"Dr. Margaret Murray was a nice, respectable Egyptologist until the day she got the notion that the medieval witch-trial records should be taken literally.  She wrote three books of progressively less mainstream scholarship: The Witch-Cult in Western Europe, The God of the Witches, and The Divine King in England, in which she links more of the English nobility to Wicca than any writer until Katharine Kurtz - and swears that Wicca is a religion stretching in unbroken practice back to the caves at Lascaux" (Edghill 1995:110).

The Witch-Cult in Western Europe is basically a description of Murray's conception of the witch religion, its assemblies, rites, organization, etc.  It contains many long quotes from accused witches' testimony, which make it less than easy reading.  For example: "the Devill thy maister, quhome thow seruis, and quha techis the all this vytchcraft and sorcerie, apperit to the, in the licknes of ane horss" (Murray 1962 [1921]:207).  That is, "The Devil thy master, whom thou servest, and who teaches thee all this witchcraft and sorcery, appeared to thee in the likeness of a horse."  And that's just an English example.  There are also a good number of French excerpts, and some German. The God of the Witches is a more accessible read, written for a popular audience, covering the above material and expanding on it.  For those reasons, this is the book I'll address in this critique.

A word before we begin.  A number of Murray's points rest on anthropological evidence, and in the introduction she expressly says that the book is intended for students of anthropology.  I myself happen to be an anthropologically-trained archaeologist, and I notice several flaws in her methodology and data.  Besides that, I have to say that Murray's work shares some of the characteristics of the writings of fringe anthropologists, making claims that aren't really supported.  This is one of my pet peeves, but I'll try not to be too harsh.
 

Back in the nineteenth century, when anthropology was young, a popular way to work in the discipline was to use the comparative method, so named because it involved comparing customs, beliefs, etc. from different cultures and noting the similarities.  See how alike we all are!  The best-known example of this is probably James Frazer's monumental work The Golden Bough.  Unfortunately, this method has some serious flaws.  Its users often assumed that similar practices sprang from the same source, either a common historical source or the similarity of the human mind everywhere, ignoring the historical development of the individual cultures and the natives' own explanations.

At the turn of the century, Franz Boas, now known as the father of American anthropology, criticized the use of the comparative method, noting

We cannot say that the occurrence of the same phenomenon is always due to the same causes, and that thus it is proved that the human mind obeys the same laws everywhere.  We must demand that the causes from which it developed be investigated and that comparisons be restricted to those phenomena which have been proved to be effects of the same causes (1988 [1896]:88-89).

Murray, however, appears to be an adherent of the comparative method.  When discussing horned gods, masks, and dances, she cites examples from Egypt around the Near East and across Europe, encompassing at least two very different culture areas.  She even links the art of the European Paleolithic to the art of the Bronze Age Near East and India.  This approach just doesn't cut it anymore.  I'd be extremely leery of using these data as evidence for a widespread religion.

An additional problem with the presentation of the material is that Murray tends to throw out statements with no backup, and the reader is supposed to take her word for it.  This might work if the statements were common knowledge, at least among a certain set, but, at least to my knowledge, they're not (though I suppose they might have been at the time), and the book's proposals are novel enough to require solid support.  Also, the works listed in the bibliography largely consist of books on witchcraft, with some additional topics such as Joan of Arc and superstitions.  The closest I can find to anything anthropological is a few titles on folklore, a surprising omission for a book whose jacket calls it "a classic work of anthropology."

Here's an example of the kind of thing I'm talking about.  Speaking of the "sorcerer" painting from the Caverne des Trois Frères, the book states,

The period when the figure was painted is so remote that it is not possible to make any conjectures as to its meaning except by the analogy of historical and modern instances.  [I agree so far; this is standard practice, though it can't be taken too far or made too precise.]  Such instances are, however, sufficiently numerous to render it fairly certain that the man represents the incarnate god, who, by performing the sacred dance, causes the increase of the kind of animal in the disguise of which he appears (Murray 1970 [1931]:24).

