With the publication of Paul Radin’s Primitive Religion: Its Nature and Origin (1937), nearly seventy years of intense evolutionary theorizing about religion had apparently come to an end. Why, then, did E.E. Evans-Pritchard decide to revisit these debates in Theories of Primitive Religion (1965)? Among other things, Theories is a trenchant critique of psychological and sociological theories of religion, all of which revolved around “primitives” and were therefore evolutionist. But why publish such a critique in 1965? Nearly thirty years had passed since anyone had proposed additions or amendments to such theories, and there had been little scholarly discussion of the various theories during that time. The issues had seemingly been settled, either because scholars had completely rejected evolutionary theories or fully accepted Durkheim’s sociological theory and its structural-functional progeny.
So how can we account for Evans-Pritchard’s apparently odd, and certainly belated, decision to publish a book revisiting these debates? There are several possible answers. The most tempting, and unlikely, is Evans-Pritchard had foreseen that evolutionary theories of religion would, in the early 1990s or thirty years in the future, make a comeback. Theories might thus be seen as a pre-emptive challenge or strike. While Theories does in fact raise methodological objections that any evolutionary theory of religion must address and overcome, we need not grant Evans-Pritchard this sort of prescience or clairvoyance.
A better answer is that Evans-Pritchard wrote Theories because he had been thinking and writing about these issues his entire life. This too seems odd given that Evans-Pritchard is known primarily for his data-intensive ethnographies of the Azande and Nuer. Anyone who has read these classics knows they are long on description and short on theory. To the limited extent that theory makes an appearance in these monographs, it is structural and functional. One gets the impression that Evans-Pritchard was simply following in the sociological footsteps of Durkheim and Radcliffe-Brown. He was fleshing out their theories with ethnographic facts.
If this were all that one knew about Evans-Pritchard, or if the classic Azande (1937) and Nuer (1940) ethnographies are the only things one reads, this impression is not incorrect. It turns out, however, that Evans-Pritchard had been thinking and writing about theory his entire life. During the course of his career, which began in the 1920s and continued through the 1970s, he published numerous articles in addition to his famous ethnographies. These articles, often appearing in obscure journals, provide the theoretical frame within which he carried out his better known work. Toward the end of his career, Evans-Pritchard seems to have sensed that this method had rendered his corpus opaque and given rise to misunderstandings. In this professional light, Theories can be seen as a summary of his scattered work, a corrective to misunderstandings about that work, and as the capstone to a career. In the more personal light of his conversion to Catholicism, it can be seen as a rebuke to atheist scholars who would explain religion away as an epiphenomena of either mind or society.
Because Theories has achieved canonical status as a critique, and is considered by some to be the definitive refutation of evolutionary and explanatory theories of religion, understanding the particulars of the critique is essential. Because most of Evans-Pritchard’s critique is methodological, his objections are relevant to evolutionary theories both past and present.
Methodological Critique of Evolutionary Theories
Evans-Pritchard argues that the primary problem with evolutionary theories arises from the fact that they are developmental and sequential. Because all developmental sequences have a beginning, evolutionary theories are always about origins. While there is nothing wrong with seeking origins, all such searches face a significant obstacle: no one was there, in the beginning or at the origin. Because evolutionary theorists lack original data, their inquiries are necessarily conjectural. The procedure goes like this: “religion” is reduced to something essential, such a set of ideas, suite of emotions, or panoply of practices. If ideas are deemed essential, the theorist asks: If I had been present at the beginning, how would I have developed this concept? If emotions are deemed essential, the theorist asks: If I had been present at the beginning, how would I have responded to this feeling? If practices are deemed essential, the theorist asks: If I had been present at the beginning, why would I have engaged in that ritual?
While there are many plausible answers to these “if I were a horse” kinds of questions, they are no more than informed or educated guesses. Things may or may not have happened that way. No one was there and no one knows. The answers, while often plausible and sometimes ingenious, amount to “just-so storytelling.” This storytelling, though it purports to be science, does not ask questions that can be tested or give answers that can be falsified.
