Speeches of Jesus (3) – human memory experiments

Again, I am  moving a comment up to a post of its own. Mark Goodacre says, referring to April DeConick, “Human Memory and the Sayings of Jesus” in Tom Thatcher (ed.), Jesus, the Voice and the Text: Beyond The Oral and Written Gospel (Waco: Baylor University Press, 2008): 135-80 and Robert K. McIver and Marie Carroll: “Experiments to Develop Criteria for Determining the Existence of Written Sources, and Their Potential Implications for the Synoptic Problem,” JBL 121 (2002): 667-87:

I am not persuaded that we should use the 15 or 16 word string criterion of McIver and Carroll. This is something derived from experiments on contemporary students that included some flawed methods and dubious inferences (some comments on McIver and Carroll are included in a post on DeConick’s similar experiments, http://ntweblog.blogspot.com/2008/12/contemporary-memory-experiments-and.html).

[For some reason that I don’t understand, if you click on the link above, it takes you to an error message saying this page doesn’t exist, whereas if you copy the link and paste it into your browser, it finds the post quite happily. This problem is now fixed.]

Mark, as I indicated in my comment on the previous post, I also have reservations about applying research on 20th and 21st century undergraduates directly and uncritically to 1st century Christians because their memories have been trained so differently. I also agree that the fact that the students in April’s research were given stories that are not culturally familiar will have had an effect on how well they remembered. I therefore don’t think that you can take the statistics in her results and apply them to the 1st century, but she doesn’t seem to be trying to do this in what she terms a pilot study. I think that the kinds of transformations are, however, relevant and would bear further investigation in a population that is more like 1st century Christians.

Although I am using a very small sample size (3) and anecdotal evidence, my friends who are blind have far better memories than do I or my sighted friends because it used to be hugely more difficult for them to keep and access written records than it was for people who could write on paper with pencils and carry the notes around with them. I suspect that PDAs and talking text is going to change this, but I think that blind people in their 40s and 50s and older would be a better example of how memory works than are younger undergrads.

Also, at the beginning of her article, she talks about the Parable of the Lottery Ticket, which she gives to her classes. This is a culturally congruent story and it would be very interesting to see the statistics on it. I have also used this parable in my own classes and bible studies, but I get the participants to re-tell it the same day and even when I warn them that they are going to be asked to re-tell it, I only ever get accurate gist, not accurate detail and not terribly long verbatim strings. Their memories really are not terribly impressive.

All this, I think, makes McIver and Carrol’s data potentially an underestimate of the length of verbatim string needed for reasonable certainty that there is a textual rather than an oral relationship between two texts. OTOH, it is also likely that Jesus’ audience was paying closer attention to him than my students or April’s students were paying to us. Neither April nor I are demonstrable miracle workers, after all! Possibly what we needed to do was to say “pay careful attention – the material I am about to present will be examined”. 🙂

At the same time, I don’t think we are dealing with the kind of material that lends itself to incontravertible proof of how it was transmitted. I think we can probably say, along with the originality software, that any verbatim string of 8 words or more that are not aphorisms or stock phrases has the possibility of being copied from a written text, but that we need a much longer string of verbatim correspondence to be able to say with any certainty that this is so. We then need to look at the string in context and analyse the kinds of changes that the material around it has undergone in order to be a bit more certain about our judgements.


12 thoughts on “Speeches of Jesus (3) – human memory experiments

  1. you’re right, I think, about being skeptical of using 21 century students to compare with 1st century folks when it comes to memory. I recall reading somehwere very recently that moder (i.e. TV/internet fluent) minds have developed very differently in terms of memory capacity to ones that do not have access to the internet. It’s apparently something to do with the way our brains now store and recall information by association.

    Anyways, I only wish I’d saved the link…. if onyone has sen published studies along those lines it’d be interetsing to read them

  2. Thanks for the interesting post, Judy. The URL doesn’t work when you click on it because it features the closing bracket ). Eliminate that and the click will work.

    McIver and Carroll’s work really needs to be read alongside Poirier’s critique. He makes several important observations about their research, one of which notes that lack of any consideration of evangelists’ creativity and redaction in their study. It is as if the evangelists were always attempting to retail stories and sayings as accurately as possible, which we know not to be the case.