Along these lines, there are also several instances where the book's rhetoric makes the veracity of certain statements appear irrefutable, when in fact it's not.  For instance, "It is certain that there was some kind of ceremony, religious or magical, in which a horned man, presumably a god, took the leading part.  It is equally certain that there must have been a worship of the female principle" (Murray 1970 [1931]:15, my emphasis).  This is talking about the Paleolithic.  We can't be so certain about anything having to do with a phenomenon that was largely taking place in people's minds, i.e. religion.  Another example: "The only explanation of the immense numbers of witches who were legally tried and put to death in western Europe is that we are dealing with a religion which was spread over the whole continent" (Murray 1970 [1931]:54).  Evidently not everyone agrees that this is in fact the only explanation, or there wouldn't be so many other explanations (see the suggested reading for a couple).  Even today people are executed for witchcraft in Africa, not for reasons of religious prejudice but because when things go wrong, people tend to look for someone to blame.  Surely this is just as plausible a model to apply to Europe.

In addition to unsupported statements, our author makes occasional factual errors.  Some of these I'll deal with later, in context, but one I'd like to mention now deals with religion in general:

The idea of dividing the Power Beyond into two, one good and one evil, belongs to an advanced and sophisticated religion.  In the more primitive cults the deity is in himself the author of all, whether good or bad.  The monotheism of early religions is very marked (Murray 1970 [1931]:14).

Before I move on to more specific issues, there's one last thing I have to mention, though it doesn't really have any bearing on the arguments (Murray's or mine), and that is the use, albeit rare, of pejorative adjectives to describe other cultures.  "Dances...survive in recognizable form only among the more backward peoples.  In Europe the details have not always been preserved, and it is often solely by comparison with the dances of savages that their original meaning can be seen" (Murray 1970 [1931]:106, emphasis mine).  Do I have to point out that this is not an acceptable attitude for an anthropologist, even one in the early part of the twentieth century?  Another of the major contributions made to anthropology by our friend Boas (who had lived all but eleven years of his life and done the greater part of his work before The God of the Witches was published - that is, his ideas were out there at the time of its publication) was the concept of cultural relativism, which means, not that anything goes because "it's all relative," but that other cultures should be looked at on their own terms, not judged by the standards of our own.  No culture is intrinsically "better" than another.

Now on to specifics.

Murray interprets certain Paleolithic cave paintings as evidence for a horned god worshiped during that period.  One is the commonly called "sorcerer" from the Caverne des Trois Frères (left), the other, another mysterious figure from a jumble of images below it (right).

The sorcerer

Her interpretation is that these are people wearing animal skins and antlers/horns in order to impersonate the god.  There's nothing invalid about this interpretation in itself.  However, there are other, more recent interpretations concerning rock art that deserve some mention.  Studies of rock art in places like Africa where people still remembered the meaning behind it, along with studies of shamanism, have led some to conclude that Paleolithic cave art is a record of shamanistic experiences.  In the earliest stage of shamanistic trance, the practitioner sees geometric shapes, which would explain the heretofore unexplained occurrence of such shapes painted in the caves.  In a deeper stage of trance the visions may include humans partially transformed into animals.  These types of imagery apply cross-culturally, because of the way the human nervous system functions in trance, although of course the exact images will differ from place to place according to culture.  For example, a San shaman in Africa might see an eland where a European shaman might see a stag.

The pictures above fit neatly into this explanation.  I also find it worthy of note that shamanistic paintings sometimes include lines of power running from a creature's nose.  Could this be the case in the right-hand picture?  Murray, among others, interprets those lines as a bow that the buffalo-man is holding, but, although it may be hard to tell from this picture, he's not actually supporting it with his hand (well, his hoof), and it definitely is connected to his nose.  Paul Bahn suggests the lines may represent nosebleeds frequently experienced by shamans entering trance, "certainly a less fanciful explanation...than the traditional one of playing a 'musical bow'!" (1988:158).  I find the shamanistic parallels intriguing.

