Gene-Culture Translation in the Avoidance of Sibling Incest (Human Sociobiology/Social Development/Ethnography) CHARLES J
Total Page:16
File Type:pdf, Size:1020Kb
Proc. Nati. Acad. Sci. USA Vol. 77, No. 10, pp. 6248-6250, October 1980 Population Biology Gene-culture translation in the avoidance of sibling incest (human sociobiology/social development/ethnography) CHARLES J. LUMSDEN AND EDWARD 0. WILSON Museum of Comparative Zoology Laboratories, Harvard University, Cambridge, Massachusetts 02138 Contributed by Edward 0. Wilson, June 20,1980 ABSTRACI A preliminary analysis is made of the relation choanalytic model, the disruption of family bonds by incest, between the epigenetic rules of brother-sister incest avoidance, and the dependence upon exogamy as a bonding device among which operate during individual development, and the fre- families. quency of occurrence of this form of incest among cultures. The likelihood of sibling incest is automatically diminished by an epigenetic rule in the development of sexual behavior. In an earlier article (1) we outlined a theory of gene-culture From the study of sexual and marital preferences in Israeli coevolution, in which genetic prescription determines the ep- kibbutzim (9) and Taiwanese villages (10), it has become ap- igenetic rules of individual behavior and is in turn altered by parent that close domestic familiarity during early childhood natural selection through the epigenetically generated cultural (in the case of the Israeli data, during the first 6 years) neu- patterns. Procedures were given by which it is possible to pre- tralizes later sexual attraction. The effect is very strong on any dict the influence on whole cultural patterns, defined by eth- children raised closely together, whether genetically related nographic probability distributions, imposed by bias in the or unrelated. This general result appears to be supported by behavioral development of individuals. anecdotal accounts of other societies, including the Trobrian- Epigenetic rules have been identified in a wide range of ders, Tallensi of Ghana, and Tikopians, as recently reviewed behavioral categories, and additional data permit the portrayal by Fox (11). of a few ethnographic distributions (2-4). The method we have There is a need to confirm and refine the ethnographic proposed has substantial power and can be applied to a wide analysis, which is directed at the ultimate adaptive value of range of cases both to predict the form of the ethnographic incest avoidance, and developmental studies, which consider curves from developmental data and to infer the properties of the existence of epigenetic rules. On the basis of existing in- the epigenetic rules from statistical ethnographic data. In formation, we nevertheless can characterize the epigenetic rule applying translation analysis to real behavior, however, it is best of sibling incest and employ the translation model developed to start with the simplest conceivable relationship, in which the earlier (1) to predict the form of the ethnographic curves. epigenetic rules strongly bias individual development but not in a manner responsive to choices made by other members of METHODS the society. Although the evidence is still less than decisive, brother-sister incest avoidance apparently possesses properties Two culturgens (1) are recognized: outbreeding and brother- that approximate these conditions and hence promises to serve sister incest. Any one of several degrees of incest can be used as an early test case of gene-culture coevolutionary theory. to demarcate these alternatives. The incest can entail marriage Many human societies tolerate and even encourage marriages or equivalent long-term bonding, full sexual activity, and the between first cousins, but nearly all forbid it between siblings procreation of children; or it can be defined at any lower level and half-siblings. A very few societies, including the Incas, of involvement down to and including the most transient and Hawaiians, ancient Egyptians, Buganda, and Bunyoro, insti- casual forms of sexual contact. The more extreme possibility tutionalized brother-sister incest in the case of royalty or other will be considered here, in which full sexual intercourse is groups of high status. Van den Berghe and Mesher (5) note that practiced with or without marriage-like bonding. the incestuous males were in fact polygynous, behavior that From the Israeli and Taiwanese examples we can tentatively resulted in outbreeding and a higher personal genetic fitness. assign a value of the decision bias U21 in favor of outbreeding Because of the general trend toward hypergamy, especially in (culturgen c1) close to unity and U12 in favor of brother-sister such patrilineal societies, high-ranking women were less likely incest (culturgen C2) close to zero. The young people in the large to marry downward in rank and hence were more susceptible Israeli sample never chose heterosexual