Crime Control in a Humane Society

In the previous post, I discussed the possibility that cooperation and pro-sociality were deeply embedded in human evolutionary history. Alongside this more positive take on human nature, I also pointed to a good deal of evidence that human beings also have the capacity for egocentrism and for the pursuit of unbridled self-interest. Among the evidence for this capacity for selfishness is the ubiquitous emergence across time and across cultures of acts that can be defined as criminal – that is, acts that go against formally defined (legal) codes of conduct, often to the detriment of other people. In this post, I would like to explore the extent to which criminal behaviour is a fundamental part of human nature. I will outline broad theories of crime in Psychology and Criminology, and examine how these relate to our nature. In particular, I will focus on whether human beings need to be prevented from engaging in criminal behaviour by punishment, containment etc. as a result of a fundamentally selfish nature, or whether human beings are drawn to criminal behaviour through social and environmental pressures, thus suggesting a more pro-social or neutral nature that can nonetheless be corrupted. I will finish by proposing ways in which the criminal justice system could be transformed, in order for it to be consistent with human nature, while also achieving goals of justice for all and protection of the community.

In order to investigate the fundamental underpinnings of criminal behaviour, it is important first to have an appropriate working definition of crime, which can constrain and inform our understanding of the processes that give rise to that behaviour, and of how that behaviour can be controlled. Most attempts at defining crime begin by outlining its relationship with deviance, where deviance is defined as any behaviour that departs, in a statistical sense, from social norms or expectations. Crime, then, is often defined as a form of deviance, as most crimes are statistically deviant forms of behaviour; murder, assault, or fraud, for example, are typically only perpetrated by a small sub-section of the population. While there is a general relationship between crime and deviance, it is important not to go too far and treat them as equivalent concepts, as some behaviours can be deviant but not criminal – for example, cis-gender male “cross-dressing” is statistically infrequent, but is not a crime anywhere in the western world – and some behaviours can be criminal but not deviant – for example, marijuana use is defined as a crime in many jurisdictions, but is often found to have been tried at least once by a majority of the adult population. As such, crime is usually a form of deviance, but is not always, and deviance can be criminal, but in most cases is not. Deriving a working definition of crime therefore requires going beyond a simple study of its relationship to deviance.

In going beyond a deviance-based definition of crime, criminologists typically analyse the historical processes by which different behaviours came to be understood as criminal. In so doing, it turns out that all criminal behaviour was characterized as such in one of three ways: based on consensus within a group, based on conflict between different groups, or based on some form of moral entrepreneurship. When a behaviour is defined as criminal based on consensus, this entails that all or most people in a society will view that behaviour as repugnant, or in violation of the moral standards of that society. Murder is one such crime, where there are very few who would disagree that killing another human being is a violation of our deepest sense of morality.

When a behaviour is defined as criminal based on conflict, this is to say that those who hold political power wield that power within the criminal code in such a way as to protect their own interests over the interests of the less powerful. In this way, behaviour is labelled as criminal not so much because the behaviour is universally thought to be repugnant or immoral, but rather because a ruling class has criminalized the behaviour of a less privileged class, in order to protect their own higher status. There are countless examples across the world of behaviours being added to criminal codes in this way, but one of the clearest concerns the way in which marijuana possession, distribution, and use became illegal, first in the United States and then across the world. In the early 20th Century, the hemp plant was grown relatively widely, and marijuana was used relatively commonly as a recreational drug. In the 1930s, however, the spread of the ‘decorticator’ (a machine for stripping bark, skin, leaves or other parts off a plant) meant that hemp became a very cheap substitute for paper. This then led to disquiet among wood pulping and paper mill operators and investors, whose profits would likely be greatly reduced if hemp production expanded any further. As such, these operators and investors (including notable people in the newspaper and chemical industry such as William Randolph Hearst, Andrew Mellon, and the DuPont family) began a campaign to link marijuana use to violent crime, and thus to have it criminalized – successfully. The tobacco industry has also been at the forefront of maintaining the criminal status of marijuana use, as marijuana is relatively easy to grow compared to tobacco, and has fewer known side-effects. In this way, marijuana possession, use and distribution came to be defined as criminal, based largely on the intervention of powerful corporations and their officeholders acting in their own interests.

