No Haute Cuisine in Africa

The brilliant, illuminating and intellectually cohesive tradition known as social anthropology has long been dominated in Britain by the thought and research styles established by Bronislaw Malinowski. Those who were close to him and who attended his seminar at the LSE were the Companions of the Prophet. Many of them are still with us, active and productive: but they are by now, all of them, in academic retirement. Of the cohort which follows them, Jack Goody is one of the leading figures. In Eastern Europe, there is a joke about the current generation of Communist leaders which says that its coming was predicted in the Bible: ‘Then came a generation which knew not Joseph.’ Jack is prominent amongst those who knew not Bronislaw.

This being so, one is bound to look in his latest book for evidence concerning which way anthropology is going. In his previous work, Jack Goody has already done a great deal to give the subject a new direction. In his Technology, Tradition and the State in Africa he showed how an anthropologist’s fine sense of the intimate texture of societies could be combined with systematic comparison and with a sense of historical context and accumulation. At the same time, his sustained invocation of the means of coercion (or, as he himself puts it, destruction) was an invaluable corrective to the Marxist fantasy of a world dominated in the end by the means of production, with coercion as merely secondary. (Marxists talk of the ‘idealist theory of violence’, meaning the allegedly erroneous supposition that political domination – purely self-serving and not the instrument of any preexisting economic class – could ever be the basis of a social system. It is ironic that an ideology which has helped to engender such a system should also deprive itself of any language in which to describe it.)

His work on the reproduction of the domestic unit has related anthropological questions to historical material, and he has been toying with a schema of European history which links it to kinship patterns. Crudely summarised, this runs as follows. First, tribal groups were perpetuated by the regulated allocation of brides. Eventually, the Church breaks up these kin units by extending the range of marriage prohibitions – in the interest of laying its hands on the land which ceases to be guarded by the now eroded kin groups. (Isolated individuals, scared about their own fate in the after-life, and with no ever-perpetuated group to worry about, eagerly bequeath their lands to the Church.) This situation, however, helps to engender an individualist society in which the Church itself eventually loses both power and land. This sequence, it seems to me, constitutes an improvement on the Hegelian formula according to which mankind proceeded from a stage when only one was free, to a second stage when some were free, and a final fulfilment when all were free. In the less starry-eyed Goody version, we go from a first stage when you must screw your cousins, to a second stage when you may not do so, and on to a final free-for-all when you do everybody.

The present volume makes another major contribution to an overall theoretical issue in the discipline. There is a piquancy about this subject-matter of food. An outsider, at any rate if of genteel background, might suppose that there is something a bit gross and vulgar about a keen interest in food. (I remember a colleague telling me how embarrassed he was in front of his grammar-school friends by his working-class father’s passionate and loud interest in food, until he went on to Oxbridge and found that at the top it is once again perfectly in order to be demonstratively concerned with it.) You might well suppose that in the debate between idealists and materialists, the former would concentrate on the role of ideas and the latter on the importance of sustenance. We must eat before we think. Grub first then morals, says Brecht. Man is what he eats, the German pun insists. So you might expect materialists to haunt the kitchen and scullery, and idealists to browse in the library. This would be a great mistake. Ideology is the favourite subject of Marxist writers. In fact, they can’t keep off it. But if you want studies of that most earthy of our conditions of self-perpetuation, you must go to our neo-Idealists. They are at it so much that they lend themselves to, and have provoked, parody. Jack Goody tells us of one such parody which appeared as early as 1968, La Langue verte et la cuite: étude gastrophonique sur la marmythologie musiculinaire by Jorn and Arnaud. I’m glad Jack warned me that this is a parody – otherwise I might have struggled to make some sense of it. It wouldn’t be the first time in modern thought that the parody is indistinguishable from its object.

