By way of a conclusion I want to summarise as succinctly as possible the general points I have been making about archaeology, about prehistoric south-east Spain, and about the study of complexity in the Mediterranean. The details can be found in the earlier chapters. Whether these points are welcomed or criticised by the reader is less important than their expression in a coherent form. Gellner's ‘diversified and uncontrollable community of scientists’ is still out there and waiting.

The four points I wish to make concern my approach to the past, and particularly the emphasis on variability, the meaning given to the archaeological record, the evaluation of models for the emergence of cultural complexity, and lastly the differences between forms and rates of cultural change in the east and west Mediterranean basins. In all cases, my aim is to look to the future of research in Mediterranean prehistory.

Variability

My approach to the past is stated clearly in chapter 1. It is an approach which centres upon the study of cultural evolution, using processual explanation, and which expects to be able to evaluate different ideas about cultural change using the archaeological record. This record is not uniform, but variable, as was behaviour in the past. Processes of change may occur at different rates and at different scales, and, as has often been pointed out, different models may have what we might call their own ‘scale-ranges’.

The world about us is in a constant state of flux. The nomad in the Kalahari Desert and the Western city-dweller in front of his television set are both repeatedly confronted by changes in their natural surroundings and in the behaviour of other members of their own species. In this respect, at least, their experience is no different from that of all other organisms.

Culture endows man with exceptional flexibility in coping with his surroundings and, in consequence, human beings regularly cope with an unusual diversity of natural and social environments. As a result, the normal lifestyle of people in different parts of the world can be radically different. Yet the inherent temporal instability of these environments still poses problems and unusually severe perturbations frequently claim human lives. Shortage of food, one of the most basic and yet least reliable of the requirements for human survival, remains a common cause of loss of life. The means by which human beings secure their food supply in the face of such uncertainty are thus as central to society as the consequences of shortage are drastic and they have far-reaching ramifications throughout cultural behaviour and social life.

Human communities have developed an impressive array of cultural mechanisms for buffering variability.

At the beginning of this volume three questions were posed concerning the ways in which societies protect themselves against scarcity: (1) How do societies buffer themselves against periodic variation in food availability? (2) How do these coping activities influence other aspects of cultural organisation? (3) To what extent can coping strategies provide the impetus for social change? We can now return to these questions in the light of the varied cases that have been presented.

How do societies buffer themselves against periodic variation in food availability?

In essence, buffering uses selected aspects of variability to dampen the effects of others – exploiting resource heterogeneity through diversification, spatial variability through mobility and exchange, and temporal variability through storage. The particular mix of responses employed, however, is influenced both by the nature and structure of variability and by a host of enabling and constraining cultural factors. So, for example, when considering storage as a buffering mechanism for high-latitude hunter–gatherers, Rowley-Conwy and Zvelebil emphasise the ecological and technological preconditions while Minc and Smith highlight the social and organisational requirements of such mass capture and storage systems.

Collectively, the examples presented in this volume suggest a series of basic relationships that characterise risk-buffering systems. In particular, they emphasise regularities relating to scale and to the way specific buffering activities are combined to form coherent coping strategies.

Hunger was never far away in ancient Greece. Conventional histories of the Greek cultural achievement rarely draw attention to the material bases of ancient society. In this chapter we aim to show how risk and uncertainty played a part in shaping the central institutions and practices of the Hellenic world, and how historians and archaeologists can use the Greek evidence to help them understand responses to risk in other complex societies. In particular, we will focus on the changing relationships between interannual climatic variability, population growth and the state from the eighth to the third century BC. During this half-millennium, two broad types of state can be distinguished within the Greek world, the polis and the ethnos. Most of our chapter will be concerned with cultural responses to variability and risk within the polis, a state formation based on the political relationship of citizenship.

The polis and politics were invented in a land where agricultural producers laboured under certain natural disadvantages, related in particular to the seasonality and variability of production of staple crops.

The subject of surplus may seem out of place in a volume concerned with scarcity, but the two issues are integrally linked. The importance of surplus lies in its problematic relationship with the emergence and maintenance of elites and other non-producing specialists. It is proposed that surplus be viewed as a normal response to the risk of scarcity in certain types of economy and certain types of environment. The changing role of surplus, and the way in which it may be appropriated by an emerging elite, are explored in the particular context of early farming communities in Thessaly, Greece. In conclusion, the Thessalian case is reviewed in a broader context.

