On a recent retreat I took, the Buddhist scholar Andrew Olendzki (head of the Barre Center for Buddhist Studies) gave a presentation on the Pali Canon, which is the scriptural canon of the Theravada Buddhist tradition. The Pali Canon is the only complete surviving Buddhist scriptural canon in its original language, and is also the oldest Buddhist canon.
Andy’s presentation included some fascinating information about oral transmission versus written transmission and the kinds of misconceptions that Westerners are likely to have about the same. It is generally assumed here in the West that a written transmission, via writing, will be more accurate than an oral one, which would be handed down by rote memorization only without written versions.
The Pali Canon was originally transmitted orally only; the Buddha passed on in ~485 BCE or thereabouts and his teachings weren’t written down for about two or three hundred years. It has generally been assumed that the period of oral transmission is a period of change, in which the canon was “massaged” by oral transmission into the form we now have it, and after the written versions were created, the canon was “frozen” at that point.
Except that it doesn’t really work that way. The underlying assumption (that oral transmission is by definition inaccurate) is easily disproven. The first misconception to overcome is that it isn’t like one of those games of ‘gossip’ in which one person hears something, whispers it in the ear of another, who then whispers it in the ear of another, and so forth. That isn’t how the Pali Canon was memorized or transmitted. If only one person is transmitting a scripture to one other person, the kind of warping and changing described above can certainly happen. But what happens when many people are transmitting the scriptures to many people, and they’re all sharing their transmitted versions with each other?
Let’s take a simple example of a group of fifty people singing “Happy Birthday”. We all know the words of the song, I should hope. Now let’s imagine that one person out of the group of fifty sings: Happy Dirthday to You, Happy Dirthday to You, Happy Dirthday dear <person>, Happy Dirthday to you. A number of things are liable to happen here: either the person singing the incorrect version realizes that he/she should be singing “Birthday” and not “Dirthday”, or else other people in the group point out (gently, I hope) that he/she is singing it wrong. In fact, “Happy Birthday” is so well known that singing it a bit differently is considered funny — try singing: “Happy Birthday to TheOneAndOnlyYou, Happy Birthday to TheOneAndOnlyYou, Happy Birthday dear <person>, Happy Birthday to TheOneAndOnlyYou”. It’s mildly funny.
So the oral tradition contains these built in correctives, like negative feedback loops that keep everything within strict guidelines.
Now imagine the same situation — i.e., one person singing “Dirthday” instead of “Birthday”, but now imagine it not as singing, but as writing down the words. Fifty people write down the words to “Happy Birthday”, and one person writes “Happy Dirthday” instead. In this case, generations later scholars might study these fifty versions and note the ‘textual variant’ in the one version. They might even postulate that this is some kind of dialectical difference — perhaps this one person came from a place where “Birthday” is pronounced or spelled “Dirthday” or, worse yet, a scholar might write some article speculating about some strange tradition in which “Dirthday” is a mispelled version of “Dearthday” and the person sings a song about having a dearth of something. Odder articles have been published…
In other words, the written transmission in this case is more likely to include a variant reading than the oral transmission, which means that instead of one version of the text coming down, there is a different version which causes critical concern (and possibly interpretational difficulties, as can happen). Again, consider “Happy Birthday” — we all know it, and we all know exactly the same version of the song. How did that happen? It happened because many people transmitted the song to many people.
The Pali Canon was memorized by many people, and was transmitted to many people by an oral transmission. Thus the same feedback loops were in place — if one monk in the group was chanting the sutta in question with imperfections, the other monks can make sure that it is the same for all. In order to assure a clear transmission, in fact, the monks responsible for the memorization used a lot of interesting techniques to ensure accuracy. For example, they memorized it with every other syllable only, or backwards, or backwards with every other syllable or every other word. It sounds like an astonishing feat of memory, but pre-literate cultures do in fact have proven, extraordinary powers of memorization. (Remember that the great epic literatures of the world were handed down in precisely the same way).
Andy even pointed out that the known variants in the Pali Canon come from the written versions of the canon, and not the oral versions. (Yes, there are still many monks throughout the Buddhist world who memorize the Pali Canon and hand it down to another generation that way. It has been only lately that scholars have begun realizing that the oral transmissions are free of the kinds of pesky, and potentially confusing, errors common in the various textural traditions of Sri Lanka, Burma, and Thailand.)
Critical to this teaching, therefore, is that the notion of the Canon being “massaged” into its current state during the years of oral transmission is much less likely. We know for sure that the Canon was assembled during the three Buddhist councils, which met respectively immediately after the Buddha’s passing on, and about a century later for each subsequent council. Certainly parts of the of the Canon were added later — such as the Abhidhamma. But the central core of the Canon is attested as having been established right at the very first council. If that is the case, then the only viable source for those suttas which were chanted and established is the Buddha himself.
Thus the Pali Canon may very well be a much more precise record of the Buddha’s actual teachings than has been thought. Andy presented a number of subsequent arguments which make a very convincing case for that possiblity.
But that’s a topic for another entry…