Comment on David Wengrow’s The Origins of Monsters

The attempt to engage with recent developments in evolutionary psychology and neuropsychology has been one of the more distinctive theoretical trends in recent histories and archaeologies of art, increasingly dissatisfied with purely cultural modes of interpretation, whether derived from the German critical tradition or more recent structuralist and post-structuralist art histories. I think in particular of Orians’ and Heerwagen’s (1992) account of aspects of landscape gardening and painting in terms of evolved dispositions to respond favourably to certain features characteristic of the savannah landscapes within which modern humans evolved; or of David Freedberg’s recent essay (2007), co-written with the neuropsychologist Vittorio Gallese, which explores how the functioning of mirror-neurons may inform our response to depictions of action, from the abstract expressionism of Jackson Pollock to Doubting Thomas’ thumb, pressed into the wound in Christ’s side, in a painting by Caravaggio. Whilst interesting, such studies have had little resonance within mainstream art history, in part because of their rather ahistorical character. Although they may address how certain features of evolved human psychology may inform the character of and responses to some forms of artistic representation, they do little to explain why such forms manifest themselves in some places and periods rather than others.

David Wengrow’s account of the origins of monsters in the iconography of early Bronze Age Western Asia takes the debate to a new level by integrating evolutionary psychology, and the epidemiology of representations, with a strong archaeological emphasis on the material technologies of visual representation, and a sophisticated account of art as an institution, embedded in social and political structures, and articulated with trade networks permitting intercivilisational exchanges. The surprising stability of composite iconography in cross-cultural transfers supports Wengrow’s claim that the composite character of this ‘monster’ iconography combines the right combination of aberration and appeal to taxonomic common sense to have the same kind of broad cross-cultural appeal as certain themes of religious symbolism famously discussed by Pascal Boyer (2002). The relative infrequency of such representations before the Bronze Age and their extraordinary efflorescence and widespread transmission thereafter are then explained in terms of the changing affinity between the socio-cultural affordances of such representations and key features of cultural ontology grounded in the rather differing social and political structures of Bronze Age, Palaeolithic and Neolithic civilisations.

These arguments seem to me largely persuasive, and certainly much more compelling than competing explanations, not least since most of them, as Wengrow points out, focus only on one part of the larger picture he describes, the reception of composites in Bronze Age Crete, for example, or in early Iron Age Greece. A particularly attractive aspect of Wengrow’s argument is the role played by comparisons across time and space both to justify his claim about the status of composites as ‘minimally counterintuitive images’ and to unpack the different kinds of mechanism – cognitive, social, and political – which inform the genesis and the chronological and spatial distribution of the images. It was here, however, that I found myself on the one hand questioning some of the claims made by Wengrow for the mechanisms operating to explain the epidemiology of composites, and on the other simply wanting more, in particular an extension of the analysis beyond the West Asian/EastMediterranean focus of the bulk of his discussion. It is against this background that I would like to raise a series of questions which I hope he may be able to address in order to clarify, and perhaps elaborate, his arguments.

Both the title of the book, and Wengrow’s analysis of composites, lay particular emphasis on the mechanical replication of composite images, primarily through the medium of sealstones and their impressions. Wengrow makes a persuasive case for a structural affinity between modularity in the construction of composites, and the standardization, through modular principles, of material culture and social life in the increasingly bureaucratically organised societies of early Mesopotamian city-states, transformations in which seals and sealing played an integral role. “Composites thus encapsulated in striking visual forms the bureaucratic imperative to confront the world not as we ordinarily encounter it – made up of unique and sentient totalities – but as an imaginary realm made up of divisible subjects, each comprising a multitude of fissionable, commensurable and recombinable parts” (p. 71). It is this affinity (if I understand Wengrow correctly) which explains the universal affordances of composites, not exploited in earlier Neolithic etc. civilisations, being taken up and exploited so intensively in the early Bronze Age. Yet in neither the archaic Greek case – where most representations of composites are singular representations (on vase-paintings or sculptures) rather than media of mechanical replication – nor the Shang Chinese (as discussed by Wengrow himself pp. 85-6) is mechanical reproduction particularly central. What does this imply about the causal significance of technologies of mechanical replication in the genesis and distribution of composites? Is the relation a necessary and internal one – no mechanical replication, no composites—or simply a contingent and facilitative one, first in the genesis of composite iconographies, and second in their transmission beyond their context of origin? As regards the issue of their genesis, how might comparison with other cases – particularly independent ones like the first New World cities and states, and perhaps also those of South Asia – clarify (or complicate) the issue?

