Drifting Along with the Tumbling Tumble Weed

The adaptive value of syntax—that is to say, the survival advantages that made syntax universal among humans—may never be known, reported Dr. Massimo Piattelli-Palmarini to a biolinguistics conference held in the Dominican Republic last weekend. In a survey of both biological and linguistic reasons to question “classical adaptionism” he suggested that if you think of understanding the evolution of sytnax as involving a 1000 details, ideas about survival advantages will only explain around 5 or 6, leaving investigators with no idea how to proceed. Ultimately, he concluded, investigators do want to make sense of the “Full Monty” (the language faculty as it is broadly understood), but that understanding may not have much to say about survival. He made a strong case, although I suspect that if he wins it he will have destroyed all hope of ever getting the general public to be interested in the origins of syntax.

The presentation appears to have been given in the spirit of a lawyer representing a side, for Piattelli-Palmarini seemed unconcerned with doing justice to both sides of the argument. For example, early in his presentation he offered “a revealing quote by an ‘unbiased’ expert,”

Natural selection has always been considered a key component of adaptive divergence and speciation, but the importance of selection has been eclipsed in recent decades by a strong focus on the geography of speciation and on the purely genetic mechanisms by which reproductive isolation evolves” (Emphasis added)

The quotation was accurate. Yet, like an attorney saying, “Thank you,” to stop a witness from saying more, the passage does not really give the thinking of its source (an editorial in the Proceedings of the National Academy of Sciences by Christopher Schneider, linked here). Doing a routine check I was startled to discover that the paragraph that opened as quoted went on to say the wind has shifted back toward selection,

Currently, there is a resurgent interest in the role of ecology in speciation. Several recent studies have emphasized the role of ecology and selection in speciation, regardless of the geographic context in which populations diverge (p. 12398)

Nevertheless, Piattelli-Palmarini presented a strong case against adaptive syntax. There are two general processes for biological change:

• Drift is any set of changes over time that happen by chance. These differences mount and you get different species.
• Selection means the drift has led to differences in fitness and that some changes have survived while others have not.

As all change begins with drift and only some changes undergo selection, the importance of drift is not a serious question in biology. Natural selection is usually suspected in cases where traits display some adaptive function. A seal’s fin was likely adapted to its purpose through natural selection, while the various differences between the spots on giraffe species may well simply reflect drift. At bottom, then, an argument for understanding the origins of syntax without looking at its adaptive role is an argument for drift alone.

That might seem like a shocking idea—we got sentence structure in pretty much the same arbitrary way the zebra got its stripes—but appeals to selection are, as lawyers say, an affirmative argument. You’ve got to prove your case. And the Piattelli-Palmarini presentation showed you cannot just assume that selection occurred.

The great stunner of biology in recent years has been the discovery that humans have so few genes (about 25,000) and that so many of them are identical with the genes of many other species. Mice have about the same number of genes. Piattelli-Palmarini reports that humans and mice have 99% matching genes, 96% of them in the same relative order in the chromosomes. Before the human genome was mapped, biologists expected to find 100,000 or more genes. The actual count, therefore, didn’t just indicate the experts were wrong. They didn’t know what they were talking about.

This finding has forced attention on some facts that were well known but whose meaning was not truly grasped. For example, caterpillars and butterflies have different organs and behavioral repertoires, yet a butterfly has exactly the same genetic makeup as the caterpillar it used to be. Cater-fly genes cannot be mere “recipes” for organs. Network operations, timing, laws of form, and principles of self-organization must have roles to play in the transformation of genetic information into organisms. Press the genes one way, you get a caterpillar; another way, a butterfly. In such a complex system, one change might lead to many changes. Thus, even if one part of the change is undergoing selection, it is hard to discover what that selection is amid the general drift that is producing all the other differences.

Interesting, you say, but what has all this got to do with evolution of language? Piattelli-Palmarini eventually got around to his point. Syntax is complicated and has functionless features. The example given was conservativity, a universal semantic property that seems to have no explanation based on psychology, logic, or social requirements. (Look here if you want a fuller discussion of conservativity.) Piattelli-Palmarini set forth the orthodox position of syntacticians that there is no reason apart from the nature of syntax for this property and therefore it cannot have evolved for adaptive reasons.

In the long run, this argument strikes me as risky. An old line of reasoning went, I can find no way to explain X, therefore it is a product of the will of God. The new line goes, I can find no explanation for X, therefore it is a spandrel (useless byproduct of something else). In the short run, however, we have to go with what we know.

I have read many a syntactician say we have to understand both how syntax evolved and why it evolved the way it did rather than some other way. Derek Bickerton in particular is often frustrated over the way non-linguists don’t seem to give a hoot about the why-this-way question. But Piattelli-Palmarini sees wondering why as futile, no more sensible than wondering why a meteor struck the Gulf of Mexico instead of the Indian Ocean.

