Remembering Without Recording

Remembering Without Recording December 31, 2013

Judy Redman has posted a reply to my own post about memory and orality yesterday. Judy writes:

I agree that having composed something doesn’t mean that you will remember it word for word. I do think, however, that when a person has composed a story to illustrate a particular point, s/he is much more likely to reproduce the significant points accurately than is someone who has just heard it. I also think that when you are retelling your own story, you are more likely to use similar wording for subsequent retellings than is someone else who is retelling it from your ‘original’ because you talk like you and they don’t. I therefore think that Jesus is likely to have retold reasonably close versions of his parables, so the disciples would have heard essentially the same thing several times.

I don’t disagree with anything she wrote, but because there is such potential for those of us living in a highly literate culture to take certain things for granted, I want to make sure that some potentially ambiguous language is understood as I intended. I also want to go into more detail on some points.

When someone composes a story (let’s say a joke) and tells it more than once, perhaps frequently, it will probably improve over time. The gist may or may not stay the same – it really depends on whether the story’s author wants to change it or not. And so we cannot presume the story’s stability as a result of multiple retellings by the person who composed it. All that we can presume is that the author’s intention is consistently present even if the story changes (although even that may be too much, since the author could well feel that an earlier version was better, but find it impossible to remember it exactly).

If the author of the story never writes it down, then there is simply no basis for the author or anyone else to talk about verbatim reproduction. This is what is meant when experts in the study of oral cultures and oral tradition say that there is “no original version” but only “multiple performances.” The author and/or others can tell the story, and the gist may well remain the same. But the original is never written down, and thus cannot be repeated verbatim even by the author. There is simply no way to do this, or to tell if it has been done, without some method of recording being utilized.

This need not lead to pessimism, except for those who hoped to be able to get Jesus’ precise verbatim words, or those of any other historical figure who delivered an unwritten speech in the era prior to audio recording technologies. We have good reason to conclude that we have the gist of a saying of Jesus in a number of instances, and where translation back into Aramaic uncovers word play and other such features which tend to be recalled, there is reason to think that we may have the very words of Jesus. But such instances will be the exception, rather than the rule, not because of any issue unique to the quest for the historical Jesus, but because of the nature of memory, orality, and history themselves.

Browse Our Archives

Follow Us!

TRENDING AT PATHEOS Progressive Christian
What Are Your Thoughts?leave a comment
  • Judy Redman

    Yes, indeed, James!!!

  • arcseconds

    I’m far from being any kind of expert on this (I’m not sure I’m even going to claim to being well-informed on this matter…) , but while I think your comments often do apply to oral cultures, I don’t think they always do apply.

    It seems that some cultures do have surprisingly accurate preservation of oral ‘texts’. The best example I know of is the Vedic corpus. The earliest attestation of writing in India (leaving aside the Indus valley script, which isn’t certain to be writing, and probably has little to do with Vedic culture) is the 3rd century BC. Yet the Rg Veda is dated to sometime in the middle of the 2nd millenium BC, and is believed to be pretty accurately preserved. The language used is a more archaic variety of Sanskrit even than the other Vedas.

    The writing used in the 3rd century was not used to write Sanskrit, but the everyday language of the time. The earliest Sanskrit inscriptions are a couple of centuries later, and the earliest manuscript of the Rg Veda is 1100 A.D. — 2½ millennia after its redaction into its final form.

    Of course, accurate oral preservation of this sort requires special circumstances, including most importantly the idea that the text must be preserved accurately.

    The question of accurate transmission you raise does have a sound criteria here: that the student recites the text just as the masters do. I don’t think this is really much different from manuscript preservation: you can check and correct your version against an older version. Both have the possibility of both error and error-correction. The only difference is that the turn-over is forced to be once a generation for oral preservation, and we don’t have any of the older version to compare against. But that doesn’t mean we’re not in a position to assess the age and accuracy of the text — we don’t have the earliest versions of manuscript traditions either, so the question is a similar one.

    I don’t think there’s any particular reason to suppose the early Christians were doing anything akin to this, though — Ian’s comments in the earlier thread are apropos: the preservation of the Rg Veda is the land-speed record, using special technology.

    • The Rg Veda is a great example. It clearly has linguistic features that show its antiquity (although not that it is as old as some Hindu nationalists have claimed), and we have no evidence that the entirety of that transmission was in writing. Yet the archaic language indicates that the storytelling did not evolve together with the language, as one might expect if it were transmitted orally in an informal manner. The closest parallel I can think of are nursery rhymes – plenty of archaisms in those! I haven’t studied the Vedas in Sanskrit and so don’t know to what extent features like rhyme and meter might have facilitated memorization.

      • arcseconds

        I haven’t studied them in Sanskrit either (barely any truth to the idea that I’ve studied them at all, in fact, although I have read selections in translation and read a bit about them), but this isn’t necessary to know that they do have a pretty strict metrical structure! One of the ‘Veda readers’ I read was pretty insistent in telling you what meter each selection was in.

        It’s not going too much out on a limb, I think, to say that rhyme and meter contribute substantially to the ability to memorize a text. Anyone who’s memorized a long poem can testify to this.

        Also, a community of specialists trained and dedicated to memorizing such texts would help a great deal. It’s not necessary that every member of the community memorizes every text (although obviously that would result in the highest probability that the text would be preserved, and preserved accurately). If you want to know what Yaska has to say about a particular construction in the Rg Veda, instead of dong a Google Scholar search, you ask your colleague who has memorized his work. It’s potentially just as fast (assuming he’s nearby) and you can also ask him ‘OK, so what does that mean, then?’ !