A parallel set of questions might be asked about the importance of the role of bureaucratic states. Wengrow makes a compelling case for core structural parallelisms between early Bronze Age states in terms of key features of social organisation, and the cultural practices associated with them, in particular bureaucratization and standardisation in practices of state administration and in the organisation of commercial life within “the large urban institutions, which acted as the religious and economic hubs of the earliest cities” (69). None of these features is really characteristic of the emergent poleis of early iron-age Greece which—in striking contrast to their Bronze Age counterparts—lacked any elaborate bureaucratic organisation of political life or commercial enterprise. Nevertheless, as Wengrow discusses, early Iron Age Greek artists and their patrons were enthusiastic adopters of composites.  What does this imply about the variable causal weight of the different factors identified by Wengrow in different contexts? For example, would it make sense to argue that once generated, the psychological stickiness of composites of counterintuitive images is sufficient to explain their transmission independently of any affinity between their composite character and the character of the receptive society? How far is it possible to disentangle the role of the strategies of local elites, emulating the practices of peer-polities, from the intrinsic potency of composites as minimally counterintuitive images? There are, after all, many other cultural practices, and features of artistic style and iconography, which are transmitted between the Near East and Greece in the same period, for which one cannot invoke the kind of evolved psychological mechanism relevant to composites: iconographies of lion hunting, animal friezes and the like. Ockham’s razor might be taken to imply that we could explain the transmission of the composites also simply in terms of sociological processes of elites emulating their Near Eastern counterparts and appropriating a range of exotic visual images to legitimate the new positions they were carving out for themselves in emergent Greek states, without needing to invoke the evolved psychological mechanisms which may be associated with composites. These two lines of criticism are of course mutually contradictory, but they do at least open up some issues of the logic of causal explanation which it would be good to see clarified.

A final set of questions addresses the passing comment that Wengrow makes about the transmission of composite iconography as occurring most intensively in ‘proto-‘ or ‘archaic’ periods, before the coalescence of the officially sanctioned styles sponsored by the ruling elites of emergent state-level civilisations, for example of Dynastic Egypt and Classical Greece. This is intriguing, and surely requires further exploration. What does Wengrow see as the relationship between the intercultural character of composite iconography – sponsored by state-building elites according to his model – and the forms of social and cultural closure characteristic of the civilizational styles developed by the same elites (cf. Baines and Yoffee 1998)? How far does the more bounded character of these civilizational styles suggest that the kind of epidemiological model, linked to evolutionary psychology, which informs Wengrow’s account of composites, is applicable only in rather special cases, rather than being a model which may be of general relevance to the analysis of ancient visual art?  How far does the Chinese case fit this model, since the dominant artistic style of the Shang elites and the composites of Bronze Age China seem to develop together, indeed in internal relationship to each other, on ritual bronzes?

It perhaps seems churlish, in raising these questions, to ask for even more wide ranging comparison from one of the few studies – at a time when there is so much empty talk of ‘World Art Studies’ –  that is genuinely cross cultural and comparative in its approach. Doubtless a full answer would require another book, maybe even a series of books, since The Origins of Monsters offers not just an intriguing set of case studies but an entire research programme which deserves much further elaboration if the fruits of Wengrow’s approach are to be fully realised.



Baines, J.  and Yoffee, N. 1998. “Order, legitimacy and wealth in Ancient Egypt and Mesopotamia”, 199-260 in G.M. Feinman and J. Marcus eds. Archaic States. Santa Fe: School of American Research Press.

Boyer, P. 2000. “Functional origins of religious concepts: ontological and strategic selection in evolved minds”. Journal of the Royal Anthropological Institute 6 (2): 195-214.

Freedberg, D and Gallese, V. 2007 “Motion, emotion and empathy in aesthetic experience”. Trends in Cognitive Sciences 11: 197-203

Orians, G.H. and Heerwagen, J.H. 1992. “Evolved responses to landscapes”, 555-579 in J.H. Barkow et. al eds. The Adapated Mind. Oxford: Oxford University Press.

