2011-07-04

Demystifying R. Joseph Hoffmann, and the war over Bayes’ theorem

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.

by Neil Godfrey

Some academic jargon is nothing more than the modern equivalent of a sorcerer’s mumbo jumbo — designed to awe while hiding the fact that there is really nothing meaningful on offer at all
Updated 5th July to add link to Richard Carrier's post taking Hoffmann to task.

R. Joseph Hoffmann has let a crotchety side to his nature show as he publicly attempts to humiliate a younger scholar who, in exchanges with the aging don, has exposed a dint of mediocrity in his intellect.

The casus belli is, at least on the surface, the place of Bayes’ theorem in historical Jesus studies.

Now Hoffmann’s writing is surely more renowned for its thick overlays of esoteric intellectual jargon and rhetoric than for its content. The reason is pure mathematics. More people can read his posts than can understand them. (Stanislav Andreski wrote that this sort of intellectual jargon as the modern equivalent of earlier efforts to bamboozle the uninitiated and impress the elite: various uses of medieval Latin, witch-doctor mumbo-jumbo, etc.)

But on the positive side, one does get a sense that he is thoroughly enjoying himself as he shows off his verbal wit, and there’s nothing wrong with that. Everyone has a right to enjoy themselves (or “oneself”, as I am sure the don would prefer me say).

But what the hell is he trying to say when he burgeons like a Baroque artist doing abstract?

I’m sure he won’t like any attempt at simplification, but then why would any biblical scholar be bothered with a blog like mine when the guild does not even consider it to be an honest discussion of the Bible and Christian origins anyway.

All Hoffmann means to say is that he thinks:

  • Bayes’ theorem is only good for assessing probabilities of what did or did not happen;
  • but the Gospel authors were not interested in informing their readers what did really happen historically and what did not;
  • so scholarly studies should only concern themselves with the thought world of those authors;
  • and therefore Bayes’ theorem is irrelevant, lacking any rightful place in the study of the Gospels;
  • and by trying to apply Bayes’ theorem to the Gospels one is falling into the same sins as fundamentalists/conservatives/literalists (I’m still a bit unclear on American usages of these terms) who are trying to use the Gospels to prove or disprove historical facts, and this is not fair because that’s not how the authors intended them to be studied.

Thus translated into plain English, his argument loses all mystique and its inadequate mediocrity becomes all too evident. So do Hoffmann’s inconsistencies in his selection of targets.

Indeed, one can even think of scholars who have written about the Gospels in just this “correct way”. Bishop John Shelby Spong has often written about what he thinks the Gospel authors were attempting to convey (building on his mentor Michael Goulder’s work), why they wrote the “myths” about Jesus that they did, and how they are not about historical events – historical facts – at all. (That does not mean Spong is a “mythicist” since he does still believe in a historical Jesus.) Thomas L. Thompson is another who has discussed his understanding of the same themes in The Messiah Myth and The Mythic Past/The Bible in History: how writers create a past (same book, different titles).

Of course Bayes’ theorem is not used to understand the texts. That is a different exercise. But it is justifiable to ask if a text, any text, really does represent historical facts or not. One does not have to disrespect the integrity or original intent underlying the texts to ask such questions. There is room for more than one set of questions to be brought into most discussions.

Surely one of the best ways to encourage interest in the Gospels as the authors intended them to be understood, and to explore what they really teach us about early Christianity (as Spong/Goulder and Thompson have in some measure begun to do) is to expose the groundlessness of the currently dominant practice of using the Gospels to learn what Jesus did in fact historically do or say. If most scholars are missing the point of the Gospels as they were really intended, then why not expose the poverty of the way they are (mis)applying them by seeking answers from them they were never meant to yield?

As pointed out in my earlier post, Carrier explains that Bayes’ theorem is useful for non-mathematical or non-quantifiable reasons.

