Monday, July 07, 2008

Just a shot away

Chris Dillow points us to Arnold Kling's statement that his most 'absurd belief' is:

that human nature has changed in the last few hundred years. If you could go back to 1708 and replace all of the babies at conception with babies conceived today, my prediction is that the alternative history from 1708 to 2008 would have less violence, more economic growth, and faster scientific progress. Conversely, if you were to replace babies being conceived today with babies conceived in 1708, they would grow up to produce staggering increases in crime and violence.

Dillow follows up by suggesting that Kling's belief might not be so absurd. He cites the alacrity with which early 19th century gentlemen entered into deadly duels with each other over trivial matters, and he points to Steven Pinker's claim that violence has declined in the last few centuries. (I'm confining my brief comments here to Western Europe, since that's the context I know best, even if the tendency is also visible elsewhere.)

There is nothing absurd about noting the decline in violence in modern history, as this is now a fairly well-established consensus among historians, based upon several decades of empirical work in various countries. (I discussed this briefly in relation to Pinker's essay last year.)

The criminologist Manuel Eisner, having surveyed nearly 80 quantitative histories of homicide rates has concluded:

the data confirm the notion, now hardly controversial among historians of crime, that homicide rates have declined in Europe over several centuries. Typical estimates referring to the late Middle Ages range between 20 and 40 homicides per 100,000, while respective data for the mid twentieth century are between 0.5 and 1 per 100,000. The notorious imprecision of population data, deficiencies of the sources, shifts in the legal definition of homicide, changes in the age structure as well as improved medical possibilities, surely have to be accounted for. But the evidence is so consistent, the secular decline so regular, and the differences in levels so large, that it seems difficult to refute the the conclusion of a real and notable decline. (Manuel Eisner, 'Modernization, Self-Control and Lethal Violence: The Long-term Dynamics of European Homicide Rates in Theoretical Perspective', British Journal of Criminology 41 (2001): 618-638, 628)

It's important to note that this discussion of violence is based on homicide rates. While they are our most reliable measure of serious violence, they do not say everything there is to know about all of the violence going on in a particular society. Personally, I think there is likely to be a meaningful relationship between the prevalence of violence more generally and the homicide rate, but it's not a simple one and this issue is somewhat controversial. (I.e., it's possible for a society to be relatively violent but not particularly homicidal.)

Also, one should keep in mind that the decline in violence occurred at very different rates with somewhat different patterns and various ups and downs depending on which region you look at.

The English decline, for example, appears to have been a relatively smooth one, going from the 13th/14th centuries (c. 24 homicides per 100,000) to the late Middle Ages (3-9 per 100,000) to the mid-19th century (c. 1.8 per 100,000) to the early 1960s (0.6 per 100,000) (Eisner, 622-23). As Eisner points out, there was a similar pattern for the Netherlands. However, things look very different in Scandinavian countries. In Sweden, for instance, Eisner finds that the data 'suggest a spectacular decline of lethal personal violence by a factor of at least 10:1 within a period of only 150 years' (624) from the beginning of the 17th century (10-25 per 100,000) to the mid 18th (below 1 per 100,000); afterwards, there was a rise in the rate and then another fall beginning by the mid 19th century.

But taking into account all these caveats, the decline of violence in Western Europe -- overall -- seems have been a fact. (As does the rise in violence since the 1950s or 1960s depending on where you are in Europe, followed, again, by a decline in many places. However significant these have been, by longer-term historical standards, the 20th century variations in Western European homicide rates -- excluding war -- have been very small scale.)

The 'decline in violence' thesis seems counter-intuitive, obviously, considering that there still seems to be so much violence about in European countries. And perhaps there is: but not by historical standards (or by comparisons with many other regions of the world). Even admitting that our perceptions of violence are partly exaggerated by the media, one might however point out that the 20th century was an exceptionally bloody one (and the next one maybe isn't shaping up to be much better).

Taking war and genocide into account complicates the above picture of increasing civility, certainly. And in terms of raw numbers, the last century can likely not be beat. (So far...)

But Lawrence Keeley, in his excellent book War Before Civilization, points out that many low-technology tribal societies achieved rates of killing that dwarfed those in the last century. The tactic of utterly wiping out an opposing tribe (so, genocide on a small scale) was also not unusual. Keeley's book is fascinating and full of details worth discussing...but that is for another time perhaps.

In short, there's nothing absurd about saying that violence has declined, even if the issue is complicated.

However, what I do find a bit absurd (or at least questionable) about Kling's statement is the idea that this (decidedly non-absurd) decline in violence has been caused by (or caused) a fundamental change in 'human nature'.

This is a complicated issue obviously, and I'm not really up for a mammoth post today (I'm heading off for a crime conference and long research trip the day after tomorrow), so this is all rather off the cuff.

However, in other contexts -- primarily in reaction to Gregory Clark's A Farewell to Alms -- I have taken a sceptical look at the idea of rapid genetic change shaping our behaviour. While a firm believer in a meaningful human nature (and advocate of the idea that historians and other humanities scholars should take it seriously) I would align myself with those who have argued that it is a rather less malleable thing than Kling and others suggest.

