Book Review: Richard Carrier’s ‘Proving History: Bayes’s Theorem and the Quest for the Historical Jesus’

Book Review

Proving History: Bayes’s Theorem and the Quest for the Historical Jesus (New York: Prometheus Books, 2012), by Richard Carrier

Table of Contents (Reviewed chapters are bolded)

  1. The Problem
  2. The Basics
  3. Introducing Bayes’s Theorem
  4. Bayesian Analysis of Historical Methods
  5. Bayesian Analysis of Historicity Criteria
  6. The Hard Stuff

1. Some Caveats:

I’ve had the pleasure of knowing Richard Carrier for the past seven years or so.  Over the course of our friendship we’ve exchanged a lot of ideas and had some interesting conversations, and at times we haven’t seen eye-to-eye (I think he drops one too many ‘F’ bombs on occasion, but maybe I’m just a bit more sensitive after hanging around with septuagenarians and Baptists in the academy).  I’ve read a lot of his work and, suffice it to say, with the exception of perhaps Thomas L. Thompson, no one has had more influence upon my academic pursuits than Richard.  I hope he takes that as a compliment (he’s also read a lot of my work, so goodness knows he probably doesn’t want that blame)!  So when he told me some years back of his two-volume project–one volume handling historical Jesus criteria and the other dealing with the historicity of the figure of Jesus–I couldn’t wait to get my copies.

To be clear, my enthusiasm for the release of these volumes has nothing to do with blind or uncritical devotion to a particular individual or agenda.    I don’t belong to a school (though I do consider myself a minimalist and had the pleasure of working with others who call themselves minimalists as well), I don’t belong to any label (I am not an atheist like Richard, nor am I out to disprove Christianity or any such nonsense), and whether or not the figure of Jesus has a historical core is of no serious consequence to me or my research.    But I do have my own biases.  And this is where I think the Richard’s preface to his work needs to be highlighted.

I like the Jim West Method of Book Reviews, so I shall continue to update this page as I read through and review each section rather than creating a new post for each section I review.

2. Initial Impressions of Proving History and a Brief Review of Carrier’s Preface

This book appears to be a triadic tour de force; in this ambitious volume alone, Carrier educates his readers (1) on the nature and value of doing proper critical history, (2) on the value of Bayes Theorem and logical analysis, and (3) on the current methods of Historical Jesus scholarship.  No wonder he had to split the content into two volumes!  I am not as math-inclined as I should be–a flaw among many in the field of Biblical Studies–though I have some standard knowledge about the basis of Bayes Theorem.  My  grasp of it is still pretty basic and I’m looking forward to reading those chapters dealing just with the general applications of the theorem (For those curious, Carrier has a brief explanation of its use here, a blog post here with all sorts of nifty links, and, for those not easily offended, a video demonstration from Skepticon IV).

Now onto the Preface.  He is open and honest about the purpose of the book, which he writes:

All historians have biases, but sound methods will prevent those from too greatly affecting our essential results. No progress in historical knowledge, in fact, no historical knowledge at all, would be possible without such methods. Hence, the aim here is to develop a formal historical method for approaching this (or any other) debate, which will produce as objectively credible a conclusion as any honest historian can reach. One need merely plug all the evidence into that method to get a result. That’s a bold claim, I know; but the purpose of this book is to convince you, and if in the end you are convinced, provide the background necessary to implement the method I propose. All I ask is that you give my argument a fair hearing.

As noted above, Richard admits he is not without bias, as none of us–no matter what we want to believe–are without it.  One of my favorite lines in his preface, and one I have used on occasion, is when he outlines presuppositions within historical Jesus scholarship.

I have always assumed without worry that Jesus was just a guy, another merely human founder of an entirely natural religion (whatever embellishments to his cult and story may have followed). I’d be content if I were merely reassured of that fact. For the evidence, even at its best, supports no more startling conclusion. So, I have no vested interest in proving Jesus didn’t exist. It makes no difference to me if he did. I suspect he might not have, but then that’s a question that requires a rigorous and thorough examination of the evidence before it can be confidently declared. Believers, by contrast, and their apologists in the scholarly community, cannot say the same.

