“Oy vey, have you got the wrong vampire!” A reply to Frans de Waal

I am used to being attacked by fellow anthropologists for having a naturalistic approach and for arguing that cognitive science, experimental methods, and evolutionary theorizing are highly relevant to anthropology’s pursuit. Some of these attacks have been quite violent (one, in l’Homme 1983 [1] concluded with the suggestion that, in order to show me the irrelevance of what is in the skull, I should be given a blow on the head); few if any have paid much attention to my precise claims, but at least they were quite right in targeting me as a naturalist. I am also used to having to work harder in order to get evolutionary biologists and comparative psychologists to pay attention to what I have to say than I would have to if I were one of them. That is understandable.

However, what happened in the past few days was a novel experience.

I posted on PLoS ONE a comment [2] expressing mild reservations on an interesting article by Darby Proctor, Kristin E. Bonnie, Andrew Whiten, and Frans de Waal, “Prestige Affects Cultural Learning in Chimpanzees” (PLoS ONE 5(5): e10625) (to which I had drawn attention here at ICCI). The authors replied and I have now posted a rejoinder to their reply. So far, business as usual. Then Eric Johnson, who runs the excellent blog Primate Diaries, joined in with a comment attributing to me weird views that I do not hold, and I have now replied to his comment. Johnson put an expanded version of his comments on his blog under the title “Anthropology, Primatology, and the Definition of Culture: Reply to Sperber.” It would have been nice to have my work really discussed at Primate Diaries. Johnson might have for instance discussed, or at least reported our paper: Nicolas Claidière and Dan Sperber (2010) “Imitation explains the propagation, not the stability of animal culture.” Proceedings of the Royal Society B: Biological Sciences 277(1681): 651-659. He might, in any case, have looked at the title of that paper and its mention of animal culture and spared himself the embarrassment of attributing to me the view that “humans alone are capable of culture.” His post is aimed at correcting me of that misconception and of other views that I do not hold! I assume that Johnson assumes that I hold such views because they are typical of mainstream anthropologists: I am guilty by association. Anyhow, Olivier Morin posted a comment [3] to correct these misattributions, and I have now replied to Johnson’s post in the form of a comment on his blog (which is not yet online as of this posting).

Things become, I believe, of at least anecdotal relevance to our cognition and culture blog with Frans de Waal joining in the discussion. He first objects to Olivier’s point that while “most animals are able to survive without receiving information from their conspecifics, we can’t …” writing: “This is in fact not true. … I’d say most smart social animals are as culture-dependent as us humans.” As Olivier points out in his rejoinder, most animals are not smart social animals. After all, let me add, ‘animals’ includes for instance a vast number of species of insects. Among insects, social learning “is currently known only from a few well-studied examples in social Hymenoptera” (Reuven Dukas (2008) Evolutionary Biology of Insect Learning, Annual Review of Entomology Vol. 53: 145-160).

Then de Waal turns to my case:

“Sperber believes our evidence is weak, whereas in fact we feel it is the other way around. We have no idea what kind of evidence cultural anthropologists bring to the table. This field seems to get by without any empirical evidence, let alone controlled experimentation.
Once anthropologists start collecting the sort of data we collect on primates, we can compare notes and see if there are substantial differences. But until then anthropologists should think twice before asking others for evidence they themselves never produce.”

Reading this brought to mind Roman’s Polanski’s film, The Fearless Vampire Killers: The Jewish inn-keeper who has recently turned into an undead enters his Christian servant’s room by the window. When she tries to fend him off with a cross, he laughs: “ Oy vey, have you got the wrong vampire!”

Oy vey, Frans de Waal, have you got the wrong anthropologist!

I happen to be also an experimental psychologist. As such, I have worked and published on a variety of topics, some of which – theory of mind or epistemic vigilance – are or clear comparative psychology and evolution-of-culture relevance. I also have – for those who care – all the conventional credentials, publications, academy memberships, and so on, not to be told that I should “think twice before asking others for evidence.” I myself think the only credential you should need is a reasonable argument.

Now, having vented, let me point the more general relevance of the anecdote. Much of Frans de Waal’s work is of clear anthropological importance. Whether they are convinced by his specific claims or not (sometime I am, sometimes I am not), anthropologists and other social scientists should definitely pay attention to his work. I for a long time, and we now at the Cognition and Culture Institute have been trying to help bridge the gap between anthropology and the natural sciences. The greatest obstacles have come from anthropologists most of whom are opposed to naturalistic approaches. But telling anthropologists, as does de Waal, that they should shut up, that they have everything to learn from natural scientists but nothing to contribute, that they “seem to get by without any empirical evidence” (our years of fieldwork must be some kind of paid vacation) is, to put it mildly, neither fair nor smart. De Waal’s reaction illustrates the fact that a serious obstacle to an integrated approach to cognition and culture is to be found also on the side of at least some natural scientists who are blind to anything other than their own contribution.

