Thought
In a previous post, it was mentioned that anthropology is concerned with that which takes place between people, not with their innermost feelings and thoughts. How can it then be that this chapter is going to be about… thought? The answer is not simple. It may justly be said that thought has an important social aspect; in different societies, the inhabitants think differently because of differences in the circumstances of learning, different experiences etc. At the same time, thought has an undeniable private and personal dimension, which cannot be studied directly with the methods available to anthropologists.Fortunately, thoughts are usually expressed in social life, for example when people say what they think or express it through their acts, in rituals and other public performances. Therefore, thought can be explored, if often obliquely, through the field methods available to anthropology – participant observation, questions and answers, and common curiosity.
The rationality debate
Studies of thought and modes of reasoning have been central in the history of anthropology from the nineteenth century to the present day. The most famous (and possibly most voluminous) anthropological work from the years before the fieldwork revolution was James Frazer’s twelve-volume The Golden Bough (1890/1912), a comparative work about myth, religion and cosmologies among virtually all the peoples the author had heard about. Frazer shared the evolutionist views of his contemporaries and had little faith in the ability of ’savages’ to think logically and rationally. A younger contemporary of Frazer, the philosopher Lucien Lévy-Bruhl, was less impressive in his use of empirical materials, but as a compensation, he was more analytically lucid than Frazer. Lévy-Bruhl described traditional peoples as representatives of what he spoke of, in an unfortunate turn of phrase, as a ‘pre-logical mode of thought’. However, Lévy-Bruhl emphasised that the term ‘pre-logical’ did not necessarily refer to a developmental or evolutionary line of progress, but rather that the unhampered, metaphorical and symbol-laden way of thinking he associated with traditional peoples was more fundamental, and logically prior to, logical thought. Contemporary moderns may have retained their ability to think in a ‘pre-logical’ way, but a logical rationality has been superimposed on it, as it were. Lévy-Bruhl was criticised sharply by several of his contemporaries, who pointed out that the empirical foundation for his lofty generalisations was weak to say the least. However, it would nonetheless be Lévy-Bruhl’s books from the years around the First World War that set the stage for one of the most exciting theoretical debates in anthropology, where contributors from several academic fields have discussed (and still do) to what degree there are fundamental differences in thought styles between peoples, and conversely, to what extent it may be said that a common human rationality exists.
One of the first to criticise Lévy-Bruhl on an empirical basis was Evans-Pritchard. In the 1930s, he had several lengthy periods of fieldwork in the Sudan. His Nuer research has already been mentioned, but his 1937 book about the Azande is no less important – some would argue that it is much more important – than The Nuer. Whereas Evans-Pritchard’s first Nuer monograph dealt with politics, ecology and kinship, Witchcraft, Magic and Oracles Among the Azande is a book about the system of knowledge and belief in a traditional people, and as such, it was one of the first of its kind. One would in fact have to wait for Kluckhohn’s Navaho Witchcraft (1944) for another study of comparable depth.
The Azande live right in the middle of the African continent, only a few hundred kilometres south of the Nuer; but in terms of culture and social organisation, they are very different from the nomadic peoples to the north. They are sedentary crop growers, politically relatively centralised with aristocratic clans and princes. At the time of Evans-Pritchard’s research, they had been incorporated into the British empire, and the power of the traditional rulers had been reduced considerably.
The Zande belief in withchraft, and their use of various remedies to control it, are in the foreground of Evans-Pritchard’s book. Witchcraft, as it is defined in anthropology, is distinguished from magic in that it is an invisible force. Accordingly, it is difficult to decide who is responsible when someone is struck by witchcraft. Magic is, on the contrary, the result of rites and technologies which are known, and one may consult recognised magicians for assistance with one’s problems. In societies where witchcraft is assumed to exist, it is thus necessary to develop methods to expose the witches. When a Zande experiences a ‘mishap’ (Evans-Pritchard’s term), he is likely to blame witchcraft for it, and he may begin to suspect people he believes has a reason to want to harm him. (It stands to reason that like other peoples who are concerned with witchcraft, the Azande may be said to fit Benedict’s ‘paranoid’ cultural type fairly well.)