Surplus: problems of definition

The traditional view of surplus, and of its relationship to the development of social complexity, is that: ‘Society persuaded or compelled the farmers to produce a surplus of food over and above their domestic requirements, and by concentrating this surplus used it to support a new urban population of specialised craftsmen, merchants, priests, officials, and clerks’ (Childe 1954:30–1). An agricultural surplus was needed to support the full-time specialists who created ‘civilisation’, and in this sense surplus was a necessary precondition of the development of social complexity.

The early civilisations occurred in rather similar natural environments.

This contribution distinguishes three temporal scales of resource fluctuation (seasonal, interannual and long-term), and examines the responses available to higher-latitude hunter–gatherers, concentrating in particular on storage. The specific environmental contexts in which storage is likely to be a major risk-buffering mechanism are defined. Storage will cope only with the seasonal and interannual scales of resource fluctuation, not with the long term. Direct evidence of storage is unlikely to survive in the archaeological record, but (a) resource specialisation, (b) more permanent settlement, and (c) mass capture technology are put forward as indirect evidence. These features are found among prehistoric European hunter–gatherers in the locations predicted by the model. Social storage will be important when resources are stored and when local spatial resource variability is considerable. Planning for the worst likely situation leads to surplus storage in most years. This, coupled with local resource imbalances, provides a context in which some groups or individuals may be able to acquire and retain more prestige, and hence status, than others.

Storage among hunter–gatherers has been discussed in a number of recent publications. These view storage as (1) one among a number of risk-reducing mechanisms, developed usually as a response to gaps in the subsistence cycle (e.g. Cashdan 1983, 1985; Wiessner 1982), and (2) as contributing to, if not causing, the development of social and economic complexity (e.g. Hayden 1981; Testart 1982a and b; Rowley-Conwy 1983; Price and Brown 1985).

This chapter is based on ethnographic data from a modern, predominantly subsistence-orientated, agricultural community in southern Greece. The community's environmental background is summarised with special reference to the inherent variability of factors affecting crop production, and several superficially ‘inefficient’ behaviours are identified as buffers against environmental variability. It is argued that these mechanisms do not all operate simultaneously or at the same level, but represent a hierarchised set of culturally specific responses to hazard. The interconnectedness of social and economic factors in these buffering mechanisms is discussed, and it is argued that several of these primarily economic behaviours are forces which discourage change in the social sphere.

It goes without saying that climatic – and, more broadly, environmental – factors play a crucial role in affecting agricultural communities and how they make a living. In the past, most writers on the subject of traditional or ‘primitive’ agrarian economies have tended to treat discussions of the environment in general, and climatic factors in particular, in a normative fashion, frequently not looking further than averages of rainfall, temperature and the like. This equilibrium-centred thinking frequently assumed that ‘the climate’ of a locality is essentially static from year to year, though accepting that it is punctuated by occasional hiccoughs or ‘crises’.

The Wodaabe of Niger are pastoral ‘nomads’ inhabiting the West African Sahel, a region characterised by low and uncertain rainfall. Drought poses a recurrent threat to the well-being of livestock and adaptations to seasonal, year-to-year and long-term cycles of drought permeate Wodaabe economics and social organisation. The first part of this chapter explores traditional Wodaabe mechanisms for coping with drought. Responses to individual years of low rainfall are often an extension of responses to the more predictable seasonal cycle of aridity but, in the long term, more severe cycles of drought are encountered which necessitate radical disruption of Wodaabe behaviour. The second part of the chapter explores further sources of long-term disruption–the fundamental economic changes which have taken place since colonisation of Niger by the French. The impact of these changes has widely undermined traditional mechanisms for coping with drought.

The Wodaabe of Niger are one of the groups of nomadic pastoralists inhabiting the Sahelian region of West Africa. This region is characterised by low and uncertain rainfall, and drought exercises a profound influence on the well-being of livestock and on the behaviour of the Wodaabe. Strategies for alleviating the effects of seasonal and interannual variability in rainfall permeate many aspects of Wodaabe economic and social organisation, and the potential risks of longer-term cycles of drought are vividly illustrated by the prolonged famine of 1968–74.

This brief paper presents an example of a refuse-filled pit from a rural settlement of the Middle Uruk Period (mid-fourth millennium bc) in south-western Iran, to illustrate the importance of, and potential for, recovering very fine-grained temporal evidence from the archaeological record. The pit provides evidence of seasonal variation in production, consumption and administration. This information is used, in turn, to document the differing responses to successive good and bad years by the rural community.

