Bart Ehrman’s book Jesus Before the Gospels is really quite excellent. It’s a must-read for anyone looking into how stories became Gospels, about memory, memorisation, distorted memories, and transmission. I am listening to it now, but let me pick out a few sections for you.
In Chapter 4, Ehrman summarises the work of British psychologist FC Bartlett, and his work in the seminal Remembering (8.21-8.24, ePub):
Of even greater importance for understanding the transmission of tradition in early Christianity—during the period when all stories of Jesus were being passed along not in written Gospels, but by word of mouth—is another set of experiments that Bartlett did involving something he called “serial reproduction.” In this experiment, rather than a person observing something and then repeatedly trying to reproduce it, one person makes the observation and relates it to another, who relates it to another, who relates it again to another, and so on. The object that is observed can again be a short story, or a descriptive prose passage, or even a picture. As Bartlett conducted the experiment, Subject A would, for example, read a very short passage twice. After an interval of doing something else for fifteen to thirty minutes the person would be asked to recall the passage in writing. Subject B would be shown the account as A wrote it, read it twice, do something else for fifteen to thirty minutes, then write down what he remembered of it. And so it would go through, say, ten subjects.
In this case the alterations made during the serial recollection of the material tend to be very serious and get worse with each recollection, so much so that if you look at the original story (or description, etc.) itself, and the reproduction of it by the tenth subject, you would be hard pressed indeed to say that what she was reproducing was the story you had read. The differences tend to move in the same direction. Material gets omitted, from one reproduction to another; the accounts tend to become increasingly coherent, as links between thoughts are provided that were not in the original; and details get changed all over the map. As Bartlett summarizes:
It is now perfectly clear that serial reproduction normally brings about startling and radical alterations in the material dealt with. Epithets are changed into their opposites; incidents and events are transposed; names and numbers rarely survive intact for more than a few reproductions; opinions and conclusions are reversed—nearly every possible variation seems as if it can take place, even in a relatively short series.6
Bartlett goes on to point out that the results that emerged in these controlled experiments, with unusually intelligent and highly educated undergraduates at Cambridge University, would almost certainly be far worse in the real world with average people, who were not the elite students at one of the greatest universities in the English-speaking world. What, one might wonder, would happen to serial reproductions of, say, sermons of Jesus, or accounts of his life? One should not urge that these would not change much given the presence of eyewitnesses to guarantee their accuracy, in light of what we have already seen in chapter 3. Nor should anyone think that a predominantly “oral culture” such as found in the early Roman Empire would effectively preserve traditions without changing them, for reasons we will see in chapter 5. For now I want simply to emphasize the point first demonstrated by Bartlett more than eighty years ago: “The one overwhelming impression produced by this more ‘realistic’ type of memory experiment [i.e., as opposed to remembering nonsense syllables] is that human remembering is normally exceedingly subject to error.”7
Ehrman goes on to talk about the case of John Dean, and one that sparked much further research into memory (8.41-8.51):
A famous example can demonstrate my point. There is a much-cited study done of both detailed and gist memories of a person who claimed to have, and was generally conceded to have, a very good memory: John Dean, White House counsel to Richard Nixon from July 1970 to April 1973.
During the Watergate hearings Dean testified in detail about dozens of specific conversations he had during the White House cover-up. In the course of the hearings he was asked how he could possibly remember such things. He claimed to have a good memory in general. But he also indicated that he had used later newspaper clippings about events in the White House to refresh his memory and to place himself back in the context of the events that were described. It was after he publicly described his conversations with Nixon that the White House tapes were discovered. With this new evidence of what was actually said on each occasion, one could look carefully at what Dean had earlier remembered as having been said, to see if he recalled both the gist and the details correctly.
That’s exactly what the previously mentioned Ulric Neisser did, in an intriguing article called “John Dean’s Memory: A Case Study.” Neisser examined two specific conversations that took place in the Oval Office, one on September 15, 1972, and the other on March 21, 1973, by comparing the transcript of Dean’s testimony with the actual recordings of the conversations. The findings were striking.17 Even when he was not elevating his own role and position (as he did), Dean got things wrong. Lots of things wrong. Even big things.