Some may argue that rigorous and thorough examinations have already taken place, but remember that the purpose of Proving History is to analyze the current criteria to determine if they are even sound!  Time, reading, and careful analysis of Carrier’s arguments will determine if he has been successful in that regard.  So far I believe the work has a lot to accomplish.  Looking forward to getting into the first chapter soon enough.

3. Chapter 1: The Problem

Carrier follows up his preface with an overview of the problems, that is to say, the issues that currently plague historical Jesus research.  Rhetorically, this is a great place to start.  While most scholars are probably already aware that problems exist, I doubt many laypeople are aware of them, and even scholars might not know just how deep seeded the problems are and how much their conclusions are hindered by them.  But Carrier aply points out, and argues this convincingly that none of these problems are new, nor have they been avoided in the past.  I’ve argued similarly, along with Thomas L. Thompson, in our forthcoming edited volume ‘Is This Not the Carpenter?’.

He also notes that scholars have done little more than develop additional methods and produce additional portrayals of Jesus which, if anything, have countered any attempt for clarity and any call for a more narrowed consensus.   But Carrier presses the point even further, stating quite directly:

When everyone picks up the same method, applies it to the same facts, and gets a different result, we can be certain that that method is invalid and should be abandoned. Yet historians in Jesus studies don’t abandon the demonstrably failed methods they purport to employ. This has to end.

And his conclusion to this point is one I found particularly refreshing:

If historians can’t agree on what that method [i.e., the best method – ed.] should be, then their whole enterprise is in crisis, because agreement on the fundamentals of method is the first essential requirement for any community of experts to deem itself an objective profession.

I find that last bit to be dead on.  Indeed, his point is two-fold here: if the methods used are coming to different results then clearly the methods are determined subjectively, not objectively.  While many historians might agree with that claim, Carrier’s solution is simple and obvious; if subjective methods aren’t working, and they clearly aren’t, then the next step must be more objective.  And I’m sure that many of my colleagues would agree that subjectivity is indeed dangerous when one is making an objective type of claim (i.e., that ‘Jesus existed historically’, rather than say ‘Jesus may have existed historically’–but more on this when I review a later chapter).

Carrier’s solution, as you already know, is Bayes’s Theorem.  He makes it clear that in the coming chapters he will defend various aspects of the structure of the Theorem, its application, and its value overall not just for historical Jesus scholarship but for historical research as a whole.  But before he can even do this, he must make it clear that everyone is on the same page.  This makes sense, since carrier’s method appears to rely heavily on logical constructs and statistical math–to bring everyone to the same level of understanding before proceeding with the meatier data is a no-brainer.

4. Chapter 2: The Basics

The function of Chapter 2 serves as a means to get everyone on the same page with how historical research, that is proper historical research, is done (or perhaps, specifically, how carrier views historical research). Carrier lays down what he considers to be the 12 axioms of professional historical investigations.  He writes that these axioms should be accepted by everyone in the field, and they aren’t at all that hard to accept. Notions like ‘All conclusions muist logically follow from evidence available to the observer’ to ‘overconfidence is fallacious; admitting ignorance or uncertainty is not’ are general common attributes of all historical inquiry and most of Carriers colleagues will have no trouble
agreeing with them.

Next he produces the 12 rules.  The rules are not axiomatic, but Carrier recommends these rules be followed to the best of human ability (though he admits that everyone fails to follow them from time to time, including him).  These rules include the following of the 12 axioms, continuing to better ones understanding of the periods they study, confirming all interpretations using the original languages, and so on.  Again, these may be idealized rules but, overall, Carrier writes that to better oneself as a historian, we should strive to follow these rules to the letter.  I don’t think that many scholars would disagree with these either.  Though I can understand when Carrier urges us not to assume what isn’t in evidence.  It seems very simple, but it is a very easy rule to break.

