The Dilemmas of a Libertarian



The rate of crime is an inverse function of the chances of going to prison. What a simple formula and how easy could be the life of the criminologist if it were true! For the rates are, on the face of it, easily measurable and the nostrum which Charles Murray presents us with has a heavy leavening of common sense. Indeed, the graphs show quite clearly that crime rose in England and Wales as risk rates declined whilst the obverse occurred in the United States. The American experiment is thus seemingly a success story and the prison boat that arrived across the Atlantic, as I write this article, to be docked somewhat controversially in Portland Harbour is a tangible export of American know-how and initiative when faced with the intransigent problem of crime.

Charles Murray is a brilliant polemicist, he has a knack of taking the conventional wisdoms of libertarianism and dramatically 'proving' the converse. He has done this throughout his career: in Losing Ground (1984), he lambasted the Welfare State for generating 'welfare dependency', in The Emerging British Underclass (1990) he argues that such dependency has generated an attendant culture where accountability for behaviour is undermined and the disciplines of the family and community disintegrates, whilst in The Bell Curve (1994), written with Richard Herrnstein, he provocatively endorses the existing social structure as increasingly reflecting differences in intelligence between classes and races rather than inadequacies in the meritocracy. How compatible these theories are with each other I am unsure. In the first book he has the poor as rational calculator evaluating the benefits of idleness as against work or crime as against honesty, in the second, in Murray's words, their 'habits of virtue' inculcated as youths are seriously eroded, in the last the poor are the lowest 'cognitive class' incapable of accurate assessment of risks and rewards. I sometimes test my students by calling them Murray I, Murray II and Murray III and ask them to try and draw lines of logic between them. On the face of it, they are not compatible. Each, for example, employs a divergent theory of crime, in the first crime occurs because of rational calculation, in the second it is because of lack of suitable upbringing (because of single mothers and feckless fathers), in the third delinquency occurs out of stupidity. In one the poor are rational calculators just like you or I (only poorer), in the second they actively endorse cultures which flirt with risk, in the last they are congenitally less aware of risk. Human nature may not have 'miraculously changed' in the Twentieth Century, as he argues in his conclusion, but it seems to change within the corpus of his own work. But even more miraculous is this new book. For here, let us note, he starts off with a very strong statement of the inverse relationship between risk of imprisonment and crime, which is premised on the rational calculator (Murray I) but by the middle of his text his argument crumbles and Murray II emerges. (Murray III, thankfully, does not set foot in the argument.) For 'habits of virtue' have been irretrievably eroded and the threat of imprisonment, by page nineteen, is no longer a magic nostrum. It does not affect the root causes of crime, it will never result in a return to 1950s levels of virtue, it is rather "a standoff" an amelioration of a dire situation rather than a solution. For it is, he acknowledges, only by tackling the problems at source that crime rates can substantially be reduced.

But let us turn back to the strong version of his thesis, for this is the message that will surely grab the reader: namely, that there is a direct and obvious relationship between high risk of imprisonment and the level of crime. This formula is a classic of common sense, yet it is as incorrect as its opposite, rather irritating liberal assumption that the crime rate has nothing whatsoever to do with the imprisonment rate. Indeed, the fact that both beliefs exist side by side as self-evident in the public consciousness, allows Charles Murray the leverage to dismiss one by appealing to the, just as impossible, mirror opposite. At heart the problem is the recurrent notion that the social is simple. Namely, that the social world is a relatively simple structure in which rates of different social events (eg marriages, suicides, strikes, crimes) can be related to narrowly delineated changes in other parts of the structure. In fact the social world is a complex interactive entity in which any particular social intervention can only possibly have a limited effect on other social events and where the calculation of this effect is always difficult. Thus the crime rate is affected by a large number of things: by the level of deterrence exerted by the criminal justice system, to be sure, but also by the levels of informal control in the community, by patterns of employment, by types of childrearing, by the cultural, political and moral climate, by the level of organised crime, by the patterns of illicit drug use, etc. etc. And to merely add together all these factors is complicated but insufficient for it does not allow human assessment and reflexivity. The perceived injustice of unemployment, for instance, or the felt injustices of bad policing or imprisonment. For the social is not only complex, like the natural world (who would ever think that only one factor would explain the weather?), it is even more intricate because each factor can be transformed over time by human interpretation. To take imprisonment for example, the same sentence can be perceived by the individual as perfectly just ('I had it coming'), or as an unjust and unwarranted reaction to a petty crime which stokes up further resentment, leading to more serious crime in the future, or as a rite of passage which everyone in one's social circle goes through. Thus strange and distinctly unlinear relationships can occur. Thus, low rates of imprisonment can act as an effective deterrent when the community is in accord as to the venality of the crime and the impartiality of the criminal justice system. In contrast, high rates of imprisonment can be counterproductive where they are seen as grossly unfair with regards to the level of gravity of the offence and the degree to which the system focuses on particular sections of the community rather than others.