The reader may not be aware that Idealism is a strong, possibly a dominant element in contemporary thought. Not under that name, of course. The notion, however, that our life should be understood, not in terms of constraints, operating through fear or hunger, but as guided by symbols and meanings and systems thereof, codes etc – this notion is extremely widespread, and is associated with slogans such as hermeneutics and structuralisme. Associated with these slogans, it has constituted one of the major attempts to modify or transcend the Malinowskian tradition, to go beyond Bronislaw. Though people of this turn of mind have also applied themselves to texts, some of their most celebrated recent work is not about what comes out of our mouths, but about what goes into them. Jack Goody sums up the views of the most influential: ‘Language is therefore the model for the analysis of socio-cultural phenomena, which are interpreted in terms of communication, that is to say, of exchange. Women are like words in language systems and goods in economic systems, objects of circulation.’ He then proceeds to quote Lévi-Strauss: ‘The rules of marriage serve to ensure the circulation of women between groups, just as economic rules serve to ensure the circulation of goods and services, and linguistic rules the circulation of messages.’ Though there are indeed important parallels between these diverse fields, their assimilation to each other, the suggestion that language provides a model for the understanding of the others, seems bizarre. Verbal messages are cheap, sounds can be uttered almost effortlessly, whilst goods and women are often scarce. This totally transforms the rules of the game. There is a great difference between sending a message, which may cost me virtually nothing, and giving a daughter in marriage, which deprives the household of her labour. Communication and exchange cannot be as easily equated as Goody’s summary seems to imply, and to imply in a disturbing en passant manner, as if nothing contentious were in question.

A message can only convey information if it is one of a number of possible alternative messages. (The technical sense of ‘information’ hinges on this, and the amount of ‘information’ is in the end a matter of how many alternative constellations a given system can display: how rare, if you like, the message-conveying constellation of the system is, within the total set of possible constellations. Whether in fact the alternative states of the system are then linked to actual messages is left open.) But when goods or people are exchanged by social groups, it is not an essential part of this process that alternative goods or personnel also be available for exchange: if they are available, if another daughter is to hand as a bride to be given away, then this, significantly, does not alter the message. If any message is conveyed, it is contained in the selection of donor and recipient, rather than of the object or person exchanged, which may be rigidly prescribed. All this highlights the difference between communication and exchange, and the weakness in the recent tendency to assimilate them to each other.

Moreover, the significance of real semantic messages and of ordinary objects or persons exchanged, as evidence of status within the societies in which they occur, is quite different. It is perfectly possible for a culture to play out all its phonetic and/or conceptual possibilities in its utterances and in its myths. Hence, when these are recorded, one can work back from the recorded text to the core generating mechanism which can be assumed to be responsible for them. But political and economic possibilities are seldom, we can safely assume, played out to the full. The cost of the tokens in this game is too great, and there are too many constraints imposed by reality on their unlimited deployment. Many possible constellations of the system are never actualised, because they are contrary to the interests or inclinations of the participants. So we have no reason to suppose that the record, as it were, exhausts the social possibilities. Hence the possibilities which are realised are a very incomplete guide to the full powers and nature of the core generating mechanism, and moreover, we are more interested in those limiting constraints than we are in the hypothetical core mechanism.

No doubt cultural tokens do form systems of a kind which need to be explored, but the analogy has been greatly overplayed, and has been used without care. Take the celebrated ‘culinary triangle’ of Lévi-Strauss, discussed by Goody, which is meant to isolate the basic elements of the gastronomic language: the raw, the cooked and the rotten. Now everything I consume must be either in its original state (raw), or modified by human agency (cooked), or modified by natural process unaided by man (rotten). No other possibility exists. The classification is exhaustive. The triangle is not the hidden secret, the deep structure of our culinary messages: it simply conveys a tautology. This being so, no wonder the aim of the analysis based on it is not, as we are told on the next page, to provide ‘exhaustive knowledge’ of specific societies, but ‘to derive constants which are found at various times from an empirical richness and diversity that will always transcend our efforts at observation’. Well, yes. You can be quite sure that anything you’ve ever eaten will always have been drawn from within that triangle, which does indeed define the very bounds of our gastronomic world.

I am reminded of one of the dreariest phenomena in the Soviet Union: the fact that many restaurants, including ethnic ones such as the Georgian or Uzbek, tend to have the very same menu, evidently supplied by the Central Commissariat of Menus. This consists of a small booklet with sections for Russian food, Ukrainian food, Armenian food, and so on. Each restaurant simply crosses off those extensive parts which are inapplicable to that particular establishment. The difference between this Menu of All the Russias and Lévi-Straussian deep-structure gastronomy is that the former does indeed describe quite interestingly the very bounds and polarities of the Soviet gastro-world, whereas the triangle is purely formal. Actual ethnic recipes can of course be attached to it or to its elaborations, but it is not clear to me what is achieved by this.

