10 thoughts on “Pinker

  1. Thanks for kicking off this debate John. I have decided not to apply for more credit and to use a few minutes in more socially relevant manner.

    The nature/nurture debate has an incredible capacity to go on for ever without making progress, a characterisitic shared with the “people are altruistic vs selfish” debate and the “people are rational vs irrational’ dispute.

    It helps to discriminate, as John has done in part, between (1) characterisitics that are hard wired (eye colour, blood group, capacity for language), (2) characteristics that depend about 100% on enrivonment (our native language), (3) characteristics that depend on a combination of nature and nurture (height, intelligence), (4) scientific theories that describe how things are and how they work and (5) theories of value including ideas about personal morality, aesthetics and politics.

    It is probably useless and unhelpful to try to reduce, explain or justify ideas of type 4 and 5 in terms of nature or nurture, they need to be appraised in terms of their capacity to solve problems and stand up to various kinds of criticism including practical tests.

    Explanations of social events are best approached by a situational analysis to consider the (more or less) objective elements of the situation, the way that it is perceived by the actors and the ideas that they have in their heads about the way they are supposed to act in that situation.

    Focussing on the control of violence, this can be pursued in the short term by changing the situation (physically separating potential combattants, for example) and in the medium to long term by more fundamental changes in the situation (eliminating gross differences in wealth or access to public goods, for example) and by changing the dominant theories and value systems that tend to promote violence. For me, this means promoting free trade under the rule of law, and all those theories and ideologies that impede the implementation of free trade under the rule of law are conducive to conflict and war.

  2. Your review seems to me to make a mistake. You seem to think that Pinker either overestimates the proportion of our behavior due to genes, or perhaps makes a reasonable assessment, but buries it in his rhetoric. But it seems better to say that estimates of the proportion of our behavior due to genes are meaningless, not overestimates. When a population biologist gives an estimate of heritability, they relativise it to a population in an environment: given the facts, we can say that the proportion of variance in the height of corn (say) is 70%, or whatever. But we should not understand this as the claim that height in corn is 70& ‘genetic’. Suppose the field of corn was genetically homogenous: in that case 100% of the variation in its height would be due to environment. Suppose its environment was perfectly uniform: in that case 100% of its variation would be genetic. The same phenoptypic trait in the same plant might have a different heritability depending upon the comparison class. Heritable doesn’t mean innate, or unchangeable by environmental intervetions, or whatever. Nothing in our phenotype is due to genes alone (even if heritability is 100%), nor is anything due to environment alone (even if heritability is 0). We would be better off focusing directly on whether a trait is easy or hard to change, which is orthagonal to the ‘nature/nurture’ debate.

  3. Neil, I thought I covered these points in the review, here

    “Pinker suggests that about 50 per cent of the observed variation in individual character traits within modern societies is genetically determined. (This proportion is conditional on the amount of variation in environment for the population being considered, and would be much lower for comparisons between societies.) ” [emphasis added]

    and also in the concluding discussion.

  4. I don’t think your disclaimers get to the heart of the matter. It seems to me that you raise doubts about the reliability of Pinker’s figure; with another comparison class, the figure might be much lower. Granted. My point is that, whatever the figure is, it is irrelevant to the question at issue. Pinker is concerned whether human behavior is innate or hardwired, and heritability figures are not, as he seems to think, estimates of the extent to which traits are innate (whatever innate means; Paul Griffiths suggests that innateness is meaningless and he may well be right).

  5. I haven’t looked, but I have sometimes wondered if there have been studies done on the cumulative effect of heritability on cultural transmission. To give a simplistic example, suppose that some characteristic was 90% culturally transmitted; under a simplistic model, using successive relaxation, that would converge on a limit that was independent of the initial cultural stock because the 10% would keep working each generation to dilute it; in the end it would determine the limit.

    Obviously there are many things that could go wrong with that, so it seems unrealistic, but on the other hand it also seems to suggest that small levels of heritability might be material after all – taken cumulatively over generations. So clearly “more study is needed”, but it’s worth asking what if anything has already been done along those lines.

  6. PML,

    I don’t think anyone has done a quantitative study but the general pattern is obvious. Look at any third-generation descendant of migrants, for example.

  7. JQ, your observations don’t test my hypothesis. Third generation migrants have been receiving feedback via nurture from their contemporaries, so it includes a feedback from the ethno-cultural mix three generations before and not from their own immediate ancestry.

    Let’s see if I can bring it out with a neutral thought experiment that won’t (hopefully) tread on any unthinking reactions.

    Suppose certain types of people have difficulty with certain vocal sounds. Suppose further that a whole population of these is taught some language that makes heavy use of these sounds, and is prevented from practising their parents’ language (maybe by making them a “stolen generation”).

    Now consider what will happen to their dialect over several generations. Mostly it will somewhat resemble their parents’ version, but it will gradually shift to reflect the sounds that come easiest to their hereditary shapes of noses and throats or whatever. This shift will follow a rule something like “take 10% of what comes naturally and combine with 90% of what you heard when growing up”.

    What this will converge to will largely reflect what comes “naturally”, despite the dialect being largely culturally conditioned. In this simplistic model, that dominates whatever the starting position was.

    Of course, the real model might have several stable limits with basins of attraction, the real cultural features might be dominated by non-physical things, and any convergence might be far slower than other culturally driven evolution of the culture (which isn’t stationary after all even when left to itself, except in some cultures which have all now been destabilised from outside).

    Given all these, though, it seems equally simplistic to dismiss the role of nature in nature v. nurture questions just by seeing which of the two is dominant in a one on one comparison. The thought experiments show that even a weak “nature” influence could theoretically dominate over time. So, the important question arises, how do we distinguish and test the various possibilities? Has anyone done that, and if so how? For if not, we cannot let the matter rest with what amounts to a randomly chosen result.

Comments are closed.