Weber's 'Sociology of Religion' part 2: magicians, prophets and priests

The main reason I wanted to read this book (apart from wanting the opportunity to bitch about it on the internet) is Weber's description of a framework for categorising different kinds of religious specialist.

Systems of etic categorisation are always going to be a bit problematic in the social sciences. It doesn't take a genius to see that even in the best taxonomies there are going to be intermediate cases that don't quite fit into any category. This means that some people are inevitably going to start questioning the point of categorising the thing at all.

Every now and again you get arguments to the effect that human social constructs are inherently unclassifiable. This position generally seems to be adopted after someone has the startling revelation that People Are Complicated (this is a scientific fact). Maybe this is just me being anal-retentive and Teacherite at things, but I say suck it up, social sciences. Biologists seem able to cope with the level of complexity that comes with bog-standard taxonomy (God help them if the issue in question involves asexual species or anything at all to do with conservation). As long as people make an effort to flag up the fact that there are going to be ambiguous cases, a flawed typology is often going to be more useful than no typology at all, at least when it comes to the initial job of getting our heads around the problem.

Notice that I said 'often'.

I'm in two minds about Weber's framework. His main dichotomy is between priests and magicians, and as such is basically similar to the priest vs shaman distinction which you see in a bunch of other people's work*. There's also the 'prophet' category, which doesn't seem to have received much attention in anthropology (maybe because they seem to be world-changing figures who only come along every once in a while, and ethnography tends to be better suited to studying more routine aspects of religion).

This is Weber's framework, as I understand it:

Priests: people who are permanently employed by a large stable organisation which they are expected to support. They have professional qualifications which are earned by studying to become an expert on specialist knowledge (e.g. formal theology). They are also supposed to be experts on "the practical problems involved in the cure of souls" (p.30), which means general pastoral care and also whatever rituals are deemed necessary for salvation (e.g. confession, baptism, etc). In addition, priests worship gods (i.e. they adopt a relatively humble attitude towards supernatural beings which are seen as being worthy of respect). Oh, Weber also labels their training as "rational" (p.29)- as I said in last post, I have no idea what we're meant to assume that word means in this context.


Magicians: are generally "self-employed" (p.29), and tend to act alone in rituals which are not held according to any fixed schedule. Their qualifications are based entirely on personal charisma, which is made manifest through the ability to perform various types of magic. They're emically deemed to gain this charisma via a ritual rebirth plus training in "purely empirical lore" (p.29). Weber doesn't mention if he thinks they do pastoral/salvation work. They focus not on worshipping gods, but on coercing and charming demons. And yes, Weber has decided that the way they operate is in some sense "irrational" (p.29).


Prophets: a person who makes it their personal mission to spearhead a campaign for religious change (whether this is subsequently viewed as an attempt to start a new religion, or just to revitalise an old one). They may focus on preaching and giving advice on how people can live better lives (in which case Weber calls them 'ethical prophets'), or on just making themselves a paragon who others can try to emulate (in which case he calls them 'exemplary prophets'). They generally don't have steady jobs or live in a stable household; they may travel around a lot, beg or do a series of temporary jobs.


'Magicians' seem to be the least exclusive category in Weber's eyes. Priests may do magic (p.28)- e.g., given what Weber has to say about magicians and demon-coercion, I'd guess that Catholic exorcisms would count as 'magic' under this system. Successful prophets almost always do magic, or at least have magic attributed to them: they have so much charisma that people would tend to present them as BEST MAGICIAN EVER even if they tried to deny the title. I suspect Weber felt that Christ's resurrection may fall into this category (p.78).

I'm sure that there are going to be at least some things that don't fit into any of these categories, and the specific words 'priest' and 'magician' are probably unhelpful here (especially since, as is often the case with these issues, we start running into serious theoretical difficulties when we try to work the modern-day Pagan movement into the model). The question is, does this describe any sort of general tendency? For example, do people who have fulltime jobs as religious specialists have a greater tendency to present their supernatural patrons as uncoercable?

It would be interesting to see if anyone's tried to systematically test that idea- sadly, since Weber isn't phrasing his ideas in terms of testable hypotheses, it sounds like it would be quite a lot of work.



*For those who care, I've so far seen a binary priest/shaman typology come up in: W.A. Lessa and E.Z. Zogt's Reader in Comparative Religion (1958)

Victor Turner's essay 'Religious Specialists' (originally 1972, but republished in Magic, Witchcraft and Religion: an anthropological study of the supernatural: fourth edition, 1997)

Morton Klass' Ordered Universes (1995)

A somewhat similar typology is proposed in Fredrik Barth's 'The Guru and the Conjurer' (1990, Man, 25, 4, pp.640-653). This article went on to inspire Harvey Whitehouse to write extensively on the same topic; I think his most persuasive article on the issue is still 'Memorable Religion' (1992, Man, 27, 3, pp.777-797), but if you don't have access to that journal then Arguments and Icons is probably the next best bet.