If a Zande walks on the forest path, stumbles and hurts himself, only to discover that the wound won’t heal, he blames witchcraft. If one objects that occasional stumbling is normal, he might respond that yes, it is normal, but I walk this path every day and have never stumbled before, and besides, wounds normally begin to heal after a few days. When a group of Azande sit under an elevated granary on poles (to protect the cereals against wild animals), which suddenly collapses and hurts them badly, the immediate cause is that termites have slowly perforated the poles until they were no longer capable of keeping the granary stable. But the Azande will say that it was extremely unlikely that they should sit beneath their granary just as it fell, and thus witchcraft had to be involved somehow. Deaths among Azande are always caused by witchcraft, Evans-Pritchard reports; disease is usually caused by it.
The Azande have at their disposal a range of techniques enabling them to explore whether or not a suspect is actually a witch. (The term witch is, in anthropological usage, gender neutral.) Most commonly, they consult so-called oracles, that is spiritual beings who talk to them through mediums. One popular medium is a kind of sounding board, and there are others, but the most expensive and famous is the poison oracle. To make it communicate, one needs a strong plant-derived poison and a chicken. The chicken is fed the poison, and the oracle is asked whether a certain person is a witch or not. If the chicken dies, the answer is yes; if it survives, the accused is innocent.
In the old days, Evans-Pritchard says, witches were regularly executed. Under the ‘indirect rule’ of the British, implemented from the early 20th century, the princely power was reduced, and judicial power was transferred to the colonial courts of law. Therefore, Evans-Pritchard himself never witnessed executions of witches. In his time, many in fact believed that the very witchcraft institution would gradually disappear thanks to ‘progress’.
The oracles were not infallible. When a witch was dead, one would cut their belly open to establish whether it contained a certain ‘witchcraft substance’, described as a dark lump of flesh. If a witch had been convicted and killed, and no such substance could subsequently be found, the relatives of the dead person could demand compensation.
Evans-Pritchard describes the witchcraft institution in a sober and morally neutral way, skilfully showing how the Azande think and act rationally and logically, given their cultural context. If one were to ask an educated Zande if it might not be the case that bacteria, not witchcraft, made him ill, he might respond that yes, of course, but this so-called explanation said nothing about the reason for his illness right now: the bacteria were around continuously, so why wasn’t his neighbour ill, and why didn’t the illness occur last year? The logic is, as we see, impeccable. Unlike medical science, the witchcraft institution offers answers to the pressing questions ‘Why me?’ and ‘Why now?’.
The book on witchcraft is a remarkable read, and it has rightly been praised as one of the few books that set an agenda for research and discussion which lasted more than half a century after its publication. The book offers rare, deep insights into the knowledge system of a traditional people, and shows how it is coherent, gives meaning to the world, and explains unusual events. Had Evans-Pritchard been ideologically bolder, he might have compared the institution of witchcraft with religions such as Christianity.
The book also shows how the witchcraft institution is functional in the sense that is socially integrative. Usually, the people accused of witchcraft belong to politically weak lineages (nobody would dream of accusing a prince), and he points out that the institution functions as a security valve by channeling discontent and frustrations away from the social order (which would have been exceedingly difficult to change anyway) towards individuals who become scapegoats. Much of the later literature on witchcraft in Africa, especially that published in the 1950s, is purely structural-functionalist, and strongly emphasises that those who are accused of witchcraft are often women, who, in virilocal societies are outsiders without strong political support locally. Evans-Pritchard offers a richer picture, supplementing the functional analysis with a vivid description of local life-worlds.
Unfortunately, many of those who have never read the book itself have heard about it through secondary sources, and therefore believe that it is a condescending, functionalistic description of a primitive people that believes in phenomena that do not exist. A main culprit in creating this distorted view of the book is the philosopher Peter Winch. In 1958, he published the very challenging book The Idea of a Social Science and Its Relation to Philosophy, where Evans-Pritchard appears as one of his main opponents. Winch refers to a number of intermittent remarks in the Azande book, where the anthropologist expresses the view that witches obviously do not exist. In an appendix to the book, Evans-Pritchard distinguishes between three kinds of knowledge: Mystical knowledge based on the belief in invisible and unverifiable forces; commonsensical knowledge based on everyday experience; and scientific knowledge based on the tenets of logic and the experimental method. The middle, quantitatively largest category is common to Azande and Englishmen; the latter exists only in modern societies, whereas the first category is typical of societies where one believes in witchcraft.
