Do Genes Bring Up the Rear?

The Masai of central Kenya and Tanzania use tones to distinguish between the subject and object of a sentence.

My previous two posts were about genes and language. One (about the FoxP2 gene, here) was about the kind of thing we take for granted these days; a gene has been found that causes speaking problems when it mutates. If you have the mutation, you have a problem articulating properly. This process makes easy sense to us because we expect genes to affect the individuals who carry them. The other post (about tonal language, here ) reported the much more surprising idea that individual behavior can be affected by the genes of an entire population. Thus, for example, if you live among a people where the ancestral  variety of the Microcephalin gene  predominates, you probably speak a tonal language, even though you have a more recent variety of the gene. Meanwhile, another person with that gene who lives among people where the ancestral variety is rare, probably does not speak a tonal language. I have the feeling that this discovery is one of those things, like Robinson Crusoe observing a footprint, that changes many expectations. The Dediu-Ladd paper on the relation between tones and genes (see here) calls for a new way of thinking about thinking. The nature-nurture discussion just took a new turn.

By itself, the idea that the distribution of genes in a population matters is not new. A classic example is the sickle-cell gene. One of them in a person helps resist malaria. Two leads to the sickle-cell disease. Optimally ,there should be some of these genes scattered around the population to help keep down the malaria rate, but not so many of them that it is overwhelmed by sickle-cell. In this case, although the overall population’s health is reflected in its genetic structure, the fate of individual members are still governed by their own genes. If you have one sickle-cell gene you are directly protected against malaria; if you have two, you have sickle-cell disease. If you have none, you are at risk for malaria, but the risk is reduced by the general reduction of malaria parasites in your area.

The Dediu-Ladd paper says something else. The speech of all a population’s members is reflected in its genetic structure and the speech of each individual reflects the population’s genetics rather than their own individual genes. Imagine how confused Gregor Mendel would have been if his short peas came up tall when planted with mostly other tall peas.

The endless nature-nurture debate seems to have taken another jolt. Fifty years ago, the ball was almost entirely in the nurture side of the dispute. Language was treated as a human invention that was learned reflexively, i.e., without conscious thought or internal processing. When the weakness of that approach was demonstrated, arguments for the nature side of the debate became predominant. Language was taken to be a biological invention that is acquired computationally, i.e., through internal processing but without conscious thought.

A typical nature-predominant account of language today refers to an innate capacity. We are born with (i.e., our genes provide) the ability to distinguish, say, subjects from objects. If, however, we are subject to some terrible fate and are left alone in a closet until age 15, we may never learn that distinction. Even if we are raised normally, our manner of expressing the distinction will depend on what language is spoken around us. So, in this view our genes set the bar and then experience determines how close to those limits we come. Genes determine one’s potential; experience determines how much of that potential is realized. Nobody ever exceeds their potential.

Daily experience provides a mechanical analogy. For the past many decades, the Intel corporation has regularly turned out computer processors with improved capabilities over previous products. The rate of software production that takes advantage of these capabilities has been much slower, and most people today have computers whose potential greatly exceeds their performance. But nobody’s computer ever performs beyond the level of its processors. Nature limits nurture.

This nature-predominant model promotes expectations about what the kind of research conducted by Dediu-Ladd should find. If all languages are learnable by all normal people, then every language is within a person’s potential and there should be no relationship between gene variation found between normal populations and the characteristics of their languages. For the most part, Dediu-Ladd did confirm this expectation. They studied 983 alleles and 26 linguistic features for 49 world populations, and found no relation between genetic structure and 25 of those linguistic features. The “black swan” in this study was tonality. There is a link between population genetics and using pitch to make minimal distinctions between words or syntax.

The nature-predominant model expects that if a genetic relationship is found, then normal people from other populations should not be able to learn that feature of a language. For example, if some populations could hear higher frequencies than the rest of us and used that range for some characteristics of language, the rest of us would not be able to learn their language. However, this expectation is not satisfied by the Dediu-Ladd study. We can all learn to speak tonal languages, or at least we could have if we had been introduced to one at an appropriately early age.