For example, the hearing that involved the September 15 conversation occurred nine months later. The contrast between what Dean claimed was said and what really was said was sharp and striking. In Neisser’s words:
Comparison with the transcript shows that hardly a word of Dean’s account is true. Nixon did not say any of the things attributed to him here. . . . Nor had Dean himself said the things he later describes himself as saying. . . . His account is plausible but entirely incorrect. . . . Dean cannot be said to have reported the “gist” of the opening remarks; no count of idea units or comparison of structure would produce a score much above zero.18
It should be stressed the Neisser does not think Dean was lying about what happened in the conversation to make himself look good: the conversation that really happened and the one he described as happening were both highly incriminating. So why is there a difference between what he said was said and what was really said? Neisser argues that it is all about “filling in the gaps,” the problem I mentioned earlier with respect to F. C. Bartlett. Dean was pulling from different parts of his brain the traces of what had occurred on the occasion, and his mind, unconsciously, filled in the gaps. Thus he “remembered” what was said when he walked into the Oval Office based on the kinds of things that typically were said when he walked into the Oval Office. In fact, whereas they may have been said on other occasions, they weren’t on this one. Or he might have recalled how his conversations with Nixon typically began and thought that that was the case here as well, even though it was not. Moreover, almost certainly, whether intentionally or subconsciously, he was doing what all of us do a lot of the time: he was inflating his own role in and position in the conversation: “What his testimony really describes is not the September 15 meeting itself but his fantasy of it: the meeting as it should have been, so to speak. . . . By June, this fantasy had become the way Dean remembered the meeting.”19
Neisser sums up his findings like this: “It is clear that Dean’s account of the opening of the September 15 conversation is wrong both as to the words used and their gist. Moreover, cross-examination did not reveal his errors as clearly as one might have hoped. . . . Dean came across as a man who has a good memory for gist with an occasional literal word stuck in, like a raisin in a pudding. He was not such a man.”20
And so, whether Dean had a decent gist memory probably depends on how broadly one defines “gist.” He knew he had a conversation with Nixon. He knew what the topics were. Nonetheless, he appears not to have known what was actually said, either by Nixon or himself.
In this instance we are talking about an extraordinarily intelligent and educated man with a fine memory, trying to recall conversations from nine months before. What would happen if we were dealing with more ordinary people with average memories, trying to recall what someone said maybe two years ago? Or twenty? Or forty? Try it for yourself: pick a conversation that you had two years ago with someone—a teacher, a pastor, a boss. Do you remember it word for word? Even if you think you do (sometimes we think we do!) is there any actual evidence that you do? It is important to emphasize what experts have actually learned about memories, and distorted memories. Leading memory expert Elizabeth Loftus and her colleague Katherine Ketcham reflect on this issue: “Are we aware of our mind’s distortions of our past experiences? In most cases, the answer is no. As time goes by and the memories gradually change, we become convinced that we saw or said or did what we remember.”21
These comments are dealing with just our own personal memories. What about a report, by someone else, of a conversation that a third person had, written long afterward? What are the chances that it will be accurate, word for word? Or even better, what about a report written by someone who had heard about the conversation from someone who was friends with a man whose brother’s wife had a cousin who happened to be there—a report written, say, several decades after the fact? Is it likely to record the exact words? In fact, is it likely to remember precisely even the gist? Or the topics?
Jesus’s Sermon on the Mount in Matthew chapters 5–7 was recorded about fifty years after he would have delivered the sermon. But can we assume he delivered it? If he did so, did he speak the specific words now found in the sermon (all three chapters of them) while sitting on a mountain addressing the crowds? On that occasion did he really say, “Blessed are the poor in spirit, for theirs is the Kingdom of Heaven,” and “Beware of false prophets, who come to you in sheep’s clothing but inwardly are ravenous wolves,” and “Everyone who hears these words of mine and does them will be like a wise man who built his house on a rock”? Or did he say things sort of like that on the occasion? Or did he say something sort of like that on some other occasion—any occasion at all? Which is the gist and which is the detail?22
Of course, as I talked about in my book The Resurrection: A Critical Examination of the Easter Story, any account that relays speech – particularly from the period before accurate speech recording – should set off alarm bells. Think of the private interactions between Jesus and Pontius Pilate. There is no way that the accounts in the Gospels represent accurate history – an accurate account of what was actually said. Indeed, we already have huge variations amongst the Gospels, especially with John.
If the Christian claims that the accuracy is bestowed upon the Gospels on account of them being written or sourced by eyewitnesses, then they have to deal with the heavily researched data of eyewitness testimony accuracy. As Ehrman shows in early chapter so the book, eyewitness testimony is not accurate. In fact, many people claim to have been eyewitnesses by inadvertently creating false memories!