These are potent methodological objections. It is ironic that it fell to Evans-Pritchard, who was a Christian, to point out the methodological shortcomings of theories that were being presented as science. While Evans-Pritchard’s faith surely motivated him to find flaws in reductive or explanatory theories of religion, this in no way detracts from his methodological and science-based critique. His own bias, which was against any theory that would explain religion away as error, illusion, mistake, or epiphenomena, was no greater than the bias of the theorists he opposed. On this point, Evans-Pritchard observed:
[T]he persons whose writings have been most influential have been at the time they wrote agnostics or atheists. Primitive religion was with regard to its validity no different from any other religious faith, an illusion. We should, I think, realize what was the intention of many of these scholars if we are to understand their theoretical constructions. They sought, and found, in primitive religions a weapon which could, they thought, be used with deadly effect against Christianity. If primitive religion could be explained away as an intellectual aberration, as a mirage induced by emotional stress, or by its social function, it was implied that the higher religions could be discredited and disposed of the same way.
Religious belief was to these anthropologists absurd, and it is so to most anthropologists of yesterday and today. But some explanation of the absurdity seemed to be required, and it was offered in psychological or sociological terms. It was the intention of writers on primitive religion to explain it by its origins, so the explanations would obviously account for the essential features of all and every religion, including the higher ones. Either explicitly or implicitly, explanation of the religion of primitives was made out to hold for the origins of all that was called “early” religion and hence of the faith of Israel and, by implication, that of Christianity which arose from it.
[These theorists] may to some extent have ignored the higher religions to avoid controversy and embarrassment in the somewhat delicate circumstances then obtaining, but it was chiefly because they wanted to discover the origin of religion, the essence of it, and they thought that this would be found in very primitive societies. However some of them may have protested that by “origin” they did not mean earliest in time but simplest in structure, the implicit assumption in their arguments was that what was simplest in structure must have been that from which more developed forms evolved. This ambiguity in the concept of origin has caused much confusion in anthropology.
It was precisely because so many anthropological writers did take up a theological position [i.e., atheism], albeit a negative and implicit one, that they felt that an explanation of primitive religious phenomena in causal terms was required, going, it seems to me, beyond the legitimate bounds of the subject. (Theories, pp. 14-17).
There can be little doubt that Evans-Pritchard’s complaints about bias are correct. Most evolutionary scholars of religion, past and present, assume at the outset that the primary stuff of religion – spirits, souls, powers, and gods – are not real. Within the empirical purview of science, there is nothing wrong with this assumption. But this assumption can often lead, as Evans-Pritchard correctly observes, to explanations that do not adhere to scientific standards.
This does not mean, as is often assumed, that Evans-Pritchard rejects these theories out of hand or dismisses them as futile. To the contrary, he takes them quite seriously and considers some to be better than others. As his critique proceeds, one gets the distinct impression that Evans-Pritchard has mastered the details and nuances of each theory he discusses. The range of scholars he covers is impressive and were I to provide the list here, it would be overly long and probably arcane. Suffice it to say that Evans-Pritchard displays a range of erudition, and a gift for compression, that is truly rare among scholars today. Also rare is the fact that Evans-Pritchard had taken the considerable amount of time, surely measured in years, that was required to master these materials in their original forms. He did not read these scholars in summary or through the interpretive lenses of others. On this issue, Evans-Pritchard felt it important to state: “I shall embark on a general review of anthropological theories of religion. Permit me to say that I have read the books I shall criticize, for one finds only too often that students accept what others have written about what an author wrote instead of reading the author himself” (17).
Critique of Psychological Theories
Evans-Pritchard begins his critique with “psychological” theories, which come in two forms: “intellectualist” and “emotionalist.” Whatever their form and however presented, all these theories suffer from the same methodological problems: they are about origins, there is no data, the questions cannot be tested, and the conclusions cannot be falsified. These familiar objections aside, Evans-Pritchard takes psychological theorists to task for focusing on thinking and feeling individuals. While individuals do indeed have thoughts and emotions, these are never sui generis and do not occur in pristine psychological space: they are socially conveyed and culturally patterned.