    An important fallacy in McIver and Carroll is their use of the wrong model with reference to the Q theory that they assume, comparing the students’ texts with the source text rather than with one another (see Poirier). They also apparently fail to notice that one of their top nine texts in which copying has definitely occurred is Matthew // Luke in triple tradition (Matt. 8.1-4 // Luke 5.12-16), which of course stands against the Two-Source Theory (and another of their top 23 passages is similar).

    The 18 figure you mention here, Judy, comes from their second article and appears to be an arbitrary adjustment from the 16 figure in the first article.

    I agree with you that McIver and Carroll’s 16 figure is an underestimate. The briefest glance at other dependent texts from antiquity confirms this.

  3. Mark, it wasn’t the closing bracket causing the problem. I fixed that four times. There was some other thing in the tag, a “no follow” comment that seems to have been the problem. I have never seen this before, but the link to your blog now works.

    McIver was at SBL Auckland in 2008 and I asked him about the change in string length. He said that they had looked at their data again and decided that 18 was a more rigourous interpretation of it, so it wasn’t arbitrary, but they don’t explain it.

    Thanks for reminding me about Poirier’s article. I just started reading it but it’s a bit late for me to get my head around the finer points of his argument. However, I agree that there is a need to factor in the creativity of the authors and their redaction. Realistically, if they had access to another version of the text and felt that it was a totally accurate and acceptable account, there would be no need for them to write another version of it. The work I did on psychological research into eyewitness testimony and human memory for my JBL article makes me quite convinced that some of the variations between parallels are the result of human memory rather than deliberate redaction. (NB this link will only work if you are a member of SBL and can log in to their site, but it’s Volume 129, Number 1 / Spring 2010, pp 177-197.)

  4. Thanks for posting on this and for sharing your conversation on this topic with a wider audience.

    It seems to me that the limit on verbatim recall with and without textual aids ought to be tested in a few different cultural contexts. But ultimately we will certainly discover a limit that is imposed by the typical human brain rather than cultural training in memorization.

    The most important thing to remember (in my opinion) is that, unless we have some sort of text or recording, we cannot assess verbatim recall and reproduction even if we wish to.

    It is certainly the case that, in societies where either literacy, literature, or both are less widespread, extensive memorization has occurred and does occur. But such instances of the use of texts and repetition to assist in memorization are not to be confused with the situation of a purely oral tradition, in which there is (at least at first) no written document to allow for repeated exposure to the same exact words as an aid to memorization. In the latter cases, people may well have believed that they were remembering something “word for word.” But neither they nor we have any way to assess the precision of recall, and we have every reason to think that any inherent limits of human memory not aided by texts would have applied to such situations.

    And so I think there is certainly room for more research on this topic, but I think that the work of people like McIver, Carroll and DeConick is on the right track. Without a text, what could be remembered were the gist of longer stories, short memorable sayings or expressions with mnemonic aids like rhyme and word play, and precise wording in specific instances when it was important to the story (as e.g. in the punch line of a joke).

  5. Thanks for the interesting comments, Judy and James. I enjoyed your JBL article, Judy.

    Poirier’s article in JBL is a bit terse, but it really repays careful study — he makes some very strong points. There are additional problems with McIver and Carroll that I should write up some time myself. One of them both you and James touch on here. It is the crucial distinction between memory of a single written text and the transmission of multiple oral traditions. McIver and Carroll appear to assume that the writing-up of a short-term memory of a single text in some way correlates to the evangelists’ use of oral traditions. What McIver and Carroll are comparing are different kinds of use of single-fixed text, not the difference between oral tradition and written text.

  6. That’s a good point, Mark. I’ve tried to think of a situation or experiment that might allow for testing of a scenarion more like that in early Christianity. Maybe we should get some volunteers to agree to watch Doctor Who, and try to remember what the Doctor says – without DVRing it and watching it again!

  7. Pingback: Dr Who as a test case for human memory? « Judy’s research blog

  8. My personal experimentation is to ask people to recall JFK’s assination. Eyewitnesses have somewhat varying memory. However they all agree on the major points and when they are sure, they are sure. Lately I have moved to the moon landing and 9/11. My experience seem to suggest that Judy is correct that memory degrades quickly till about 5 years. Then it coalesces. People seem to get JFK as correct as 9/11.

    What is most interesting, however, is that when an eyewitness is certain of a rememberance, even other eyewitnesses are generally unsuccessful at disabusing him of that memory. What that means is that an eyewitness author will insist that what he rembers is faithfully reproduced.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s