I also have to mention regarding the bow interpretation that it's quite possible that the bow hadn't been invented yet at the time the picture was drawn.  Both of the figures above probably date to the Magdalenian period, and archaeological evidence for the bow in Europe doesn't appear until after the Magdalenian (Foster 1994:679).  Before that time the atlatl (spear-thrower) would have been in use.

Some time in The God of the Witches is spent discussing the relationship between witches and fairies.  The latter supposedly were the still-present but hidden descendants of the Neolithic occupants of the land, a pastoral people who naturally practiced the old religion and whose diet of milk and occasional meat accounted for their short stature:

The immense difference in physique caused by the introduction of grain into the regular diet of mankind is hardly yet realized except by the few who have studied the subject.  It is not improbable that the small stature of the fairy, the stunted size of the changelings, the starved condition of the 'mortal' captive among the fairies, may have been due to the diet (Murray 1970 [1931]:51).

I'd like to mention as an aside that, while it's not implausible that ideas of the fairies were based at least in part on memories of the earlier inhabitants of the British Isles, this is not the only possible or logical explanation.  Many cultures have little - or not so little - people.  The Hawaiians, for instance, had stories about the menehunes, and there were no previous inhabitants of Hawaii to base them on.

The fairy "hills" presented in this book are the semisubterranean turf houses built by these people, resembling the remains of Bronze Age houses found in Britain.  However, when describing these houses, along with the "elf-bolt" arrowheads found in association with them, Murray only mentions them dating to the Bronze Age, and that more than once.  One would think that if people truly still were living in this manner throughout the Middle Ages, there would be physical evidence to support the statement.  To my knowledge, there is no archaeological evidence for this concept of the fairies.  (I'm not sure there are even any folkloric indications toward this idea.  Certainly Turlough O'Carolan sat on an actual hill to compose "Si Beag, Si Mor" (about the fairies).)

By the way, the elf-bolts are linked back to the Paleolithic buffalo-man figure by means of their size, "so small and slight that they could have been used only with a small and light bow, such as that carried by the masked dancer" (Murray 1970 [1931]:61).  I have to question this statement too.  Late Woodland points of eastern North America are not particularly large (see examples here and here), but that didn't stop their makers from using long bows, as can be seen in John White's sixteenth-century depictions of Algonquians (on the right of this page).  This isn't a major point, but it does show how Murray tends to fit everything into her schematic, even when problems or other potential explanations should be obvious.

It's not just the use of facts that don't fit but also some of the logic in this section that I find puzzling.  In a passage about a sixteenth-century woodcut showing a man consulting with fairies, in which the fairies' abode might possibly be construed as a turf hut (although to me it looks like the fairies are looking out from an opening under a hill), we find the following statement:

This then is clear evidence of the belief in elves and fairies at the date of the picture, i.e. 1555, and is proof not only of the human nature of the fairies and of their close resemblance to the Neolithic people but also of the survival of the Neolithic and Bronze Age folk and their civilization as late as the sixteenth century (Murray 1970 [1931]:52, my emphasis).

One last thing about the fairies, and that is that the portrayal of their society in The God of the Witches is inconsistent with the working of the kind of society it resembles.  I get the impression of a society made up of small independent groups (what anthropologists classify as a tribal social structure), but such a society wouldn't have a queen and king; it would lack that hierarchy of unearned rank.  The royal couple could of course be explained away as the misunderstanding of the position of a tribal chief, but what about the "fairy ladies of rank" (Murray 1970 [1931]:60)?  And why do the fairy knights wear gold and silver armor?  These people eking out a marginal existence mined and made armor out of such militarily useless metals?  All of this sounds more like the projection of European feudalism and of romanticism onto the supernatural.

So much for the fairies, who, unless new evidence appears, appear very unlikely to have existed in the form presented in this book.
 