None of this means, and Evans-Pritchard does not suggest, that either intellectualist or emotionalist theories are without value. It simply means these theories cannot account for the origin of religion or the subset of ideas and emotions that are supposedly essential to religion. With this limitation in mind, psychological theories should be seen as aids to understanding and explication. They can, when done well, shed light on already existing ideas, emotions, and practices deemed to be “religious.”
The other major problem with psychological theories comes, ironically, from evolutionary theory itself. All psychological theorists assume that the various ideas and emotions essential to religion result from some combination of faulty perception, mistaken reasoning, and erroneous attribution. “Religious” ideas and emotions may be intelligible or explicable, from a scientific and rationalist perspective, but they are ultimately wrong. If this is so, Evans-Pritchard repeatedly asks, then how could these errors have persisted over evolutionary time? The sine qua non of evolutionary theory is of course that natural selection will eventually eliminate traits, whether they be ideas or emotions, that are disadvantageous to survival and fitness. Like Durkheim, Evans-Pritchard finds it significant that those particular ideas and emotions supposedly essential to religion have persisted, among all peoples, up to the present. Moreover, these ideas and emotions are not unique to religion: they cut across all domains of human thought and are experienced in all aspects of life.
Critique of Sociological Theories
The various shortcomings of psychological theories led, in due course, to sociological theories that were, at least on their face, not evolutionary. Evans-Pritchard recounts these in considerable detail, with a major focus on Durkheim. This focus is almost entirely negative. While Evans-Pritchard agrees with Durkheim’s contention that society or culture determines which ideas are formulated, how emotions are experienced, and why actions are taken, the agreement ends there. Had Durkheim been content with these insights and used them to explicate, rather than explain, he would have performed a valuable service. But Durkheim was not so content. As a rationalist and atheist, Durkheim felt compelled to explain the origins of Aboriginal religion and, by extension, all religion. By way of this compulsion, Durkheim committed all the methodological sins of previous evolutionary theorists. The cardinal sin, however, was Durkheim’s brazenly humanist and metaphysical conclusion that religion was nothing more than society worshipping itself. Like all previous theorists, Durkheim was explaining religion away in a manner not acceptable to Evans-Pritchard.
Making matters worse, Evans-Pritchard finds that Durkheim has kicked psychological theories out the front door but has brought in a new psychological theory through the rear. In his critique of emotionalist theories, Evans-Pritchard had poignantly observed that these had a strong pragmatic or utilitarian flavor. This was not surprising, given that most of these theories had been formulated during the pragmatic era of John Dewey and William James. Pragmatic philosophy and psychology was instrumental: human experience could be understood, and human action explained, in functional terms. In Durkheim’s case, he understood and explained religion as the product of social psychology. The emotional effervescence generated by group rituals supposedly gave rise to “sacred” experiences and hence religion.
Like all emotionalist or psychological theories, this “just so” story was speculative and conjectural. Instead of imagining himself as a horse, Durkheim imagined he was an Aborigine. He then asked how he would have felt had he participated in intense Aboriginal rituals. Because Durkheim imagined he would have become emotionally aroused, he imagines that everyone would have the same effervescent experience. Having “discovered” this feeling in his mind and christened it as the “sacred,” Durkheim then projected it onto others and into an explanatory theory. Evans-Pritchard rejects this attempt on both methodological and empirical grounds. With respect to the latter, he observes that many rituals are often performed privately and others are rote, involving little or no emotion. All that Durkheim has done is describe social effects; he has not explained original causes. Durkheim’s sociological theory, in the end, is no better than any other evolutionary theory. It suffers from the same methodological defects.
Critique of “Primitive” Ethnographic Data
Evans-Pritchard did not raise only methodological objections to these theories. In fact, his most important contribution to these debates revolves around the critical issue of data. All the theorists he criticizes thought they had data. This data came, of course, from “primitive” ethnography. Evans-Pritchard’s assessment of that data is deeply profound. He rightly recognized that previous evolutionary theories of religion were driven by an Enlightenment, atheist, and rationalist agenda. This agenda, and the accompanying story, required a progressive sequence with a starting point. Primitives thus became pawns placed at the imaginary origin. If those pawns had to be sacrificed in the service of a secular, progressive, scientific agenda – or war against religion, then so be it. And sacrificed they were. Knowing that “primitives” played a key sacrificial role in all evolutionary theories of religion, Evans-Pritchard spent the majority of his career studying “savages” and testing the various theories that relied on them as data.
