Darkness in El Dorado - Archived Document
Anthropological Niche of Douglas W. Hume
Home | Darkness in El Dorado | Contact

Internet Source: email
Source URL: none

My friends, Clay and Carole Robarchek asked me to post this response to the UCSB report on the Tierney-Chagnon debacle, as they are aware that the UCSB report was posted to this list.

We are reluctant to introduce a red herring into this discussion of Tierney's dishonest allegations, but we feel it is necessary to correct the erroneous characterization of our research presented in the recently circulated UCSB Preliminary Report on the Neel/Chagnon allegations. While we have been very critical of sociobiological explanations, including Chagnon's, of human social behavior, we are delighted that some of the data from our Waorani research has been useful in rebutting Tierney's charge that Chagnon "cooked" his data. (Our quarrel is not with his data but with his interpretation of them [Robarchek and Robarchek 1998; Robarchek 1989, 1990].)

Both the University of Michigan Report and the the UCSB report correctly point out that, as Chagnon claimed for the Yanomamo, our data on the Waorani also showed a positive correlation between the numbers of killings that men had been involved in and the numbers of their wives and children.

Contrary to assertions the UCSB report, however, we are neither "very uncomfortable" with that fact, nor do we believe that we have "essentially replicated Chagnon's finding." Rather, our study reveals the essential weaknesses of many of these sociobiological explanations, specifically the assumption that correlation equals causation and the reluctance of sociobiologists to consider alternative hypotheses.

What we found among the Waorani was that the appearance of correlations among the numbers of a man's killings, his wives and his children results from the fact that all three are functions of two other variables: when a man lived and how long he lived. Moreover, the latter was heavily influenced by the former: if a man lived out his entire life in the pre-contact period of intense raiding, he might continue fathering children as long as he lived, but his life was likely to be significantly shorter than was the case for those who were still alive when the large-scale killing ceased. Those who survived into the contact period when large-scale raiding ended averaged more than 50% more children than did those whose lives were cut short by spears. In other words, the longer a man lived, the more children he had, and the cessation of warfare dramatically increased killers' fitness.

It was not the case, however, that those who survived were the most proficient killers; in fact, precisely the opposite was the case: the more killings a man was involved in, the more likely he was to be a homicide victim. Of those with 3 or more killings to their credit, nearly 70% were killed. Of those with 2 or fewer killings, 32% were killed.

Moreover, women and children were also killed in these raids, and the high rate of homicide for women was the primary reason why men who survived had more wives: the longer a man lived, the more killings he would participate in, the more likely it was that his group would be raided in retaliation, that he and/or his wife and children would be killed and that, if he survived, he would remarry. A man involved in 3 or more killings was nearly three times as likely to have his wife killed (and thus to remarry if he survived) than was a man who had been involverd in two or fewer killings (Robarchek and Robarchek 1998).

As in the case of numbers of children sired, the number of a man's wives was a function of his longevity, given that his reproductive life was lived in the context of this warfare complex. Thus we were able to show that the covariations among numbers of wives, children and killings--precisely the kinds of data offered in support of the sociobiological paradigm--are, at least in the Waorani case, incidental consequences of an entirely different set of causal relationships that derive from the sociocultural context and that are entirely irrelevant to sociobiological assumptions.

As we said at the outset, we are delighted to be able to help rebut Tierney's allegations against our colleagues, but we also feel that there has already been quite enough misrepresentation in this affair.

Clayton A. Robarchek
Carole J. Robarchek


Robarchek, C.A.

1989 Primitive Warfare and the Ratomorphic Image of Mankind. American Anthropologist 91(4):50-67.

1990 Motivations and Material Causes: On the Explanation of Conflict and War. In: The Anthropology of War, J. Haas, ed., pp. 56-76. Cambridge: Cambridge University Press.

Robarchek, C.A. and C.J. Robarchek.

1998 Waorani: the Contexts of Violence and War. Fort worth: Harcourt Brace.