This finding was borne out by one of the most interesting recent experiments undertaken outside the laboratory. Three psychologists at Wesleyan University—John Seamon, Morgan Philbin, and Liza Harrison—wanted to see if imagining a bizarre experience could later lead to a memory of it. They titled the results of their study, “Do You Remember Proposing Marriage to the Pepsi Machine?”14
The study involved forty students who were taken to a variety of locations around campus. In each location they were instructed either to perform an action, to imagine performing it for ten seconds, to watch the experimenter performing the action, or to imagine the experimenter performing it. The actions were either normal or bizarre. For example, if they were in the library, they were asked to look up a word in a dictionary; or they were asked to pat the dictionary and ask how it was doing. Elsewhere they were asked to check the Pepsi machine for change or to go down on one knee and propose marriage to it.
Two weeks later the participants were interviewed and asked if the action had been imagined or performed. The conclusions were clear. Whether the action was normal or bizarre, participants who imagined it often remembered doing it: “We found that imagining familiar or bizarre actions during a campus walk can lead to the subsequent false recollection of having performed these actions.”15 In this instance the researchers found that imagining the action vividly, but just one time, could produce the false memory. Moreover, imagining someone else performing the action led to just as many false memories as imagining doing it oneself.
To sum up the situation, consider the words of one of the world’s leading experts on false memory, Daniel Schacter: “Numerous experiments have demonstrated ways in which imagining events can lead to the development of false memories for those events.”16
Does such research have any bearing on the memories about Jesus, a great teacher and miracle worker, by eyewitnesses or by those who later were told stories by eyewitnesses—or even those told stories by people who were not eyewitnesses? Can imagining that a great religious leader said and did something make someone remember that he really did say and do these things? It might be interesting to address that question by looking at another famous Jewish teacher. For my example I have chosen a person from the modern period known as the Baal Shem Tov. He was the eighteenth-century founder of Hasidic Judaism. [7.23-7.28]
Pertinent to the Jesus story are the supposed eyewitness accounts of the miraculous events of Israel ben Eliezer – Baal Shem Tov, or “the Besht” – tales that came from a mixture of eyewitness accounts and other reliable sources within the community. But no one outside of the community of believers in him believe the outlandish claims. As Ehrman points out (7.42-7.43):
There are many, many tales such as these throughout the account. And what is my point? Do I think the Besht actually had supernatural powers to do these things, to be transformed into a divine, glowing presence, to cast out and imprison demons, to ignite trees with his finger, to raise the dead, and all the rest? No, personally, I don’t believe it. But are the stories based ultimately on eyewitness reports? Writing some fifty-four years after the events the author claims they were indeed based on eyewitness testimony. Does that make them reliable? Even if devoted followers of the Besht say yes, virtually everyone else realizes that these allegedly eyewitness reports are anything but historical.
What then about the Gospels of the New Testament? If they are based on eyewitnesses are they necessarily accurate? Do they in every instance represent accurate memories? Given what we have seen in this chapter, I think the answer has to be no. They are not necessarily reliable. And, of course, they are not necessarily unreliable either! All of them have to be examined historically to see whether and how far they preserve accurate memories of Jesus and distorted memories. But before we proceed to do that, what can we say about the relationship of these canonical accounts to eyewitness memories?
Eyewitness testimony is not only not what it is cracked up to be, but it is also a problematic claim in and of itself. This is something I have recently come to understand concerning the Gospels. To say, for example, that Matthew was an eyewitness of Jesus (and forget the arguments surrounding this – just go with it), and is therefore an accurate source for the claims about Jesus on account of this eyewitness-ship, is really problematic. Because what the Christian is trying to say is that Matthew knew Jesus; he was a contemporary follower.
But the inference is skewed. It can’t mean – in any pragmatic sense – that Matthew was present for each and every story of Jesus he recounts. Matthew can’thave been present at Jesus’ birth, at his baptism, when he went alone into the wilderness, at each and every miracle he performed, was there for every single event reported of Jesus’ life, including every moment of the death and Resurrection story, including the private conversations with Pilate and the trial.
The claim that someone was an eyewitness to Jesus is not – cannot be – to say that they were eyewitness, therefore, to everything Jesus was ever reported as doing.
The oft-heard claim from Christian apologists is that oral memory in cultures without a strong literary (or any literary) tradition meant that oral transmission was particularly accurate. The claim goes that because oral transmission was the only way of remembering things, cultures with oral transmission (such as the Jewish context of the Gospel stories being told and retold countless times before being written down) produced particularly accurate transmission of the original stories.
There are several problems with this.
- It’s not true.
- There’s no real way of being able to test the accuracy of oral traditions over time without first recording or writing the first one down and comparing it with the most recent.
And that’s why the John Dean case is so interesting, though not in an oral culture, so to speak.
The question is, then, do we have any data about oral transmission from such cultures that can lead us to the conclusion in point 1.)?
Yes. Yes, we do. And I’ll discuss that in my next post.
Stay in touch! Like A Tippling Philosopher on Facebook: