go home Genetics

American Political Science Association conference
Washington DC, 1 September 2000

ON GENETIC ETHNIC INTERESTS

by Frank Salter

Today I shall argue for four propositions.

FIRST, that inclusive fitness is the ultimate interest that prioritizes all other interests.
SECOND, that in certain contexts ethnic groups constitute a large quantifiable store of inclusive fitness for their members, which I call Agenetic ethnic interests.@
THIRD, that genetic ethnic interests can sometimes be defended by individuals in ways that do not reduce their overall inclusive fitness.
FOURTH, that Iit can be ethical knowingly to pursue one=s genetic ethnic interests.

Proposition 1:

Inclusive fitness is the ultimate interest.

Subjectively the needs and wants of the phenotype are interests, and conventional political theory adopts this perspective. Vital interests are prioritized by introspection to be the survival and well being of ourselves as individuals and for emotional reasons that of our kith and kin and other individuals and groups for whom we have sympathy. Resources and rank are two interests usually considered vital, once survival has been secured. As compelling as these may feel, and despite contributing to genetic fitness, they are actually proximate interests of secondary importance to ultimate interests. Since Ainterests are seen as reproductive, not as individual survival,... pleasure and comfort are postulated to have evolved as vehicles of reproductive success@ (Alexander 1987, p. 81). Proximate interests are means, not ends, in the evolution of life. Richard Alexander expresses this well:

Humans, like other organisms, are so evolved that their Ainterests@ are reproductive. The interests of an individual human (i.e., the directions of its striving) are expected to be toward ensuring the indefinite survival of its genes and their copies, whether these are resident in the individual, its descendants, or its collateral relatives.... We need not be concerned with the possible argument that interests are only definable in terms of what people consciously believe are their interests or intentions. Biologists continually investigate the life interests of nonhuman organisms while lacking knowledge on this point and nonhuman organisms live out their lives serving their interests without knowing in the human sense what those interests are (Alexander 1995/1985, pp. 182C3).

From an evolutionary perspective, then, inclusive fitness is the ultimate interest of all life whether or not perceived by phenotypes; it has priority over other interests. People are prone to risk their lives for close relatives in emergencies, testament to the power that ultimate interests can have to motivate human action, including self-sacrifice. Valuing of proximate interests such as self preservation evolved to the extent that they enhanced the ultimate reproductive interest. E.O.Wilson put proximate interests in neo-Darwinian perspective thus:

In a Darwinist sense the organism does not live for itself. Its primary function is not even to reproduce other organisms; it reproduces genes, and it serves as their temporary carrier.... The individual organism is only [the genes=] vehicle, part of an elaborate device to preserve and spread them with the least possible biochemical perturbation.... [T]he hypothalamus and limbic system are engineered to perpetuate DNA (E.O. Wilson 1975, p. 3).

Proximate interests have become fallible guides to ultimate interests because modern humans live in a rapidly changing world, and our environment no longer provides all the extra-genetic information needed to optimize investment in our inclusive fitness.

Proposition 2:  Genetic ethnic interests are often real.

Pierre van den Berghe=s (1981) idea that ethnic groups constitute extended families recognizes the fact that ethnies are lineages and that kinship can become highly diluted without losing all its genetic flavour. Or as Dawkins colourfully puts it:

Individuals do not, in an all or none sense, either qualify or fail to qualify as kin. They have, quantitatively, a greater or less chance of containing a particular gene.... [T]he post Hamilton >individual= .. . is an animal plus 1/2 of each of its children plus 1/2 of each sibling plus 1/4 of each niece and grandchild plus 1/8 of each first cousin plus 1/32 of each second cousin . . . Far from being a tidy, discrete group, it is more like a sort of genetical octopus, a probabilistic amoeboid whose pseudopodia ramify and dissolve away into the common gene pool (1978, p. 67).


There are two measures of genetic ethnic fitness, gene frequencies and shared genes by descent. The latter qualification was recognized by Hamilton as a strategy for countering free-riders. The rule for allocating altruism is not that another individual shares one=s genes, but that those genes shared genes are due to common descent. This might appear subtle, but it can have real consequences for solidarity. Consider what is evolutionarily rational for the members of two societies with the same gene frequencies. By simply counting gene frequencies, as Cavalli-Sforza does, one might conclude that there is no evolutionary reasonCno inclusive fitness stakeCin directing altruism towards one=s own society in preference to the other. But the web of kinship is usually much denser within than between societies. On this basis Hamilton predicted much greater altruism within than between tribes, even though gene-frequency differences are tiny. Hamilton=s definition of inclusive fitness nullifies most objections from crude gene frequency measures, such as come from Cavalli-Sforza and others. Note that this geneticist=s works carry no references to altruism; no mention of sociobiological theory; no citations of Hamilton, Dawkins, E. O. Wilson, Richard Alexander, etc. etc. Yet without such input a theory of ethnic solidarity and competition can hardly be started from a modern biological perspective.

Iceland is a useful example for assessing the possible scale of genetic ethnic interests. That country has remarkably complete lineage records going back five centuries, is also highly homogeneous, and has kept thorough medical records since the early 20th century. This combination of factors makes Iceland a rare natural laboratory for tracking and isolating genetic diseases (New York Times 16 February 1999). In personal correspondence Fridrik Skulason (21 July 1999), a provider of genealogical software for Decode Genetics Inc., the company licensed to analyze the country=s genetic data, reports that based on a sample of 13 individuals, a common ancestor appears on average only seven to eight generations back. If this is correct, then two randomly chosen Icelanders share, on average, a great great great great great grandparent. If this were the only common ancestor, which is highly unlikely in Skulason=s judgment, they would share 1/512th of each other=s genes, or a coefficient of relatedness of 0.00195. It follows that Iceland's population of 270,000 holds a genetic stake of 527 copies of a randomly chosen citizen=s genes, which is the equivalent of 1054 children or siblings. And this is before counting immediate kin. For the average Icelander, Iceland's population represents a large but dilute store of inclusive fitnessCgenetic interestCfor each of its members. Most other ethnic groups are not as homogeneous as Icelanders, so the concentration will be less. But depending on the contrast being made, many ethnic groups and groupings of ethnies are much larger, counted in the millions, tens and even hundreds of millions, or between one and three orders of magnitude larger than Iceland. It would appear possible that genetic ethnic genetic interests are typically as large as Iceland's.

Some group members will have less-than-average genetic stakes in their groups, and some more. The group has little or no genetic interest for those individuals whose characteristic genes occur at low frequencies in the population. But for the majority, the interest is real, vis a vis other groups, so long as the group is a descent group, which most ethnic groups are (van den Berghe 1981).

[From here, for simplicity, I resort to gene frequencies rather than the relatedness measure. Note that this probably underestimates genetic ethnic interests.]

The following hypothetical reckoning illustrates the above point in general form. Imagine two ethnic groups, X and Y, of one million members each. These groups are competing for territory or other resources. Assume for the sake of argument that pairs of individuals chosen at random from within either population have a coefficient of relatedness equivalent to ten generations of descent (i.e. the relatedness of an individual to his or her great great great great great great great grandmother). These pairs share one 1,024th of their distinctive genes. A population of one million will hold about 1,024 copies of the average member=s distinctive genes, not in the form of 1024 identical twins but spread throughout the population. That is the equivalent of 2048 siblings or 2048 children, a large nuclear family indeed, making this hypothetical ethnic genetic interest far more valuable in terms of inclusive fitness than any individual has achieved through personal reproduction. But the difference in concentration of genetic interest is spectacular: one=s child or full sibling holds 1024 times the inclusive fitness of a randomly chosen fellow ethnic.

All humans belong to the one species and thus share most of their genes; we all have a genetic interest in any randomly chosen group of humans. Were all groups to be genetically identical, noone would have a genetic interest in group competition. But there are group genetic differences, allowing for some fitness payoff from engaging in group competition.