The interesting part of this chapter is that Carrier uses his background as a professional philosopher to bring to the table a set of ethical clauses for the historian that I think are often taken for granted.  While scholars know they should be following the logical path of evidence, some don’t actually verify they are but presume it.  I see this mistake made consistently by maximalists (particularly those who argue for, say, a historical Moses, or a historical united monarchy, or a historical patriarchal period) and also with historical Jesus scholars (they’ll make the fallacious mistake of ‘possibly, therefore probably/certainly’ as Ehrman makes in his book Did Jesus Exist [Introduction, p. 4]).

After Carrier is done setting out these axioms he then goes into explaining the basics of Bayes’s Theorem.

5. Chapter 3: Introducing Bayes’s Theorem

For this section, I recommend a piece of scrap paper.  Though Carrier does a fantastic job of simplifying everything, it may be useful to write down the formula so you can make notes around it as you read through this section.  This will be extremely helpful as you go through the rest of the book as a reference and a good way to use the formula as you go (and in turn to gain experience using the Theorem).  I know most historians probably do not need scrap paper to read, but one of the values of Carrier’s book is not simply that it educates you about this new method, but engages you to think about history while utilizing it.  So it is a bit of an educational tool in this regard which, in all honestly, makes me feel like Carrier should have charged more (as it is a tutoring device in the use of historical statistics as much as it is a guide for the historian). That said, let’s move on to the meat of the chapter.

Carrier starts off his chapter with an account of the disappearance/eclipse of the sun from the synoptic Gospel accounts which takes place during the passion narratives.   He analyzes in typical historical form why this account is improbable and unlikely.  He notes, rather brilliantly, that the Gospels talk of this event as having an effect over the whole world, yet even with how unlikely it is that it occurred (indeed, that is probably didn’t), not a single person argues against it or points out how erroneous the Christians were for including it in their narratives:

This entails, in turn, that the Gospels, even from the very beginning, contain wildly unbelievable claims of inordinately public events that in fact never occurred, yet were never gainsaid by any of the millions of witnesses who would surely have known better.  I’ll consider the significance of that fact in my next volume.

We return to this event at the end of the chapter, but before that he explains some rather important details to the reader.  Carrier’s followup section deals with the intersection of science and history and raises some very important points, like the fact that science and history are in many ways symbiotic systems.  Science is dependent upon history to maintain and catalog events of the past (especially studies) and history is dependent upon science for a strong and useful method with which to maintain accurate information about the past for future scientific studies.

A point I agree with completely that is worth noting in this review is his statement that essentially we are analyzing theories which address now just what happened, but why it happened.  In effect, he argues, the function of the historian is not that different from the function of a scientist in that we seek to understand and explain three basic questions of our subject of interest:

(1) If our theory is false, how would we know it?

(2) What’s the difference between an accidental agreement of the evidence with our theory, and an agreement produced by our theory actually being true—and how do we tell the two apart?

(3) How do we distinguish merely plausible theories from provable ones, or strongly proven theories from weakly proven ones?

Carrier states that Bayes’s Theorem is what is needed to effectively answer them in the best manner possible.

And he starts the next section off by directing readers to important studies in Bayes’s Theorem as a method for historical research which I find very useful.  But he also makes the very astute point that we actually intuitively already use Bayes’s theorem without even knowing it.  When we make the claim that something is ‘improbable’ or ‘plausible’ we are actually using a very basic form of Bayes’s, but we often fail to account for the rest of the formula which is necessary for determining if what we’re stating from intuition is arbitrary or sound.  This, Carrier argues, is why actively using Bayes’s is far more superior, and produces superior results, than simply guessing.  Even when we make educated guesses they are still merely guesses at best, which is what we do when we arbitrarily apply a statement of probability–like saying something is ‘likely’–to an event of the past without using the theorem.  Which overall is a very strong motivation for using the theorem in the first place. Interestingly enough, he raises another point which I had forgotten, which is that Bayes’s theorem is already used in archaeology to analyze the data.