Finally, behaviours can be defined as criminal as a result of some form of moral entrepreneurship. This typically involves a powerful group campaigning for the criminalization of particular behaviours because members of that group hold particular moral convictions, opinions or preferences. In this case, the act of defining a behaviour as criminal does not protect the interests of the powerful, but rather flows directly from the fact that their social status gives them the power to frame the criminal code in line with their attitudes. A clear example in many jurisdictions in the western world (including in Australia) is the Catholic Church’s campaigning for the continued criminalization of abortion. Abortion is viewed as a right rather than a crime by the majority of the population in many jurisdictions, but remains a crime in many of those jurisdictions because powerful groups campaign for its continued inclusion in criminal statutes on the basis of personally held spiritual or moral beliefs. In most of these cases, the criminalization of abortion does not benefit the status interests of the powerful group campaigning for it, but rather simply aligns with their own values, which then come to be enshrined in law because of the relative power they hold.

So where does all this leave us with regard to defining crime? In order to derive a working definition, we must, first, recognise the generally deviant nature of most criminal behaviour. Second, we must integrate the insights derived from historical analyses of the processes by which certain behaviour became criminalized. Third we must include the role of criminal codes in delineating criminal behaviour from non-criminal behaviour. And, finally, we must take into account the consequences that are defined in criminal codes for violating them, including stigma, and the loss of freedom and social status, as these give the violation of criminal codes a special kind of significance that other non-criminal violations do not have. Combining these insights, then, we can define crime as: a violation of societal rules of behaviour, which results in a variety of defined social sanctions, as interpreted and expressed by the criminal law, which reflects public opinion, traditional values, and the viewpoint of the people who hold social and political power, often for the purpose of protecting their own self-interest. As such, crime is multi-faceted, but has relatively universal key features, most notably that (a) it always involves the violation of societal rules, and (b) there is a special significance afforded to criminal rule violations that is not afforded to other similar, non-criminal violations. This then entails that a science of crime, emerging from Criminology and Psychology, is a science of when and why societal rules will be violated despite the social consequences that result from that violation.

In attempting to answer the question of when and why societal rules will be violated – and thus when and why people will engage in crime – four main approaches have emerged in Criminology and Psychology: Choice approaches, Trait approaches, Social Structure approaches, and Social Process approaches. Choice approaches assume that criminals are rational actors who freely choose to engage in crime. Potential offenders choose to commit crimes or not based on a cost-benefit analysis, in which they consider personal (e.g., money, revenge, thrills, entertainment) and situational (e.g., target availability, security measures, police presence) factors, as well as the risk of apprehension, seriousness of expected punishment, value of the criminal enterprise, ability to succeed and need for criminal gain. Choice approaches argue that anyone could commit a crime given the right personal and situational circumstances, and therefore implicitly (or sometimes explicitly) assume that human beings have a tendency toward selfish gain unless curtailed by other costs to the self. As such, choice approaches suggest that crime is best controlled by punishment and deterrence, as without these the costs of engaging in crime would not be high enough to mitigate against the fundamental selfishness that could motivate any one of us to engage in crime.

There is some evidence in favour of Choice approaches. First, it is well known that crime can be deterred by increased punishment. It is also known that, when interviewed, people who have committed crimes will admit to having chosen targets of crime based on availability, likelihood of success, size of expected gain, and probability of arrest or incarceration. For example, the high rate at which drug dealers are robbed relative to other members of society is likely due to the fact that drug dealers often carry large amounts of money (i.e., the rate of return from a robbery is likely to be high), they are unlikely to report the robbery to police (i.e., the likelihood of arrest is low), and they are common in areas where would-be robbers operate (i.e., they are more easily accessible and available than other potential, high-value targets). This suggests that rational choice does indeed come into play in the criminal enterprise. Importantly, however, most of the evidence points to its role after an offender has decided to commit a crime; what is less clear is whether rational choice plays any role in the decision to commit a crime in the first place. It could very well be that the distal causes of crime have no relation to rational choice, but rather come about either through social and environmental pressure, or because some individuals have an in-built tendency for crime. Overall, Choice approaches do well in explaining how specific targets might be chosen when a decision to commit a crime has been made, but do less well in explaining whether someone will commit a crime at all. And, importantly, this lack of ultimate explanatory power means that Choice approaches do not provide much in the way of insight into our fundamental nature.