1 Comment

  • comment-avatar
    David Wengrow 15 January 2016 (15:57)

    I’m delighted to have a view from Jeremy Tanner, who is able to put my book in the context of recent developments in the history of art. Many of the people to whom I first presented these ideas were in fact historians, especially of ancient art. What they forced me to do, among other things, is work towards a kind of technical precision in my description and classification of forms and images that was otherwise lacking, so when I talk of “composites” it has some internal consistency across time and space. I think that sort of precision is essential in a study of transmission. If the subject matter is permitted to broaden out towards such loose and general categories as “chimerical representations” (see the earlier post by Martin Fortier) the whole discussion loses focus, and the comparisons become highly subjective.

    Art historians – or at least some them – can be usefully pedantic “scientists of culture”. Archaeologists are similarly and rightly insistent about tight control over the spatial and chronological parameters of cultural analysis. I don’t want to dip my ladle too deeply into the porridge of disciplinary stereotypes, and it’s wrong to generalise, but I do wonder sometimes whether socio-cultural anthropology has moved too far from the kind of classificatory rigour it once had. Hence my initial interest in epidemiological approaches, and their willingness to revisit intriguing and unresolved problems such as the ‘mother’s brother controversy’ in kinship theory.

    In responding to Jeremy’s substantive points and questions, I find it necessary to invoke a three-headed “David-Monster” (myself, Napier, Graeber); but before revealing this unholy trinity, some thoughts on mechanical reproduction. Taking the counter-examples of archaic Greece and Shang China, Jeremy asks me to clarify my views as to the causal significance of mechanical replication (e.g. using moulds or seal impressions) for the genesis and spread of composite figures in the Bronze Age. I should say at the outset I think the whole issue of mechanical reproduction – and its impact on society and visual design – has been horribly underplayed for these early periods, and here it is perhaps the art historians, and their insistence on treating such images as singular art objects, that is mostly to blame. The topic merits an entire book in itself.

    Jeremy’s question about causality takes me back to a point in my research (around six years ago) when I was pursuing a particular critique of the epidemiological approach. I subsequently abandoned this critique because it was over-simplistic, and based on an out-of-date and unsupported psychology. Something of this original train of thought, and its rejection, is preserved in chapter 5 of the book. But I agree it is worth clarifying.

    In The Sense of Order, Ernst Gombrich (1984: 256) proposes that pictures of imaginary composite animals – far from being cognitively infectious – confront unusual obstacles to transmission: “there is nothing to hold on to, nothing fixed, the deformitas is hard to “code” and harder still to remember, for everything is in flux”. This now seems completely wrong to me on almost every level, but seduced by Gombrich’s marvellous prose (see p. 82 of my book for the whole passage) – and at this early stage in my research – I wondered whether we had an explanation here for the apparent relationship to mechanical reproduction.

    Could it be that the transmission of counterfactual images on a spatially and temporally extended scale poses cognitive challenges, and needs support from some kind of extra-cognitive (i.e. technological) scaffolding, such as mechanical reproduction? This would provide an economical and attractive explanation for the sparseness of composite figures in prehistoric art, and tallies reasonably well with their later spread in the Bronze Age of western Asia and the Mediterranean. But I abandoned the idea of a strong causal link between mechanical dissemination and the spread of composites, for two reasons: Gombrich, and China.

    The issues with Gombrich are already clear – he offered no supporting evidence for his psychological assertions. China, and more specially the case of Chinese ritual bronzes, was brought in to my study partly because it was central to Rostovtzeff’s earlier work on the subject, and partly because debates over the significance of composite figures are perhaps nowhere more central to the reconstruction of Bronze Age society than in China (see, for example, the many works by K.C. Chang and their subsequent critiques and modifications; e.g. L. von Falkenhausen [2006], Chinese Society in the Age of Confucius. Los Angeles: Cotsen Institute). The interpretation of early Chinese bronzes and their ornamentation is highly controversial, and consulting with some of the leading art historians, epigraphers, and archaeologists of this material made me intensely nervous about venturing into it.

