| Home | About Edge| Features | Edge Editions | Press | Reality Club | Third Culture | Digerati | Edge:Feed | Edge Search |

The Two Steves
Pinker vs. Rose: A Debate

Patrick Bateson on The Two Steves


From: Patrick Bateson

Comment on the debate between Steve Pinker and Steve Rose

For many years I have amused myself by noting when one of my colleagues calls another a fool (usually behind his or her back). By degrees, I have assembled triangles or circles of these highly intelligent people (A calls B a fool, B calls C a fool and C calls A a fool). I react in somewhat the same way when two very clever people, as both Steves undoubtedly are, demonise each other, as both undoubtedly do. I reckon that it is possible to find ways of bridging their positions in ways that are likely to be highly productive scientifically and helpful socially.

Some of the advocates of evolutionary psychology seem to want to revert to the brand of old-style sociobiology which, in Ed Wilson's phrase, "decoupled" individual development from the project to link evolutionary biology and behavioural biology. It is quite plain, though, that Steve Pinker does not want to sink back into a nothing-but genes position. However, I am not sure quite where he stands on instinct. At least eight different ways of characterising it have been used over the years. These are: present at birth; a behavioural difference caused by a genetic difference; adapted over the course of evolution; unchanging throughout development; shared by all members of a species; present before the behaviour serves any function; not learned; and a distinctly organised system of behaviour driven from within the body. These are separate dimensions and they don't necessarily hang together. If we take a particular case, evidence for one of the characteristics of instinct shouldn't imply that evidence for all the others will be found. What is particularly important, when assessing the more extreme claims of evolutionary psychology, is that behaviour that was adapted to its present function during evolution may itself be learned in the course of individual development and highly labile when environmental conditions are changeable rather than stable when the behaviour evolved.

It seems to me that two agendas have to be disentangled. One is simply to have a language that describes the variety of ways in which a given pattern of behaviour may be characterised in terms of its origins and development. The other is to have ways of understanding the processes of change during development. The first agenda may be met provisionally by simply describing what is known about the characteristics of the behaviour in terms of the various ways in which instinct has been defined. The second requires hard thought about the orderly cooking processes of development. It is true that the many different approaches to the old instinct problem have helped such a program because they have shown how a great number of developmental processes combine in a profusion of ways.

Like many others who are interested in evolution, development, brains and behaviour, I do not see a great deal that is new in evolutionary psychology. It would be a great pity if the admirable project to bring different bodies of thought and knowledge together failed because excessive claims were made for the value of one of those bodies. A sense of proportion must be preserved about the value of a Darwinian approach to homicide, let us say. Sure, there are sex differences and age differences, but the differences between cultures account for much more of the overall variance. The old distinction between statistical significance and effect size needs to be brought into play.

So, I too have sensed what Steven Rose has detected, namely a messianic tendency among some of the would-be Darwinists. Nevertheless, I feel much more comfortable than he does about bringing together the four approaches to behaviour advocated by Niko Tinbergen, thereby bridging the gap between the why and the how questions. It is easy to make fun of those who spin plausible functional stories, but the half-lives of the majority of the stories are very short. They collapse in the face of the evidence - as undoubtedly would any of Steven's functional explanations for the preponderance of Stevens and Richards who are active in this debate. There are many good examples now of alternative functional explanations being run against each other with lethal consequences for most of the hypotheses. Indeed, one of the jewels in the crown of sociobiology, namely parent-offspring conflict theory made predictions about the course of relations between parent and offspring that turned out to be simply wrong. That being the case, the theory can't have been vacuous; if it had been, no test would have been possible.

If the functional and evolutionary approaches are to be helpful to those who work on mechanisms of behaviour, they must make it is easier for us to organise and understand the data which we have already available to us. In my own field of behavioural development, explanations in terms of current utility have helped to clear the decks by distinguishing between behaviour that meets the needs of the young and the precursors of adult behaviour. And attempts to uncover the adaptive regularities of learning have proved illuminating, even though nobody should underestimate the difficulties of doing this. It seems very likely that the initial rules for learning are themselves unlearned, universal and the product of Darwinian evolution. Does that mean all human behaviour is predictable? The answer is emphatically "No".

The point is made obvious by taking a rule-governed game like chess. It is not possible to predict the course of a chess game from an ability to distinguish between king, queen, rook, bishop, knight and pawn and knowledge of the game's rules. The players are constrained by the rules and the positions of the pieces, but they are also instrumental in generating the positions to which they must subsequently respond. The range of possible games is enormous and virtually impossible to predict. In other words, simple underlying rules can generate surface behaviour of enormous complexity. Inferring the underlying rules from watching a lot of instances is possible, but is much more intellectually demanding and much more open to equally plausible alternative proposals than say offering an adaptive explanation for a dark skin in a hot climate.

The development of behaviour in humans, which has a lot of the same characteristics as cooking, has yet another dimension. As in the kitchen of a large restaurant, many different dishes are being cooked at the same time. Sometimes the behavioural dishes are thrown together and something quite novel - and useful - is serendipitously produced. Humans, opportunistic as they are, and aware of at least some of the things they do, are perfectly capable of appreciating the value of these experiments. A combination of spoken language, which has obvious utility in its own right, and manual dexterity in fashioning tools, which also has its own utility, combined at a particular and recent moment in history to generate written language. The discovery of written language took place several times and in several forms in different parts of the world with ideas represented by pictures or spoken sounds represented by symbols. The techniques, once invented, were quickly copied and became crucial elements of modern civilisations. Attempts to factor out the role of evolution in this cooking process make no more sense than using those massively misleading heritability ratios. Analysing the precursors is not the same as understanding how the cooking works.

The impact of evolutionary thinking on human can be highly beneficial. Some of the apparent support for social injustice, seemingly provided by sociobiology, was based on a muddle about what happens in the course of individual development. As this was straightened out and genetic determinism fell away as a serious issue in the debates, I believe that the biological knowledge has helped the understanding of social issues by showing precisely how human potential is expressed in some conditions and not seen in others. However, what is needed in approaching such problems is constructive collaboration between biologists and social scientists and a proper respect for the insights that the different disciplines can provide. I suspect that, outside the debating chamber, both Steves would agree with that.


Back to The Two Steves

| Home | About Edge| Features | Edge Editions | Press | Reality Club | Third Culture | Digerati | Edge:Feed | Edge Search |

| Top |