IMGP3678
Image by mattbuck4950 via Flickr

1. It helps to tell if your theory is probably true rather than merely possibly true

2. It inspires closer examination of your background knowledge and assumptions of likelihood

3. It forces examination of the likelihood of the evidence on competing theories

4. It eliminates the Fallacy of Diminishing Probabilities

5. Bayes’ Theorem has been proven to be formally valid

6. Bayesian reasoning with or without math exposes assumptions to criticism & consequent revision and therefore promotes progress

Of course one does not need Bayes’ theorem to work at achieving any of the above. But if the theorem offers some formal discipline in one’s study, and assists in balancing a host of probability variables, then it hardly deserves cruel censure.

Hoffmann appeals to a mathematics prodigy to support his attack on Carrier’s application of Bayes to the field of history. But if Hoffmann or the mathematician had enough knowledge of how history really works, what historical enquiry really is, they would be embarrassed to argue as they did.

Most historical enquiry is an effort to understand known facts, and to uncover additional verifiable facts in order to increase one’s understanding of events in order to construct a meaningful historical narrative. By “facts” historians usually mean primary evidence, or evidence that physically belongs to the time and place under investigation: diaries, realia, artefacts, telegrams, official reports, newspaper reports, statistical records, etc. Criteria are used to help the historian assess the bias within the tangible evidence for the facts. So it is simply not true to say that attempting to apply Bayes’ theorem to historical studies generally would lead to pages of mathematical equations.

Historical Jesus studies are not like that at all and here Bayes’ theorem does have a place. HJ studies begin not with primary evidence or tangible facts, but with the assumption that there is a historical Jesus to study, that the Gospels are the closest thing available by way of evidence for him but that they also depict not the historical Jesus but a Christ of faith. So the task of the historical Jesus scholar is to work out what he or she can decide is a “fact” by applying certain criteria to the Gospels. What other historians use to detect bias within the evidence, historical Jesus scholars use to manufacture “facts”. As one would expect, this method leads to scholars all disagreeing with each other over what are the facts of the historical Jesus.

The results speak for themselves. There is only one Julius Caesar in history (he’s a military and political leader), only one Socrates (he’s a philosopher), though historians may disagree over some of the motivations and details of their lives. But there are many very different “Jesus’s” – everything from a rabbi to a revolutionary.

Well, what does one expect from a “discipline” that begins with an assumption and seeks to find new ways all the time that will enable them to find “facts” to pair with that assumption! Is there any reason to wonder why it is said that each historical Jesus scholar tends to discover a Jesus in his own image?

I’m all for Hoffmann’s approach

So I’m all for studying Christian origins the way R. Joseph Hoffmann wants. It’s the way I attempt to explore Christian origins myself. This is why I have been in two minds about calling myself a mythicist. The label makes it sound as if I am primarily focused on disproving the historicity of Jesus. I’m not. I don’t see such a question as a historical one. A historical question, in my mind, seeks to understand the origins of Christianity, and does so through learning to understand the nature of the evidence at hand. It is invalid to start with the assumption of a historical Jesus when all we have to begin with is a story about Jesus, that is, a literary Jesus.

Indeed, it is this approach that I have advocated over and over on this blog and that has earned most of the criticism from mainstream scholars who insist– contra Hoffmann — on using the Gospels to “prove” this or that historical fact about Jesus!

But is Hoffmann really interested in following it through?

But Hoffmann exposes his own lack of genuine interest (or courage) in challenging the mainstream approach to historical Jesus studies by targeting certain mythicists, comparing them with fundamentalists or conservative scholars, while letting the mainstream, the real guilty party, off the hook.

In his recent blog post attacking Carrier, one commenter quoted a mainstream scholar, Dr Maurice Casey, explaining that he (Casey) was interested in assessing probabilities of something being a fact, and even quantifying a probability at not more than 51%. But the commenter did not explain that the quotation came from Casey, and Hoffmann waded in by agreeing with the commenter that “no real scholar” in historical Jesus studies would approach the exercise that way.

At the time of this post, I understand that the commenter is still waiting for Hoffmann to allow through his response explaining that he (Hoffmann) had just criticized the words of Dr Maurice Casey himself.

Mainstream scholars are not taken to task. Why not? Dr Maurice Casey has even written a book titled Is John’s Gospel True? He is said to be currently working on a book to attack mythicists. Presumably the book will be (like Bart Erhman’s) an argument for why we can believe the Gospels as testimony for certain historical facts about Jesus. Will Hoffmann challenge these scholars for falling into the same error as mythicists themselves, for perpetuating a misuse of the texts, and failing to focus on the meanings the Gospels were originally intended to convey?

