Clearly, anthropological case studies cannot be used as simple analogies for societies studied in history. The logic that 'X is true in a certain African village and so is probably true for seventeenth century English villages' is inadmissible, even when there is a lack of any alternative 'conventional' evidence: X's truthfulness in England has to remain a mystery in this case. "[A historian's] anthropological training prompts him to ask many interesting questions, and often...these are questions which historians have neglected to ask. It does not, however, equip him to answer these questions, except in so far as he submits them to more orthodox tests of historical evidence." Not only does the particular nature of anthropological fieldwork make generalisations on the basis of the example of a handful of villages dangerous, but the differences between an African village and a seventeenth century English village frequently outweigh any similarities. The two might share elements such as size and mode of production, but the English village is shaped by many elements which differ from the African case, such as its religious system, a different seasonal cycle, and having fewer lions to contend with. Occasionally, the differences are so great that even analogy qualified by proper historical research become worthless: "Study of a more historical kind at once reveals such enormous differences in circumstances and situations that the value of such comparisons - even their capacity to suggest new questions and insights - becomes very problematical."
Fortunately, Thomas did not use simple analogy in Religion and the Decline of Magic, despite his 1963 opinion that, "...anthropological studies of primitive mentality could provide valuable reinforcement for historians confronted by a paucity [or even an absence??] of evidence for the mental life of the lower reaches of the distant society which they are studying." The first two of the witchcraft chapters do not draw directly upon individual anthropological surveys, taking a more conventional line of using documents and other works of history (although some of the conclusions may owe something to anthropology). When references to anthropological works do appear (and nowhere are they overwhelmingly numerous), they are always backed up through reference to documentary evidence. When, in chapter 16, Thomas asserts that, "Like other primitive peoples, contemporaries believed that curses worked only if the party who uttered them had been unjustly treated," he instantly underlines this point with documentary evidence from a Puritan preacher, John Selden, and William Shenstone. In the following chapter, a suggestion made by anthropology is rejected after comparison with 'traditional' evidence: "In many primitive societies virtually all deaths, save those in advanced old age, are attributed to witchcraft or ancestral spirits or some similar phenomenon. But in England at this period men were fully accustomed to the possibility of accident and misadventure."
Thomas, through his continuous use of a scattering of examples to back up each point he makes, certainly avoids the charge of untypicality, but the consistent use of 'thin description' leaves him open to other accusations. "To boil down 20 instances to a line or two apiece must...entail much selectivity and the suppression of much attendant evidence. The reader must still place his confidence in the historian who has decided that this feature only (and not all those others) of the evidence shall be singled out for remark..." While the citing of a large number of examples helps to reassure the reader that these examples are not exceptional, the failure to explicitly explain the various facets of a piece of evidence and the processes by which Thomas distilled it to a single sentence, arouses suspicions that the examples might have been distorted. Thomas provides little to refute G. R. Elton's charge that, "When the externally obtained scheme becomes doctrine, as too often it does, it stultifies the study of history by reducing history to a repository of examples selected or distorted to buttress the scheme."
Whether or not the single sentences Thomas uses accurately reflect the essence of the sources (or even an element of them), it is extremely doubtful whether somebody's actions can be written off as being indicative of a single motivation as Thomas frequently does. "In anthropology, or anyway social anthropology, what the practitioners do is ethnography...What the ethnographer is faced with...is a multiplicity of complex conceptual structures, many of them superimposed upon or knotted into one another, which are at once strange, irregular, and inexplicit..." The credibility of Thomas' examples would have been enhanced had he taken the occasional case and exposed it to 'thick description', examining the whole variety of influences behind it, and, taking these into account, explaining why the interpretation he places upon it is valid.
Further questions concerning Thomas' selection of evidence can be asked if one considers that the sources he draws upon were not written with the systematic analysis of social science in mind. E. P. Thompson sees Thomas' use of example after example as being a result of his desire, caused by his interest in social science, to collect enough evidence to satisfy statistical criteria that a conclusion can be proved by the number of examples, rather than the quality of them. If this is the case, then distortion of the examples is almost inevitable as Thomas attempts to force varied accounts to say the same thing, mentioning only the qualities they share and saying nothing of the many elements which differentiate them.