A few pages further on, Goody observes that ‘other factors too suggest a modification of the linguistic model used in the seminal work of Lévi-Strauss ... he argues that the language of cuisine, unlike the language of ordinary life, “translates” unconsciously; it is not used to communicate between men as much as to express a structure.’ Once again, we get a terrifyingly casual implied identification – this time between that which is unconscious, and that which expresses some total structure rather than a specific message. In fact, the two oppositions – conscious/unconscious, and individual message/expression of shared structure – cut across each other. A collective ritual may express loyalty to a shared community and its hierarchy and indicate its boundaries, but the message can be blatantly overt and conscious; conversely, in any well-constructed play, characters will convey highly specific and individual messages to each other by means of tone, posture or context, and yet not be consciously aware of these messages.

But there is more to it. All this cannot possibly be described as merely a modification of some early model. What we have here is the conflation of three radically distinct and probably incompatible programmes: the pursuit of the universal features of the human mind, as manifested in the ordering of food or speech; the interpretation of a cuisine as the ‘symbolic’ expression or ratification of an entire social order; and the use of elements from a given cuisine to convey specific messages within a given social context. These must be clearly separated, not treated as ‘modifications’ of each other. ‘La place à table ne ment jamais’ (Mauriac). Et le plat, lui, est-ce qu’il ne ment pas non plus? If man speaks so as to hide his thoughts, as the Frenchman said, does he eat so as to betray them? I think not.

Jack Goody does not necessarily endorse this approach, but his doubts and criticisms are expressed with astonishing mildness: ‘While all theories obviously require a testing of some kind, the nature of functional and structural hypotheses, with their assumption of fit and homology combined with the initial plausibility of suggestions that give an all-embracing unity to the diversity of experience, places them in a special position. The danger can perhaps be gauged ...’ He illustrates what happens when testing is attempted by summarising some evidently valuable work by Adrienne Lehrer. But it would greatly help further testing if those who propound these ideas displayed more conscientiousness in formulating them, rather than being content with ‘initial plausibility’ plus rhetoric plus a shower of ethnographic illustrations. Only at the very end does he go a little further: ‘An understanding of a set of activities needs to take into account the hermeneutic dimension whenever it is relevant ... But the sociological analysis of meaning must also take into account the social dimensions of the problem of their roles and relations, their position in the hierarchy, their membership of one society as against another and their position in the world system. The concern with culture must not exclude the social ...’ And he writes: ‘I am aware that too little attention has been paid to the ritual, symbolic and cosmological aspects of food.’ Why on earth is he so apologetic about his findings and about his stress? Neither food nor society is primarily a code. They have conditions and consequences, which are far more important than their role as tokens. I have never doubted this, but my unwavering faith is greatly strengthened by Goody’s data and conclusions. The only thing that puzzles me is his almost shamefaced tone in reaching them. This must, I suppose, be a sign of the strength and authority of current neo-idealist trends. People ought to be helped to overcome them. Perhaps one could found a Hermeneutics Anonymous to give them strength to do so.

Society is not a language. Language isn’t a language either, if it comes to that. Virtually nothing is a ‘language’ in the structuraliste sense, except perhaps systems such as packs of playing cards. In a real language, we say complicated and specific things: we do not exhibit and wave objects at each other, objects selected for their position at the limits of the spectra which define our world, in the interest of reminding each other of those limits, lest we forget. Nor is our alleged habit of polarities-waving somehow the basis of our capacity to say specific things and to conceptualise the world: on the contrary, structuralisme naively takes those capacities for granted – though they are far from self-explanatory – in its simplistic accounts of how our world is endowed with its structure. The idea that beating the binary conceptual bounds is the essence of communication is a structuraliste myth, even if it does provide a facile recipe for sparkling interpretations of texts.

