Almost everyone is interested in murder, in theory if not in practice, and literature would be quite at a loss without it. Suppose, for example, that Hamlet’s father had died of old age, or Banquo of a pulmonary embolus; one cannot help but feel that the subsequent drama would have been rather less compelling.
In my time as a prison doctor I have met quite a few murderers, one or two of them world-famous in their field; and I have found that anecdotes of such murderers, suitably anonymized of course, seldom go amiss at dinner parties when conversation flags. Indeed, they seem to revive flagging appetites also, for there is nothing like a nice asphyxiation to go with a crème brûlée.
It has to be confessed, however, that interest in murder is mainly of the prurient kind. By contrast, the academic study of unlawful killing can be rather dry — the academic, after all, can suck the juice from any subject — and anyone looking for cheap thrills in Prof. Randolph Roth’s book is likely to be disappointed. It contains many vignettes of individual crimes, but they are cursory and over in a couple of lines, with none of the lingering over detail so beloved of readers of crime fiction.
The fact is that most homicides are merely sordid rather than intriguing, a fact noted long ago by Thomas De Quincey in his famous essay “Murder Considered as One of the Fine Arts.” In any case, Professor Roth is a historical sociologist rather than a novelist, and treats homicide as an aggregate social phenomenon, rather as Emile Durkheim treated suicide. He is more interested in statistics than in stories, the latter being for him but corroboration of the former.
Homicide, like suicide, appears at first sight to be the most individual of acts: the outcome of a unique response to unique personal circumstances. However, it is subject on a population level to statistical regularities and fluctuations, and this leads naturally enough to a search for influences or forces that go beyond those of individual psychology. Even the fact that most murderers are male, a constant in all societies, suggests a hormonal component to murder.
Professor Roth examines the history of homicide in America from the colonial period to just before World War I, with only a brief excursus into more recent times, mainly through the lens of fluctuating rates in time, region, and race. Class does not come much into his analysis, perhaps because the statistics on class through most of that history are lacking, and would be very difficult to calculate.
His conclusions from the statistics cannot, of course, be more certain or credible than the statistics themselves. At times, indeed, he shows a rather naïve confidence in the correspondence of his numbers to reality. He thinks that his statistics can reveal the “final,” “real,” and “root” causes of a social phenomenon such as homicide, as if any such finality could exist.
There are other problems. He uses the homicide rate per 100,000 inhabitants as his preferred standard of comparison, but this is not legitimate in comparisons of very large with very small populations, or of different small populations. In very small populations, unusual events can give a misleading statistical impression. One would not say that in the Smith family the murder rate in a certain year was 75,000 per hundred thousand because in that year Mr. Smith killed his wife and two children, or that, over a ten-year period, it had been 7,500 per hundred thousand.
And, if it is true that homicide is everywhere and always disproportionately committed by one age group rather than another, the age structure of a population, which can and does vary over time, alters the validity of comparisons made without appropriate adjustments. Furthermore, homicide is the endpoint of a process whose outcome varies across ages. For example, it has been calculated that if the same resuscitation and surgical techniques were used today as were used in 1960, the homicide rate would be up to five times more than it is. People whose lives are now saved would in those days have died, and this gives a false impression of the relative violence of society then and now.
And if it is true that it is easier to kill with a handgun than with a handsaw, say, a lesser degree of homicidal intent will be necessary to result in a fatality with the former than with the latter. (Of course, it is possible that the presence of lethal weapons might make everyone more careful.) There is therefore no necessary connection between homicidal intentions or emotions and the rate of homicide, as there needs to be if you are going to draw the kind of conclusions that Professor Roth draws.
Of course, some statistical changes are so startling that they overwhelm any possible caveat. For example, in 1950 there were 200 violent crimes recorded in the whole of New Zealand; in 1999 there were 70,000. Even allowing for the doubling of the population, and for differences in the way crimes were classified and recorded, the difference is so great that it must surely reflect a change in reality: a conclusion borne out by everyone’s personal experience.