Winch argues that the two systems of knowledge – the English one and that of the Azande – cannot be ranked in this way; they can in fact not be ranked at all. All knowledge is socially produced, he continues; and mentions the widespread ’superstitious’ belief in meteorology as a modern equivalent to Zande witchcraft beliefs. In other words, Winch regards scientific knowledge as a kind of culturally produced knowledge on a par with other forms of knowledge.
The criticism of Evans-Pritchard is not based on fabricated evidence, but as I have shown, it does not do justice to his pioneering, and largely non-judgemental exposition of a non-Western knowledge system.
Be this as it may, Winch’s book gave the impetus to a broad debate about rationality and relativism. It would give the initial inspiration for several books, dissertations and conferences in the 1960s and later. Both anthropologists, sociologists and philosophers contributed.
The criticism against Evans-Pritchard contains several independent questions, at least three. The first and second concern methodological possibilites and limitations. The third concerns the nature of knowledge and is anthropological in a philosophical sense.
Firstly: Is it possible to translate from one system of knowledge to another without distorting it by introducing concepts initially alien to that ‘other’ world of representations?
Secondly: Does a context-independent or neutral language exist to describe systems of knowledge?
Thirdly: Do all humans reason in fundamentally the same way?
There are, perhaps, no final answers to any of these questions, and yet (or perhaps therefore) they remain important. We should keep in mind here that Evans-Pritchard himself criticised Lévy-Bruhl’s dichotomy between logical and pre-logical thought, and emphasised time and again that the Azande were just as rational as Westerners, but that they reasoned logically and rationally from premises which were, at the end of the day, erroneous when it came to witchcraft. Winch’s question was whether general, unquestionable criteria exist to evaluate the premises or axioms, and he replies that this is not the case – since the axioms themselves are socially created and therefore not true in an absolute, ahistorical sense.
It should be noted here that a research area which has grown rapidly since the 1980s is the so-called STS field, that is the sociological study of technology and science. In this research, Western science and technology are studied as cultural products, and most of its practitioners adhere to the so-called symmetry principle, which entails that the same terminology and the same methods of analysis should be used for failures as for successes; in other words, that what we are doing is looking at science as a social fact, not as truth or falsity. Similarly, most anthropologists would argue that our task consists in making sense of ‘the others’, not judging whether they are right or wrong.
Classification and pollution
Unfortunately, it is necessary to leave the fascinating controversies about rationality and the rich anthropological research tradition dealing with witchcraft here. Another, no less interesting, way of approaching other knowledges and thought systems, points the searchlight towards classification. All peoples are aware that different things and persons exist in the world, but they subdivide them in different, locally defined ways.
Already in 1903, Durkheim and Mauss published a book about primitive classification, which was to a great extent based on ethnography from Australia. They there argued that there existed a connection between the classification of natural phenomena and the social order. This connection has been explored by later generations of scholars, but historically, there has been a difference here between European social anthropology and North American cultural anthropology. The latter tradition is generally less sociologically oriented than the former, and often explores symbolic systems as autonomous entities, without connecting them systematically to social conditions. Geertz once wrote that whereas society was integrated in a ‘causal-functional way’, culture was integrated in a ‘logico-meaningful way’, and could thus be studied independently of the social. In social anthropology (and, in all fairness, to many American anthropologists), such a delineation is unsatisfactory, since a main preoccupation in this tradition consists in understanding symbolic worlds through their relation to social organisation. Power, politics and technology inevitably interact with knowledge production in a society.
Of the many books about classification and society that have been published since Durkheim and Mauss, two have been especially influential. Researchers and students continue to return to them, and although both were initially published in the 1960s, they do not appear dated even today.
Mary Douglas studied under Evans-Pritchard, and carried out fieldwork among the Lele in Kasai (southern Congo, then Belgian Congo) in the 1950s. She published a monograph about the Lele, but she is far better known for her later theoretical contributions. Especially Purity and Danger (1966) has exerted an almost unparalleled influence on anthropological research dealing with thought and social life.