As sacrificial pawns and supposed exemplars of the evolutionary past, “primitives” had been variously portrayed as simple, child-like, ignorant, superstitious, illogical, irrational, and above all, mystical. Their alleged inability to reason caused them to believe in myths and magic. These were, of course, the characteristics that evolutionary theorists deemed essential to religion. If these “supernatural” precursors could be explained and exposed, then everything that followed – including modern or “civilized” religions – could be similarly explained and exposed.
From the beginning, Evans-Pritchard questioned these characterizations. He poignantly observed that none of these theorists, excepting Lowie and Radin, had ever lived among or done fieldwork with “primitive” people. Despite this significant fact, they presumed to know what natives thought and why natives did what they did. They got inside “primitive” minds by reading reports that ranged in quality from bad to worse. These reports, compiled by travelers, explorers, administrators, and missionaries, typically focused on the exotic and strange: if natives said or did something that struck them as bizarre or absurd, they noted it. The reports almost never said anything about ordinary life or the routine things people spend most of their time doing each day. Using scissors and a “cut and paste” method, evolutionary theorists selectively culled these misleading reports for information that would confirm their a priori ideas about what primitives must be like. The result was a total distortion and complete caricature. Primitives became what these theorists needed them to be in order to expose religion as error, falsehood, and illusion.
All these things were on Evans-Pritchard’s mind when he first went to Africa, in the 1920s, to study the Azande. They were also on his mind for his later studies of the Nuer. In fact, it can be said that the entirety of Evans-Pritchard’s ethnographic output, from articles to books, amounts to a test of ideas and theories previously put forward by evolutionary theorists of religion. When Evans-Pritchard went to the field, his goal was to test the idea that “primitives” were steeped in superstition, magic, and myth because they were deficient in mind, perception, logic, rationality, and reason. It is, therefore, no accident that his ethnographic work deals primarily with magic, witchcraft, sorcery, spells, ritual, cosmology, and ultimately, religion.
The Cairo Articles and Theory
The curious fact about all this is that Evans-Pritchard never says, in his ethnographic work, that this is what he is doing. It was apparent only to those who either knew him or who read a series of theoretical articles that he published in succession during the 1930s. Few were aware of these articles (and few are aware of them today) because Evans-Pritchard puzzlingly published them in an obscure journal, Cairo University’s Bulletin of the Faculty of Arts. Evans-Pritchard was teaching there at the time, so this may explain his decision. Regardless, these three articles constitute the theoretical frame and essential context for understanding his considerable corpus of ethnographic writing. In fact, I would go further and say that his ethnographic work, while having intrinsic merit and independent value, can only be understood in light of the three Cairo articles. Though the articles appeared separately, they are linked and build one upon the other. They could easily have been published as a single, coherent set. There is a sense in which they were, only at a much later date: Evans-Pritchard revisits these same scholars and issues in Theories of Primitive Religion.
In the first Cairo article, “The Intellectualist (English) Interpretation of Magic” (1933), Evans-Pritchard closely examines the work of Tylor and Frazer. While critical of these scholars for the usual evolutionary reasons, he is also deeply impressed with them. He finds, in particular, that Frazer has accurately described the psychology that supports magical thinking. Frazer’s mistake was in thinking that because he had mapped these associationist operations (i.e., the “law of similarity” and “law of contagion”), he had explained the origin of magical thinking. All he had done, Evans-Pritchard argues, is explained how magical thinking is consistent with entirely normal psychological processes. These processes, he observes, are not unique to magic: the same associationist reasoning is at work in all domains of thought. Moreover, complex magical systems are never conjured out of individual cognitive air: these ideas are always socially transmitted and culturally learned. It would indeed be odd, Evans-Pritchard observes, if that which is taught was not consistent with that which is thought. The two are mutually supporting, and just because Frazer has accurately described the thought processes involved, he has not thereby explained how magical ideas originated.