Some of Roth’s statistical differences are of this order, but many are not. It seems beyond doubt, for example, that the Gold Rush days in California were very violent: The Wild West really was wild. But quite a number of his comparisons are based on statistics whose margin of error might easily eliminate the putative differences on which he relies. There is not much discussion of these problems in what is a very long book.
But let us take the figures at their face value. What does Roth say that they show?
In essence, he says that the homicide rate is a reflection of the state of social and political solidarity of a population. Social solidarity is not necessarily good by our current moral standards, however. For example, where whites felt it important and necessary to dominate blacks, and were actually able without difficulty to do so, they were less inclined to kill one another (or blacks for that matter). But once their dominance was threatened, they started to kill, if not with abandon, at least more frequently, and often over trifles.
Why should this have been? Here Roth has a plausible psychological explanation that accords with my own experience of murderers. (It is one of the strengths of his book that he is able to articulate both individual psychology and larger social and economic forces without in any facile way seeing crime as the inevitable outcome of poverty.) People who live in a very hierarchical society — as was the South — need to feel superior to someone for the sake of their own self-regard. Membership of the dominating race conferred this feeling automatically on whites when blacks were officially held inferior. But when the old hierarchy was overturned, even only partially, this comfort was no longer available to poor whites. They therefore became irritable over supposed matters of honor, such as perceived insults and slights, and ready to kill or to die to defend that supposed honor.
The author explains, sensibly, that in such matters people’s perception of reality is more important than reality itself. If people think that society is just and founded on a rock, it matters not, at least for a time, that from another standpoint it is unjust and founded on sand. With their illusions, they will be peaceable. What mattered about FDR, for example, was not that he tackled unemployment during the Depression, but that people thought that he did: He and his government were doing their best and cared for the little man. As a result the little man was not so frustrated by the Depression that he killed his neighbors out of pique — even if, in fact, the New Deal actually prolonged his poverty and unemployment. The homicide rate did go down in the Roosevelt years before World War II.
But here, it seems to me, the author misses a trick. It is curious that as Western societies have become both richer and more just in the Rawlsian sense, so their crime rates have risen rather than fallen, precisely the opposite of what one might have predicted. Why is this?
The answer to this seeming paradox lies in meritocracy. Irrespective of how meritocratic our societies actually are, we believe we live in meritocracies. The good news about meritocracy is that we can attribute our successes to ourselves, like Josiah Bounderby of Coketown; but the bad news is the corollary that we attribute more of our failures to ourselves also. Since it is very difficult for human beings to do this, and failure is more common than success, many people take refuge in a prickly resentment, and a proportion of them are ready to resort to violence.
A meritocracy, therefore, is in urgent need of a repressive apparatus, for where else is control of criminality and violence to come from? Religion and ethical teaching are by themselves likely to be insufficient. On the matter of the criminal justice system, however, Professor Roth is ambivalent. He several times states that punishment has had no discernible effect through history on the homicide rate in America, though he can’t fully believe this since he also states several times that the homicide rate rose dramatically wherever the state and the criminal justice system were weak or nonexistent.
Moreover, the recent falls in homicide in New York and California do suggest that if you lock up people with a propensity to violence, there is, unsurprisingly, less violence all round, including homicide. One important aspect of homicide that the author scarcely considers is the violent histories of most killers. Rarely is homicide the first occasion on which a man breaks the law with violence.
The author has clearly labored mightily over this book; he has gathered an immense amount of information. He writes clearly; he never descends to hiding his meaning behind a smokescreen of obfuscatory jargon. It seems churlish then, and it gives me no pleasure, to say that his book is neither a joy to read nor enlightening in proportion to its length. Rossini’s judgment of Wagner kept recurring to me: The book has its good moments but its bad quarters of an hour.
– Mr. Dalrymple, a doctor, is the author of several books, including The New Vichy Syndrome: Why European Intellectuals Surrender to Barbarism, forthcoming from Encounter Books in March 2010.