If one can bring oneself to trust Thomas as far as his evidence is concerned (and the criticisms outlined above are quite harsh - simply because there are areas where his accuracy might be doubted does not mean that the evidence he uses is necessarily bad or irrelevant), then the witchcraft chapters of Religion and the Decline of Magic are quite persuasive. He shows, among other things, that witch beliefs served as a brake on anti-social behaviour and helped people make sense of phenomena which they were ill-equipped to understand by other means. However, it is questionable whether his anthropologically-informed conclusions can tell the full story, and Thomas seems to be too confident that his picture of the place of witchcraft, where almost everything serves a social purpose, is all there is. "Preconceived notions are a much greater danger to historical truth than either deficiency of evidence or error in detail, and nothing entrenches them so deep as the approval of some other form of study which seems to rest on independent thought." Thomas does not often take into account other explanations of the conflict between religion and magic, for example that part of the social role of religion may well have been to impose the values of the elites upon the people, and the result is a skewed survey, drawing upon anthropological findings and seeming to confirm them, while failing to mention other interpretations, even to dismiss them. Potentially, he could be guilty of Elton's charge that, "[Social scientists] establish 'models' which they test by supposedly empirical evidence. To an historian this seems a very dangerous procedure: far too often the model seems to dictate the selection of facts used to confirm it." The number of examples which Thomas uses would seem to limit this charge's relevance to him, but had he considered different conclusions, it is not inconceivable that his selection of evidence would also have been different.
One of the main weaknesses of Thomas' general argument, which his anthropological study no doubt led him to, was his view of man's psychology as being essentially pragmatic. "He seems to assume that, although the origin of specific pratices is historical in that we inherit each assemblage of customs and beliefs from our predecessors, their continued enactment and transmission to the next generation is to be explained in terms of the satisfaction of needs, either psychological or sociological. Since in this view, societal "needs" are mediated by individual agents - responding, holding attitudes, and acting - such a need fulfillment theory reduces to a psychological one. Every important custom has its pragmatic or emotional value for some members of the society, or else it falls into disuse and is forgotten." To reduce all actions to satisfying needs seems slightly simplistic. Many customs are observed today despite their original function now being long obsolete: no doubt many convinced atheists still throw spilt salt over their left shoulder, but in no way believe that this is because Beelzebub himself is standing behind them. Furthermore, conclusions based upon the psychological makeup of people in the distant past are problematical because it is very hard to make the jump from what the sources tell us to what assumptions about the world actually lay behind them. As much as the anthropologist, studying cultures which are distant in space, the historian studying cultures distant in time must overcome a considerable difference in language, where the subtler meanings of words may be lost on the outside observer. Thomas does attempt to recreate the mental world of witch belief, and probably gets close to how it actually was, but there is no way of absolutely verifying his ideas. As such, with his anthropologically-informed conclusions resting at least in part upon his reconstruction of past mentalities, Thomas' conclusions seem to have a potentially weak foundation.
Religion and the Decline of Magic undoubtedly profits in some way at least from its borrowings from anthropology. Some of its main conclusions owe much to anthropological conclusions, and these are not simply forced upon history without the cross-examination of historical evidence. In the last resort this is a work of history drawing upon anthropology, not of anthropological conclusions imposed upon the past as part of some general positivist scheme. It is, however, far from perfect. The examples used, while numerous, are given in a sentence or two, which cannot hope to relate the full implications of the source being employed, and might simply be distortions of the facts to back up the scheme Thomas attempts to describe. The whole survey seems slightly lopsided as a result of the consideration of mainly anthropologically-informed conclusions, and some of the conclusions which he does draw seem to have their share of weaknesses. All of this is harsh criticism, though, especially as very few books (and this would be one of them) claim to tell the whole story. Thomas does show that anthropology can open up a number of routes of enquiry which may well have been obscure to 'conventional' historians, and indeed to shed light on historical events. He may have gone a little too far in his borrowings, but he does not write bad history by any means. In the opinion of E. P. Thompson, Religion and the Decline of Magic is, "...an immensely important and stimulating book..."
BIBLIOGRAPHY
G. R. Elton, The Practice of History (London 1987)
C. Geertz, The Interpretation of Cultures (London 1993)
H. Geertz, 'An Anthropology of religion and Magic', Journal of Interdisciplinary History, 6 (1975)
K. Thomas, 'History and Anthropology', Past and Present, 24 (1963)
K. Thomas, Religion and the Decline of Magic (London 1971)
E. P. Thompson, 'Anthropology and the Discipline of Historical Context', Midland History, 1 (1971)