intercourse or marriage to matches with their brothers. with unrelated children with whom they had been raised, even Inbreeding at the level of siblings and half-siblings leads to though the adults were permissive and encouraged matches a substantial increase in genetic defects among the offspring within the kibbutzim. In 19 Taiwanese families that adopted (6, 7). From a recent cross-cultural survey, Ember (8) concluded girls as small children for the purpose of marriage to the hosts' that the avoidance of these deleterious effects is the only ex- sons, the young couples later refused to go ahead with the match planation that fits the detailed patterns of same-generation in 15 of the cases, despite parental pressure and a general con- incest aversion. It accounts not only for the inhibition of mating sensus in the community favoring the arrangements. In two between brothers and sisters but also for the detailed patterns cases, one member of the pair died in childhood; the two re- observed in the variable tolerance of cross-cousin marriages. maining couples married. The competing hypotheses that were tentatively discarded or The circumstances in the two societies suggest the develop- assigned secondary explanatory power are the Freudian psy- ment of a strong sexual inhibition during close domestic asso- ciation (10). In the Taiwanese case particularly, the pressures exerted in favor of the sibling-like marriage are a reasonably The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "ad- potent substitute for a surrounding culture in which sibling vertisement" in accordance with 18 U. S. C. §1734 solely to indicate incest is generally practiced. Thus the epigenetic rule appears this fact. to be relatively insensitive to social context. The assimilation 6248 Downloaded by guest on September 26, 2021 Population Biology: Lumsden and Wilson Proc. Nati. Acad. Sci. USA 77 (1980) 6249 1-1 Q- U, C. cow 0 P(t) 0 0a r. cow 25 20 15 10 5 0 Number of nonincestuous members, k FIG. 1. Ethnographic curves derived from the brother-sister incest model in the case of a group of 25 members. The P(Q) fork = 0, 1, .... 25 were calculated using Eq. 2. Only t values corresponding to these k values are realizable. functions vt(4), which relate the individual transition proba- master equation for cultur en dynamics (see equation 4 of ref. bilities to the choices made by other members of the society, 1) is the binomial density can be approximated as constants. The variable t is defined as n2/N - n /N, in which n2 individuals possess culturgen C2 and P(n1, N - nj) = pn'(l p)N-ni, [2] nlI individuals possess culturgen cl in a population of N mem- bers. The translation model can then be applied according' to in which the remaining procedures outlined in ref. 1. A p (1 + r2UI2/TIU21)'. [3] RESULTS When the assimilation functions v0j are such that v21 > Nv12, the most probable ethnographic state is pancultural avoidance The average number of outbreeders in an array of societies each composed of N members is expected to be of brother-sister incest. In the limit of large N, the variable t is effectively continuous and for Vl2 << v2 Eq. 2 is conveniently i = Nv21(v12 + V2)'1 [la] approximated by the Fokker-Planck density (see equation 5 of ref. 1) -N(1 + r2u12/r1u21)1, [lb] P(t) = C(3 - [4] in which ujj are the individual transition probabilities at each ¢)NA-eNf/3tb, decision point, and rI and r2 are the mean lifetimes between in which C is a normalization constant given by decision points for outbreeders and inbreeders, respectively. e Changes in the values of u0 are mapped through the ethno- C-1 = I () EyNA, N3(# 3-)) graphic curve P(t) into shifts of the mean toward = +1.AIn the reverse direction, observation of in arrays of societies if, - y (NA. NGa \)) [5] provides estimates of the epigenetic parameters from ethno- graphic data. such that The simple and robust form of the epigenetic rules of sibling incest generate a correspondingly simple structure ,for the X = V12- V21 ethnographic curves P(s), in which by definition = I1- A-. V12 + V21 2n1/N. The exact steady-state solution P(nj, N - nj) of the A = (#2. -2)/02 Downloaded by guest on September 26, 2021 6250 Population Biology: Lumsden and Wilson Proc. Natl. Acad. Sci. USA 77 (1980) I 1 if V12 <V21 translation yields a correspondingly clear and unsurprising -l if > relation between the degree of bias in the epigenetic rules and V12 V21 the mode of the ethnographic curve. But it also illustrates the and the y(.,) are incomplete gamma functions of the first kind. very general principle that even fixed epigenetic rules, provided Fig. 1 presents a series of ethnographic curves based on Eq. 2. they do not favor one culturgen absolutely, can be expected to In this case we have taken the transition values to be indepen- yield substantial cultural diversity as measured by the variance dent of the culturgen possessed by the individual; in other of the ethnographic distributions.