 |
 |
|
|
|
||
|
|
||
|
|
|
|
|
|
|
|
||
|
|
||
|
|
Among the many varieties of popular conceptions of Darwinian evolution, the most common is probably the one that relies on the metaphor of a struggle. According to this conception, evolution teaches that life is a competitive struggle between individuals, where the victory of one is the defeat of another (a zero-sum model).
The technical language of evolution provides a very different picture. The most immediate and striking difference is that the unit of selection is no longer the individual, but the gene--or more specifically, an allele. An allele is defined as a suite of genes that are required for constructing a specific structure or behavior; to the extent that genes are selected on the basis of their phenotypic results, natural selection cannot act on genetic units smaller than the allele. The measure of evolution has not received a definitive formulation; I will consider a comparatist and an independent view.
According to the comparatist view, evolution is a change in the distribution of gene frequencies in a population. This view can be characterized as a modified zero-sum view; it is zero-sum in that the gain of one allele is necessarily at the expense of another, in terms of their relative frequencies; it is modified in the sense that the absolute numbers of both can rise.
The independent view, on the other hand, considers evolution to be measured simply by the absolute number of copies of a certain allele. It should not matter, by this reasoning, what other alleles are doing; success is a function of replication pure and simple. Since this implies that two alleles that occupy the same site in the genome can both be successful, the independent view is not a zero-sum game.
Unless we attribute complex goal-seeking behavior to alleles, it makes little sense to say they struggle with each other, or with their environment. Dawkins' metaphor of "the selfish gene", although it may be useful in drawing attention to the gene as the unit of selection, continues to draw on the construal of a zero-sum competitive game.
See Diamond p. 343, Ridley p. 152. Red Queen joke.
Further reading:
Zahavi, Amotz (1975). Mate selection--a selection for a handicap. Journal of Theoretical Biology 53:205-214
Zahavi, Amotz (1977). The cost of honesty (further remarks on the handicap principle), Journal of Theoretical Biology 67:603-5.
Hamilton's proposal was convincingly elegant and simple, and can be even more easily explained today, with the additional genetic knowledge and theory available. Since the functional unit of evolution is the allele, selection pressures do not strictly speaking act on individuals. An allele will tend to spread if it codes for structures and behaviors that make it spread--this is the basic fact of natural selection. However, if the allele underwent a mutation that made it code for behavior that in addition made copies of itself spread, this mutated allele would have a good chance of being even more successful. This, in brief, was what Hamilton suggested had happened wherever we observe altruistic behavior.
Kin selection assume a fairly complicated genetic machinery in an allele. First of all, the allele must code for some adapted trait--this is the basic feature that makes it spread. Secondly, it must then evolve additional coding for picking out cues that it is present in another individual. Finally, it must evolve structures that inclines or directs the organism to behave in ways that have the effect of increasing the reproductive fitness of the other individual. In this endeavor, the allele is surrounded by other alleles that are evolving or have evolved to do the same thing. The cumulative effect of all the detection devices leads to the measure of the degree of relatedness, and thus to a decision about the degree of investment to apply in aiding the other.
How does the allele code for recognizing copies of itself? This could in principle be done in a variety of ways, and they are not all fool-proof. There are examples, for instance, of warrior ants in Africa that occupy another hive and somehow convince the conquered ants to work for their new masters in the same unself-interested way that they previously worked for their own queen. The ants use a chemical signal that is normally a reliable cue for genetic relationship, but the intruders have adapted to mimic it. Among animals, smell may be used to determine relationship. Humans appear to depend heavily on being reared together (cf. the Westermark effect). Visual similarities probably would not work; in the ancestral environment, one may not know what one looks like, and kin often show such a large variety of appearances that they cannot be categorized reliably on that basis.
Hamilton defined kin selection in terms of what he called "inclusive fitness"--a term that still echoes the old paradigm that the basic unit of selection is the individual. Dawkins (1982:185-86) points out that inclusive fitness is a measure of an individual's effect on the reproductive fitness of his relatives, and more properly simply a measure of the probable future frequency distribution of the alleles that contribute to the altruistic behavior. Formally, inclusive fitness is defined as an individual's own reproductive fitness, together with its contribution to the reproductive success of the relative in question, discounted according to their degree of relatedness. If you are an identical twin or a eusocial ant (where the workers are all genetically identical), the logic of kinship selection dictates that your twin's, or fellow worker's, reproductive fitness is just as important for your inclusive fitness as your own reproductive fitness. If you have two full siblings, their reproductive fitness together equals (in terms of your inclusive fitness) your own reproductive fitness, a logic that prompted J.B.S. Haldane to quip that he would lay down his life for more than two brothers, four half-brothers, or eight cousins. It is not clear, however, that he would really have done so--organisms, after all, are adaptation executers and not fitness maximizers.
How well do the predictions of kin selection explain human behavior? This is an empirical rather than a logical question. In the case of twins, for instance, kin selection theory would predict pure altruism, but this does not seem to be borne out by the facts (Daly and Wilson 1988a:11). The explanation for this is that twinning may have been sufficiently rare in the Environment of Evolutionary Adaptedness (EEA) that there is no specific adaptation for how to treat a twin. More generally, kin selection provides a genetic basis for the known human proclivity towards nepotism.
See Nicholas Humphrey ( 1997). Varieties of Altruism - and the Common
Ground Between Them. Social Research 64: 199-209. Full
text (external).
|
|
|
|
|
|
||
|
|
