A comment on The Origins of Monsters

The beauty of the epidemiological approach came from its theoretical strength. Its weakness came from the difficulty of demonstrating occurrence in actual, non-historically trivial cases. David Wengrow’s book is a serious attempt to do this, it is probably the only such serious attempt. Whether he is successful in this enterprise seems to me less important than the courage of such an attempt in an anthropology which is becoming ever more pusillanimous. We can only move forward by taking risks. In the case of this book the totally admirable risk taking also involves mastering and discussing completely different types of disciplines. This is, after all, what Darwin did when he combined ideas coming from people such as pigeon fanciers with those of his more scientific work.

The theoretical strength of Dan Sperber’s epidemiological approach was that it recognised the necessity of taking into account species-wide human characteristics without running the risk, so evident in previous approaches of this kind, of self-fulfilling reductionism such as we find, for example, in Malinowski’s need theory or in Freudian psychology. The approach also avoids the common mistake of ignoring the obvious fact that representations are more usually borrowed than the fruit of individual creation. The whole point of the epidemiology of representations is that it predicted recurrences within populations without invoking over-strong explanations of specific cases. These could only be explained by the coincidence of general human predispositions and specific historical circumstances.

David Wengrow attempts to do this in The Origins of Monsters with a fascinating wealth of examples. Wengrow therefore explains a coincidence. That is the coincidence of cognitively equipped humans with a specific development occurring very late in the history of mankind and only in certain places. These places are principally Mesopotamia, Egypt, and China. Only the first two geographical locations are discussed in the book in any detail.

Since we are talking of the coincidence of two types of very different phenomena it is important to be clear what these two sides, so to speak, are. I would like Wengrow to tell me if I am correct in what I take these to be.

On one side the author accepts the existence of certain general human characteristics from the work of a number of writers such as Scott Atran, Dan Sperber, Pascal Boyer and, more recently, developmental psychologists such as Susan Carey and Liz Spelke. The problem which much of this recent literature considers and which is most difficult for the anthropologist wanting to take this work into account is how far very young children’s predispositions are modified, changed or lost through development—especially development in specific historical contexts. (This second aspect is raised by the points made in the citations by Tomasello which form the basis of the conclusion.) (Clearly, if these predispositions were to get obliterated through ontogenic development in specific historical contexts they would have no explanatory value for the kind of things considered in Wengrow’s book.).

These general human characteristics taken on board by Wengrow are three. Firstly, there is the proposal that infants display a privileged interest in human as well as non-human animals. This privileged interest is especially focussed on self-generated movements, eyes, and certain facial features. There is by now a mass of evidence for this which, even if I was competent, I will not review.

Secondly, Wengrow takes on, mainly from Atran, the idea that there is a general human tendency to classify the natural world in essentialist ways, thereby creating a non-Darwinian understanding of categorical differences between species.

Thirdly, Wengrow takes on from Sperber, Boyer and others the suggestion that certain systematic limited contradictions of how the world is intuitively conceived become interesting and, as a result, super catchy. It is obvious that representations of the composite creatures discussed in the book would be candidates for this kind of catchiness.

(I will say nothing here of subsequent work which has questioned the specific modular view of the mind proposed by the writers on which Wengrow relies.)

Now for the other side of things. According to Wengrow the historically specific occurrences which stimulate and encourage the type of composites discussed by Wengrow are, above all, the growth of urbanisation and/or states (surprisingly, given Wengrow’s subsequent work, these two factors are not much differentiated here).

Associated with the growth of towns and state three further factors are emphasised by Wengrow: standardisation, mechanical reproduction (mainly through seals) and cultural exchange.

The coincidence of these two sides is what (Wengrow argues) causes the salience of composites. This is a bold and fascinating thesis. The composites from dynastic Egypt and Mesopotamian states cannot but unsettle and interrogate any visitor to the great Museums of the world. The reasons for their existence here proposed are much more powerful than the alternatives reviewed by Wengrow.

Obviously there may well be some particularist objections to the argument such as whether there is something truly distinctive about the composites discussed. Are these truly distinct, as Wengrow argues, from other famous examples of the presence of composites such as those from the west coast of north America or Island Oceania? There is also the question whether exclusively concentrating on material culture, as the archaeologist must, while ignoring, for example, mythology, is theoretically legitimate. (Discussing these niggly points does not seem appropriate in a forum such as this.)

Putting these problems aside, I am left with the central question which the book raises. Is the coincidence that the book argues for a compelling explanation for the occurrence of the types of composites Wengrow discusses?

Some elements seem to me unproblematic. It seems right and very important to stress the connection between the state, standardisation of all sorts, and mechanical reproduction.

More challenging is the proposed connection, or affinity, between composites and urban states. As a first step we can accept with James Scott (cited on page 110) that a characteristic aspect of the development of states is the fact that they impose incompleteness on the much more coherent constitutive communities they absorb.

However, accepting this point is still a long way from seeing a causative connection between this political fact and the presence of relatively standardised types of images or carvings of composites occurring in these states. Further elements along the causation trail connecting  states/towns and composites are proposed in chapter 6. There Wengrow talks of three modes: transformative, integrative and protective. I do not, however, see how these connect up with the general theories of the epidemiologists Wengrow discusses. Am I missing something there? Probably, but, if that is so, I would like Wengrow to spell this out in further exchanges in the forum.