The implications of this surprise are very radical and I suspect that in the end the nature-predominates model will have to be greatly modified. We will never go back to the old behaviorist view in which nature provided a blank slate on which nurture could write whatever poetry it wished. But culture is very likely to recover some of its claim to power. And I suspect the role of consciousness will at last get some of its due. That’s what I suspect, although of course time will tell.

It wouldn’t be much fun writing a blog, however, if I just stopped when I got to the point where the traditional commentator shrugs time will tell and says no more. What do I think time is going to tell us?

Perhaps instead of marking the maximum capacity available to an individual, many genes, especially those that affect linguistic performance, provide no more than a bias. We can see this role easily in something as relatively trivial as a language’s tonal features because ultimately it probably does not matter whether a language is tonal or not. One kind of language as good as another, so the presence or absence of tonality bears a simple relationship to the genetic bias of the population. As infants it may take a bit more work for those without the tonal bias to master tonality, but the culture pushes them, and they push themselves, and master it. Cultural pushes and deliberate pushing of the self make the individual’s genes irrelevant to the story of whether one individual or another speaks a tonal language.

And what about all those many cases where linguistic details are not trivial? A language with a way for distinguishing subject from object is definitely better than one that cannot make that distinction, so much superior that all known languages can distinguish between actor and recipient of the action.  The old, nature-predominates assumptions led many to assert there must be some genetic basis for the distinction. Despite the high frequency of such assertions, the evidence is based on theory only, without supporting data. You can be skeptical and empirical at the same time.

If not nature, what then? Cultural and self-conscious pushes may be the way. I know many artists and talented artisans and all of them have taught themselves many little techniques and had many insights into this or that detail of what they work with. Most of them are not hailed as geniuses because others before and after them have pushed themselves to gain similar knowledge. But the ordinariness of the experience does not change the fact that they deliberately pushed themselves to a new level of skill. Perhaps half a million years ago one of the ordinary experiences of Homo was to spot the subject/object distinction and develop a routine for expressing it.

Similarly, I don’t know a single skilled artisan who did not learn the bulk of a craft by watching others. You want to write? Read! Write well? Read plenty! You want to make music? Listen! It is by paying attention to the culture and traditions already available that artisans learn most of their tricks. Perhaps half a million years ago, the insightful few were providing examples for the many of ways to express object/subject.

So Alley Oop asked, What’s she doing with that he saw her business? Ohhh! I get it. And having gotten it Mr. Oop plagiarized it.

That kind of recognition is standard today. Why shouldn’t it have been going on for the whole of Homo history, with genes bringing up the rear then, as they are today? I see no a priori reason to deny that the push of culture and deliberate attention is older than Homo sapiens. Indeed, now that I have read the Dediu-Ladd paper, I feel no need to assume that genes have been anywhere close to leading the way.

Comments

[Michael]

Thanks for the grat and very illuminating post.
Just one thing:
"if you live among a people where the gene known as Microcephalin-D predominates, you probably speak a tonal language, even though you do not have the gene."
Wasn't Microcephalin-D the allele predominating in non-tonal societies, so that you'd rather be more likely to speak a non-tonal language if you lived among a people with the derived version?

Michael
----------------------------------
BLOGGER: Eek. How did I make that mistake? I'm correcting it now, so further readers will not see the error, but this comment records the goof for posterity. I guess I should not make a post just before going off on vacation.

[Erasmussimo]

I'd like to offer some very cranky thoughts on the whole nature-v-nurture silliness. This is a political argument, not a truly scientific one. People have emotional investment in this issue, and that makes it almost impossible to discuss rationally. But I'll offer two basic observations:

1. It's NOT "nature versus nurture"; it's "nature and nurture". It is patent that both genetics and culture together determine human behavior. Arguments about which is more important are just silly political posturing. Some behaviors are more dependent upon genetics, some are more dependent upon culture. Trying to assess which of the two is more important overall is about as silly as talking about intelligence as if it were a one-dimensional trait.

2. Genetics and culture do not operate in parallel; genetics provides the foundation of behavior and culture builds upon and modifies that foundation. In some cases, such as socialization of males towards females, the cultural component completely overrules the genetic foundation. In other cases, such as gender-based differences in promiscuity, cultural factors either reinforce or have no effect upon behavior.

Ultimately, I see the whole nature-v-nurture debate as in the same league with, but more respectable than, studies of the relative differences in intelligence of different races.