The following two tabs change content below.

Neil Godfrey

Neil is the author of this post. To read more about Neil, see our About page.


If you enjoyed this post, please consider donating to Vridar. Thanks!


23 thoughts on “Demystifying R. Joseph Hoffmann, and the war over Bayes’ theorem”

  1. “I understand that the commenter is still waiting for Hoffmann to allow through his response explaining that he (Hoffmann) had just criticized the words of Dr Maurice Casey himself…”

    It appears that someone has been walking into an amusing trap. By concealing the true source of a statement cited as the basis for a claim, it is sometimes possible to extract an honest opinion free of bias from cultural and educational experience.

    Rhetorical snobbery is the subject of Ludvig Holberg’s play, “Erasmus Montanus” (1723). An English translation can be found at Project Gutenberg (www.gutenberg.org/ebooks/5749).

  2. History is a sequence of unique events. Probability theory simply does not speak to unique events. Probability theory is only applicable to a statistical ensemble.

    In other words, it seems to me like apologists have taken the statement, made by Robert Price and Bart Ehrman, that history tries to interpret what “probably” happened and used this abuse of the term “probably” to introduce nonsense math into the discussion in order to co-opt the legitimacy mathematics has. It’s a form of name-dropping, one of the favorite tactics used by Christian mountebanks.

    1. vorpal: “Probability theory simply does not speak to unique events. Probability theory is only applicable to a statistical ensemble.”

      First off, I am not a mathematician, and stats always gave me a brutal headache in grad school, so you can take what I say with a grain of salt. However, I think what’s missing in the above assessment of probability theory is the difference between frequency probability (which is the foundation of most university introductory statistics courses) and Bayesian probability. That’s to be expected, since if you’re a biologist or a chemist taking stats, you need to know how to apply statistical models to repeated experiments and make sense out of large data sets (the larger, the better).

      Without going into the excruciating details, Bayesian probability is amenable to the analysis of unique events, as long as you have a way to establish prior probability as well as some set of evidence to stack up against the prior probability. (I probably didn’t say that the way a real statistician would say it.)

      Essentially, Carrier is calling out NT scholars for saying that some event (like the Baptism or the Cleansing of the Temple) “probably happened” or “almost certainly happened.” Whether they like it or not, such a statement is an assertion of probability — they’re making a “math-y” statement based on gut feelings and flimsy criteriology.

      I’m for anything that will encourage NT scholars to use a formal, structured approach that will provide some basis for their assertions of historicity. I get very frustrated when I read works by scholars like E.P. Sanders who present evidence and then follow up with probability assertions. Is it really too much to ask them to “show their work”? How did we get from multiple attestation to “secure fact”? — via the plausibility catapult?

      1. The problem with all of this is specifying what the “events” are. You can always cherry pick what events count and what events don’t count.

        The number of ways for the statistics to be completely wrong is well beyond the capabilities of any historian. Consider the difficulty of getting accurate statistics today, even when the events are well delineated. Any reputable institution that publishes statistics probably has multiple PhDs in statistics on the payroll. Humans didn’t start to get this sort of thing right until the 20th century. It’s that easy to make an incorrect assumption.

        Carbon-14 dating is an appropriate use of statistics in history. Unless the use is something akin to this, it is almost certainly flim-flam. And, IMHO, calling it flim-flam is the only appropriate response to this sort of exercise.

        I think William Lane Craig peddles flim-flam. I think anybody who argues a ‘historical’ case along the lines that WLC does, is also peddling flim-flam.

        1. Selection of events is the subject of much philosophical discussion among historians. It is inevitable and necessary. No two historical narratives will be the same. History is not something that exists in some objective reality — it is a human construct. Real events do happen, but it is impossible to record everything that happens. Selection of certain events is necessary to create a narrative out of the mass of everything available.