Since this is a story of wild misinterpretations, let us indulge. Let’s pretend that, when de Waal writes, “We have no idea what kind of evidence cultural anthropologists bring to the table,” this is not the contemptuous remark it sounds like but a commendable admission of ignorance.


[1] L’Homme, 1983, tome 23 n°1

[2] Prestige Affects Cultural Learning in Chimpanzees _ PLOS ONE

[3] Anthropology, Primatology, and the Definition of Culture_ Reply to Sperber _ ScienceBlogs

5 Comments

  • comment-avatar
    Radu Umbres 23 June 2010 (18:42)

    … couldn’t I still join the debate? After all, I come from a research tradition with a rich empirical dimension (albeit one without the benefit of a unified methodology), which also contributed with a rethinking of some central concepts. EMJ quotes quite selectively anthropologiccal works which associate age with prestige in human societies, yet Maurice Bloch argued that a decrepit elder Malagasy has some kind of prestige, while a succesful Big-man has another kind. It seems that, at the very least, anthropology (some parts of it for sure) can be a worthy partner for dialogue, if only at the conceptual level. Moreover, it proposes mechanisms which have empirical validity. Surely the difference between age-based and skills-based prestige is analytically salient (although in many human societies they go hand-in-hand) and can be employed in empirical research, maybe even for primates. Of course, this would require us to compare notes, results and also ontologies and both parties could benefit.

  • comment-avatar
    Christophe Heintz 24 June 2010 (11:51)

    Dan has a an online paper, [url=http://www.interdisciplines.org/interdisciplinarity/papers/1]Why Rethink Interdisciplinarity?[/url], on the difficulties related to interdisciplinary research. He recounts for instance the respective frustrations of anthropologists and cross-cultural psychologists when the latter presented their work to the former. [i] The anthropologists fail to see the relevance of experimental evidence in favour of a thesis they feel confident has already been amply demonstrated with ethnographic data [cultural differences in modes of thought]. They object to what they see as the artificiality of experiments collected outside of an ethnographic context. Moreover, they find the psychologists’ view of culture, exemplified by the fact that they are talking about Western and Asian cultures in general, far too crude. The psychologists feel that the anthropologists are just blind to the importance of experimental evidence, that they criticise experimental methodology without understanding it, and that they fail to appreciate how much their work might contribute to a fruitful exchange between psychologists and anthropologists. In the end, the thesis itself is not given any discussion.[/i] The current situation is therefore very familiar to Dan. In his ICCI post, however, Dan wants to draw attention to a new difficulty, which stems from the interdisciplinary research that takes culture as its subject. Surprisingly, some researchers who brought new methods and ideas to the study of culture have paid little attention to the decades of work of social anthropologists (I am thinking of the most famous naturalist approaches, from socio-biology to memetics). But is it really surprising? Integrating social anthropology in one’s naturalistic research, it seems to me, introduces specific difficulties in addition to the usual ones of interdisciplinary research. This is because social anthropology has — at the moment — no clear synthetic agreed upon take home message about culture. And this probably leads to play down of participant observation as the anthropologists’ method. This said, Frans de Waal does not seem to be really guilty of ignoring the anthropological literature. But in the heat of the discussion, he revealed a value judgement about social anthropology that is probably quite common among those adopting a naturalist approach to culture. Let us take this “cultural fact” as a standard challenge of interdisciplinary research. Work by Frans de Waal and Dan Sperber are, in any case, good reasons to think that interdisciplinary research about culture is possible and fruitful.

  • comment-avatar
    Dan Sperber 24 June 2010 (18:18)

    Frans de Waal sent me this reply, and I am grateful for it. He has agreed to have it posted here: “Thanks for all of the comments, Dr. Sperber. Nothing personal was intended, and I would be happy to stand corrected if there in fact is detailed quantitative anthropological data on social role models. I don’t see such evidence presented in your comments, however, so must assume that in fact it doesn’t exist. This means that my initial comment still stands: it is unreasonable to treat our study as flawed if there are no human studies that meet your own criteria. As for dominance in chimpanzees, there is in fact a double-layered social system that I described already in Chimpanzee Politics of individuals having more influence than their dominance rank suggest and others having less influence, i.e. a young, muscular alpha male who is not taken seriously except in situation where force is required. Hence the disconnect between rank and prestige described for humans cannot be ruled out for chimpanzees. I hope this helps, — Frans de Waal”

  • comment-avatar
    José-Luis Guijarro 25 June 2010 (10:55)