1 Comment

  • David Wengrow
    David Wengrow 13 January 2016 (17:26)

    It’s good to begin with Maurice’s thoughts on the book, since as I note in my précis the whole thing originates in a conversation with him. In that context I should also mention my UCL colleague David Napier, with whom I began discussing these topics some years ago, and whose work (especially ‘Masks, Transformation and Paradox’ [1987], ‘Foreign Bodies’ [1990] and ‘The Age of Immunology’ [2003]) is highly relevant to the kind of issues we are discussing. It would be great if David could also be lured into the forum.

    Back to Maurice, who stresses the exploratory – and potentially risky – terrain we have entered. He notes, for instance, that my whole argument about image transfer is based on a specific modular view of cognition that may yet be refuted (all the more reason, then, for putting its assumptions to the test). I think this generally risky state of play with regard to culture and cognition has been well summarised by David Graeber, in his recent introduction to the English edition of Carlo Severi’s ‘Le Principe de la chimère’, which should enter our discussion at some point.

    The dilemma, as Graeber puts it, is that cognitive science has by now demolished the analytical basis for many classic studies of thought, symbolism, and communication (Lévi-Strauss, Vernant, Goody, etc.), without yet providing tools from which to construct new studies of comparable ambition or sophistication. The epidemiology of representations probably represents the most concerted effort at building such a tool-kit. Partly what we are doing in this forum is a “health-check”, assessing its current fitness for purpose, in relation to what is hopefully an appropriate dataset.

    Maurice characterises the ‘two sides’ of the dataset in a way that is completely consistent with the aims and arguments of the book, as I tried to develop them. Providing they fulfil certain minimal common-sense requirements, we are innately predisposed to perceive disjunctive anatomical forms (like the coquettish unicorn decorating his forum post) as being nevertheless organic and animate, in ways that are special and arresting. This is an evolved mental capacity, sporadically attested in prehistoric art dating back at least to Upper Palaeolithic times, and no doubt earlier examples will one day be forthcoming. But the epidemic properties of the composite figure as image – and on my limited definition of such – were unleashed only through a combination of factors that came into play much more recently; factors institutional and technological, coinciding broadly with the emergence of large-scale, centrally organised societies around six thousand years ago.

    If my book really is, as Maurice suggests, the first serious attempt at a case study in the epidemiology of culture (sensu Sperber), then that’s both exciting and gratifying. I’m not sure it is. Others will no doubt have their own ideas of what constitutes a serious attempt, or in fact what constitutes a truly epidemiological approach. More specifically he asks how my three suggested ‘modes’ of image transfer – described in chapter 6 as ‘transformative’, ‘integrative’, and ‘protective’ – relate to the general principles of such an approach.

    I can best try to answer this with reference to a passage from Maurice’s own study (with Dan Sperber) on the mother’s brother controversy:

    ‘… the task of the epidemiology of representations is not to describe in any detail the actual causal chains that stabilize (or destabilize) a particular cultural representation (although in some cases it is of great historical interest to be able to do so) but to identify factors and processes that help explain the existence and effect of these causal chains. For instance, showing that a particular folktale has an optimal structure for human memory and that there are recurring social situations in a given society in which people are motivated to tell it or to have it told helps explain why the tale is told again and again with little or no distortion of content in that society.’ (Bloch & Sperber 2002, ‘Kinship and Evolved Psychological Dispositions’, in Current Anthropology 43[5]: 727).

    My ‘modes’ are stepping stones towards the ‘recurring social situations’ of this formulation. Neither macro- nor micro-scale phenomena, they attempt to describe something in between: processes such as state formation and the expansion of trade networks, unfolding over generations, that seem to recurrently frame and stimulate the kind of visual experimentation I’m concerned with. My colleague Stephen Shennan suggested a possible parallel here with the ‘cycles of contingency’ discussed by developmental systems theorists like Russell Gray: ‘contingent cycles of interaction among a varied set of developmental resources, no one of which controls the process’ (see Oyama, S. et al. 2001, ‘Cycles of Contingency: Developmental Systems and Evolution’. Cambridge, MA, and Boston: MIT Press).

    I would like to think a bit more about how these middle-range phenomena relate both to the larger and smaller scales of analysis required by an epidemiological approach. My last comment for now concerns one of Maurice’s “niggly” points, because I don’t think it’s all that niggly. That’s the one about different types of figuration, and whether the composite images discussed in my book are truly distinct from other famous examples, such as those of the American North West Coast or Island Oceania?

    Other contributors to the forum are better placed to answer this, but the general point is quite central to my argument. Hard as they sometimes are to pin down, I don’t think these kinds of contrasts in the way images are put together are necessarily vague or trivial. As I see it, the Descola/’Fabrique des images’ exhibition aimed precisely to show how different modes of figuration – different ways of building images – imply wider differences in concepts of relationality, linking up with other areas of practical activity, associative reasoning, social learning, and speculative thought. Barbara Stafford’s work, also referred to in my précis, does this in a different way.

    My thanks to Maurice for comments that give me a lot to think about – I will carry on thinking about them in light of the unfolding discussion.