Is there any substantive difference between natural, sexual, and artificial selection? Or is it just semantic sugar, useful for humans in our own cognitive bookkeeping? I lean toward the latter proposition. To some extent I would think that this is an irrelevant issue, selection is selection, but I have encountered folks who seem surprised at analogies between “artificial” and “natural” selection quite regularly. Of course Charles Darwin famously elided the distinctions across the two categories in his original works in the 19th century (this was later a subject of controversy, insofar as Darwin’s conflation of the properties of artificial and natural selection may have misled him in terms of the weight of factors shaping evolution in the wild).
These are the questions which bubble to the fore of my mind when I encounter reports such as Elizabeth Pennisi’s in Science, On the Trail of Brain Domestication Genes:
Researchers have proposed that bonobos evolved domesticated behavior to encourage group living. By isolating a group of 40 putative brain domestication genes in the prefrontal cortex and comparing their expression in humans versus chimps and bonobos, researchers found that the activity of that gene group in bonobos was clearly “domesticated” compared with chimps, they reported at the Biology of Genomes meeting.
The full piece is gated, so here’s the relevant section in the details:
He looked for genes whose activity was increased or decreased in the three domesticated species compared with in the wild ones. He found 60 such genes and then sought confirmation by doing a similar comparison of the domestic guinea pig and its close wild relative, a cavy. About 40 of the 60 genes displayed the same boost or drop, he reported at the meeting.
With these 40 putative brain domestication genes in hand, Albert compared their expression in humans versus chimps. No clear pattern of human domestication emerged. But the activity of that gene group in bonobos was clearly “domesticated” compared with chimps, Albert reported. “What causes this very dramatic behavior in bonobos is somehow mechanistically similar to what was selected for when the dog and pig were domesticated,” Pääbo says.
This was reported a meeting, so it should be taken with a large crystal of salt. We already have a theory, a preconception, which these preliminary results slot into. Therefore there’s a natural bias to fixate on the “wow” factor here. Why else would you be reading this on this weblog? More seriously I’ve seen previous reports in Science on what went down in conferences turn out to be less interesting when a further statistical analysis was performed. I don’t generally blog the follow ups, often because those results sometimes don’t even see the light of day because like many negative findings they’re not publish-worthy for the scientists themselves.
With that caveat out of the way, this is highly exciting. It reminds me of the story of “domesticated” silver foxes. There are two tracks through which one can filter these results. First, on a phenotypic level it looks like certain clusters of traits may go together. One of the major properties of organisms domesticated by humans is their tolerance of social density. Even domestic cats exhibit this tendency insofar as feral “colonies” are not uncommon. They may not hunt cooperatively, but they do seem to tolerate cohabitation reasonably well. This reflects I think a parallelism with a property necessarily associated with human beings; we’re incredibly social beings whose inter-personal networks exhibit a far greater level of complexity than seen amongst other mammals (perhaps even more than social insects, though this might be an apples to oranges comparison). Wildebeest and bison may roam in massive herds, but they are lacking in the many diverse levels and varieties of social units found in our own species, the “platoons” which taken together form a “society.”
The second track is one of genetics. Are the same genetic regions the targets of natural selection in divergent lineages? This tells us something about the role of contingency and constraint on the genetic architecture of living organisms. Consider a population which is “moving” through “adaptive landscapes.” The direction and speed of that movement is obviously going to be shaped by exogenous parameters of selection and drift. But these dynamic properties of evolutionary trajectories may also be impacted strongly by endogenous parameters of the genetic properties which need to vary in response to selection or drift. That is, there is a limit to the flexibility which collections of genes can exhibit over the short term to output a particular phenotypic result. Given infinite population sizes and infinite time to evolve one presumes that the full space of genetic and phenetic possibilities could be explored. But for the purposes of our analysis the genetic variation and the time required to explore the full space of possibilities is limited.
Finally, going back to the distinction between types of selection, the whole argument is necessarily anthropocentric. I am a fan of anthropocentrism to some extent. I’m a human, and you are probably a human. Our nature is to look at the world from our own human perspective. The idea of domestication naturally requires a human reference, ergo, the “self-domestication” of humans. But the example of bonobo chimpanzees and their possible “self-domestication” suggests the limitation of the very framework of human directed artificial selection. Domestication and artificial selection are just subsets of the possibilities of selection, and by bracketing our thinking into the idea of “domestication” we may miss broad trends toward convergence across the tree of life. There may be a lot more “domestication” out there that we simply don’t pay close attention to because it lacks a human actor.
Razib Khan’s degrees are in biochemistry and biology. He has blogged about genetics since 2002, previously worked in software development, is an Unz Foundation Junior Fellow and lives in the western US. He loves habaneros.
May 26th, 2011 at 5:57 pm
For the notion of self-domestication in human evolution see also Bednarik’s work on the Mid-to-Upper Paleolithic transition. See, e.g., http://www.springerlink.com/content/l6522807742p8163/
May 26th, 2011 at 6:26 pm
Worth noting that one of the bigger systemic problems in doing research with laboratory animals is the strong tendency of lab animals to develop domesticated animal tendencies. Lab rats start to differ in a lot of systemic and regular ways from wild rats after not all that many generations of a wild rat source. You don’t have to try very hard to do it to get that tendency going.
IIRC, however, the traits involved are not so much tolerance for population density as they are docility (which certainly fits the bonobos v. chimps).
Also, just to reiterate: “No clear pattern of human domestication emerged.” So, human patterns aren’t a very good example of a domesticated species. We’re still wild!
Now, what would cause bonobos to select for docility rather than wildness? The biology and climate of chimps and bonobos is quite similar. One lives on one side of a river in Africa in the Congo, one lives on the other side in the Congo (and elsewhere in tropical Africa).
Perhaps bonobos were isolated from their top predators by the same circumstances that isolated them from chimpanzees, and when your only other potential predator is you, you want to be docile? Did chimpanzees have a habitat that overlapped with hostile human predecessor species while bonobos did not, forcing Red Queen struggles in one place but not the other? Was bonobo territory more nutrition poor requiring a lifestyle that conserved energy relative to chimps? Did human ancestors literally keep captive populations of bonobos and literally domestic them in the process? It is hard to come up with a story that clearly is superior to other post hoc explanations.
May 26th, 2011 at 6:33 pm
I fully agree with your starting point here. This also seems to be at the conceptual heart of the kin selection debate. “Kin selection” is a conceptual hanger for us to use about one aspect of standard genic selection. Selection is, indeed, selection.
This is a problem that pervades biology–the illusion that our mental categories are really how life is organized. You can see this in the bullshit “wow” factor attached to finding new functional roles for “known” genes. One of my favorite examples was the brouhaha that developmental “morphogens” could also function as “cues” for axon guidance. Of course they do! They provide spatial information in embryos, and are therefore useful to axons. We decide that whatever the first thing we catch a protein doing is somehow its essential nature. Therefore sperm can “smell” because they express a few “olfactory receptors.”
May 27th, 2011 at 3:39 pm
[...] a sharp evaluation of the report, read this blog post by Razib Khan, a biologist and biochemist who, unlike me, actually understands the [...]