        Although not as old as the Rg Veda, it’s also impressive that they not only preserved the sacred texts themselves orally, but also highly academic commentaries on them, including grammatical treaties.

        The Vedic/Hindu apparent preference for memorization even after writing was definitely available reminds me a little of the scepticism Plato expresses about writing in the Phaedrus.

        • The question I still have is whether completely oral texts can be memorized anything close to verbatim without the use of a written text at some point in the process to aid in repeating word for word.

          • arcseconds

            How did you learn your ABC? Certainly not by reference to a written text 🙂

          • I disagree completely! A language does not have an “ABC” until it develops a writing system, and that is how the writing system is taught. There is no “ABC” in a purely oral culture.

          • arcseconds

            Sorry, I wasn’t being very clear.

            How did you learn the order of the letters? If you had a normal anglophone upbringing, you learnt the alphabet song. You can’t have learnt this with reference to a text, because you didn’t know how to read at that point. Yet I’m sure you managed to learn the song at least ‘close to verbatim’.

            If I’m to take your statement literally, that no-one can learn anything close to verbatim without reference to a written text, that seems plainly false. Non-literate people (including pre-literate children) learn songs pretty accurately all the time. The alphabet song is a good example: you get drilled until you get it right. I suppose you might argue that the parents can refer to a text, but does this even ever happen? I might have seen it written down once or twice, but I certainly didn’t learn it by comparing my recitation to a text, and I’m pretty sure I (and practically every other anglophone adult) remember it just fine and could teach it to a child without using a text. In fact, I think that’s how it normally happens.

            What I think you mean is that you can’t have the concept of learning verbatim without a written text because a written text gives you an ideal of reproduction that otherwise you don’t have.

            I agree that writing makes the notion of verbatim obvious and that oral cultures don’t necessarily have this concept, value the notion, or find it useful, and we tend to make all sorts of mistakes by projecting our literate biases onto cultures that don’t use written language

            (While there’s definitely a class of symbol-systems that aren’t writing and in some cases pre-date writing, calling them ‘proto-writing’ is an excellent example of the problems we get into here. ‘Pre-literate’ is similarly problematic.)

            But I don’t think that means the ideal of perfect reproduction requires writing. Also, I think to some extent it’s a red-herring. If we are interested in how ‘texts’ are preserved orally over time, it doesn’t matter too much whether they have the concept of verbatim reproduction. It’s also not helpful to think of the matter in terms of ‘verbatim reproduction’ versus ‘non-verbatim reproduction’.

            Instead, we should concern ourselves with the error rate (or mutation rate, if you prefer). No method of reproduction has an exactly 0 error rate (including copying a file on a computer) and as you know the error rate in written manuscripts is actually reasonably high and quite noticeable over time. So all reproduction exists on a spectrum with the best forms of computer reproduction at one end (close to perfect) and joke and stories told largely for entertainment at the other (can be mutated at whim by the teller, and no-one minds too much).

            I could say more, but I’m wondering whether I’m interpreting you correctly. Do you agree at least it’s possible to learn something close to verbatim (i.e. with a low rate of error, say less than 1% chance of mis-reciting a word) without reference to an individual text, even if this is something that only someone from a literate culture would think of striving for?

          • Verbatim learning of songs, especially where there are other features such as rhyme or alphabetic patterns, is certainly possible. Psychologists of memory have estimated that we can recall about 17 words in a row verbatim without the aid of a text or other elements such as meter.

            I think parents have the alphabet as a text as a memory aid that has been part of their education, which helps with passing the song on to the next generation. But that is certainly a different scenario, memory-wise, than remembering a parable or a bit of ethical teaching, isn’t it?

          • arcseconds

            It is a different scenario. I’m less and less certain I know what your position is, and I think we may be talking at cross-purposes.

            Let me sum up mine. I agree with the way you characterize oral culture in general: it doesn’t make a lot of sense to talk about ‘the original’ and ‘verbatim reproduction’ when it comes to most stories etc. passed on orally.

            However, I don’t think this is universally true. Oral texts can be passed on very accurately in some cases. This often (and always for lengthy texts) involve special techniques such as rhyme and meter, members of the society who specialize in memorizing and reciting texts, special institutions set up to train them, etc.

            (Speaking of specialist techniques, check out the description of jaṭā-pāṭha on this page )

            In these cases, the intent is definitely to preserve a particular text as it is, and there’s clearly a difference between an error-free recitation and one with errors.

            I don’t think there’s any real difference in kind between a manuscript tradition and a culture that cares about the fidelity of the transmission of its oral texts. Of course, if we can find early manuscripts it allows us to more easily assess how accurate the transmission has been, but it’s not like we’re helpless without this. A lot of interpolations and so forth have been detected purely on textual grounds.

            This remark of yours

            The question I still have is whether completely oral texts can be memorized anything close to verbatim without the use of a written text at some point in the process to aid in repeating word for word.

            Sounds like you don’t believe that this is possible, and that the Rg Veda and any other oral text that’s accurately reproduced from one generation to another must have writing involved in its transmission. Is that really what you believe, or am I reading this remark with greater generality than you intended?

          • I honestly don’t know how the Rg Veda was transmitted, and whether texts as aids to memorization were used at any point in the history of it. What I do know is that, without texts by which to verify accuracy, until such time as there was a community that agreed upon wording, which had highly accurate recall, and which oversaw performances, then there was no way to know how accurate any performance of the text was by any meaningful standard. I’m sure we’ve all substituted one rhyme for another in trying to recall the words to a song or poem. And so even that doesn’t guarantee accuracy, but only helps, and constrains the possibilities for changes.