If Piattelli-Palmarini’s line does become the standard one with syntacticians, I suspect the result will not be the one they hoped for. Most people who are interested in speech origins are curious because they hope it will explain something important about human beings. If it becomes generally accepted that the rules of syntax are spandrels, their evolution becomes a technical question like the evolution of the immune system: interesting, if you’re interested, but not compelling. Meanwhile, the functional role of speech origins gets more and more intriguing.

Comments

[TLTB]

It is interesting you mention the immune system since it is the perfect example. Before its true origins were discovered, there were dozens of natural selection stories about how it came about. Of course, they were all very wrong, illustrating the problem with the general adaptionist approach I've mentioned before: it can explain anything, and so explains nothing. The only way to get at the true root of origins is to do actual science.

The comparison with creationism is often made and never fair. Massimo is simply pointing out, as many have, that syntax doesn't look like an adaptive trait. Spandrels, punctuated equilibrium, horizontal gene transfer, and other processes do occur and are well-recognized in the literature. There are no gods here. So it is only a fair scientific question to ask whether they might be responsible for language. Everyone evolutionary story - every science story - is a technical question. Science is about the details and following the evidence whereever it may lead. We cannot concern ourselves with whether the results will be philosophically or spiritually satisfying.

As for Massimo's (and Chomsky's) position on syntax, this blog seems to indirectly support this position narrowly since I haven't seen a story here yet for why syntax evolved (though many other components of speech have been dealt with fairly). But as Marantz has put it, "you can't get away from syntax." In a sense, it's the whole ballgame when it comes to evolution since its the most difficult to explain.

[Dave]

I'm glad you made the point about negative arguments ("I can't think of an adaptive explanation for X") not being positive arguments ("There is no adaptive explanation for X"). All a negative argument shows is precisely what it states in its hypothesis: we simply don't know yet. I think that, in science, it's very very difficult to show that a certain theoretical framework cannot explain some observed phenomena, because the efficacy of theoretical frameworks is contingent on the correctness of their predictions as determined by carefully controlled observations--experiments. Thus, it's a fallacy to ask "Is such and so a scientific theory right or wrong?"

With respect to the specific paradigm of evolution and natural selection/adaptationism, it seems to me that this is a particularly difficult area to do good science in, simply because we can't do experiments on things that happened in the past. So, unless great care is taken, adaptationist arguments tend to be highly speculative, and ultimately there is no real criteria for determining, say, whether or not it is good enough to say "well, seals are genetically quite closely related to land-dwelling mammals, now live in the water, and seem to use their fins for swimming--thus, the fins on a seal were selected for at some point." In this case it seems to make sense, but there's no way to go back and check to see if it's true: it itself doesn't make any predictions that can be empirically tested.

I'm not an evolutionary biologist myself so I don't know what kind of assertions are fair game in evolutionary biology, but it seems that the popular conception of "good" questions to ask in that domain is not very accurate. I think that this is what the thrust of Piattelli-Palmarini's argument...but then again, I wasn't there, so you might want to take my thoughts here with a grain of salt.
-------------------------------
BLOGGER: I like this comment a lot, but I want to caution against the rejecting natural selection grounds on the argument that “we can’t do experiments on things that happened.” While this is true, and it is also true that it is easy to switch from science to imagining things, the fact that we cannot do experiments is not an argument against saying the fins of a seal were naturally selected. The process of natural selection itself has been tested with many many experiments. (Why do you think we know about the evolution of fruit flies?) It has also been observed many times in the wild. If it weren’t for the theory of natural selection, we would have no explanation for why antibiotics, anti-malarial drugs, and pesticides cease to work. So, while it is true that we weren’t there to observe the evolution of seal fins, the natural-selection explanation of seal fins is as scientific as accounts of the Big Bang, a study of the geology of the Grand Canyon, or investigations of the physics of catapults in Roman times.

[Vijayachandra Ramachandra]

I recently stumbled upon this blog and have been a fan ever since! I had a couple of questions and I would appreciate if someone could answer this (may be this question has already been answered somewhere in this blog!).
1. What is the role of "bigger brain" in the quest for language evolution....I always took it for granted that the bigger brains that we possessed was the answer! Our bigger brains lead to enhanced social cognition and then language. I know this is a simplistic view but I want to know the current thinking regarding the role of brain?
2. Why is it that people like Massimo and other nativists completely ignore the role of social cognition in language evolution?
-------------------------------------------------------
BLOGGER: The "bigger brain" issue was discussed in a number of consecutive posts a few weeks back. You might want to start with Supersizing the Brain (http://ebbolles.typepad.com/babels_dawn/2007/02/supersizing_the.html) and follow the posts after that.

[AquaLink]

Hello,

Very good post you have made here.
Good reading.

Keep up the good work.

Grtz.