"Historical ethnography: it's a killer...a killer."  - my documentary archaeology professor (who will remain nameless for privacy)

So, supposedly the witches who were tried and/or executed were those who had kept this faith alive, passed down through their families, and the real reason for their persecution was the fact of their belonging to the religion.  I have to ask, then, why there is no mention in the official records that this was the reason behind the trials, if indeed it was.  Surely the judges were not unaware of the actual crime committed?  And besides, why should the mere fact of belonging to another religion have been such a great crime?  This is not to say that non-Christians didn't experience persecutions of other sorts (hate crimes, being driven out of countries, being restricted to certain occupations), but just belonging to another religion wasn't generally a crime (with some exceptions, such as when the Jews had to either leave Spain or convert).  Someone who had started out a Christian could certainly get in trouble for heresy or apostasy, but one would think that people born into witch families should have been all right, if this were truly a recognized religion.  Of course, it being pagan instead of monotheistic might have made a difference, but that still doesn't explain the problem with the historical records.

Murray's treatment of the witch-cult in antiquity falls prey to the use of the comparative method again, as well as to her tendency to associate things with the cult on the slimmest of evidence.  For instance, she interprets a Paleolithic painting in Spain as a ritual dance, with the central male figure wearing the garters that she believes are so important to the witches.  The scene could be a dance, but there's nothing to say that those are garters except Murray's belief that this is a religious scene and that garters are important in this religion.  The "garters" could just as easily be the folded-over tops of high boots, or something else entirely.  Even dealing with periods less obscured by the mists of time, we run into problems.  One gets the impression that Norse religion was the same as Gaulish religion and so on; basically, that people everywhere were practicing the same religion.  Romulus had twelve lictors and the Danish hero Hrolf had twelve berserkers, therefore, they must both have belonged to covens.  Never mind that they were separated by centuries and lived in different cultural spheres.  The comparative method conquers all.

The descriptions of the witches' rites bring to mind the above quotation about historical ethnography.  Ethnography is the study of cultures; historical ethnography is the study of cultures in the past, usually through documents.  It's not easy to form a picture of a culture from historical records.  The warning holds equally well for the witch-cult.

Cotton Mather wrote that the witches "have a Baptism, and a Supper, and Officers among them, abominably resembling those of our Lord" (Murray 1970 [1931]:65).  Indeed, I find it curious that this other religion so closely resembles Christianity.  Why should another religion practice baptism - and why should names like Caterine and Mary, the names of Christian saints, be given to the witches at these baptisms?  One would expect something more along the lines of Alfreda (elf-counsel).

Murray calls the supper, consisting of bread and wine, "[t]he main part of the religious rite...comparable with the Mass" (1970 [1931]:120).  Again, it seems surprisingly close to Christianity.  And what are we to make of the following passages?

The Chief ... took the sacred wine and sprinkled it on the kneeling people, while they cried out in chorus, "His blood be on us and on our children" (Murray 1970 [1931]:121).

She also related, that the said Magician did sprinkle the consecrated wine vpon all the company, at which time euery one cryeth, Sanguis eius super nos & filios nostros (Murray 1962 [1921]:149).

I think it likely that the accused, when asked for details of the witches' rites, based their testimony on what they were familiar with, that is, Christianity.  And naturally witchcraft, being associated with the diabolical, would pervert the rites of the "true" religion.

A lot of the evidence against the accused witches was gathered by means of torture.  I've seen objections that some people confessed voluntarily, so that must mean the witch-cult really did exist, since the confessors weren't being asked leading questions.  It should be remembered, though, that the accused knew that they would be tortured if they didn't talk (and the torture would continue until the authorities heard what they wanted to hear).  Is it surprising that some people opted to avoid this experience?  I've also seen the assertion that the similarities of all the stories prove the truth of the cult's existence.  Could it not simply be that everyone knew what activities went with witches?  If someone today were made to describe an alien abduction, the description would probably resemble other abduction stories, because everyone knows what elements go into such tales.  Just because people in the past didn't have mass media doesn't mean they didn't have a shared body of knowledge.

The connection between Murray and Becket (my reason for addressing her work here in the first place) is the naming of Becket among the divine victims put to death as stand-ins for the god.  The whole idea of the victim comes from The Golden Bough and is pretty iffy.  I'm not going to go into detail on the whole idea of the divine victim or on the particular victims named here.  I'll just mention a couple of things regarding Becket.