    One thing, however, struck me as both relatively uncontroversial, but also fascinating from a comparative perspective. Unlike their counterparts in western Asia and the Mediterranean, the Bronze Age craft workers of China seem to have systematically avoided mechanical techniques for replicating complex visual designs, despite being aware of their possibilities. Bronze manufacture was intensely modular, but resolutely non-mechanical (the classic study is L. Ledderose’s [2000] Ten Thousand Things: Module and Mass Production in Chinese Art. Princeton: Princeton UP).

    This raises all kind of fascinating questions about the ritual circulation of images and how that might relate to things like courtly shamanism, which are far beyond my competence to discuss (but see M. Puett’s [2002] To Become a God. Harvard-Yenching Institute). It also offers proof that mechanical reproduction is in no way necessary for the dissemination of standard templates for the visualisation of imaginary beings (see p. 87 of mine). In western Asia and the Mediterranean, mechanical reproduction formed an important environmental factor in the “epidemiology” of images throughout the Bronze Age, so it has a correspondingly important place in my study, which focuses on these regions. But as the case of China shows, it cannot by itself be accorded any strong causal or independent value for the cognition of images, or at least not the sorts of images I’m concerned with.

    What about early Iron Age Greece, with its competitive city-state system and enthusiasm for composites? Jeremy points out that when discussing the genesis of composite figuration in Mesopotamia, I ascribe causal value to the development of complex bureaucratic systems based on modular forms of reasoning. Yet these features are only weakly developed in Archaic Greece (or Etruria), which are nevertheless enthusiastic adopters of composites from larger civilisations and empires of the east. This would indeed weaken my argument, were I proposing a uniform set of environmental expectations for the popularity of composites. But I think this neglects the spatial and temporal dimensions of argument (especially chapter 6), which not only allow for but also explicitly define a variety of institutional contexts for the take-up of composite figuration.

    The most relevant ones here are my ‘transformative’ and ‘protective’ modes, which try to address spatial and cultural dynamics between large, centralised polities and the smaller societies on their edges. The underlying model of state formation here is familiar to experts on South Asia as Stanley Tambiah’s (1973) ‘galactic polity’, recently reproduced in the pages of Hau (2013). It relates partially to the spread of ‘new ideas and concepts’, along with various material resources, and how these are mobilized on a local scale to establish and demarcate new types of political horizon – a process involving selective adoption as well as calculated rejection of certain traits between expanding economic-political centres and their margins.

    It is precisely here that I see the recurrent links between exotic imagery and those formative periods called ‘proto-‘ or ‘archaic’, when the self-image and public persona of elites is taking form, and is also therefore at its most vulnerable (and see also my earlier comment on ‘modes of transfer’ and ‘cycles of contingency’ in response to Maurice). The relevance of this model for archaic Greece has been discussed in much greater detail by David Napier (the second component of my David-monster), with specific reference to the Perseus-Gorgon story and its associated imagery of self, non-self, and supernatural protection derived from heroic conquest of the foreign (‘Greek Art and Greek Anthropology’, in his [1990] book of essays called ‘Foreign Bodies’).

    Napier stresses that the foreign encompassed here is not the proximate foreign of expanding commercial interactions, but a more distant and exotic one: ‘the most foreign of all imaginable places’. So he argues, more controversially perhaps, that the Greek iconography of the gorgon contains traces not only of Near Eastern but of still more remote (South Asian) influences. It is worth noting, in this context that – in terms of remoteness – the relationship between Mesopotamia and Egypt in the fourth millennium BC, and between Egypt an Crete in the early second, was probably roughly equivalent to that between Greece and India in the Iron Age. This is of course a function of “ancient globalisation”, as wonderfully described in Philippe Baujard’s (2012) ‘Les mondes de l’océan Indien’ (Paris: Armand Colin).

    In such contexts the focus on composites for the host society has much to do with their provenance and established functions combined – as always – with their innate appeal and cognitively rooted capacity to absorb and attract new (e.g. ritual, mythological) associations and inferences. General things often have specific functions, and I see no reason why it should be otherwise with images. Napier’s suggestion for archaic Greece is that the extreme foreign becomes a ‘medium for making social change permanent’ (or at least relatively stable) in Greek society, leading to the kind of ‘canon formation’ – and symbolic closure to the outside world – that Tanner asks about, and which characterises classical Greece as much as pharaonic Egypt or dynastic Mesopotamia.