Goody’s summary, and his much too cautious evaluation of recent trends in this field, only take up the early part of the book. He then proceeds to a detailed account of the whole process of food production and consumption amongst two Northern Ghanaian societies of which he has intimate direct knowledge. One of them, the LoDagaa in the north, were stateless, uncentralised and livestock-oriented; the Gonja, a little further south, had a state organisation and a formal social stratification with an aristocracy, a Muslim clerical stratum, commoners and slaves. They were more disrupted by colonialism than the relatively amorphous LoDagaa, and in any case they were less populous.

Goody remarks that this last feature is puzzling to anyone starting from Middle Eastern data and expecting states and populousness to go together. I’m not clear whether he included North Africa in the Middle East, but it has been my impression that the stateless, or rather state-resisting, Maghrebi populations of regions such as the Atlas, the Rif or Kabylia, and probably the Aures, were thicker on the ground than many of the governed areas of the plain. Why should oppression lead to populousness? Some of the sedentary valleys of the Atlas, if one only counts the actually cultivated and inhabited valleys themselves (and not the surrounding hills intermittently shared by the inhabitants with surrounding pastoralists), must have had extremely high population densities. I do not know whether these impressions have been confirmed by demographers.

From the viewpoint of testing theories about gastronomy as conveying social messages, the results of Goody’s inquiry are entirely negative, interesting though they are otherwise: ‘the actual shape of the cuisine in both societies was surprisingly similar.’ It seems that there were no significant differences, either periodically or in terms of social strata, between the various levels and occasions of Gonja social life. As for the LoDagaa, it seems they had few differences to express. So if a local Rip Van Winkle woke up in the middle of a meal, the dish in front of him would offer him no clue whatever as to either season or occasion or social standing. The most he could hope for, if Lévi-Strauss is right, is some insight into the nature of the human mind.

It is this fact – the absence of culinary stratification in sub-Saharan Africa – which led Goody to the central question which, as he himself stresses, underlies this book. What are the conditions for the emergence of a high and low cuisine? Why is political differentiation, which was certainly present in some African societies, insufficient to engender gastronomic stratification? His accumulation of relevant data takes up the major part of the volume. The data are extremely rich, diverse and interesting, and it would be impossible to summarise them. But if the details elude summary, the conclusions do not.

Neither food production nor political centralisation as such are sufficient for the emergence of high cultures. So there is another, complex and subtle, but immeasurably important, Great Divide in the history of mankind, posterior to food production and the state, but closely linked to writing, and to social differentiation by access to literacy, and to domination by literate specialists. Intensive agriculture and literacy seem linked to that sustained elaboration of culture which marks off Asian pre-industrial civilisations from the societies of sub-Saharan Africa. No high cultural products without differentiation, specialisation and stratification ...

Friedrich Engels thought that the subjection of women marked one of the great social revolutions of human history. Goody, by contrast, thinks it was the sign of a crucial transformation in human culture when cooking for great men and occasions was handed over to men. That showed it to be serious. It was then that it was linked to a codified and regulated division of labour, going beyond that of the domestic unit. So gastronomy is not so much a code, whether universal or specific, as a reflection of the level of development of the forces of production and coercion. When writing engenders clerisy, it also makes possible a codified, recipe-borne cuisine. Culinary codification accompanied systems of legal and moral recipes. It is only when the tools are available to chain man’s spirit as well as his body that you can really get a decent meal. So in order to understand the new cookery, one must look to its preconditions, rather than to the messages allegedly encoded in it. Back to pre-hermeneutic sociology. Societies are systems made up of concrete people and their activities, and the interdependence and preconditions of these activities require exploration: they are not codes, used by a mysterious and all-powerful central mind, whether culture-specific or universal, to emit cryptic messages, whose decoding can then keep our latterday structuraliste clerisies in business.

Thorstein Veblen ironised and inverted the once-favoured anti-egalitarian theory which had taught mankind to accept a Leisure Class as the price of cultural advance. The spirit of craftsmanship, Veblen thought, which could be self-generating, would suffice, and we could dispense with the parasites. But apparently that spirit has complex social preconditions too. Using a greater range of ethnographic data than was available to Veblen, and without that special American bias which takes work ethic and political order for granted (Veblen’s book doesn’t even consider the state), Jack Goody has stood Veblen on his head once again. With great force and persuasiveness, he has in effect rewritten The Theory of the Leisure Class without, this time, the irony intended in its title.