Barth, Fredrik, 1990, 'The Guru and the Conjurer: transactions in knowledge and the shaping of culture in Southeast Asia and Melanesia', Man, 25, 4, pp.640-653
So, cards on the table: I really like this essay, and I agree with a lot of the things Barth says here. But the more I think about this topic, the more I think that the issues Barth brings up here need to be addressed in a much more thorough and rigorous way. This isn't a criticism of Barth per se- I think this essay's meant to be more about asking the first few speculative questions, so while there are areas where he's not as clear as I'd like, I don't think it's an insult to say that these ideas need to be taken up and developed further. One of the people who's done a lot to try to develop these ideas is Harvey Whitehouse. I have some disagreements with Whitehouse's approach, so I'm going to try to express those in a later post, but I think it's worth starting off by going and looking at the person who kickstarted an important part of the debate.
Barth, unlike most social anthropologists, has done fieldwork in several different countries. In this essay, he tries to articulate some of the differences between his fieldwork with the Baktaman of Papua New Guinea vs his experience of Bali and Bhutan. He proposes that some of the differences between these societies comes from the fact that the transmission of religious knowledge is based around very different sets of assumptions. He proposes that the mode of transmission used in Bali and Bhutan could be referred to as the 'Guru' mode, whereas that found among the Baktaman is the 'Conjurer' mode.
The basic difference between the two modes hinges on different strategies of cultural transmission. In the Guru system, people gain prestige by continuously broadcasting religious information (theology, religious history, discourses on ethics, etc) in a verbal form to anyone willing to listen. You see this happening in religions like Islam, Buddhism and Christianity.
By contrast, in the Conjurer system, people gain prestige by having a reputation for knowing many great secrets. This means that, rather than publically broadcast their knowledge, religious experts will only impart their wisdom slowly to a selected few (who are often first required to undergo some sort of taxing initiation ritual in order to prove their worth). This knowledge is often not conveyed in a verbal format, but via ritual, or visual symbols.
These differences have consequences for the way the religion develops. As Gurus compete with each other for prestige, they strive to portray themselves as erudite, learned and logically consistent. This leads to the kind of material that we usually think of when we use the word 'theology': complicated verbal arguments about the relationships between text, or how to apply a particular text to a specific situation. By contrast, a Conjurer may not have spent a great deal of time working out the minutiae of their beliefs about the divine, and if they have then they may not share this with anyone else. There's a good chance that they haven't ever articulated their beliefs in a specifically verbal way, rather than in terms of the associations between specific symbols (like, say, totem animals). This leads to a religious system which doesn't look much like how Western people generally expect it to.
Another upshot of this is that Gurus are much better at invading* populations of Conjurers than vice versa. This is because Gurus have gotten into the mindset whereby they see it as a good thing for them to spread their knowledge to as many people as possible, therefore it's relatively easy to persuade them to start travelling around and infecting* new populations. Also, broadcasting verbal information is a pretty efficient way of reaching a large number of people. By contrast, if a Conjurer for some reason decided that they wanted to convert a new population to their belief system, then any initiation requirements would make it a much slower process with a higher degree of initial commitment required from converts. Also, the Conjurer would probably struggle to convey the largely non-verbal elements of this religion to people who came from a different cultural context. As a result, religions like Islam, Buddhism and Christianity have spread around the world, whereas religions like the Baktaman initiation system have stayed very local.
I think this is a really interesting idea, particulary because if it's true then it has big implications for ethnography. Since Westerners frequently assume that absence of verbal information means absence of information, period, this theory implies that in cultures like the Baktaman ethnographers are frequently going to miss what's going on with the local religious system.
It's also interesting in terms of what it implies for religious history. I'm wondering whether anyone's tried making an AI model of something similar to see if it seems to capture the sort of behaviour you see in real life conversion situations.
One thing I would have liked to see Barth clarify, though: he never specifies what sort of scale he sees most individual Gurus and Conjurers as working on. In his initial examples of Gurus (p.642), he seems to be describing people who focused on preaching to entire communities whenever they got the chance. But elsewhere (e.g. p.643) he describes Gurus as "mystifying" things in order to make his students feel more dependent on him, and to "exclude outsiders from the circle of disciples". The latter strategy doesn't sound like a particularly good way to convert new populations. I'm wondering whether Barth is using the term 'disciples' to refer specifically to the closest students of a guru who also preaches to the masses (as was the case with Jesus), or whether he's describing two strategies which are distinct from a transmission perspective but may be used interchangeably by members of one culture depending on the situation. Either way, I think it would help if he'd clarified what sort of numbers we're talking about in terms of the number of 'disciples' a Guru is likely to have, versus the number of 'initiates' per Conjurer. I expect Barth elaborates on the latter in one of his books, but it would have been helpful to get some idea of what sort of approximate transmission rates we're talking about.
My other criticism is that the paper's a bit androcentric- by which I mean: I think it's fine to decide that you want to focus exclusively on the ways in which men in some cultures transmit information (particularly if you're talking about religious transmission in a context where women don't usually become preachers/ ritualists/ etc), but I think you ought to make a point of saying explicitly that this is your focus. Otherwise you risk giving the impression that when you talk generically about 'people', you actually mean 'men'. It's possible, of course, that the cultures he's referring to have both male and female Conjurers and Gurus, in which case I've jumped the gun, but that's not the impression I've gotten from my (admittedly non-specialist) reading.
*When I say 'invading' or 'infecting', I'm stealing the terms from population genetics, where they're used to talk about how a genetic variant arises in one population and then spreads to another one. I'm not intending to imply that an 'invasion' or 'infection' of Gurus is necessarily going to be military, unfriendly or damaging.