Let me start by clarifying, as I think was well-taken, that my initial remarks were not intended as a rebuttal of Foucault per se, but against the notion of making things “appear against a background of emptiness”, as I felt that this phrasing neatly encapsulated what I felt had become a default position among anthropologists. This tends to result in relativism, it is true, but I did not mean to cast relativism as the central problem, but rather an epiphenomenon of the underlying issue, which is an underlying assumption among many anthropologists and, I suspect, critical theorists more generally, that thought is language-like.
Let me start by distinguishing two approaches to thinking about cognition that seem to encompass most of the thinking that people who identify as anthropologists do. One I would call evolutionary cognition, the other, as the dedicated LSE Msc program calls itself, 'Anthropology of Learning and Cognition'. It is my feeling that the arguments raised by Twinglebrook-Hastings, in particular, apply much more strongly to the former than to the latter.
The evolutionary-cognitive approach tends to set the question up thus: what is the explanation for [apparently irrational behaviour/belief/cognitive trait]? Answer: [irrational cognitive trait/belief/behaviour] arose in response to [particular threat to survival of species] on the Savannah. Porblem disappears. This is not even slightly a caricature. The Avatar of this approach is of course Pascal Boyer, Cognitive Twat in Residence, Paris Institut des Conneries. Boyer built his name on the argument that all religious behaviour arises as slippage between cognitive domains which arose for straightforward evolutionary reasons. Principally, that in order to predict the actions of enemies/women/sabretooth tigers we needed to have a 'theory of mind', and that this 'theory of mind' gets misapplied well beyond its 'intended' evolutionary domain, leading us to believe that rocks and rivers and badgers are alive and intimately interested in our personal destinies. All religious belief/activity is spandrelly epiphenomenon, porblem solved.
(I must admit I think that the comparative study of, say, witchcraft, could be greatly enriched by attention to possibly innate understandings of intentionality. I also don't think the theory of domain specificity is totally without promise, as I shall briefly suggest below. it's the reduction of the problem to an evolutionary question that is objecitonable, or, as Prof T-H so cogently put it, the reduction of the problem to itself.)
The learning-cognitive approach is significantly different. It's central question picks up from Bourdieu, whose main concern could be summarised as: “how is it that people make culture, while culture makes people”. The observation which arose in the 80s, partly under the influence of the Selfish Gene, and partly among those troubled by what we call the 'writing culture' movement, was that without a richer empirical understanding of how children actually learn a culture, we were at something of a dead end. That is, if you wanted to see how inculcation, enculturation worked, including for major concerns of critical theory like the construction and maintenance of ideology, then you needed to look at how children learn. This is where my problem with the 'background of emptiness' becomes clear. nonetheless, I think my views, and the better parts of this school of anthropology, are entirely consistent with Twinglebrook-Hastings's conclusion that
“Foucault's formulation need not imply that humans are blank slates. Rather we would like to have a look at the slate, at where it was mined and manufactured, at the chalk, the schoolroom and how it gets written on.”
An anthropology of learning and cognition then, is an attempt to do exactly this. And if done properly, far from performing a vanishing act on the contingent and variant content of cultures, should add enrichment and nuance to our understandings, one could happily say our descriptions of them. I will try and give an example:
Rita Astuti of the London School of Economics wrote her PhD thesis on the Vezo of Southwestern Madagascar. One of the most interesting things about the Vezo is their decidedly fluent and contingent account of ethnicity and person-making (one which is, as it happens, shared to some extent by the Amhara). In short, the Vezo are fishing people. They live on the coast and are skilled makers and pilots of Malayo-Polynesian style canoes. This is as opposed to the Masikoro, who are cattle herders and ploughsmen, and know little of fishing and canoeing and are thus slightly inferior. However, Astuti argues, all a Masikoro needs to do in order to become Vezo is to move to the coast, and learn to fish and to build canoes. This is pretty unusual, as generally even the more open small societies require you to at least marry someone in order to gain group identity.
Anyway, after this fairly standard enthnographic thesis, Astuti becomes increasingly interested in cognition and conducting field experiments. One of her questions surrounds Vezo ideas of heredity. It seems that Vezo insist that children gain their physical features not form their birth parents but from the people who are around them most as they grow up. Vezo tend to downplay ideas of parental 'ownership' of children, and rather see it as the part of the whole village to raise every child. Astuti's question is, to what extent can the Vezo actually be said to 'believe' that children get their physical characteristics from pater/mater rather than genitor? If you ask them about children, they reply almost without exception according to the cultural norm. However, when she conducts a serioes of experiments asking people about animal reproduction, results are strikingly different. She asks questions like – a chameleon with thirty teeth gives birth, and the child is raised by a chameleon with twenty teeth. How many teeth will the child have? She is careful to use weird animals rather than domestic ones, so as to try to lead people to reason based on some find of organising principal rather than refer to specific ideas. Adults invariably say that the child chameleon will have thirty teeth, in line with Western expectations. So people appear to have some kind of understanding of heredity which does not conform to their – universally agreed-on – stated beliefs on the matter.
