Nature v Nurture

There is a good diary over on Kos concerning the so-called nature vs. nurture debate. This is one of the more contentious, and in my opinion silliest, debate going on. There is a strong tendency to believe that given that each of these two things (nature and nurture, genes and environment) contribute to the characteristics an organism has each factor can be assigned some ranking, or percentage, of importance in the characteristics seen. This view is completely unfounded. I go into this in more detail here, but both genes and environment are 100% necessary and their role cannot meaningfully be divided into degrees of responsibility.

One comment to the diary, by aludlam, had this to say

But (and I'm being theoretical here, so don't slam me too hard)

If we could say with some level of confidence that one side of that equation is definitely less involved than the other, is it not prudent to come to the conclusion that the opposite factor is dominant?

eg - we look at 20 pairs of twins raised apart, and 19 of the pairs wind up having similar tastes in music, based on a survey of songs. We obviously don't have all the data on their development, but the end result leads us to believe that genetics plays a significant role in predisposition toward certain types of music - perhaps so much so that genetics determines the ear formations that determine what melodies are appealing absent some aversion event (eg, one of the 20th pair of twins got dropped on his head while listening to Mozart as a baby).

We can't measure these things except in the objective. Doesn't mean the objective is a worthless measurement. Just that more care is needed in interpreting the results.

To which I responded

Part of the problem is, what range of environments are you considering. See "environment" is not a single thing, but a near infinite range of possible things. If you could have a measure of the ability of identical twins to correctly identify faces at a given age and at 30 feet, say, taken throughout human history. You would find an exceedingly strong correlation with genetics, because that ability is largely determined by the shape of the eye, which is set by genes. But you would also have a large group of folks (since the invention of eyeglasses and in places where glasses are available) for whom the correlation would disappear. How does one classify that in terms of nature vs. nurture?

There is just no meaningful way for us to describe the degree to which "environment" can affect a characteristic when there are so many possible "environments".

I occurs to me that another analogy might help as well. Trying to assign a degree of responsibility to genes and environment is about as useful as trying to assign degree of responsibility to air, fuel and spark in the operation of your car's engine. Each is essential and the running of your engine is a result of the interaction between the three. Changing, or removing, any of the three will affect your engine's performance, but we cannot say that it's running is 33% due to the fuel, 33% due to the air and 33% due to the spark. Nor can we make amy more sense by assigning different percentages. It just is not a sensible way to understand how these components relate. A similar relationship holds for genes and environment.

But also note that I am in no way saying that because we can't assign some particular percent responsibility to genes that therefore they do not matter. This makes no more sense than to say that because we cannot assign a particular percentage responsibility to fuel in your car's engine, that therefore fuel does not matter. It again is just not the way these things work.

Larry Summers's Comment

This is a bit dated now, but I have to come down on the side of those who walked out and protested the comments by Larry Summer concerning genetic factors in the number of women in the sciences.

The first thing to understand is that the debate, at least as it is carried out in the popular press, is very confused. The way that genetic and environmental influences are discussed clouds the issue and makes for a great deal of silly commentary. People are often talking about characteristics and whether or not they are determined by environment or genes. This is a very misleading way to think about the role of genes and environment on the ultimate characteristics of an organism. Essentially every characteristic of an organism is, in some way, controlled by genes. However, some characteristics are highly sensitive to environment, others are not. For example, blood type is certainly determined by certain genes. There is, as far as I'm aware, no environmental influence that will change blood type at all. Blood type is extremely insensitive to environment. On the other hand, characteristics such as adult height or quality of vision are certainly determined by genes, but they are also extremely sensitive to environment. The typical nature vs. nurture distinction, unfortunately suggests that if you can identify a clear genetic component that there is therefore no significant environmental influence. This is just not the way that the universe we actually live in works.

So to bring this to relate to Larry Summers's comments, there is no doubt some genetic components, that might well we sex related to how one will perform as a scientist. But the important question then is, how sensitive to the environment are these factors. Now we already know that there is a great deal of sensitivity to some environmental factors. Having no graduate programs that will consider admitting a woman is an environmental factor which it has been well demonstrated will reduce the number of women scientists. But the question still remains as to how much of the difference today in the number of women and men in science is due to factors that are sensitive and insensitive to the environment.

Now to demonstrate the proposal that a given characteristic is insensitive to the environment is to try and demonstrate a universal negative. As far as I know, someone might well find a compound which if added to a mother's diet will change the blood type of a fetus. I doubt it, but I can't prove that no such thing can happen. The problem of the universal negative. For this reason, scientists will sensible investigate the proposal that a given characteristic is sensitive to the environment. This, it seems to me, was the goal of the conferees to whom Dr. Summers made his comments. Given what we know of sexism, women in the sciences in this century, etc. it seems to me to make a great deal of sense to continue to investigate the possibility that there remains some considerable environmental impact on the discrepancy between the number of men and women in science. Furthermore, if the discrepancy is sensitive to environmental influences, there is a great deal of social benefit to modifying environments so as to reduce the current imbalance. The social benefits flow not just to women who could have successful careers in science but are diverted away, or to those who are forced to put unnecessary effort into overcoming environmental obstacles. In addition, the process of science depends critically upon having a diverse group of independent critics to review scientific developments. The diversity of the group is necessary to insure that as many errors as possible can be caught in any given scientific proposal. To unnecessarily exclude a substantial number of individuals with potentially novel outlooks on scientific issues can be expected to impede scientific advances. So it is certainly sensible to investigate the potential existence of environmental factors influencing the disparity between numbers of men and women in science, which the conferees, by all accounts were in the process of doing.

But if what I've said so far is correct, Larry Summers's comment was at best of no value at all, and at worst intellectually vacuous and deeply insulting. Certainly there may well be genetic factors, but the only meaningful question is 'are there environmental factors'. The only sensible approach is to ask if there are environmental factors and search for them. Larry Summers's comment served to say no more than, maybe you shouldn't bother to look. No matter what genetic factors may be shown to shape performance as a scientist, the important question still is 'are these factors like blood type or are they like adult height or eye sight?'.

Larry Summers's comment strikes me as being very much like the ID movement, and intellectually vacuous for the same reason. In that case, there is considerable evidence that all species are linked by ties of common ancestry and the scientific community is sensibly searching for possible evolutionary mechanisms that link the different lineages and continuing to do so, so long as research continues to reveal promising lines of investigation. The ID movement is providing no new lines of investigation, it is simply trying to say stop looking, give up. Larry Summers's was doing the same thing, and given the potential social costs of the current arrangement, to advocate surrender is both an intellectual failure and insult.