Supposedly the king of England was occasionally required to act as the sacrificial substitute for the god, but sometimes he could get the Archbishop of Canterbury to take his place.  This serves as the explanation for the often stormy relationship between king and archbishop, for instance the quarrels between William Rufus and Anselm (Rufus ending up as the victim) and Henry II and Thomas Becket (Becket ending up as the victim).  "The death of Thomas à Becket presents many features which are explicable only by the theory that he also was the substitute for a Divine King" (Murray 1970 [1931]:171).  As with the previous statement regarding the witch religion, I have to say that it should be obvious that this is not the only explanation.  To my knowledge, no reputable historian has ever looked at the story and decided the facts didn't fit the usual straightforward interpretation.  Besides, why would a Christian archbishop let himself be killed for a pagan cause?

What are these features of Becket's death that can only be explained by this theory?  Murray mentions Becket's expectation of approaching death, his resistance to the monks' attempts to save him, the condition of his body after death, the fact of his death being known abroad on the day of it occurrence.  None of these is explicable only by the divine victim idea.   Neither is the following, assertions to the contrary notwithstanding:

That Becket was regarded as a Divine Victim is seen in the comparisons between his death and that of Christ, which are found in all the contemporary biographies, comparisons quite impossible if the death of the archbishop were simply murder...[quoting William of Canterbury] "As the Lord, his passion being imminent, approached the place of suffering, so Thomas, aware of coming events, drew near to the place in which he should suffer.  They sought to seize, as Jesus, so Thomas, but no one put a hand on him because his hour was not yet come.  The Lord went in triumphal procession before His passion, Thomas before his.  The Lord suffered after supper, and Thomas suffered after supper..." (Murray 1970 [1931]:172-173).

I'm afraid that throughout her works Murray is seeing too much what she expects to see.  Not all of her ideas are without merit, but they need to be presented in a better-documented, better-reasoned manner before they can be seriously considered.  However, I'm not sure it's possible to do so - I suspect much of her argument would crumble upon such close examination.  As it is, too much rests on ifs - if one accepts this explanation, then that follows; if this really means that.  Unless and until someone expands upon her work in a truly scholarly manner, I would advise taking her theories with a grain of salt, or perhaps with the Dead Sea.
 

Addendum:  Since this section was written, there has been a change in archbishops of Canterbury.  George Carey retired on October 31 - one of the great Sabbats of the year!  And his successor is named Rowan, a wood with magical properties!  And he's been accused of dabbling in paganism!  Is this connected to the witch-cult?  What does it all MEAN?

It just goes to show how people with an agenda could find meaning whether it's really there or not.
 
 

This isn't an exhaustive list of sources that could have a bearing on this discussion; just what came to mind first.  Some of the books have later editions now.

All about Ice Age art.
 

Boas, Franz
    1896  (1988) "The Limitations of the Comparative Method of Anthropology."  In High Points in Anthropology, 2nd
               edition.  Paul Bohannan and Mark Glazer, eds.  Pp 85-93.  New York: McGraw-Hill, Inc.

The title is self-explanatory.
 

Boyer, Paul and Stephen Nissenbaum
    1974  Salem Possessed: The Social Origins of Witchcraft.  Cambridge, MA: Harvard University Press.

Why did witch hysteria break out at that particular place and time?
 

Clottes, Jeanne and David Lewis-Williams
    1996  The Shamans of Prehistory: Trance and Magic in the Painted Caves.  Translated by Sophie Hawkes, 1998.
              New York: Harry N. Abrams, Inc.

The shamanistic explanation for cave art.  Written for the layperson, and with good photographic illustrations (it's a coffee table book).
 

Cohn, Norman
    1975  Europe's Inner Demons.  New York: Basic Books, Inc.

This contains some critique of Murray's hypothesis.
 

Diamond, Jared
    1987  (1994) "The Worst Mistake in the History of the Human Race."  In Applying Cultural Anthropology: An
              Introductory Reader, 2nd edition.  Aaron Podolefsy and Peter Brown, eds.  Pp 105-108.  Mountain View, CA:
              Mayfield Publishing Company.