All of this being said it would be truly remarkable if there were to be found a simple relationship between risk of imprisonment and the rate of crime where there was a one to one relationship, with all other factors fading into the background. This is even more so given the timescale (1950-1990) which Murray takes to contrast the United States and Britain. This is a period in which all social commentators agree has involved a battery of changes, constituting perhaps the most significant structural transformation of industrial societies in the last hundred years. Patterns of employment have changed, family and marriage has altered ineradicably, the role of women has been transformed, youth cultures have flourished, illicit drug use has spread, communities have disintegrates, the mass media has taken on a pivotal role in our lives, and, even factors which Murray has himself stressed, as relating to crime rates - the changing role of the Welfare State and the emergence of an underclass structurally unemployed, have occurred. And yes, also the rate of risk of imprisonment has changed. Surely it would demand a miracle for one factor to override all these others, for all the factors to change exactly in parallel in two very different cultures, and for a simple inverse law to be so easily and unambiguously detected? Yet this is what Charles Murray asks us to believe in the strong version of his thesis: with his dimmer switch theory of crime, turn the risk of imprisonment up and the crime rates fade.

Let us look at some of the evidence. At the very least one might expect countries with high risks of imprisonment to have low crime rates. The international comparison of crime rates is notoriously difficult because of differences in definitions of serious crime and the size of the hidden figure in each country. But homicide figures are reasonably indicative, at least of levels of violence. What are we to make then of the fact that the imprisonment rate in the United States is over six times that of Britain, and the risk of imprisonment per recorded crime over eleven times, whilst the homicide rate is seven times that of Britain, or indeed it is of that Switzerland which has an even lower rate of imprisonment than Britain (40% less), and, incidentally, widespread gun ownership. The response to this must, of course, be to move rapidly to a weak version of the thesis: to say that the high crime rate of the United States relates to many other factors rather than imprisonment and that the latter is only one factor out of many.

But perhaps it is wrong to make comparison between countries given the flaky nature of the data. Isn't it better to look at the impact of changes in risk of imprisonment over time within countries where the definition of serious crime remains relatively constant and all the other social factors are held, over a short time, fairly constant? Alas, such comparisons give even less credibility to Murray's thesis. For example, between 1987 and 1995 the United States increased its prison population by 124% and achieved a 2% increase in crime, whereas Denmark increased its prison population by 7% and had a 3% increase. Denmark has maintained a very low, stable rate of incarceration in this period, low risk rate 0.6 per 100 recorded crimes and a low crime rate, whereas the United States has maintained an extremely high rate of incarceration, a high risk rate and a high crime rate. At the moment the United States incarceration rate is just under 20 times that of Denmark. Why, we might ask, are not hordes of American scientists and senators going off to Denmark to find out how they got it right, and how they can find means of getting out of the vast and costly prison experiment which they have got themselves into?

And such examples could be multiplied: the Netherlands, for example, doubled their rate of imprisonment between 1987 and 1995 and the crime increase was only 8%. Surely, a success? Not really. Scotland, over the same period, had a 4% increase in imprisonment and only 4% increase in crime - half the increase of the Netherlands. Certainly this is a result which satisfies the virtues of value for money and efficiency which my countrymen are well known for!

Finally, let us look at an international comparison of European countries in the period of 1987-1995 in terms of risks of imprisonment and rates of recorded crime.






% Change in Risk of Imprisonment by 100,000

% Change in Recorded Crime

England and Wales





















Risk of imprisonment is measured in terms of prisoners per 100,000 recorded crimes.

Note first of all there is a built-in tendency for the risk of imprisonment, when measured in terms of the number of inmates in proportion to the volume of crime, to decrease with any dramatic increase in the crime rate. The prison capacity can simply not keep up with the increase and, as Murray notes, this is particularly true when politicians are unconvinced with regards to the effectiveness or prison. The causality is, however, the reverse of Murray's: crime rises because of a host of factors and inevitably reduces the risk of imprisonment rather than a reduction in the risk of imprisonment bringing upon us the rise in crime. Such a tendency is particularly evident whenever the rise in crime is considerable, as in the case of England and Wales, the Republic of Ireland, France and Austria. But where crime rises are small, the failure of Murray's Law can be clearly observed. He focuses on England and Wales where it might seem self-evident, but in neighbouring Scotland the risk remains the same yet crime rises and in Denmark the risk rises and so does the crime rate, whilst in The Netherlands, to finally put the nail in the strong version of the thesis, the risk rate goes up exponentially but the crime rate continues to rise.

Let us begin to reformulate the thesis. Imprisonment and its risk is certainly one factor that determines the crime rate, but it is one factor amongst many. Furthermore, it interacts with other factors: the cohesiveness of the community and the degree of legitimacy of the criminal justice system being obvious examples. Its effect, no doubt, varies by crime and by the subset of the community. Ironically, the white collar criminal is more likely to be deterred by prison than the lower working class youth with nothing to lose, who commits the sort of conventional crimes we send people to prison for. Risk rates, then, are important, but different groups will perceive the same risk differently. The prudent politician will hold such factors in mind just as he or she will be aware that to use one point of interaction as the major source of crime control is bound to have at some point declining marginal returns.