The other mistake made by Frazer was to think that because magical thinking creates ideal rather than real connections between things and events, this will eventually be discovered and lead, by virtue of experience and experiment, to scientific thinking. Given that this clearly has not happened and magico-religious thinking has persisted across time and space, there must be some explanation for it. In Tylor’s work, Evans-Pritchard finds a brilliant explanation for this apparently puzzling aspect of magic. Tylor contends that magical thinking persists because:
(1) magicians use art, skill, cunning, and knowledge to achieve real things that their spells and rituals do not actually achieve, (2) the things that magic allegedly achieves will often or sometimes occur as a matter of course or probability, (3) when magical rites fail, as they often do, the failure is attributed to improper observance of prescriptions or prohibitions that supposedly render the magic efficacious, (4) failure can also be attributed to other and sometimes hostile magical forces that are always at work and which may frustrate or counteract the attempted magic, (5) magical ideas are usually flexible and general enough to cover most possible outcomes; this plasticity enables the magician to claim at least partial success for anything that happens and most people will remember a single success rather than multiple failures; and (6) the weight of historical and cultural authority is always behind magic; this makes it easy for people to accept apparently confirming evidence and reject disconfirming evidence.
While Evans-Pritchard lauds this analysis and thinks it correct, he does not explain why. It seems, however, that what appeals to him is Tylor’s use of intellectualist or psychological insights to explain why socially transmitted beliefs in magic are reasonable, even if they are in fact mistaken. Tylor’s approach, in other words, sheds useful light on magical thinking even if it does not identify or explain the origin of such thinking.
In the second Cairo article, “Levy-Bruhl’s Theory of Primitive Mentality” (1934), Evans-Pritchard examines Levy-Bruhl’s thesis (developed over the course of decades and many books) that there is a fundamental dichotomy between “primitive” and “civilized” modes of thinking. Levy-Bruhl characterizes the former as being “prelogical” and governed by what he calls the “law of participation.” Because people, things, events, and ideas are tangled together in participatory webs of causation, natives typically explain things in terms of ultimate “mystical” causes. This does not mean they do not recognize proximate, empirical causes because they do. It simply means that these proximate causes are not sufficient: there is always an ultimate explanation that westerners would characterize as “supernatural.” While Evans-Pritchard agrees with this assessment, he contends they do this not because they are enmeshed in mysticism, but because these ultimate causes have social significance and relevance.
Evans-Pritchard observes that anyone who has lived among “primitives” is struck, first and foremost, by the relative ordinariness of their lives. They spend the majority of their time doing things, and talking about things, that have nothing to do with spirits, magic, mysticism, powers, rituals, and what westerners categorically set off as the “supernatural.” They are, in other words, normal people leading mostly mundane lives. Most of what they say and do is driven by empirical reality and commonsense experience. Levy-Bruhl has, therefore, radically overstated and overemphasized supposed differences. Moreover, he has not specified what he means by “civilized” modes of thought, so it is impossible to evaluate these supposed differences.
Having said all this, Evans-Pritchard finds much to admire in Levy-Bruhl’s work. He was deeply impressed by the fact that Levy-Bruhl took native thought seriously and read “primitive” records sympathetically. In so doing, Levy-Bruhl was the first scholar to evaluate native worldviews in philosophical terms and treat them as coherent philosophical systems. Though it may not have been his intention, Levy-Bruhl’s analyses of native thought painted a picture of epistemological, ontological, existential, and phenomenological complexity. He sensed that native ways of thinking were categorically different, in the Kantian sense, from other or western ways of thinking. The Kantian manifold, supposedly a universal human attribute that governs perceptions of time and space, seemed to be different among natives.
