Return to top
This feature began life as a light-hearted discussion in early editorial meetings, when we thought that there must be all kinds of established academics out there who desperately wish they could recant their earlier work - take back something they published years ago and just pretend it never happened.
We have yet to find any! (Next time, we'll get out the firewood.) But, two much-respected archaeologists with interests in the Neolithic have offered us some reflections on their past work and provided insight into the nature of archaeological publishing and the author:text relationship. Read on!
I've recently had the chance to review a large chunk of my work in archaeology, in order to produce a volume of collected papers for Edinburgh University Press. The book is called Economy and Society in Prehistoric Europe: Changing Perspectives, and it should be out this year. It's rather like one of those near death experiences where a whole life time is supposed to flash through your mind in seconds. All the golden oldies are there, from the 1970s ('That's the year I was born in', as undergraduates now chirpily remark when I cite some of them) down to some pieces still in press in Eastern Europe - and all with updates, comments and re-evaluations. Since only my copy-editor has read all this stuff more carefully than I have, I can now claim to be a world authority on my own intellectual development. Here are some conclusions.
It all still looks pretty convincing to me. Yes, of course there have been some silly mistakes. The island labelled 'R¸gen' in Sherratt (1976) Fig. 5 is in fact Usedom, and 'Zobten' amphibolite actually comes from the Ardennes (ibid. Fig. 2); but even professors can get islands wrongly labelled (John Collis has a file on one of my Oxford colleagues), so I reckon that kind of mistake as a peccadillo. Yes, I have suppressed many crutch-aching pages on General Systems Theory from the early 1970s; but even Chris Tilley wrote a whole BAR on it so I don't feel too bad. I really feel sorry for people who missed out on that formative experience, since being a post-processualist is no fun if you have no sins to confess, and came into academic life after it had all been worked out. Also, it must be deeply unsatisfactory to see the past simply as ammunition for some presentist agenda. The fashion for choosing a 'designer past' to suit one's current lifestyle is too flippant an attitude to be willingly espoused by anyone who believes that the past is real, knowable and interesting in itself . . . but that doesn't imply an unchanging commitment to a hard-line processualist view of the world, or even a soggy liberal compromise between P and PP. Let me explain, in a roundabout kind of way.
My main problem in editing these papers into a single volume has been that they are all about the same thing, or at least that their subject matter is so overlapping that they are hard to tell apart except by date. The date doesn't need to be explicit, however, because anyone with a sense of style and chronology can phase them pretty accurately to quarter-decades (which, like the limits to the resolution of radiocarbon dates, is inherent in the statistical uncertainties of publication times). It's not just the changing styles of illustration, from hand-lettering to laserprinting, but the vocabulary and expression, and even more fundamentally the story-lines. Archaeological explanations are always underdetermined by the evidence (Sherratt's Law), so that they provide a fairly accurate mirror of the fantasy life of the contemporary academic archaeologist. Are your hypotheses testable? (How quaint!) Are there concealed sexist assumptions? (How reprehensible!) Do you know what phenomenology is? (I don't and I bet most of you don't - it's like that law about funerary elaboration which starts 'To the extent that ...' and you think you are just beginning to grasp it when it ends... 'and conversely', and the sense escapes you again). And so on: that particular set of presence/absence criteria gives the decade, and there are then finer tests for the developmental subdivisions, bifurcations and schisms. Although the accumulation of evidence gives the illusion that these changing interpretations are in some way progressive and caused by improved understanding, they would arguably have taken place even if our factual knowledge had actually been shrinking, since they are largely artefacts of the world in which we live. This sense of real time explains my choice of title, 'Economy and Society' - a deliberate period-piece, since they don't write books with titles like that any more, and it has all kinds of echoes of Grahame Clark and of the intellectual milieu in which I grew up in the 60s -- and also the subtitle, 'Changing Perspectives', which can either be taken as implying a modernist improvement in the angle of vision, or a postmodernist evanescence, like shifting sands. Either way, there has been a consistent alteration in the manner in which we talk about the past, that is reflected in my articles even if they are ostensibly on the same set of related topics.
I hope that this property of changing whilst staying the same is one of the interesting features of them as a collection, and that 'Changing Perspectives' is a fair description. Norman Hammond once edited a conference on Mesoamerican archaeology, with the usual variety of papers offered, and very honestly proposed to sub-title it 'Diverse Approaches'. Nonsense, said the publisher, that will never sell -- you must call it "New Approaches", and so it was. But while this title probably sold more copies, it also gave a hostage to fortune, in that a hostile reviewer from Ann Arbor used his devastating chatty, narrative style to ridicule its claims to novelty ('I saw a student reading this book, so I asked him 'what's new'? ... 'Well, not much'...'). So Changing Perspectives might uncharitably be taken as a promiscuous swapping of objectives: in Jim Lewthwaite's immortal words, 'a string of sordid little affairs with every hustling paradigm in sight, a very library of one-book stands' (1986, 52). Is it any more than this?