Murray's decision to focus so monopolistically on prison as a method of crime control stems possibly out of a fascination for the polemical, but it also arises out of a certain fundamentalism. He correctly notes that his present favoured formula fails to touch the root causes of crime: whether it be the left's belief in social inequality as a cause, or his belief in the Welfare State and its attendant 'culture of dependency'. It matters not who is correct, he observes, because socialism is scarcely around the corner, nor the will to dismantle the Welfare State as he would wish. I cannot see the logic in this, outside of the belief that without fundamental drastic change, nothing can be achieved. For there are many short-term and intermediate goals which can be realised: social inequality is not written ins tone, it has decreased in the past and will decrease again in the future. And as for the Welfare State, I doubt if advanced industrial societies are feasible without the underpinning of the Welfare State, but our experience over the last twenty years has certainly shown that such a bedrock of decency can be undermined by determined forces of the right. What is most paradoxical is Charles Murray's fixation on the prison, given the publication of his recent lengthy personal document What it Means to be a Libertarian (1997). Why, we might ask, is a doughty libertarian like Murray advocating the American penal experiment when he also believes that it is not laws regulating automobile safety which affect the level of accidents, but rather the infrastructure of engineering improvements and that, amongst other things, air traffic control, drug regulating agencies, and anti-discriminatory measures should be taken out of state control? Why should such a staunch opponent of the State put his weight behind the greatest intrusion into the lives of citizens that has occurred in a mature liberal democracy.

Let us look at the enormity of the American experiment. The United States prison population has more than doubled in the ten years from 1985 to 1996, there are now 1.6 million inmates, a population which, if brought together, would be the size of Philadelphia (Bureau of Justice, 1995). Furthermore, the imprisoned population is surrounded by a rapidly expanding penumbra of probation and parole. Now, 1 in 37 of the American adult population is under some form of correctional supervision, a city which, if gathered together, perhaps for some administrative convenience, would consist of five million adults, and easily be the second largest city in the United States (see E Currie, 1996). It is not part of the social contract which underpins liberal democracy that it should imprison and oversee so many of its citizens. Nor is it that it should do so whilst being so palpably incapable of protecting them. The rates of violence in the United States are exceptional amongst the stable democratic societies: its overall homicide rate is seven times that of England and Wales, its murder rate for young men is a staggering fifty two times higher, whilst large tracts of its great cities are no-go areas for its citizens, whether men or women.

The United States crime rate has, of course, declined in the last few years, although in terms of violent crime, particularly homicide, it would be more correct to talk of a levelling out at a very high level. Thus the overall crime rate declined by 3% between 1993 and 1996 but then the overall rate declined in twelve out of seventeen advanced industrial countries during this period (See Criminological Statistics, England and Wales, 1995). It is certainly nothing to write home about and even if this was due to the unlikely premise that such a decrease was predominantly due to imprisonment it is difficult to imagine what size of a prison population we would need to achieve Newt Gingrich's dream to bring the American rate down to European levels. Already some zealots, such as John Dilulio are talking of doubling the prison population in order to combat crime (see Mauer, 1997). It is difficult to see where this would end: a prison population the size of New York City, a correctional population the size of Los Angeles? Already some of our most perceptive critics speak of Zyment Bauman's terms of a 'totalitarian solution without a totalitarian state' (1995).

Widespread alarm bells are set off by the extraordinary racial focus of this cercarial experiment. Thus, 1 in 9 of African-American males aged 20-29 are in prison at any one point and 1 in 3 are either in prison, on probation or parole (Mauer, 1997). These are staggering figures, even more so when we consider that they would increase drastically if we were to look at 'have ever' figures and if we were to focus on poor blacks rather than include the large middle class African-American community. There must be sizeable slices of the ghetto areas of the United States where a young man is regarded as odd if he were not under some form of correctional supervision and any stigmatisation would probably be negative, indeed it would be those untouched by the law who were regarded as strange. And yet despite this extraordinary level of risk of imprisonment, violence continues to decimate the black community and indeed increases. Homicide is the foremost cause of death amongst young black men in America, just as imprisonment is the most likely career move in their lives; can there be any greater indictment of Murray's thesis than this?

Those of us in Europe who are friends of American democracy need to make clear our misgivings rather than import their mistakes. Our politicians, of course, of all perplexions, flit across the Atlantic to learn about this gross cerceral experiment. But to attempt to learn crime control from the United States is rather like travelling to Saudi Arabia to learn about women's rights. The one lesson to be learnt is not to travel down this path of punishment, to realise that if it takes a gulag to maintain a winner-takes-all society, then it is society that must be changed rather than the prison expanded.