The point I'd like to make is that the kind of question Astuti is asking here, while still 'cognitive', is radically, radically different from the kind of question asked by Boyer. Astuti's results challenge me to a more nuanced approach to what a cultural 'belief' might be, and hints at possible different levels of meaning and understanding on which the Vezo can call in order to make sense of life. It unquestionably enriches our description of the Vezo world, and Vezo ideas, as well as provoking more general questions on the relationship between ideology and belief, or even on the viability of a notion like 'belief' in general. This sort of question is very similar to the old ding-dong about whether Trobrianders actually 'believe' that fathers have nothing to do with childbirth.
Now, the patchiness of my knowledge of the Western philosophical tradition is well-known, and embarrassing, but it seems to me that Wittgenstein would not disapprove of this kind of work, and it does seem to me that for good anthropologists, Wittgenstein is the man with the plan. Indeed, Malinowski's early work about 'phatic communion' (otherwise known as maintenance speech, or language not meant for the transferral of propositional content) would seem to conform very closely with Wittgenstein's foundational stuff on speech acts (I beg forgiveness for the clumsiness of this analysis; I don't know a better way to raise this connection but it seeeeeeeeeems to me that good cognitive anthropologists and Wittgenstein have a lot in common and I'd like to hear esteemed colleagues' thoughts, be they ever so brutal.).
So, in a lot of really important (and possibly fairly obvious) ways, thought is not like language. Now, at least in anthropology, this to me is by far the most important knock against the culture-as-text school, and specifically against Geertz. And I don't think t is too much of a step to say that it is the assumption that culture is language-like (even among those plenty familiar with Chomsky who should know better) that allows 'stupid-foucauldians' to perpetuate stupid-foucauldian relativism.
But to bring it back to domain specificity, and as an example of how culture does not float free in the way language does, there is evidence suggesting that cognition is more compartmentalised than was once thought, and that humans may have things called 'modules', of which one would deal with 'folk physics', one with 'folk biology' (the names are hideously unfortunate), one with face recognition etc. All I want to point out for the moment is that it appears to be a universal fact that children at a young age make a distinction between things that are alive and things that are not. this distinction would appear to be genuinely innate, along with certain other aspects of concept-formation, to do with fairly well-known experiments on how many parts of an elephant you have to take away before a child no longer recognises it as such. Now I am not being too precise here, because I only want to make a fairly general point here, which is that if all children innately think differently about animals and other objects, then whenever we look at cultural ideas surrounding living things, it would be incorrect to treat those beliefs against a 'background of emptiness', because the background is a significant difference in the ways humans cognise stuff. So it's not a search for the 'steady hand' of cognitive science, just towards a more accurate account of the mechanism which makes and is made by culture. That is my problem with cognitive relativism, and it is one which colours a lot of the work to which I am exposed. Whether it represents a problem with Foucault himself, I do not know.
I think the best parallel I can offer to make it clear what I am advicating is with literary studies. I hope it is becoming clear that the cognitivism I support has little or nothing to do with literary Darwinism, or the search for evolutionary 'explanations' of cultural forms, which, as Bremselhäcker notes, inevitably results in the impossibility of differentiating between different kinds of cultural content, and so ironically introduces a supreme sort of relativism. Rather, a literary parallel to the cognitivism I like would be precisely that advocated by Bremselhäcker in his post 'Further thoughts on Foucault': “a refocusing of humanistic inquiry in which regarding literature as a social construct does not imply regarding it as merely a social construct, that is to say in which charting the social forces which brought it into being as an object and an experience does not imply its hollowing out, its loss of specificity, or its reduction to a vehicle of identity construction”.
And thence to Latour. I first properly attempted to read We Have Never Been Modern one year ago. I didn't get much of it. After teaching one term of advanced social anthro, focused around demolishing the nature-culture divide, I go back to We Have Never Been Modern and suddenly it all starts to fit together in a way not incomparable to the descent of fleets of heavn'ly angels singing praise. If I can put the question somewhat flippantly as Social Construction+Real World=What?, then it is tremendously heartening to see the kinds of answers that are emerging, even if I personally needed some time to grow into them. For that, for the beginnings of a new kind of approach, I am eternally grateful to my friends and co-travellers. It's been Real, let us keep it so.