At times, the complexity of native thought drove Levy-Bruhl to near despair; he worried that native ways of categorizing and conceptualizing were so different from western ways that he could not begin to understand them: “In his more radical moments, Levy-Bruhl maintained that the translation of primitive beliefs into European languages was ‘equivalent to treason,’ because when Europeans use their own concepts to express beliefs, precisely because they are concepts, they distort what they find” (Schmaus 1996:428). Based on his own experiences, Evans-Pritchard agreed. Throughout his career, he emphasized the methodological problems surrounding the “translation” of primitive languages and concepts. These problems were much worse, he argued, than the contentious and unresolved translation problems posed by something as relatively simple as the Bible. With the Bible, at least scholars were dealing with well-known languages and familiar concepts. With native worldviews, this was rarely and sometimes never the case.
In the third Cairo article, “Science and Sentiment: An Exposition and Criticism of the Writings of Pareto” (1936), Evans-Pritchard turned his attention to the critical comparative issue that Levy-Bruhl had neglected: What constitutes “civilized” thinking? To answer this question, Evans-Pritchard mined the obscure, arcane, and sometimes brilliant work of Italian polymath Vilfredo Pareto. Best known for his work in economics, Pareto was also a sociologist and philosopher. In 1916, he published a two-volume work, Trattato di Sociologica Generale, which appeared in English translation in 1935 as The Mind and Society. As many know, Pareto was obsessed with optimality in economic transactions; the cultural corollary of this is that he was also obsessed with rationality and what he called the “logico-experimental” method. When he analyzed western or “civilized” societies, he found that logic and rationality were in rather short supply. In fact, few western people interrogated their ideas or actions using scientific methods; rather, the majority simply held irrational beliefs and rationalized about them. While this reasoning was superficially rational, it did not accord with scientific ways of thinking. In Pareto’s analysis of irrationality, Evans-Pritchard finds an interesting and informative contrast:
[This paper] aims at showing what in Pareto’s work is directly relevant to the methods and observations of anthropology. It is further the chronicle of an attempt by Pareto to apply to documents about civilized peoples the same comparative analysis as was applied to documents about savages in the great classics of social anthropology, [Tylor’s] Primitive Culture, [Frazer’s] The Golden Bough, [Levy-Bruhl’s] Primitive Mentality, etc. When we realise that Pareto reached the same conclusions about “civilized” behaviour as Levy-Bruhl reached about “savage” behaviour it will readily be granted that his writings are of concern to anthropologists and that if the rigid division of social studies into those that deal with civilized peoples and those that deal with primitive peoples is to be maintained it can only be as a temporary convenience.
Just as Levy-Bruhl leaves us with the impression of savages who are continuously engaged in ritual and under the dominance of mystical beliefs, so Pareto gives us a picture of Europeans at all periods of their history at the mercy of sentiments expressed in a vast variety of absurd notions and actions….Pareto’s work is an amusing commentary on Levy-Bruhl’s books. Levy-Bruhl has written several volumes to prove that savages are prelogical in contrast to Europeans who are logical. Pareto has written several volumes to prove that Europeans are non-logical. It would therefore seem that no one is mainly controlled by reason anywhere or at any epoch.
Indeed one of the reasons why I have chosen to analyse Pareto’s treatise is to bring out the fact that a study of unscientific thought and ritual behaviour cannot be restricted to primitive societies but must be extended to civilized societies also. Pareto allows to common-sense notions and empirical behaviour about as much place in Greek, Roman, and modern European communities as Levy-Bruhl allows them in central African, Chinese, and North American Indian communities….If Pareto’s analysis is correct it would seem disadvantageous to maintain studies of primitive societies and of civilized societies as separate disciplines as is the present scientific policy.
In science the validity of premises and the logical co-ordination of propositions are everything and the scientist aims always and above all to test his thought by observation and experiment and to avoid contradiction between his propositions. Outside the field of science a man does not trouble himself whether thought is based on observation and experiment and is not seriously inconvenienced by contradictions between his propositions. He aims always and above all to ensure that his notions and conduct shall be in accord with sentiments and if he can achieve that end their scientific value, and to some extent their logical value, are of little importance. Sentiments are superior to observation and experiment and dictate to them everywhere save in the laboratories of science. It is only in the technological field that science has gained ground from sentiment in modern societies.
