Mike the Mad Biologist has a post up, A Biologist Confuses Artificial and Natural Selection:
There's a really interesting article in last week's NY Times magazine about global warming and the spread of weeds....
...Artificial selection occurs when the fitness criterion--that is, what trait or phenotype will have higher survivability or reproductive output--is directly chosen by the experimenter. In the case of the weeds, if we were delibrately trying to grow better weeds--that is, mowing down rice biovars that aren't sufficiently 'weedy'--that is artificial selection. Simply changing the environment such that weedy biovars will do better is natural selection because we are not specifically trying to enrich for rice that have the trait of 'weedy-ness.'
Keep in mind that one can use natural selection for very applied purposes, and the boundaries will get fuzzy, but in the case of global warming, this is clearly natural selection. Besides, in a materialist sense, we are just another species of critter.
Obviously, context matters. That being said, count me as in the group which is mildly skeptical of distinctions between artificial, natural, sexual and social selection. I think the distinctions reflect systematic differences between these forms of selection, but I also think that they are themselves somewhat artificial insofar as the fundamental dynamic is the same, and the semantic separation has a tendency of deemphasizing this. Remember that Charles Darwin was influenced by the insights derived from the cultivation of plants and breeding of animals.
In particular I think that the artificial vs. natural selection distinction has a strong dose of homocentrism embedded within it as a background assumption. I'm not opposed to homocentrism, but there is normative baggage which undergirds it which does not hold when one contrasts selection with random genetic drift. Of course, the term random genetic drift itself reflects biocentrism, as it is nothing more than a particular manifestation of sample variance. In the end my attitude toward most of these definitions is guided by instrumental calculations. I happen to believe that breaking the distinction between artificial and natural selection has a positive yield in terms of insight as to the role of selection as a general evolutionary process, while a term such as random genetic drift has a specific and precise clarity which sample variance may not.
- Log in to post comments
Totally. Once you start really thinking about it, categorizing selection under these rubrics is not that useful.
I'd agree that selection is selection... but I do think there is some value in recognizing when humans have had an intentional hand in the "selection" of a pedigree, with a specific goal in mind.
Consider how some people will precisely breed animals (like dogs) for specific characteristics. Picking and choosing which animal breeds with which, generation after generation.
Perhaps the term to use isn't "artificial selection"... but I think having some word to capture that concept would be useful. I believe "selective breeding" is the term for that.
Notice though that it is not "artificial selective breeding" :-)
-- Charles Iliya Krempeaux
http://changelog.ca/
but I do think there is some value in recognizing when humans have had an intentional hand in the "selection" of a pedigree, with a specific goal in mind.
Consider how some people will precisely breed animals (like dogs) for specific characteristics. Picking and choosing which animal breeds with which, generation after generation.
grant point #1, but re: #2, what's so special about any of this aside from from the part about people. e.g., there might be an island environment which results in a new population being precisely driven by selection toward specific characteristics. as mike noted above: Artificial selection occurs when the fitness criterion--that is, what trait or phenotype will have higher survivability or reproductive output--is directly chosen by the experimenter. i think the key is the *experimenter*, the person, human agency.
This distinction between artificial and natural selection made by Darwin was illustrative, not descriptive. Most people had direct experience with human driven selective breeding. Darwin leveraged this experience to explain that the same sort of thing happens absent specific human intent. All that was required was that an environment influence the reproductive rate in a breeding population and that the traits which influence this differential reproductive success in the environment be heritable.
The fact, if it is a fact, that only humans engage in artificial selection is contingent; i.e., it's not essential to the definition of artifical selection. If apes or ants or ... engaged in activities with the effective intention of influencing the distribution of particular heritable traits in some (any) population, it would be artifical selection. The effective intention to influence the distribution of heritable traits is the key.
Experiment, in and of itself, is not the key. In experiments where the experimenters do not intend for particular traits to change in their relative frequencies in particular ways, the selection that takes place is not artificial. That's why it's possible to experimentally test hypotheses about the operation of natural selection -- think of tests of the Lotka-Volterra model of competition.
so i guess we need to talk to cognitive psychologists and ethologists to figure out what is, and isn't, artificial selection....
Well, yes -- insofar as cognitive psychologists and ethologists can help us to understand what is distinctive about the processes and products of artifice.