Trait approaches to crime argue that there are biological reasons why human beings in general, or certain individuals specifically, are likely to engage in crime. These approaches focus on biochemistry, neurophysiology, evolution, and genetics. There is some evidence that certain patterns of biochemistry are associated with crime; for example, higher testosterone is associated with more aggression and therefore higher rates of crime, while lower levels of the neurotransmitter monoamine oxidase in children who have been abused is associated with anti-social behaviour, including criminal behaviour. Similarly, mental health conditions that may have a biochemical substrate, such as ADHD, schizophrenia and Alzheimer’s disease, have been linked to crime. With regard to neurophysiology, it is well known that many crimes, but particularly crimes of passion, are associated with decreased functioning in executive areas of the brain; that is, in those areas responsible for impulse control, moral reasoning, and general self-regulation. Evolutionary models of crime generally focus on sex differences in crime, based on the fact that the rate of crime in males is often up to 9 times greater than that in females. Cheater theory, for example, proposes that undesirable males (those who are unlikely to offer parental support to females who they are pursuing) have to resort to deception, and illegal acquisition of resources to impress females, such that deception and criminal behaviour arise from male intrasexual competition. Finally, twin studies suggest that there might be a genetic component to crime, in that monozygotic (identical) twins have more similar criminal histories than do dizygotic (fraternal) twins.

While Trait approaches have accumulated an impressive array of findings with regard to the biological basis of crime, the evidence underlying these findings is often quite poor. First, in all cases the evidence is correlational; that is, we know that crime and some biological features are associated, but we have no evidence that the biological features are the cause of crime. Second, many of the correlations are between some biological process and some behavioural tendency, rather than between the biological process and crime itself. The clearest example in this category is the relationship between executive dysfunction and increased risk-taking; it may be that executive dysfunction is associated with crime through an increased desire for thrill-seeking and risk-taking, but we also know that increased risk-taking need not necessarily result in crime (it may result in something as innocuous as riding a rollercoaster). In such cases then, there is a possible biological tendency, but there is then likely an environmental trigger that leads the tendency to manifest in criminal behaviour specifically. With regard to evolutionary theories such as Cheater theory, these necessarily rely on constructing a narrative for a past event (i.e., the evolution of a criminal tendency), which means the core causal aspects of the theory can never be directly tested. And, finally, with regard to evidence for the genetic basis of crime, studies in this area (a) rely on correlational research; (b) often explore relationships between genes and traits that may or may not be necessary for criminal behaviour (e.g., risk-taking); and (c) only indicate the heritability of particular traits in the context of specific environmental influences, such that heritability estimates for criminal behaviour can vary wildly to the degree that the environment varies. Overall, Trait approaches may indicate that there is a biological basis to crime, deeply embedded in our nature, but the evidence is equivocal. What Trait approaches perhaps indicate more clearly is that some individuals will likely be more predisposed to crime from birth, but that environmental and other influences will play a large role in the emergence of actual criminal behaviour later in life. Trait theories, therefore, suggest that we are unlikely to be characterized at the species-level by selfish and antisocial tendencies, but rather that some of us may have these tendencies to a greater degree than others.

Social Structure theories suggest that almost all crime is a result of stratified societies where some people control resources and power and others do not. Under such conditions a ‘culture of poverty’ is said to develop among the less powerful (i.e., the poor), which is characterized by apathy, cynicism, helplessness, and mistrust of institutions. The violation of societal rules that characterizes crime then emerges from feelings of disconnection from the political bodies that make the rules, and from a drive to satisfy basic needs – such as food, safety, shelter, stability etc. – which are inaccessible because of low social and economic status. One particularly prominent Social Structure theory, Strain theory, elaborates on these ideas. It proposes that stratification of society, which results in poverty, causes feelings of inadequacy and a siege mentality among the poor. Despite this siege mentality, the poor are thought to remain loyal to the values of the dominant (so-called ‘middle class’) culture, including (in addition to their basic needs) desires for such things as employment, money and material possessions. When those basic needs and ‘middle-class’ desires cannot be met because of a lack of resources or access to suitable opportunities, feelings of strain and frustration will result. Then, antisocial and criminal acts are thought to arise as a means to achieve success and also to relieve feelings of strain. In this way, crime can become enduring as long as the strain endures, and therefore as long as the poverty and social stratification endures.