    2. I don’t see history as an attempt to find out what happened, but to explain what happened. Finding out what happened is the job of an archaeologist, reporter, archivist, jury, etc. When a theologian speaks about “digging beneath texts” to find out what’s there he is not doing history at all, but is doing exactly what scholars do with Hamlet or James Joyce. They are doing literary analysis. The irony is that McGrath even says literary analysis is not history, yet that’s exactly what he is doing in another form with another set of questions himself. The Shakespeare scholar is trying to get into the psychology of Hamlet, or the mind of Shakespeare — to find the “real Hamlet”. So we have many different kinds of Hamlet on offer — from Freudian to existentialist and Marxist. You don’t see that sort of “digging behind the texts” to produce comparable results about the “real Julius Caesar”.

      The only fundamental difference between the Hamlet and Jesus studies is the understanding of sources. Shakespeare scholars understand the literary sources of Hamlet and that the play has nothing to do with (a/the) historical figure, while historical Jesus scholars, as the term indicates, presume the source of the narrative is a historical person. So they think they are doing history. But they are not doing anything fundamentally different in principle from what the Shakespearean scholar is doing.

      Bayes theorem is not really about “doing history” but about exposing the fact that the Gospels do not lend themselves to doing history at all. Hoffmann is right about the Gospels. But in his personal dispute with Carrier he has rejected a valuable tool that really was waiting to be used to his advantage.

      1. In high school, I didn’t enjoy history very much, since the emphasis was on what, who, and when. In fact, way back when I was attending college and I told my mother I was switching my major to history, she immediately said, “Oh, I could never remember all those dates.”

        My estimation of history as drudgery abruptly changed when I found that the emphasis in the “real” study of history is on significance. Professors would grade your work primarily on whether you understood the historical significance of an event, and secondarily on whether you could spell everyone’s name right and remember the exact dates.

        To your point, Neil, focusing on a theoretical historical Jesus has had marginal impact on “explaining what happened.” (Cf. the many Jesuses reconstructed by NT scholarship, all supposedly explaining the same set of evidence.) The greatest significance, from a historian’s perspective, of the New Testament surely must be the ways succeeding generations of Christians interpreted, understood, and applied what they found there.

        1. The reason historians have said history is an art is because selecting data and weaving information into a narrative is a creative and evaluating process. Every generation re-writes its history. Different groups fiercely debate their versions of events. But they are arguing about the significance or impact or meaning of the facts, or even what constitutes a relevant fact.

          HJ scholars don’t have facts of an empty tomb or crucifixion record, but they do have unprovenanced narratives of these. So those narratives are the data – not “an empty tomb”, for example. Carrier uses the same example. There is no fact of an empty tomb. Only a fact of a narrative of one. Ditto for Jesus himself.

          For a narrative to be thought to contain references to historical facts we need to have some context that can only be satisfied by provenance, external verification and genre/literary analysis. But I won’t repeat the details here for a zillionth time. We have all three for Socrates but only one for Jesus. So we cannot think of Jesus as someone outside the literary narrative as we can Socrates.

    1. Hoffmann is trying to avoid Bayes’ theorem by jumping to the conclusion to which Bayes’ theorem would, I believe, lead. That is, that the Gospels are not intended to be documents about historical facts.

      1. Right, that’s what I’m saying. Although he may want to leave the issue as a “we can’t know” kind of thing. His disagreement would be then that Carrier thinks he can argue towards the hypothesis that there is significantly better reason (or many reasons rather) to believe mythicism than agnosticism. But then we’d have to know how Carrier goes beyond where Hoffman starts. And they aren’t interacting at that level. But then again, Hoffman is obviously just venting some half-baked stuff and so I agree with you that there won’t be any meaningful “following it through.”

  3. While Hoffman has refused to allow my comments through moderation, he has allowed Steph’s through where she accused me of making up these quotes and did not identify the author, who disowns ever writing them.

    Aramaic sources of Mark’s Gospel – page 110 and page 165.

    It’s all there in Google Books, even if Steph refuses to identify the author, and even if Hoffman allows malicious slander by Steph through moderation , but refuses to allow sources to be identified and references.

    1. This is almost beyond belief. Thank you for explaining what has happened. One can only hope that, in the end, your comments will be let through moderation.