In this book, Douglas combines influences from her native British structural functionalism and French structuralism, which she became familiar with early on, partly due to her fieldwork in a part of Africa where most of the researchers were French. The main argument is inspired by Durkheim and Mauss, and states that classification of nature and the body reflects society’s ideology about itself. However, her main interest consists in accounting for pollution, classificatory impurities and their results, and one of the central chapters of the book is devoted to a discussion of food prohibitions in the Old Testament. Animals which do not ‘fit in’ are deemed unfit for human consumption, and include, among others, maritime animals without fins and, famously, the pig. The pig has cloven hoofs but does not chew the cud, and there is no category available for this kind of animal. This is what makes it polluting.
Douglas’ theory is as far removed as conceivable from Marvin Harris’ interpretation of sacred cows, and indeed, Harris has argued that the impurity of the pig in West Asia is caused by objective factors, notably the disase-inducing germs which can be present in badly cooked pork. Douglas’ views on this kind of explanation are of the same kind as Lévi-Strauss’ views on Malinowski. According to Lévi-Strauss, the practically oriented Malinowski saw culture as nothing more than ‘a gigantic metaphor for the digestive system’.
The connection between the order of society and the order of classificatory systems is crucial to Douglas’ theory. Among other things, she refers to holy men and women in Hinduism and Christianity, who invert dominant perceptions of pure and impure in order to highlight the otherworldly character of their lives. She mentions a Christian saint who is said to have drunk pus from an infected wound since personal cleanliness is incompatible with the status of the holy woman; and Indian sadhus are famous for their transgressive practices, such as drinking from human skulls, eating rotten food, sleeping on spiked mats and so on.
Phenomena that do not fit in, anomalies, must be taken care of ideologically lest they pollute the entire classificatory system. If this is not done efficiently, they threaten the order of society. There has to be order in nature, just as there is order in society. Douglas’ most famous anomaly is taken from her Lele ethnography, namely the African pangolin. This original forest animal is a mammal, but it has scales like a fish and gives birth to only one or two offspring, just like a human. The Lele have circumscribed the pangolin with a great number of rules and prohibitions to keep it under control; it can be eaten, but only under very special circumstances, and one is usually well advised to avoid close contact with it.
A subgroup of anomalies are the phenomena known as matter out of place, that is objects, actions or ideas which appear in the ‘wrong’ context. The typical example is a human hair, usually far from unaesthetic when it grows out of a head, but repulsive if it floats in a bowl of soup.
Douglas does not write about humour, but one must be allowed to point out that virtually everything that is funny belongs to the same category as the hair floating in the soup: jokes nearly always derive their punchline from wrong contextualisation. Perhaps that is why Geertz once wrote that understanding a different culture is like understanding a joke. When one is able to laugh at the natives’ jokes, one has internalised local norms about correct and wrong contextualisation. This indicates that one has understood a great deal.
Douglas has been criticised for placing too much emphasis on integration in her analyses. Just as Geertz’ concept of culture seems to presuppose that all the pieces of the jigsaw puzzle of culture fall perfectly into place, Douglas assumes that both society and knowledge systems are ordered and fit together.
On the other hand, one should not rule out the possibility that she may be right. Classificatory systems change – there are many secularised Jews and Muslims who eat pork – and there is clearly a greater variation and more direct contestation, especially in complex societies, than Douglas is prepared to admit. But this very variation also seems to confirm the validity of Douglas’ model. When university educated North European Marxist-Leninists took manual jobs in the 1970s, loyal to the principle of self-proletarianisation, they turned dominant classifications on their head in their attempt to change the very ideological foundations of society. In a racially segregated kind of society as the American South, few actions are more radical, both politically and in terms of classification, than to marry across the colour line. Both these examples show that conscious transgressions serve to confirm the essential validity of the dominant mode of classification.
Douglas’ ideas about matter out of place, anomalies, pollution and the analogies between the body, nature and society, have been exceptionally productive. The next chapter will briefly indicate how some of these ideas may be transposed to studies of multiethnic societies, just to illustrate their fruitfulness.