Like Trait theories, Social Structure theories rely heavily on correlational data to test their core proposals, suggesting some caution is required when making conclusions about causal paths. Nonetheless, Social Structure theories have a weight of historical evidence behind them. First, one of the strongest and most reliable predictors of crime is relative deprivation – that is, one’s psychological sense of being disadvantaged socially or economically compared to others in one’s society. Second, measures of objective wealth inequality (such as the gini coefficient) are good predictors of crime rates across the world. Third, crime rates track income and wealth inequality across time, even for crimes such as murder, sexual assault, domestic violence etc., which do not immediately appear to be directly related to poverty or the inability to satisfy basic social and economic needs. Finally, and perhaps most importantly, social groups that were characterized by high rates of crime in the past – such as the Irish in the United States, and various migrant communities in Australia – have become less crime-ridden as they achieve greater access to resources and opportunities previously afforded only to the dominant, higher status groups in their societies. This latter point indicates that any shared genetic inheritance that may characterize particular groups and that may predispose those groups to crime (for which there is little if any evidence anyway) can be overcome by increasing economic equality and diffusing social power more widely. As such, whatever predisposition for selfishness and crime that human beings may have in their nature, or that individual human beings may have in their genetic make-up, is not necessarily deeply rooted.

In the same vein, Social Process theories suggest that criminal behaviour arises from socialization and social influence. The precise ways in which this socialization occurs is explained in different ways in different theories, but all Social Process theories suggest that there is likely no in-built tendency for crime in any one of us, or in the species as a whole. Labelling theory, for example, proposes that crime often results from stigmatizing children for certain kinds of behaviour, which would otherwise remain only mildly deviant and relatively unproblematic. The stigmatization is thought to lead to a person being labelled as delinquent, which in turn leads them to self-label as delinquent. Once a person has self-labelled, they will then be likely to take on attitudes and roles that reflect that assumption, eliciting a deeper commitment to deviance. Once a person is committed to deviance they will be more likely to join a criminal gang which reactively or deliberately create social norms in opposition to the dominant culture; from there, it’s a small step to entrenched criminality. Social Process theories, then, treat human beings as a blank slate of sorts, where crime results from the social environment, including parents, peers, teachers, and society more broadly.

Social Process theories have the advantage over other crime approaches of being testable under controlled experimental conditions. For example, experiments have shown that stigmatizing someone causes that person to negatively self-label and to conform to attitudes and behaviours associated with the label. Social Process theories also have an advantage over Social Structure theories specifically, in being able to explain those crimes that do not result from social or economic deprivation, such as white-collar crimes. Nonetheless, Social Process theories often fall short of the broader goal of explaining the causes of crime both within the individual and beyond the individual, as they do not take seriously some of the biological influences apparent in Trait approach research, and they do not adequately address the large-scale social character of most crime. Overall, Social Process theories imply that human nature is neither good nor evil, and that any one of us could be influenced to engage in antisocial criminal acts given the right social and environmental circumstances.

So what does all this tell us about crime and human nature? Taken together, these four major approaches to crime suggest a few important things. First, they suggest that most crime seems to result from social stratification and from the strain felt by people affected by poverty. Second, they suggest that there are probably some biological underpinnings to criminal behaviour, but that these are unlikely to be at the species-level – rather at the individual level – and they are unlikely to be associated with crime per se – rather with behavioural tendencies, such as risk-taking or impulse control, that could manifest in any number of (non-criminal) ways. Third, they suggest that there are social influences on crime beyond those of social stratification, such as stigmatization, self-labelling, and conformity to antisocial norms. And finally, they suggest that crime can be rational, in that an individual can be more or less likely to engage in a particular crime as a function of an internal cost-benefit analysis. This all suggests that most crime results from social and environmental forces, but that there may be a sub-set of crimes that arise from genetic, biochemical or neurophysiological causes, which may interact with the social causes and with rational decision processes, or which may act on their own to bring about criminal behaviour in particular people and not in others. Crime then is only minimally related to a fundamentally selfish nature that needs to be controlled, and then most likely only in a small sub-set of people.

In light of this view of crime and human nature, it is possible now to reflect on what a society can do to most effectively control crime, while also living up to principles of justice, community protection, and collective well-being [1]. The existing criminal justice system in much of the western world prescribes public order control, policing (i.e., surveillance, threat of punishment), punishment, and incarceration for the purpose of public protection (i.e., separating delinquent or dangerous individuals from the rest of the population). This system is based primarily on a Choice approach to crime, which, as discussed above, assumes some inherent selfishness in all of us, which needs to be controlled through increasing the costs of engaging in crime. It seems, however, that inherent selfishness is not a primary cause of criminal behaviour, and that rational choice only comes into play once a decision has been made to engage in crime. As such, the current system could be largely ineffective in achieving its own aims, perhaps diverting crime from some areas into others, without actually controlling it or protecting the community at large. Worse still, it is well known that punishment decreases intrinsic motivation NOT to engage in the punished behaviour (in a phenomenon known as the ‘Overjustification Effect’), which means that a system based on increasing costs for engaging in crime could actually be having the opposite effect of encouraging law-breaking in the longer term.