Put quickly, there is no reason to think that a stable set of average innate psychological attributes (or potentialities) across a given population and over time could not generate widely differing behaviours as a result of changing conditions. If Scandinavians can reduce their homicide rates at least 10 fold in a century-and-a-half, this suggests that changing social relationships can have a powerful effect on this behaviour without there having been any kind of plausible underlying genetic shift.

Martin Daly and Margo Wilson have usefully examined how different kinds of social cues can have a strong influence on variations in risk-taking and social competition (especially among men, and the general decline in violence noted above is primarily the result of the pacification of males), and they have connected these to homicide rates. Furthermore, we can see the ways that previously pacified societies can become aggressive again, when the social factors that seem to have driven the decline in violence in the first place (increasing social exchange, more effective 'state' policing, the social encouragement of self-discipline, the provision of relative material security) are withdrawn.

Clearly, people in Western Europe seem to be behaving in a very different way than they did centuries ago. (Or even less than a century ago: in Germany the ritualised duel known as the Mensur lasted into the twentieth century.)

But you can see relevant changes on a smaller time-scale.

In my book on nineteenth-century violence (parts of which can be viewed via Google Books), I took a more qualitative look at the issue, examining the values and norms that surrounded that behaviour in England.

What you see during that time is a shift from a relatively shared culture of violence in the 18th century: violence was -- compared to later times -- far more acceptable, and social class was not so decisive in determining what you thought about it. This began changing in the late 18th and early 19th centuries, when some people (predominantly some parts of the upper and middle-classes and many of them religious) and the state started taking a much less sanguine view of disorder and interpersonal violence. 'Violence' itself was -- rather quickly -- identified as a 'social problem' in a way that was new and that has since remained important to British culture. It also became increasingly something that 'respectable' people did not engage in, and became associated with the lower classes. By the end of the century, even the 'respectable' working class were distancing themselves from behaviour that only a generation or two earlier would have been normal.

So, on the level of attitudes (admittedly somewhat less solid than that of statistics) there seems to have been a meaningful change in values in England over a century or less. (There were many continuities, and this was an ambiguous process in many ways, but I believe the general outline holds.)

I find it difficult (and unnecessary) to credit this change to genetics (other than the obvious fact that genes serve to shape the common human psychological structure that allows us to think, feel or recognise social cues, or provides us with various innate motivations).

In any case, there is no inherent contradiction between being a highly enlightened or intelligent person and engaging in violence.

Kling's claim, though, seems to rest on the idea that violence is something that only less-intelligent and less-rational people enjoy or might potentially engage in. However, there is no reason to think that violence is merely some sort of atavistic survival of primitive times indulged in by social failures. As Daly and Wilson have succinctly observe in their book Homicide, 'poor young men with dismal prospects for the future have good reason to escalate their tactics of social competition and become violent' (287). In a different context, Elijah Anderson made a similar point in his book The Code of the Street, arguing ‘a cultural adaptation to a profound lack of faith in the police and the judicial system – and in others who would champion one’s personal security’ (34).

For the vast majority of human history and prehistory, being good at violence (or at least suggesting to others that you were) was a factor in heightened rather than lowered social status.

And I think Kling is being a bit optimistic in suggesting that that is no longer the case.

Kling's formulation also makes the implied assertion, it seems, that, in a sense, we (however defined) are now much better people than we were, that, in fact, we could loosen up the very dense network of social controls, material prosperity, economic interaction and psychological training developed in, say, Europe over the last four hundred years and everything would still just fine...because of something to do with our genes.

This is not an experiment I'd like to see undertaken, to be honest. Most of Europe has indeed become a quite pacified place, and has been so for some time now. But I think we should be anything but complacent about that. Things, as we know, do fall apart.

(Moreover: why send twins back from 2008? The British homicide rate reached its nadir in the early 1960s. It's now about twice as high. If we're going to dream social science fiction dreams, why not go all the way?)

Speculations about genetics and human nature and behaviour are certainly worthwhile. And Kling's comment was offhand -- and he did self-label it 'absurd' -- so I don't want to go on about this too much ('too late!' came the voice from off-stage...). However, as my discussion, linked to above, of Gregory Clark's book suggests, these kinds of explanations seem to be becoming more popular. And I don't know why, because even a rather quick closer look seems to show how shaky they are.

Sadly, without access to a time machine and a lot of twins, it would seem to be impossible to design a test that could clearly disentangle the issues of nature and nurture across historical time.

(Moreover consider the Swedish example noted above: presumably Kling would have to argue that the same now-and-then relationship would also apply when exchanging babies born in Stockholm in 1600 and 1740? Remember, there appears to have been at least a 10-fold decline in violence in the intervening time. Were those in 1600 at least ten times more genetically bad-ass?)

But given the speed with which we have seen the prevalence of serious violence change in Western Europe (along with a set of quite reasonable if incomplete or still controversial explanations for why that was so), it would seem to me that the reach to genetics to explain complex behavioural changes is not only often implausible (and often made without seriously demonstrating how particular behaviours are genetically generated) but also unnecessary.

They may turn out to be relevant in some way at some point.

But I remain unconvinced.

1 comment:

hysperia said...

You've managed to articulate what I'm sure is very complex material in a way that someone unfamiliar with your field can understand - a rare gift. Thanks. I look forward to watching your posts.