An individual=s genetic interest in situations of inter-group competition is the surplus of inclusive fitness within the ingroup over that in the outgroup. An average member M of group X will also have an inclusive fitness stake in group Y, by virtue of the two groups belonging to the same species and sharing most of each other=s genes. But if group Y carries fewer copies of M=s genes than will group X, and when that deficit is significant there will be a real inclusive fitness interest in the home group, shared by most of its members. So, what is the deficit in Y (or the surplus in X)? It should be computable from measures of genetic distance, but I am not aware of this analysis having been done for any pairs of ethnic groups. It is relevant to note, however, that with large populations, the genetic distance need not be great for the genetic interest in the home population to be significant. Consider the example from the previous paragraph, but where group Y has 10 percent fewer of the genes in question than group X. Then a randomly chosen member of group X will have about 100 more aggregate copies of his or her genome in the home group. That is the equivalent of 100 identical twins or 200 siblings or 200 children. That might not be an unrealistic estimate if 15 percent of all human genetic variation exists between groups (a statistic widely cited to criticize the utility of the race conceptCLewontin 1972). Even if the intergroup difference in gene frequencies were only one percent, the surplus inclusive fitness would be the equivalent of 20 siblings or children. Remember that this is in addition to members' own families. That is a large genetic interest, a large ethnic family to use van den Berghe=s (1981) analogy.


Proposition 3:  Genetic ethnic interests can sometimes be defended by individuals in ways that do not reduce their overall inclusive fitness.

The threat of free-riders and over-allocation makes investing in a population of low but large aggregate inclusive fitness riskier than investing in close kin, though the general principles are the same. When resources are limited, the main challenge is striking a balance between investing in kin and ethnie. The balance point shifts radically with contingency. General criteria for prioritizing altruism include: (a) confidence of relatedness; (b) group characteristics (concentration of inclusive fitness, group size, and genetic quality); (c) the cost to the altruist; (d) the salience of intergroup versus inter-individual competition; (e) the genetic distance between ingroup and competing groups; and (f) the availability of means by which the individual can contribute to group competition.

Taking just the last of these factors, the existence of collective goods indicates that it is possible for individuals to make a difference to the fitness of their groups. This idea derives from David Goetze=s recent analysis of ethnicity over evolutionary time. This is due to collective goods, resources and services that can be drawn-on by any number of group members without diminution of value. Example include defence, provision of adaptive information, and assets such as public infrastructure.
Comparing normative and actual strategies, the match is weakest where modern society differs most profoundly from the environment of evolutionary adaptedness. In the evolutionarily novel environment of the urbanized and centralized state humans are imperfectly adapted to recognize and defend their ultimate interest, and increasingly rely on cultural strategies to substitute for innate mechanisms.

Proposition 4:  It can be ethical knowingly to pursue one=s genetic ethnic interests.
Genetic interests are a nonmoral good that can be defended using moral and immoral means. Alexander=s (1987) contractual ethics states that attempts to frustrate the pursuit of genetic interests are intrinsically aggressive except in self defence.

I argue that genetic ethnic interests are an ethically neutral goal even when compelling, but that the means used to pursue them can be morally acceptable. In other words, it is neither right nor wrong to pursue genetic interest; but the means used in that pursuit do vary in moral quality. It is well known that ethnic interests can be pursued using immoral means. However the contemporary intellectual emphasis on the evils produced by ethnic sentiment should not blind us to the positive role of ethnicity in providing a sense of belonging, social cohesion, and continuity (Fishman 1996/1980), in addition to the ultimate genetic interest vested in the tribe. As Alexander (1987, p. 81) states, the ethics of pursuing genetic interests are bound up largely in the means used to secure them: A[T]he rules consist of restraints on particular methods of seeking self-interests, specifically on activities that affect deleteriously the efforts of others to seek their own interests.@

I still find genetic interest counterintuitive as an emotional priority except by thinking about it in phenotypic categories, as follows. First I conclude that there is only a quantitative difference between kin and ethnic genetic interests. Then I focus on the most intuitively appealing type of genetic interest, that residing in close kin. I ask myself what I or anyone could say to criticize a person who claimed a genetic interest in kin and did the sort of things a loving parent, aunt or uncle would normally do? Is parental care administered in awareness of the genetic interests thus served less moral than the naive variety? If there is no difference between genetically informed kin altruism and the naive variety, is not criticizing the informed variety tantamount to criticizing parental love? And if being genetically informed increases parental motivation, is such extra care or the belief that causes it immoral? While utopian socialists and advocates of the therapeutic state have tried to abolish the nuclear family and the parental favouritism it involves (Heller 1988; Shepher 1969; Spiro 1995), attempts to randomize parental investment have usually given way to popular demand and parental favouritism remains almost universally accepted. This inevitably contributes to inequality among children but nevertheless the parental bond is so strong that this primary form of discrimination is considered morally unexceptional.