Why the development of agriculture wasn't really such a great thing.
 

Edghill, Rosemary
    1996  The Bowl of Night.  New York: Tom Doherty Associates.

    1995  Book of Moons.  New York: Tom Doherty Associates.

    1994  Speak Daggers to Her.  New York: Tom Doherty Associates.

Mysteries featuring Bast, a Witch of the modern sort.  I highly recommend them.
 

Feder, Kenneth L.
    1999 Frauds, Myths, and Mysteries: Science and Pseudoscience in Archaeology, 3rd edition.  Mountain View, CA:
              Mayfield Publishing Company.

Most of the book deals with specific frauds and pseudoscientific theories, but the first couple of chapters explain how archaeologists know what they know and how to evaluate claims along those lines.
 

Flannery, Kent V. and Joyce Marcus
    1993  (1996) "Cognitive Archaeology."  In Contemporary Archaeology in Theory, Robert W. Preucel and Ian
               Hodder, eds.  Pp 350-363.  Malden, MA: Blackwell Publishers Ltd.

This article is an overview of cognitive archaeology, which is the study of the mental parts of ancient cultures (cosmology, religion, ideology, iconography).  It's not an easy thing to do when no written records survive.  "...[W]hen we see ancient religion reconstructed from a handful of figurines or the red dado painting on the wall of a shrine, we have a right to be sceptical" (p 361).
 

Foster, Malcolm F.
    1994  "The Origins of Weapon Systems."  In Current Anthropology 35(5):679-681.

A short summary of the dates of various weapons around the world.
 

Harris, Marvin
      1974  Cows, Pigs, Wars, and Witches.  New York: Random House.

Harris sees the witch craze as partly the creation of the authorities, who used the fear of witches to turn the people against each other, keeping them from turning against their social superiors in rebellion or revitalization movements.

I don't agree completely with all Harris's conclusions in this book, and I would feel remiss in my duty to the reader if I failed to offer some alternatives.  His explanation of Jewish dietary laws makes sense for pigs but in my opinion works less well for other animals.  I prefer Mary Douglas's discussion in Purity and Danger (London: Routledge & K. Paul, 1966) , which explains the entire system of kashrut within an anthropological framework.  Also, Harris's discussion of Jesus is noticeably dated in its anachronistic use of the Zealots.  For another view of the topic by people with more expertise in that field,  I recommend Bandits, Prophets, and Messiahs: Popular Movements in the Time of Jesus by Richard A. Horsley and John S. Hanson (Minneapolis: Winston Press, 1985).  For a more specific discussion of Jesus in relation to the Zealots, see John P. Meier's A Marginal Jew, volume 3 (New York: Doubleday, 2001).
 

Leakey, Richard and Roger Lewin
    1992 Origins Reconsidered: In Search of What Makes Us Human.  New York: Doubleday.

The chapter "Windows on Other Worlds" has a few pages on therianthropes (human-animal figures) in shamanism, relating to cave paintings.  The rest of the book, about Homo erectus and the Turkana Boy, is worth reading too.
 

Renfrew, Colin and Ezra B.W. Zubrow, eds.
    1994  The Ancient Mind: Elements of Cognitive Archaeology.  Cambridge: Cambridge University Press.

From the abstract: "One of the most troubling problems in archaeology is to determine about what or in what manner did prehistoric people think."  There are several articles on the archaeology of religion, but the book's geared toward anthropologists and isn't easy reading.
 

Walsh, Roger N.
    1990  The Spirit of Shamanism. New York: G.P. Putnam's Sons.

An examination of shamanism - altered states of consciousness, shamanistic experiences, how shamanic episodes differ from mental illness, etc. - by a medical doctor.  A fair-minded treatment that is neither New Agey nor closed-mindedly skeptical.
 
 

A summarizing essay on Murray's ideas.

The Witch-Cult in Western Europe online.

The God of the Witches online.

The Golden Bough online.

Cave paintings

The cave of Trois Frères