      1. Steph has graciously agreed that I was not making up quotes. She now declares that they are ‘out of context’- something that never bothered Hoffman when he was attacking those comments. Hoffman just lashed out, and never asked for context or sources. Why should he? It is only mythicists who have to put comments into ‘context’ before commenting on them,

        Steph is now accusing people of misrepresenting numbers ‘It should be blindingly obvious from this paragraph as a whole that his comment on not more than ‘51% probability’ is English English for ‘don’t really know’ or ‘haven’t a clue’, and has nothing to do with maths at all, let alone the misapplication of Bayes’ theorem to historical probability. He has been misinterpreted by people who cannot read Aramaic or Hebrew, and are not much good at Greek either, which is typical of the damage being done by mythicists on the internet.’

        So now we know. If an independent scholar uses phrases like ‘51% probability’ , then you are misrepresenting them if you don’t realise this ‘has nothing to do with maths at all.’;

        What’s a poor maths teacher supposed to do when leading New Testament scholars declare that ‘51% probability’ has nothing to do with maths at all? I shall have to go and tear out all those pages in my maths textbooks which discuss percentages….

        1. Oh, and Steph doesn’t dare apologise to me for claiming I made up quotes.

          When caught red-handed in malicious slander, she never dreams of apologising for her slanderous remarks, no matter how foolish they make her look.

  4. I assume Steph is unaware that , as an amateur, I actually take the trouble to look at sources, and rarely get my facts wrong.

    no wonder she has nothing left except malicious slander. No arguments, just slander, slander, slander….

    1. Steph sadly cannot even demonstrate that she understands the discussion. She fails to see she supports my point when she declares Casey wrote “Is John’s Gospel True?” to argue it is not factually true. That is the very debate Hoffmann says we should bypass, but I am not surprised Steph herself fails to understand Hoffmann’s post that she says she fully agrees with.

      On a more amusing note I do find a smile on my face each time I read remarks by the likes of Steph, Tim and others that I am in effect a “mere librarian”. Most academics owe a lot to reference and other librarians. It is clear Steph has absolutely no idea what I do or the nature of my professional standing nationally and beyond, but that is understandable in a world where many jobs defy simple labels nowadays. I actually never see or deal with books at work. My job is to assist with developments of new technologies and metadata schema and ontologies to enhance the accessibility of cultural resources online and to help build systems that enhance the availability of academic research publications and datasets. I am also on two national metadata advisory committees. So when students and scholars belittle my profession they truly have no idea they are insulting one of their profession’s major supports for the present and future.

      1. Stephanie Louise Fisher has kindly posted a big chunk of Maurice Casey’s writings so that people can confirm just how greatly mainstream Biblical scholars rely on arguments from silence, which they deploy as though no ink need be wasted justifying an argument from silence.

        Who can the doubt the validity of arguments from silence, when renowned scholars deploy them routinely?

        ‘On pages 208-9 of Is John’s Gospel True?, Casey wrote (and obviously this is only part of his argument),
        “The Lazarus story is a Johannine composition from beginning to end (n.15. see pp. 55-7). The narrator tells us that many of `the Jews’ believed in Jesus because of this miracle (11.45). The reaction of the chief priests and the Pharisees is remarkable. They convened a sanhedrin and said, `What are we doing? – for this man is doing many signs. If we let him go on like this, everyone will believe in him, and the Romans will come and destroy our place and people’ (11.47-8). Widespread faith in Jesus would not have given the Romans cause to do this. This is an extraordinary perception, formed by the Neronian persecution, which showed genuine Roman hostility to Christianity, and by the destruction of Jerusalem after the Roman war of 66-70CE.

        Some Jews attributed this to failure to observe the Torah, and Christians did not observe it. From this perspective, everyone having faith in Jesus could indeed lead to the destruction of the place and the people. This perspective has however no place in the Judaism of 30 C.E.. It leads through the prophecy of Caiaphas to the decision to have Jesus put to death. This is also profoundly ironical. Jesus has been presented as the Resurrection and the Life, and the source of life to those who believe in him. His gift of life to Lazarus is now presented as the reason why the chief priests and Pharisees seek to have him put to death.