    As I note in the book, Napier’s model of cultural transmission is ‘immunological’ rather than ‘epidemiological’. It addresses the ‘tactical use of the foreign in manipulating cultural canons’ – the modification of cultural content through ‘assimilation of what one takes to be foreign’ (Napier 1990: 108); roughly parallel to the biological action of antibodies in identifying and neutralising pathogens. This to mind offers an alternative way of approaching the relationship between individual psychology and cultural identity, which picks up on various points made by Rodney Needham and others.

    An intriguing slant on these issues is provided by David Graeber – the third head of the David-beast under construction – in his recent (2013) Strathern Lecture, ‘Culture as creative refusal’ (Cambridge Anthropology 31: 1-19). There is a degree of recursiveness here, because his argument is partly based on my characterisation of the relationship between ‘bureaucratic’ and ‘heroic’ societies – the latter as originally defined by Hector Munro Chadwick (1926. ‘The Heroic Age’. Cambridge UP). Heroic societies, like those of archaic Greece, are ‘drawn into the trading orbit of commercial-bureaucratic civilisations, and thus accumulating vast quantities of new material goods, while at the same time rejecting the ultimate values of the societies with which they were in contact’ (Graeber 2013: 5; and see also ch. 6 of my [2010] ‘What Makes Civilisation’, Oxford: OUP).

    Under such conditions, the situation described by Jeremy Tanner for archaic Greece is just as might be expected: the uptake of exotic iconography in a heroic society, and simultaneous inversion its functions as first established in the more bureaucratically oriented world of the Neo-Assyrian Empire. This is not, by the way, a simple or binary contrast between ritual and administrative functions, or between heroic and bureaucratic values. The differences are subtler. Neo-Assyrian bureaucrats were often ritual specialists of another kind (more ‘guru’ than ‘conjurer’, in Fredrik Barth’s famous terms), and they too had their heroes. But it is, nevertheless, about the movement of an established repertory of images between different kinds of ritual milieu, and between societies organised on quite different scales and principles of integration.

    A final observation, more archaeological, brings the discussion back to issues of transmission, memory, and replication. Jeremy notes, rightly of course, that most surviving representations of composite figures in archaic Greece are ‘singular representations’ on things like vase-paintings or sculptures, rather than images repeated through media of mechanical replication. That fits the contrast between ‘heroic’ and ‘bureaucratic’ contexts of image use in more ways than one. The point could be expanded if we consider the likelihood that another medium – i.e. complex figured textiles – played a central role in the transmission both of images and stories.

    The problem of course with ancient textiles is that they hardly survive in the archaeological record. But as Elizabeth Barber points out in her brilliant ‘Prehistoric Textiles’ (1991, Princeton UP) we can nevertheless infer something of their appearance in archaic Greece from the content and organisation of painted designs on pottery, which often depict people wearing and using textiles, and also replicate something of the textiles’ decorative syntax in the division of register lines and in the distribution of figural and geometric motifs around a vessel.

    ‘Could it be’, asks Barber (1991: 365), ‘that old, figured textiles with mytho-histories were the sources from which the less-destructible arts – vases, terracottas, and later sculptured reliefs – were deriving their forms and concepts in archaic Greece?’. Links between weaving, storytelling, and heroic values, as well as being near-universals, are also clear enough in the Greek case. The Iliad (125-7) describes Helen at her loom as she ‘wove in the many struggles of the horse-taming Trojans and bronze-armored Achaens’. So Barber concludes, in terms I find suggestive for our wider discussion, that:

    ‘… heirloom “tapestries” recording the earlier mytho-history of the Greeks may have survived from Mycenaean times through the Dark Age into the Archaic Greek period when Homer lived. Such a survival might help to explain the astonishing tenacity and detailed-ness of some of the Greek traditions. That is, Homer and the other bards may have had considerable help in remembering the content of their epics. Could they indeed have glanced up at the patron’s palace walls and seen hanging there a graphic reminder of the episodes in their songs, among the scenes on precious heirloom tapestries – of Troy – Helen – and Penelope’. (Barber 1991: 382)

    The image evoked by Barber would, I think, also be my counter-response to anyone who suggests we can usefully study cultural transmission by separating out the verbal, haptic, material, and visual aspects of human cognition.