In his autobiography, Bastard Tongues, Derek Bickerton said that most science papers don't lead to anything, either for the author or for others. I suppose that's so, but papers as a class remain the seeds of science. That's where ideas and facts are passed around. I'm not sure how many papers I have discussed on this blog, but it is in the hundreds and some few have turned my thinking around. Others have opened horizons I didn't even realize were out there. Here's my list of the must-read papers for understanding language origins. I thought I would whittle the list down to the top ten, but I couldn't shave it quite that close.
I'm growing more excited about my book, due out this September. I've begun to get some good blurbs and hope for more. As part of the prologue to publication I think from time to time I will post a video on YouTube in which I discuss an aspect of the blog and book. Today's post is the first. Readers through a subscription or RSS service may not be able to start the video from this site. You can find it here.
Note: Once before I posted a video and received a protest from a deaf visitor that she could not hear what I was saying. So below is the text of my video.
I was always an obedient kid, not much given to swiping things from candy stores or talking sass at my parents, so it might seem odd that since childhood I have pursued a forbidden line of research. Of course, I didn't know it was forbidden when I first got interested. I only learned its disreputable nature in college. One evening there I attended one of those after-hours discussions that the school used to hold for students and visiting scholars. In this case the scholar was a distinguished literary scholar, a Jesuit priest named Walter Ong. He said something, from this distance I have no recall what, and I asked, "If that's so, how could language have ever begun?"
Orchids adapted to the needs of butterflies and vice versa. Coevolution fits things together. How did that work in the coevolution of human biology and culture.
As I reported last week, the latest issue of Human Biology is devoted to language origins, with a particular focus on "genetic and cultural" approaches. In an afterword to the issue Tecumseh Fitch addresses a theme running through the whole issue, coevolution. Fitsch sees coevolution as the way to get past relying on an overly sharp distinction between biological and cultural explanations for language origins:
Did James Thurber's famous drawing hold the clue to speech origins?
New theories of language origins are not rare. Here's one that focuses on the "need" to make excuses. What is rare is an interesting one that reflects new information rather than a new bit of whimsy. Human Biology's latest issue is devoted to "Integrating Genetic and Evolutionary Approaches to Language" and among its many notable papers is one that presents an intriguing, new (to me) idea on speech origins. I expect to devote at least one more post to the special issue, but today I want to concentrate on William M. Brown's "The Parental Antagonism Theory of Language Evolution: Preliminary Evidence for the Proposal" (paper here).
The theory is new because it comes at the issue in a new way, not from linguistics or the study of language functions. Its approach reminds me of things that go on in the study of physics, where somebody notices an implication of an equation and the next thing you know people are hunting for black holes. In this case, the focus is on the basic equation of evolution, Hamilton's equation of inclusive fitness. The equation includes an adjustment for relatedness and relatedness is normally treated as one thing. My brother is a closer relation than my first cousin.These relations appear reciprocal. My brother is also related to my first cousin.
However, I can be related to people in two ways. I have cousins "on my father's side," and a whole different set of cousin's "on my mother's side." Thus, not all my cousins are related to one another. If my father's genes could be distinguished from my mother's, then my father's genes would have different "interests" from my mother's. Normally, these two conflicts would cancel each other out, there being no reason for an offspring to prefer one parent's relatives over the other. Yet conflict is possible if a species has:
Sex-biased dispersal: e.g., female elephants stay in the group they were born into while males go off into the larger world; and
Relatedness asymmetries: e.g., adult females are much closer kin to the elephants they encounter than adult male elephants are to any males they encounter.
Therefore, the closely-related sex is likely to be more cooperative than the sex that disperses.
Female elephants, in the example I just gave, are famously cooperative, whereas bull elephants are not. Likewise, among lions, it is the male who enters a pride and is uncooperative, while the lionesses who grew up together cooperate in the hunt.
Does it work the other way too? In chimpanzees it is the male who stays around while the female goes off. And, what do you know, it's the male chimps who hunt together while the females stand aside.
I'm tossing things off the top of my head here, but whenever I find this much under my hat to support an idea, I'm interested and ready to listen.
There is among most mammals, including almost all primates (present company excepted), an obvious relatedness asymmetry: the offspring is related to the female doing the upbringing, while the father is totally absent and doing none of the upbringing. Therefore, the mother has an interest in enjoying some cooperative behavior on the offspring's part, whereas the father, who is unrelated to the mother, has no such interest in cooperation. It is in the father's interest that the offspring be as much of a parasite on the mother as possible without killing her (and leaving the offspring to die soon after), so that the offspring grows up strong. It is this relationship of conflicting interests that the author refers to as "parental antagonism."
Again, however, these interests should cancel each other out as the offspring is equally related to both parents. But there might be a way around this impasse if one parent's version of a gene could silence the other parent's version. In that case, if the mother's gene is silent, the offspring will display parasitic behavior, whereas in the father's gene is silent, the offspring will display cooperative behavior.
The theory is clear and elegant. Genes from the mother (aka matrigenes) promote cooperation; genes from the father (patrigenes), parasitism. In communications, for example, patrigenes would promote signals that make demands and language that backbites, whereas matrigenes would promote signals that end demands by expressing satisfaction and gossip about another's virtues.
Elegant as the idea is, however, I do have some doubts. Functionally speaking, are there really such things as matrigenes and patrigenes? The usual theory is that half a child's genes came from one parent, half from the other. How is the DNA molecule to know which gene came from which parent? In sex chromosomes, yes; in mitochondria too (all from mom). But these don't have anything to do with language.
Brown writes as though a small number of genes have been definitely identified as silenced ("imprinted") because of parental source, but I want to know how this works. Imprinted genes may be as real as black holes, but for the moment they are still only speculative possibilities based on mathematics, not on anything about the mechanisms of inheritance. Any theory of origins based on parental antagonism seems, at best, premature.
Nonetheless, Brown has a theory of "language" origins that depends on parental antagonism and which, remarkably enough, ties into an existing idea, although coming at it from a new angle. A couple of years ago I had a series of posts on Dean Falk's theory (see: Where Shall I Put the Baby; Crying with an Accent; The Oldest Form of Speech.) that pre-linguistic vocalizations began in response to a loss of body hair. Over three million years ago (for reason's unknown) the human lineage appears to have lost most of its body hair, presenting mothers with a new problem. Primate offspring stay with mother by hanging onto her body hair. Once that hair disappeared, the mother had to carry the baby in her hands, and when she had to do something else with her hands—like pick food—she had to set the baby on the ground. This change occurred before the lineage brain began to increase in size and birth became burdensome. Falk's theory is that babies began to vocalize so that the mother could maintain contact while she was on the ground.
The parental antagonism theory enriches this idea a bit. The baby's vocalizations don't just maintain contact; they reassure the mother that the baby is okay, so she does not have to keep coming back and look. I'm not sure that we need parental antagonism to account for such evolution, but the theory does focus directly on point and certainly merits being taken seriously. As Brown says about Falk, "[her] hypothesis is most obviously linked to parental antagonism, in that it involves parental investment and mother-infant bonding."
Brown is not much of a literary stylist, but his paper is well worth wading through.