Now the analogy with the family comes into play. If there are types of nationalism that roughly defend ethnic inclusive fitness (and surely many types do) and that are considered morally acceptable, then on what basis can we morally condemn a genetically informed nationalism that resembles the naive type? If defensive nationalism is justified, how does it matter that the defenders know they are protecting their genetic interests? And even if the defenders= awareness of their shared genetic interests changes their behaviour, for example by inspiring greater self sacrifice or more efficient killing of the aggressors, does that make genetically informed defensiveness less justified than the naive type? The same argument applies to ethnic minorities struggling to escape discrimination. Would their cause be any less just if minority activists or their majority supporters understood that they were defending minority genetic interest? Genetic ethnic interests are not morally good ends; but neither are they bad. But they can be compelling, and the means used to secure them can be just or unjust.


This runs counter to universalism, according to which it is immoral to discriminate in favour of one ethnic group. The ethical intuition behind this is that favouring family and ethnie are forms of selfishness that compete with broader solidarity (of class or humanity in general). This intuition can be expressed in evolutionary terms by the following rule: altruism must be maladaptive to be moral (Williams 1988). Alexander (1987, p. 177) suggests that this is an unconsciously self-serving moral sentiment that, when expressed, influences some susceptible individuals to show indiscriminate altruism and thus benefit the moralist. By definition such behaviour will tend to reduce the relative fitness of the genetic altruist. Alexander is implying that universalism is often a competitive tactic, and concludes thus: AIf morality means true sacrifice of one=s own interests, and those of his family, then it seems to me that we could not have evolved to be moral.@

As an evolutionist Westermarck (Salter forthcoming) also treated promiscuous alms giving with some contempt, and he too considered the universalist elements of moral behaviour to be not always genuine. He described the moral emotions as possessing the qualities of Adisinteredness, apparent impartiality, and flavour of generality@ (Westermarck 1912, II, p. 739). Nepotism both familial and ethnic is not disinterested, but it does underlie two of the most intense and extensive forms of altruism, those founded on the family and tribe. If all adaptive behaviour were to be ruled as amoral, then most forms of altruism, such as mother love and self sacrifice for one=s nation, would be ruled as nonmoral behaviours because they serve the actor=s Aselfish@ genetic interests. Such altruism would be designated immoral if withheld from non-kin. Rigorously abstract criteria of morality impose inhuman standards on a species that universally exhibits particularism. Walzer (1994, p. 200) offers an apt maxim: A[t]he crucial commonality of the human race is particularism.@ Walzer continues:

Tribalism names the commitment of individuals and groups to their own history, culture, and identity, and this commitment (though not any particular version of it) is a permanent feature of human social life. The parochialism that it breeds is similarly permanent. It cannot be overcome; it has to be accommodated, and therefore the crucial universal principle is that it must always be accommodated; not only my parochialism but yours as well, and his and hers in their turn (Walzer 1994a, pp. 199C200).

All this might be taken as dodging the critical issue of whether an act reduces another=s interests. Can it be moral to pursue one=s genetic interests in ways that reduce others= genetic interests? Walzer implies not, but this position is untenable if competition is unavoidable. It is absurd to accept that unequal outcomes in economics and social status are inevitableCand I am unaware of any serious political theorist who recommends more than reducing inequalityCwhile failing to accept some inequality in reproductive fitness. There are many interests the pursuit of which entails making personal tradeoffs or engaging in competition for scarce resources. Every society has legitimized procedures for advancing personal interests ahead of others. In hunter gatherer societies the individual who procures and shares the greatest amount of food is accorded status by the band. In contemporary Western societies winning exams and undercutting another retailer=s prices are accepted methods for gaining more resources. Arguably these contests have fitness consequences. Society objects to killing Smith as a means of acquiring her resources, but accepts various types of competition with her, even though the two strategies stand to reduce Smith=s capacity to support a large family and pay for the best medical and educational opportunities for her children, all likely to reduce her inclusive fitness. Unless it can be shown that genetic ethnic interests, unlike all other interests, possess no imperative weight at all, there is no reason why defending these interests ought not compete with others, including genetic interests, even if this has differential fitness effects. Once again, it is the means by which genetic interests are advanced that must be judged moral or immoral, not the goal itself or its effects on others genetic interests.

This is entailed by Williams (1988) contention that only non-fitness-enhancing behaviour can be moral.