on the marginalised and the 'underclass' to more general forms of social regulation that are more comprehensive and more continuous.