1. Ward of course rejected the view that the Malthusian principle, that distribution of resources was according to merit, was relevant for human societies. He argued, against Sumner, that survivors of competition showed only that they were fit to survive, not that they survived due to their possession of some characteristics rendering them superior (Quoted in R. Hofstadter, Social Darwinism in American Thought (Beacon Press, 1955:79)). He then pointed out that all biologists were aware of this simple fact. Ward's point is strikingly reflected in a quite similar remark made recently in a different context by an eminent biologist Ernst Mayr. Mayr states that natural selection is unable to discriminate between reproductive success per se and reproductive success based on the possession of beneficial adaptive characteristics (Animal Species and Evolution (Harvard University Press, 1963 :659)).

2. It should not be assumed from this that there is any closer relation between social Darwinism and sociobiology than a belief in the relevance of biological principles for the analysis of aspects of human society. In `The Evolution of Social Behaviour' (Ann. Rev. Ecol. Sys. (1974), 5:374), R. Alexander maintains that

combining the modern concepts of kin selection, reciprocity, and parental manipulation in the analysis of the fabric of human society is not merely the revival of an old argument, refutable from examples and exceptions already used to that purpose, but the application of a new, basic, and comprehensive theory.

It is still a relevant consideration as to whether this new theory is subject to related counter-examples.

3. See E. R. Leach, Pul Eliya: A Village in Ceylon (Cambridge University Press, 1961) for a related approach by an anthropologist.

4. See G. Williams, Sex and Evolution (Princeton University Press, 1975) for a detailed discussion of this problem. It is Williams' contention that sexual reproduction in complex plants and animals is inconsistent with evolutionary theory.

5. His position here is I think self-consciously modelled after the arguments given by some transformational linguists regarding the genetic basis of language.

6. J. P. Scott and J. C. Fuller, Genetics and the Social Behavior of the Dog (University of Chicago Press, 1965).

7. G. Dahlberg, Mathemnatical Models for Population Genetics (New York: S. Karger, 1947).

8. See W. Bodmer and L. Cavalli-Sforza, Genetics, Evolution and Man (Freeman, 1976), pp. 683-685. Most of section 21.5, pp. 674 ff, is relevant to thfs sssue.

9. For a review of the various approaches taken toward `culture', see A. L. Kroeber and C. Kluckhohn, Culture: A Critical Review of Concepts and Definitions (Cambridge, Mass: Papers of the Peabody Museum, 1952), but for distinct critical approaches to the anlysis of the concept of culture cf. A. R. Anderson and O. K. Moore, `Towards a Formal Analysis of Cultural Objects', (Synthese (1962), XIV: 144-170) and L. A. White, The Evolution of Culture (McGraw-Hill, 1959).

10. Haldane also asked himself how a trait bringing about an individual saving a drowning child could spread throughout a population. He considered altruistic traits to be recessive. If simply recessive, then for such a trait to be exhibited by an individual, that individual would have to be homozygous for the trait. Such a trait could only become frequent in a small population, for the altruistic gene would be swamped by the selfish heterozygates in a large interbreeding population through genetic recombination and drift. (J. B. S. Haldane, The Causes of Evolution (Longmans, 1932)).

11. If the altruist sacrificed for a brother and had no offspring himself, then, in order to gain, the act must more than double his brother's representation in the following generation. If only first cousins benefited (r=1/8), the altruist which has no offspring has to increase a cousin's fitness eight times in order to gain; if an uncle (r= 1/4), he must increase his uncle's fitness four times in order for him to gain. In order for the altruist to gain, k, the ratio of gain in fitness to loss of fitness, must be greater than 1/the average coefficient of relationship, r; in the br-br case, 1/r=2, and the loss in fitness for the altruist leaving no offspring is total (= 1.0); for the shared altruistic genes to become more frequent, k, the gain to loss ratio, must be greater than two—the brother's fitness must be more than doubled. Sahlins mentions a case where certain individuals in a group may have a coefficient of relationship as low as (4)16= 1/32,768, with genetically closer relatives residing outside their own group.

12. According to M. J. West Eberhard ('The Evolution of Social Behaviour by Kin Selection', Quart. Rev. Biol. (1975), 50: 1-33), kin selection theory is an alternative to any group selection theory, for it assesses individual fitness, not group fitness, whereas Wilson considers kin selection to be a special case of group selection depending on the size of the population concerned. Hamilton's own discussion clearly indicates that classical fitness is a special case in kin selection theory, but I have not found an explicit statement concerning how it relates to group selection. It is interesting to note in this connection that even though Sahlins cites West Eberhard's paper on kin selection he directs no fire against it, restricting himself to Hamilton's theory. And this is the more surprising when it is noticed that West Eberhard claims that she is developing an alternative theory of kin selection which overcomes the defects of Hamilton's. In particular, she claims that her theory accounts for the manifestation of altruistic behaviour, even to quite distant relatives, from the point of view of any member of the population.

13. Darwin's discussion of this problem can be found in Chapter VIII, dealing with instinct, in the section `Objections to the theory of Natural Selection as applied to Instincts: Neuter and Sterile Insects'. He felt that the existence of neuter or sterile females in insect societies who could not, therefore, reproduce might prove fatal to his theory.

14. For an extensive study of matrilineal kinship systems, see D. M. Schneider and K. Gough, eds., Matrilineal Kinship (Berkeley and Los Angles: University of California Press, 1961). Schneider contends that biological relatedness has not proved to be of much utility in the analysis of kinship systems whereas the concept of descent, a socially stipulated rule of affiliation, has been useful.

15. `Simple Stereotypes', New Society (28 July 1933, p. 192).

Book Review: Sociobiology: The Investigation of the Biological Aspects of Social Behavior and Organization

Abstract

Get full access to this article