    It has always been a mystery to me why some people regularly misinterpret whole texts they think go against their own beliefs. It is true that Relevance Theory, as far as I know, describes the inferential processes by which we extract the explicatures and implicatures of what is being said. It also, therefore, throws light on some of the ways that misinterpretation of precise [i]statements[/i] might appear. What it doesn’t do, yet, if I am correct, is to give us a full description of the sort of wide (and wild!) misinterpretations that occur in [i]texts[/i] like this, and in other striking cases (like, say, many anti-darwinian or anti-chomskyan arguments that seem to run continuously off the mark). According to RT, interpretations are the product of the process that uses linguistic semantic loaded outputs together with a host of other cues which in some describable way permit humans to read the mind of the communicators and thereby extract the sense of what they wanted to make mutually manifest. But in cases such as the one pointed to above, the inferential processes seem to acquire such a mental strength that they overtake and regularly erase the decoding linguistic process that should also be part of the whole operation –at least, as a pointer to what the goal of the argument is supposed to be. When these pointers is not used at all, and totally new and invented ones are put into use, the result, seen from the outside, gives the impression that some such anti-interpretations are just the ravings of people who are incensed (or perhaps afraid –or both) by ideas that are contrary to what they think is the case, and by trying to destroy them, consider that a whole host of other (for them) related dangers have also been expressed and, therefore, deserve to be attacked and condemned to burn into ashes. It’s pathetic, to say the least!

  • comment-avatar
    Dan Sperber 28 June 2010 (03:55)

    In his reply, Frans de Waal writes: “Thanks for all of the comments, Dr. Sperber. Nothing personal was intended, and I would be happy to stand corrected if there in fact is detailed quantitative anthropological data on social role models. I don’t see such evidence presented in your comments, however, so must assume that in fact it doesn’t exist. This means that my initial comment still stands: it is unreasonable to treat our study as flawed if there are no human studies that meet your own criteria.” In my initial comments on the article, I did not describe the study as flawed. I made two points: 1) that ‘prestige’ was used in broader sense than in the social sciences and in particular in the ‘prestige bias’ hypothesis of Boyd, Richerson and Henrich, and 2) that the evidence did not strongly support the claim that prestige was the factor that caused one individual to be more imitated than another. My arguments may be good or bad, but I still find it hard to understand why de Waal thinks that it is inappropriate to make such arguments unless one has “detailed quantitative anthropological data on social role models” that would match the authors’ own study. Trying to make sense of this demand, I wonder whether it is not based on the following misunderstanding: de Waal may be interpreting me as implying that we anthropologists have a better understanding of the role of prestige in cultural transmission and as arguing on that implicit basis that their study is flawed. But I never said or implied anything of the sort. I don’t believe that anthropologists have such an understanding. The ‘prestige bias’ hypothesis in anthropology is an interesting one but I don’t believe that it is strongly supported either (and, by the way it is being developed by untypical anthropologists who are pioneers in doing both formal simulations and experimental work); it is a hypothesis that, at this stage, given the available evidence and arguments, I neither reject nor accept. (For reasons to be sceptical of this hypothesis, see Olivier Morin’s post “Do we bend it like Beckham?” [url=http://cognitionandculture.net/index.php?option=com_content&view=article&id=286:do-we-bend-it-like-beckham&catid=32:oliviers-blog&Itemid=34]here[/url]). In any case, I believe that either my arguments were reasonable, and it was appropriate to make them, or they were not and it was not, whatever my credentials and past contributions to the topic. I am glad that de Waal’s comments were not personally aimed at me, but they were definitely aimed at anthropologists whose field, he wrote, “seems to get by without any empirical evidence.” I believe that anthropologists would often gain by complementing their field observations with some experimental work, as a quite number of them have already been doing with very interesting results (and readers of this blog will remember that we organised earlier this year a mini-grant competition for experimental work in the field). But the idea that otherwise anthropologists have no empirical evidence worth taking into account is hard to take seriously. After all, nobody would say that, since the people they study are dead and cannot be studied experimentally, prehistorians and historians are incapable of contributing empirical evidence to the study of culture. The empirical contribution of anthropologists is, in many respects, similar to that of historians. True, the people anthropologists study are alive, so they could also be studied experimentally, and, I would argue, for the investigation of some specific issues, they should be. But anthropologists being told that, until they do, they have nothing to contribute are, quite understandably, going to be put off. Those anthropologists who think that exchanges and collaborations with natural scientists are, worse than useless, a kind of surrender to the enemy, will feel vindicated. Coming from someone with such outstanding scientific credentials as Frans de Waal, comments of that tenor are quite counterproductive. PS. In the second part of his reply, de Waal writes: “As for dominance in chimpanzees, there is in fact a double-layered social system that I described already in Chimpanzee Politics of individuals having more influence than their dominance rank suggest and others having less influence, i.e. a young, muscular alpha male who is not taken seriously except in situation where force is required. Hence the disconnect between rank and prestige described for humans cannot be ruled out for chimpanzees.” This is indeed an interesting suggestion that addresses one of the points I made in my comments on PLoS ONE.