        After the anointing story, things get worse and worse. At 12.9-11, many were leaving `the Jews’ and believing in Jesus, and consequently the chief priests took counsel to kill Lazarus. This begins a set of statements, according to which Lazarus was exceptionally important.

        If this were true, we would not be able to explain the omission of Lazarus from the synoptic Gospels.

        (Another argument from silence – Carr)

        Secondly, the plot is incredible. Killing someone raised from the dead is not a feasible Jewish reaction to such a miracle, and the plot is never mentioned again.

        (Yep, another argument from silence – Carr)

        It either worked or it did not. It is difficult to see how the plot against Lazarus could fail, when that against Jesus succeeded. Nonetheless, it is not acted upon , yet Lazarus does not reappear in the early chapters of Acts.

        (Another argument from silence – Carr)

        Nor does he appear again in the fourth Gospel, surviving an unsuccessful plot.

        (Another argument from silence – Carr)

        Finally, in the Judaism of Jesus’ time, having faith in Jesus did not mean `leaving’ in any reasonable sense. The fourth evangelist has imposed on the Judaism of Jesus’ time the situation of his own, when Jews converted to Jesus did indeed leave the Jewish community.

        But the narrator has not yet finished. Verse 12.12 slides into the old tradition of 12.13-15. More trouble begins at verse 16, where the disciples are to `remember’ what they had not previously known. It becomes serious in verses 17-19, where the crowd bear witness that Jesus had raised Lazarus, so the Pharisees declare, `the world has gone after him.’ Lazarus, however, is heard of no more. The Johannine narrative is thus internally incoherent, as well as inconsistent with synoptics. The decisive incoherence is that the story of Lazarus just stops. With so many Jews `leaving’ because of the raising of Lazarus, with the crowd who saw this miracle bearing witness to it, with a crowd meeting because they have heard of this sign, with a plot against Lazarus’ life, Lazarus was such an important figure that his further presence, and his fate, were bound to have been recorded.

        (Another argument from silence – Carr)

        But they are not recorded.

        (Another argument from silence – Carr)

        Why not? The only possible explanation emerges from the absence of Lazarus from the synoptic Gospels.

        (Another argument from silence – Carr)

        His fate is not recorded because he never was an important figure.

        (Another argument from silence – Carr)

        He does not turn up in Acts, and he neither wrote nor figures in any epistle, for the same reason.

        (Another argument from silence – Carr)

        This also tells us something about the way in which this Gospel has been written. The profound and real feeling that Jesus brought life and `the Jews’ brought death (cf. 16.2) to the Johannine community is presented in story mode. Hence the stress on the love of Jesus for Lazarus, as even `the Jews’ notice (11.36), and for Martha and Mary (11.5), for Jesus loves his disciples. Hence also the narrative precedents for Jesus’ own resurrection, especially the difference in the graveclothes, for Lazarus came forth bound (11.44), whereas Jesus left the graveclothes behind and vanished, a difference great enough for a disciple whom Jesus loved to come to faith (20.7-8) Such factors have quite overridden the historical inconsistincies which we can see.”

        1. Stephanie’s post quoting Casey is at http://rjosephhoffmann.wordpress.com/2011/06/06/%CF%80-ness-envy-the-irrelevance-of-bayess-theorem/#comment-3572 . In the same page are several others posts by Stephanie addressing Steven’s and my own remarks.

          It is pointless arguing with her since she takes personally any questioning of the most fundamental assumptions or logical validity of certain scholars and responds with knee-jerk non sequiturs, failing to grasp, and therefore failing to address, the central points of those she attacks.

  5. For a reconsruction of origins of the Jesus Tradition see the first 13 comments to The Importance of the Historical Jesus – The New Oxonian. It is in the form of a letter I mailed to R. Jpseph Hoffmann about the Jesus Project. After two separate mailings it became evident that CFI was not going to pass it on. I then posted it as comments. Several brief comments reflect concerns over a 5 day delay before it was published, so to be ignored – also many typos. It is based largely on quotes from the works of Scubert Ogden, James M. Robinson and Hans dieter Betz. It just may be of interest.

    published, so are to be ignored/ Also the typos.

Leave a Comment

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Discover more from Vridar

Subscribe now to keep reading and get access to the full archive.

Continue reading