Evaluating explanations of complex cultural processes often demands precise differentiation of temporal sequences of years, of seasons and – when we seek to utilise the remains of social, ritual or military action – even finer-grained event sequences. It is thus surprising that archaeologists interested in process still fail to analyse, or even to recover, such data. In fact, most archaeological records present opportunities to isolate samples deposited in short spans of time, and the techniques available for ascribing them to specific seasons or years are becoming more diverse all the time.

The Susiana Plain region is an eastward extension of the steppe and alluvial desert environments of Mesopotamia proper. This plain contains more than 1400 km2 of well-drained cultivable soils; during the early periods discussed here, before the three major rivers of the plain entrenched their beds, much of this area could be irrigated with small canals.

This chapter examines the effect of interannual variability on small-scale agricultural systems, and the use of wild resources to stabilise fluctuations in agricultural productivity. Societies based on small-scale agriculture are frequently vulnerable to severe fluctuations in food availability. For such systems to approximate self-sufficiency, it is crucial that other, highly productive food resources be available, whose structure of interannual variability is largely independent of that governing agriculture. In the Old World, this problem is neatly solved by the coupling of agriculture with animal husbandry. In the New World, where this was not an option, a similar result was achieved by the hunting of large mammals or the harvesting of anadromous fish. Two ethnographic examples of this strategy from North America are briefly examined: the hunting–farming pattern of the Pawnee in the Central Plains region, and the fishing–farming complex of the Huron of the Upper Great Lakes region. It is concluded that the coupling of highly productive wild resources with simple agriculture represents a common coping strategy in cases where large domestic animals are not available, particularly in agriculturally marginal environments. Such buffering mechanisms vary considerably in their organisation, however, ranging from multi-community or multi-ethnic systems with local subsistence specialisation and regularly functioning exchange networks, to very generalised subsistence strategies, in which the members of a given group may divide or switch their efforts from one subsistence pursuit to another, depending on current local conditions.

The idea of a volume looking at cultural responses to uncertainty and scarcity across a range of social scales and economic systems had its origin in discussions between the editors and Peter Rowley-Conwy during an extended car journey in 1982. The idea was developed in more concrete form in a symposium entitled ‘Cultural responses to risk and uncertainty’ at the 1984 Theoretical Archaeology Group (TAG) meeting in Cambridge. The session brought together many of the present contributors and established an organisation that has been retained in this volume. Following the TAG session, the participants were asked to produce expanded versions of their papers, and a number of other researchers with similar interests were approached in an effort to broaden the scope of the volume. The present volume consists of nine studies, ranging in their focus from simple hunter–gatherers to modern states. We have added to these an introductory and concluding chapter, which attempt to draw the various studies together and to consider at a more general level the potential value and limitations of the study of risk and uncertainty.

The editors would like to acknowledge a number of individuals who helped make this volume possible. Todd Whitelaw bore much of the burden for organising the TAG session and we gratefully acknowledge his contribution to its success.

The history of pre-industrial Europe shows that public authorities have almost invariably interfered with the urban food supply. In this paper, it is argued that there were sound reasons for such intervention. Fluctuations in supply would – in a free market – inevitably have led to enormous fluctuations in the price of the staple food. The mass of the urban population would not have been able to cope with these price fluctuations and would have demanded effective measures on the part of the authorities.

Theoretically a number of different types of measures can be predicted, and they can all be documented from the historical record. Specific historical circumstances, however, determine which coping strategy will be prevalent. A comparative analysis of Classical antiquity (in particular Imperial Rome) and the cities of late medieval and early modern Europe demonstrates that costs and benefits of particular coping strategies may differ considerably. It also demonstrates that a purely economic analysis is not sufficient. The nature and extent of state power is at least as important, and so are changes in popular expectations.

In the Digression concerning the Corn Trade and Corn Laws in the Wealth of Nations, Adam Smith complains:

While the basic structure of responses to scarcity is constrained by the nature of those stresses which coping mechanisms must mediate to be effective, the implementation of coping strategies is predicated on the sociocultural context, which defines the range of organisational and technological options for mediating periods of subsistence stress. In this chapter, we reconstruct the spatio-temporal scales of variability in the major faunal resources of interior and coastal Alaska for the late prehistoric and protohistoric periods from variability in relevant climatic and ecological factors. From the structure of resource variability, we predict the basic structure of coping responses, and examine how specific coping strategies were modified over the past 1000 years to adjust to changes in resource structure and sociocultural context.

At the time of European contact in the early 1800s, Iñupiat Eskimos inhabiting coastal and inland North Alaska pursued ecologically distinct ways of life. The tareumiut, or ‘people of the sea’, practised a subsistence economy based on sea-mammal hunting, with an emphasis on whaling.