The “sentiments” of which Pareto spoke are of interest to Evan-Pritchard because these are deeply held beliefs carrying emotional valence. Sentiments are, in other words, socially transmitted and culturally constructed worldviews. Such worldviews or patterns of thought are rarely, if ever, subjected to scientific scrutiny by those who hold them. This is just as true of “primitives” as it is of “moderns.”
These were the things on Evans-Pritchard’s mind when he went to Africa for extended studies of the “primitive” Azande and Nuer. He was testing evolutionary and psychological theories of magical and religious thinking. All his ethnographic writing revolves around these issues, including his most famous and brilliant book:
Witchcraft, Oracles and Magic among the Azande is in part an attempt to show that the thought of non-Western, “primitive” peoples is in many ways as logical and coherent as ours. Yet it is also intended implicitly to show that all popular, everyday human thought is profoundly inconsistent and imbued with non-logical sentiments. In short, Evans-Pritchard’s aim is both to show “primitive” thought to be more logical than earlier writers had written that it is, but also to show that all popular systems of thought, including our own Western thinking, work for people because they are inconsistent, compartmentalized, and imbued with circular thinking. Ordinary Western thought is often far less logical than we credit it being. Evans-Pritchard’s work is a study of the sociology of knowledge.
Evans-Pritchard made such understanding of his great 1937 masterpiece clear to those willing to trace back the intellectual influences upon him by examining his previous publications. No one who knows Evans-Pritchard’s work well would expect him to have explicitly spelled out the analytical and theoretical influences on his works within the actual expositions of his famous ethnographic monographs. His works on the Azande, Nuer, Anuak, Sanusi, and others almost never contain such references and credits to theoretical writers (Beidelman 1974). Yet careful reading of his other writings and of the writings of those whom he encouraged to translate and edit earlier anthropological and sociological classics, indicates what his scholarly inspirations were.
In the case of Lévy-Bruhl, Evans-Pritchard rejected the term “prelogical” as misleading. He also rejected Lévy-Bruhl’ s notion of “mystical participations.” What he did admire was Lévy-Bruhl’s deep empathy with non-Western thought and his capacity to see important similarities between all types of human thinking. Most important of all, Evans-Pritchard thought that much of Lévy Bruhl’s muddled terminology was actually the result of his unsuccessfully groping for a concept of social categories of thought.
What really distinguished Western from non-Western thought were the varying cultural categories by which things and experiences were organized. Differences in how much systems categorized experience accounted for the profound differences in how people of these different cultures judged what was logical or rational. The most important conclusion which Evans-Pritchard drew from all these writers was that human thinking is socially determined, that society, not individual psychology, is the primary factor accounting for differences in human thought. (Beidelman 1997:583-84).
It is truly remarkable that all these writers, or theorists, stayed on Evans-Pritchard’s mind and ethnographic radar for so many decades. The Cairo articles, and the Azande ethnography testing the ideas set forth in them, were published in the 1930s. These same writers and ideas are of course the subject of Theories of Primitive Religion. It is interesting to note that Evans-Pritchard converted to Catholicism in the 1940s; thus, his interest in rehabilitating “primitives” and making them “rational” may have served the interests of his faith. There are certainly hints in Evans-Pritchard’s later work that he found “primitive religion” to be rational because he believed that his own faith was rational. Consistent with Catholic doctrine, he may have seen all spiritualism and supernaturalism as springing from the same godhead or transcendent source. This is not, however, ever more than a suggestion or hint: his critique is grounded in scientific method, theory, and evidence. As such, it must be taken seriously, regardless of his motivation.
Evans-Pritchard’s Legacy and Continuing Relevance
At this point it is appropriate to step back and consider Evans-Pritchard’s legacy and relevance, not just for “primitive” theories of religion but for all evolutionary theories of religion. The first lesson we can take from him is that “essentializing” religion is a fraught procedure that often results in “just so” storytelling. This insight really highlights the importance of definitions, the choice of which is often determined by a scholar’s underlying purpose. If that purpose is to explain religion away, a narrow or minimalist definition will be chosen. Having identified some aspect of religion deemed essential to it, the theorist will seek to explain the origin of this essential characteristic. This procedure relies, implicitly, on an organismic analogy: the thing under consideration is deemed to start simple and then develop into something more complex over time. Having allegedly explained the origin of the simple thing, and presumably having exposed it as an error, mistake, or byproduct, they then assert that everything which comes later or builds on the simple thing is infected with the same errors and mistakes. The foundation, in other words, is bad. Scholars whose purpose is positive or which proceeds from a position of faith may use the same procedure to argue that the essential, original thing is beneficial and its development over evolutionary time should be seen as progressive and adaptive.