The current system for controlling crime is therefore not the most effective, given human psychology and the likely causes of crime. Taking a more reasoned approach based on what the study of crime tells us about human nature, the most effective strategy for controlling crime would likely involve decreasing social stratification, diffusing power from more powerful groups to the society at large, and eliminating intergroup rivalries based around social or economic status differences. Historically this strategy has proven to be highly effective. More can be done, however, as there remain areas rife with poverty, which continue to be highly prone to crime. Decreasing social stratification could be achieved through better education, better health care, more access to social services and welfare, and, most importantly, through the democratization of the economic sphere, where power remains highly centralized (in corporations and in the hands of small numbers of extremely wealthy investors). This latter strategy would lead to greater fairness (which is a reason in itself for pursuing it), but also to more opportunity for a larger proportion of the population, thus decreasing strain and the likelihood that people will engage in criminal behaviour to satisfy their basic needs.

Beyond decreasing social stratification, a more effective system for controlling crime could also change attitudes towards stigmatized groups, and prevent the reactive emergence of antisocial norms. This would also decrease labelling and self-labelling, and therefore conformity to antisocial norms should they arise. Changing attitudes towards stigmatized groups could be achieved, as with the poverty reduction strategies discussed above, with better education, more access to social services and welfare for those groups, and again through the democratization of the economic sphere. In this case, democratization would likely engender decreased intergroup competition – because status differences are often at the heart of such competition – and therefore less discrimination and derogation of others, including of those engaging in statistically deviant but harmless behaviours.

If after instituting these higher level social changes there remained a group of people who continued to engage in crime, as some of the Trait approach literature suggests might be the case, the next best strategy for crime control would be rehabilitation. At present, incarceration is geared towards punishment, but what evidence there is on rehabilitation suggests that rehabilitation programs can be very successful if implemented properly and if administered in conjunction with post-rehabilitation and post-incarceration welfare procedures. Changing prisons from places where people who have committed a crime are punished to places where people who have committed a crime are given a second chance would also encourage easier reintegration into society after a program is completed. This would also almost certainly decrease rates of recidivism.

If after rehabilitation had been attempted, there still remained a sub-set of offenders who posed a risk to society, then incarceration for public protection could potentially be justified as a last resort. This sub-set of offenders would represent the small portion of the population who possessed some deeply entrenched antisocial behavioural tendencies, possibly for biological reasons or possibly as a result of social or environmental causes that are difficult to reverse. In these cases, rehabilitation should never be abandoned as an option, but insofar as rehabilitation were ineffective, public protection and community well-being become the primary concerns. Prisons then would function as rehabilitation centres, but also as institutions responsible for the separation from the community of offenders deemed likely to re-offend. In this role, prisons also then have a responsibility for the care of those incarcerated, whose human rights must always be respected, such that punishment of any kind should not feature in such institutions.

Overall, then, a new criminal justice system based around what we know about crime and human nature would first focus on decreasing inequality and the concentration of power in some groups at the expense of others, as the primary causes of most crime. Destigmatizing particular groups and statistically deviant behaviours could then also be attempted, as a way of preventing the emergence of, and conformity with, antisocial norms. Rehabilitation should then be attempted with those people who continued to engage in crime despite the greater equality and decreased stigmatization brought about by the larger scale changes to the social structure. Finally, as a last resort, the small sub-set of offenders who proved difficult or impossible to rehabilitate could be separated from the community, purely for the purpose of public protection, and only for as long as they continued to pose a risk. Should a criminal justice system of this sort be developed and implemented, we would have come a long way to living up to the principles of justice for all, community well-being, and public protection that such a system is supposed to promote. We would also be much closer to realising a better, fairer, and safer society, in which crime would be a rare aberration; a society in which all would be in a better position to reach their full potential.

[1] The first way in which a society could humanely control crime would be to have a fairer system for classifying behaviours as criminal. In practice, this means classifying particular behaviours as criminal only by consensus, rather than through processes of conflict between groups or through moral entrepreneurship, both of which unfairly promote the interests or beliefs of more powerful groups over those of less powerful groups. This would result in more freedom and more opportunity for the less advantaged, whose everyday lives are currently (and arbitrarily) criminalized to a large degree (for example, illicit drug users in impoverished areas).