When I say 'it all still looks pretty convincing', I mean the basic perceptions of spatio-temporal structure. The continuing thread running through the reprinted articles is the secondary products revolution and the associated changes of the fourth millennium. I think I was dead right about the plough and wheeled vehicles arriving in Europe around 3500 BC as a result of events in the Near East, and that woolly sheep and horses spread shortly after, around 3000 BC. (This, and the important new information on horse domestication and early uses, are discussed in the book's Introduction.) I reckon I was right about the beginnings of farming, too, as a radical shift from the gathering of rainfed wild cereals to their planting in small patches of high ground- or surface-water; and also (for instance) that Pit-Grave groups penetrated into eastern Hungary into areas abandoned by Baden communities because of increasing salination after 3000 BC. Bits of the detail got misunderstood, like the settlement-pattern changes of the TRB (which reflect the incorporation of Mesolithic populations, not adoption of the plough - which only had a major effect on settlement patterns in northern Europe in Corded Ware times), or the adoption of assemblages for manipulating liquids (initially attributed to the use of milk, and then rapidly corrected to the discovery of alcohol!). But these are fine tunings of a structure which I still think has proved remarkably robust. What has changed quite fundamentally is the conceptualisation of what kind of changes these all were.
Some symptoms can be recognised in the subject-matter of the later papers included in the collection. A couple on megaliths - a subject that I avoided in the heyday of settlement archaeology (partly because I was inoculated against it by having to do a compulsory paper in it at Cambridge under Glyn Daniel, but mostly because we regarded religion as pretty epiphenomenal); a couple more on neolithic narcotics - not reflecting any initial unfamiliarity with the practices in question, but rather a growing realisation that prehistoric populations were far more interested in these matters than in maximising calorie-levels. What was happening here was a coming to terms with the-world-as-it-was rather than the-world-as-it-would-conveniently-be-for-modelling. (The latter is best represented by the 'discipline' of economics, which answers all the easy, quantitative questions and avoids all the hard, moral ones.) Messy subjects like religion are quite fundamental to understanding the past; and empires are built out of motivations like the struggle for spices and the desire to grow tobacco. More generally: what we are tempted to call 'technology' was not simply instrumental but also expressive; and the messaging often preceded the practical utility. It turns out to be precisely the attitude which informed Stuart Piggott in writing his Wagon, Chariot and Carriage: symbol and status in the history of transport. Such secondary uses of animals, and of their products (like wool) carry social messages, and these motivations are to some degree independent of their practical usefulness. The 'diffusionist' aspect to my description of the secondary products revolution comes from its fundamental relationship to Gordon Childe's urban revolution, understood as a transformation in the desire for and consumption of materials and objects. The large revolutions are the outcome of small human weaknesses, within a geographical setting where their satisfaction was made possible by small changes in behaviour. Read all about it in the new Introduction!
Does this turn prehistory simply into a realm of amusing anecdote? I think the implications are more fundamental (Sherratt 1995). The kind of prehistory written by children of the Welfare State and nationalised economy was one which tended to make progress look planned. The neolithic revolution was a rational response to desiccation and shrinking stands of wild cereals; civilisation was a rational response to managing diversity; the industrial revolution was the deliberate exploitation of fossil fuel at a time of growing population. Yes, in a way, they can retrospectively be seen as in effect achieving these things (though rarely without winners and losers); but that wasn't how they seemed at the time, and they can't simply be modelled as rational responses to challenges. Each of these changes happened in a restricted area, in a restricted set of circumstances, and then had a major impact on the rest of the world. Moreover they came about initially from human motivations (beer 'n' fags rather than bread 'n' butter). If there was a 'secondary products revolution', then it was part and parcel of a continuing series of consumer revolutions in the very unusual part of the world that we call western Asia (or, more ethnocentrically, the Near East); it is not, as perhaps I was initially tempted to believe, adequately described by the Esther Boserup model of population passing a threshold and requiring intensification of subsistence. What it means is that the people whose behaviour we are describing were human beings.
Have I become more human, too, in realising this?
Andrew Sherratt can be contacted via the Ashmolean Museum, Beaumont Street, Oxford OX1 2PH.
©Andrew Sherratt 1996
Return to topIt is not often that you get the chance to review your own work publicly, and rarer still the opportunity to consider two pieces of work at the same time. Assemblage has asked for comment on how I now view my earlier publication Neolithic Europe: a survey (1985; NE in shorthand here), and this coincides with the publication of the successor and replacement of that book, in the shape of Europe in the Neolithic: the creation of new worlds (1996, with the same publisher, Cambridge University Press; EN in shorthand here).
Perhaps I should hang a card round my neck scrawled with 'recantation', but this would not be entirely honest. I prefer to take my title from that of a poem in the most recent collection by Seamus Heaney (1996, The spirit level). Why should ideas expressed in poetry be so attractive in this context? Apart from the moral inspiration which a poet like Heaney provides, sharing in metaphor and allegory allows the imagination to float free and your focus to shift. Text tends to bind you to what you have written, the more so the bigger the text. It is important to be able to slough off earlier skins. But is it the same creature which re-emerges?
I no longer believe in the kind of Neolithic that NE advocated; I try to set out a different vision in EN, and I will say something about the source of the differences below. But both are in the same genre (as well as in the same publishers' series) of broad-canvas synthesis. Why has this not changed? After all, there are ever more obvious dangers of superficiality, as the amount of data goes on increasing, and what Julian Thomas has called 'totalisation' (for example in his Time, culture and identity, 1996) could be seen as a doomed attempt to capture whole systems and explain them 'as they really were'. However, I would not characterise my own approach as 'totalising' in this sense, and EN is certainly much more selective than its predecessor, and makes far more reference to contingency. On the other hand, EN does argue at one level for a set of values and ideals which are relevant to the whole area covered. Broad treatment also allows interesting contrasts to be set up, which would otherwise be denied. One that particularly interests me in EN is what was going on in central and western Europe in the Tisza-Lengyel/ Stichbandkeramik-earlier Erteb¯lle horizons of the earlier to mid fifth millennium BC. Here one has the appearance of tells, formalised ditched and palisaded enclosures and the beginnings of larger coastal middens. Is there linkage at the level of ideas about place, and can we argue for some kind of process of convergence between the inhabitants of woodland islands and coastal lagoons? Broad treatment also allows the tracking of histories of descent and relationship. The possible link between LBK longhouse and Atlantic long mound is familiar, but there are many other relevant historical relationships. Southern impact on the emergence of the Breton Neolithic is one example, and the context of the emergence of the LBK in the first place is another. I am therefore defending the genre, though it is not the only scale to work at. My own fieldwork has involved detailed consideration of one small region in southern Britain, and the geographical scope of recent important publications (including Thomas's Rethinking the Neolithic, Bradley's Altering the earth, Tilley's A phenomenology of landscape, or Barrett's Fragments from antiquity) is quite limited.
Why then does EN wholly replace NE ? Looking back, the first book tries to put too much in. It is for others to judge whether parts of it are unreadable! A slogan sometimes heard in the days of processual archaeology was that 'we are not handing out prizes for literature', but this has had an unfortunate legacy. I think we all need to work harder at the craft of writing.
Probably any work is a child of its times. NE portrays a Neolithic which spreads over much of the area of Europe by demographic expansion, which sees subsequent continued population growth and resource intensification, and which is characterised by a process of social differentiation: new people, then more people, working harder at being farmers, and competing ever more intensely for social position and resources. A lot has happened since the early 1980s when NE was written. The whole perspective of post-processualism has opened up. We have modelled the nature of the Mesolithic-Neolithic transition in Britain and other parts of north-west Europe in terms of the indigenous population, and we have tried to think in new ways about place, mobility and settling down, and - particularly fruitfully - about monuments and material culture. Ian Hodder's The domestication of Europe (1990) was important in emphasising the importance of conceptual schemes (though I do not believe in either the structure or tradition offered). I would like to think of EN as grounded in a lively theoretical and interpretive debate. In one sense, EN tries to apply to Europe as a whole a series of ideas about the Neolithic which have been debated in Britain over the last ten years or more. It remains for others to judge whether this is successful; many may think I have gone too far. In another way, I try to unite what I see as a series of incomplete models: about individuals, ancestors, the sense of time and descent, relations with animals and the natural world, social values and ideals, and so on. I do not regard this as 'totalising', more as making connections. We have modelled the Neolithic in a very secular way, and imposed on it rapid histories of change, both of which may be much more to do with modern western experience than with the Neolithic world and its forager predecessors.
At yet another level, EN represents for me a change of tone. NE could be seen as practising safe survey, while in Problems in Neolithic archaeology (1988) I was concerned to explore the limits of inference. This caution is easily overdone. We are trying to confront unknowable worlds, and I would now place far more emphasis on imagination and multiple points of departure.
The European Neolithic that emerges in EN is therefore quite different to that sketched in NE. Inter alia, I argue for the contribution of indigenous population right across the continent; for continued mobility, slow settling down, and the gradual building of allegiance to place; for the importance of new attitudes to time, descent, and beginnings; for much variation in how new resources were actually used; for the importance of novel material items, especially in connection with sharing food and drink; and for long continuity of shared ideals and values which mediated competition and created community.
assemblage
©assemblage1996