There is nothing, at least in principle, wrong with essentializing. It is, after all, the way that science proceeds: by way of reduction. Science isolates something simpler in a larger whole, evaluates it, and then provides a lower level explanation that sheds light on the higher level. This procedure is fine for individual organisms or other discrete phenomena, but when it comes to something as complex, variable, and amorphous as “religion” it becomes a difficult task. Evans-Pritchard thought that while many plausible stories had been produced about the origin and evolution of “religion,” there was no way to test these and, more importantly, no way to decide if one was better than another. He was thus inclined to reject all of them, probably because of his faith.
My inclination is to accept most of them, but once we do this the question arises: What have we really explained? For instance, if we combine all the various mental faculties that have been proposed for cognitive theories of religion, the combination is a near perfect description of a normal human mind carrying out ordinary operations. There is nothing inherently “religious” about these operations. The same operations are used for everything else that humans think and do. This explains, at least in part, why Evans-Pritchard rejects psychological theories: they do not take us very far. At their best, they can explain why socially transmitted and culturally patterned magico-religious thinking persists and is maintained. At their worst, they are simply speculation and conjecture. They do not, however, explain the origins of such complex thought patterns.
Evans-Pritchard’s attack on essentializing, combined with his determination to rehabilitate “primitives” and remove them from the putative origin of a progressive evolutionary sequence, leads directly to another insight. We really need to be asking: What actually evolved? In the past, this question has been posed in terms of “religion” and its various components, whether those be cognitive (ideas) or behavioral (actions). The problem with this is that “religion” is a very western idea with a western history. Thus, breaking it down into constituent elements and then searching for evidence of those in the past is a projection of our ideas about what constitutes “religion” into the past. Early evolutionary theories accomplished this by relying on ethnographic data culled from so-called “primitives.” Evans-Pritchard showed how this data was seriously distorted if not altogether mistaken. He argued, successfully in my estimation, that “primitives” were, at least in terms of psychology or cognition, just like us. In the long process of making this argument, Evans-Pritchard pointed the way toward a deeper understanding of what actually evolved.
While “primitives” surely are not proxies for the evolutionary past, it seems reasonable to suppose that historic hunter-gatherers had worldviews that were less derived than worldviews characteristic of agricultural and industrial or “modern” societies. If we assume these worldviews are more representative of the ancestral past, then any reliance on them as data requires a deep understanding of what those worldviews are. They are no doubt distinctive. Evans-Pritchard, taking his cue from Levy-Bruhl, understood this. This understanding, in turn, allows us to re-frame the evolutionary question and may lead to a revivified evolutionary theory. Though this surely was not Evans-Pritchard’s intent, the ironic result of his rehabilitation allows us to focus on what actually evolved: animist worldviews.
Modern evolutionary theorists of religion are not much concerned with these worldviews. They simply assume, as early evolutionary theorists did, that animist worldviews are a lot like the religions of today. They are assumed, in other words, to revolve around invisible agents, unseen powers, mysterious forces, and the “supernatural.” It is of course no longer polite, or politically acceptable, to refer to these ideas as “primitive” and superstitious or irrational. Perhaps sensing this, modern scholars do not typically rely on “primitive” data or “tribal” ethnography. The new data – or proxy for the prehistoric past – typically comes from psychological experiments on university undergraduates and children. These are the new “primitives” for evolutionary theories of religion. Setting aside, for the moment, some serious questions about the quality of this data, either ignoring animist worldviews or failing to comprehend them seems to me a serious mistake. In fact, I argue that animist worldviews constitute our best data set for evaluating what actually evolved and how this relates to what is called, improperly, the “evolution of religion.”