Size and Scaling in Hominid Evolution

ResearchBlogging.orgStephen Jay Gould and David Pilbeam wrote a paper in 1974 that was shown ten years later to be so totally wrong in its conclusions that it has fallen into an obscurity not usually linked to either Gould or Pilbeam. However, they were actually right in ways that they could not have anticipated. And even if they were not right, this paper still has much to contribute, including the opening words of that publication in Science, which are very much worthy of consideration for many reasons:

i-019aa89e74f87750754374ffc1a271c4-pilbeam_gould_01.jpg


It is no longer true that there are more practitioners than pieces. In fact, when I encountered this paper ten years after it was written … assigned in a class by Pilbeam along side then new work by McHenery on australopith body size, the number of pieces in the human evolution game was just surpassing the number of practitioners, by my own thumbsuck estimate. (I’m including graduate students on one side and postcranial remains on the other, to be fair.)

David Pilbeam, who was my first teacher in Human Evolution, is famous for putting forth dramatic hypotheses that rearranged how we think about the human evolutionary tree, only to be shot down by an alternative and better hypothesis usually about five or ten years later. The new hypothesis has often been authored or coauthored by Pilbeam as well. A true Popperian, Pilbeam presumes nothing about the truth of his ideas other than that they probably get more true over time (but not always). As I said, the Pilbeam and Gould paper was given out along side other work on Australopith body size that demonstrated that Pilbeam and Gould’s analysis was flawed and thus their conclusion irrelevant.

But what was their analysis? Pilbeam and Gould simply scaled the two sexiest hominid head variables available … cranial capacity and tooth size … against body size, to see if either really did change in an evolutionary sense, or if they were just following body size changes in a mundane way.

Forget about the cranial capacity analysis. I’m not interested in that at this time, and for a number of reasons, the amount of work it would take to address this paper in relation to brain size is greater than the benefit. But tooth size …. oh yes, that is interesting.

One of the prevailing hypotheses at the time was that Australopiths were divisible into two forms, the gracile and the robust, and that the robust were vegetarians and the gracile omnivores with a fair amount of meat in their diet. There were other ideas as well, including that they were all one species wiith a lot of variability, or two sexes of one species, etc. But since we now know that there was (and still are) more than one species in our African Miocene and PlioPleistocene hominid checklist, we’ll ignore those zany ideas.

This gracile hunter vs. robust plant-eater division was important because anthropologists fetishized hunting and meat eating as key features in human evolution, and at least one early form of hominid needed to have this feature in order to be interesting. Picking the more gracile smaller-toothed hominid may have made sense at the time, at least to some.

But Pilbeam and Gould wanted to examine this more closely, and they demonstrated, by putting the tooth measurements in the context of body size estimates, that tooth size for these different early hominids scaled to body size in such a way as to not suggest especially different diets. All of these early hominids were eating a lot of vegetation, they concluded.

Every major paper, especially the ones that involved a lot of pre-publication talk (at conferences, in other venues) … and I have every reason to believe that Pilbeam and Gould’s paper can be described this way … comes with a culture of interpretation, involving metaphors or examples, or analytic (and sometimes pedagogic) uses that the paper itself either does not contain or only implies. For Pilbeam and Gould’s paper, the implication was that the two types of australopiths (gracile vs. robust) were roughly equivalent to chimpanzees and gorillas … but more similar than these African apes. The larger one was a scaled up version of the smaller one. The degree of social and dietary change that accompanied this difference in size for the African non-human apes was probably overestimated at the time of the paper and over the following decade. Later, we were to discover that chimps eat a lot of leaves and some gorillas eat a lot of fruit (for gorillas).

The point is that the Pilbeam and Gould model brought studies of size scaling (known as isometry and allometry) to the forefront in human evolutionary studies, and made people think more about fundamentals rather than elaborate adaptive “just so stories.”

However, ten years after its publication, when I was learning this stuff, we were just discovering that the gracile australopiths were not small compared to large robust australopiths. There was perhaps little change across body size among the australopiths, with the heads varying in size instead. Therefore, the size scaling work done by Pilbeam and Gould had to be put aside.

(Note: This was naive, but it is what we were thinking.)

My own research, with my colleague Betsy Burr, on rodent dietary adaptations has revealed, however, that size scaling within the skull (related to body size, but not in the expected way) is useful in understanding dietary adaptations. It turns out that in some rodents, dietary patterns correlate to changes in skull features that are very similar to the differences we see among the early hominids, and living African apes. We think. For this reason, we are re-examining the Pilbeam and Gould hypothesis in new light with new information. Stay tuned for that.

Pilbeam, D., Gould, S.J. (1974). Size and Scaling in Human Evolution. Science, 186(4167), 892-901.

Comments

  1. #1 J-Dog
    August 14, 2008

    excellent post – thanks.

  2. #2 Hardly Noticed
    August 15, 2008

    This will be a big boost for the Vegans

Current ye@r *