The bulk of this section then goes into detailing the what’s-what of the Theorem, breaking down every part in lay terms.  this is where the scrap paper comes in handy.  The formula looks like this:

P(h|e.b) = __________P(h|b) x P(e|h.b)__________
[ P(h|b) x P(e|h.b) ] + [ P(~h|b) x P(e|~h.b) ]

It may look daunting (which even Carrier admits), but he does not fail to explain the meaning in a way even the most mathematically challenged could understand.

And thus he returns to the sun disappearance narrative and analyzes the same discussing using Bayes’s Theorem.  Carrier argues:

With BT, instead of myopically working out how we can explain all the evidence “with our theory,” we start instead by asking how antecedently likely our theory even is, and then we ask how probable all the evidence is on our theory (both the evidence we have, and the evidence we don’t) and how probable all that evidence would be on some other theory (every other theory that has any claim to plausibility, but especially the most plausible alternative). Only then can we work out whether our theory is actually the best one. If we instead just look to see if our theory fits the evidence, we will end up believing any theory we can make fit. And since that will inevitably include dozens of theories that aren’t actually true, “seeing what fits” is a recipe for failure. In fact, this is worse than failure, since we will have deceived ourselves into thinking the method worked and our results are correct, because “see how well the evidence fits!” That’s the result of failing to take alternative theories of the evidence seriously. That this is exactly what has happened in Jesus studies (as shown in chapter 1) should be proof enough that historians need a new method. One that actually works. And as far as I can see, BT is the only viable contender.

He then goes on to explain why Bayes’s theorem is so important and why math is exceptionally important to historical research.  And he isn’t the first do so, but he is the first to apply these methods-so far as I can tell-to Biblical Studies.  He even goes so far as to address all the main criticisms he has received about the use of the theorem and uses math to show how those criticisms fail.

This chapter is just too good to ignore.  I want to write more but if I do I’d essentially just be repeating what Carrier argues in a less functional and coherent manner.  Frankly, I am looking forward to the rest of the book.

More anon.

10 Responses

  1. Thomas, very well done introduction to Carrier’s new book. Thank you very much. I am waiting for the book’s arrival in my mailbox.

    He visited the Humanist Society of Greater Phoenix in November of 2009 and spoke on ancient Greek and Roman science, and the role of Christian church over the subsequent millennium. Well received by all.

  2. Thanks! I still have the rest of the book to blog through. As I read more I will continue to update this post to incorporate the rest of the review.

  3. It will be great to hear your views.

  4. Bart Ehrman has written a rather snarky article about Jesus mythicists over at the Huffington Post.

    Perhaps Tom could comment on it?

    (I also posted this comment on Richard Carrier’s blog to hopefully get his thoughts also)

  5. I’m not a mythicist, so I’m not the best person to respond. I would talk to someone like Neil Godfrey about that.

  6. […] such a moronic point of view because of Ehrman’s book.”Also relevant to this topic is Tom Verenna’s review of Richard Carrier’s new book on historical methodology and Bayes&#…, and Mike Kok’s post on Dale Allison’s treatment of the passion narratives, and how […]

  7. Great start! My copy should be in the mail soon…

  8. […] Book Review: Richard Carrier’s ‘Proving History: Bayes’s Theorem and the Quest for… […]

  9. Excellent review so far, Thomas! Please keep updating! :-)

    I would be very interested if you mentioned any spots where you disagreed or thought Carrier makes any weak points or could have done a better job making any points. It just makes it a bit easier to get a sense of a book if a review contains at least something critical (or at least says, “I can’t think of anything to be critical about!”), so that one can get a sense of the balance of the review, and hence the balance of the book. :-)

    So far, though, your review is making me very glad to hear that he’s tackling things in the right order and dotting all the i’s and crossing all the t’s. I so wish I had any money to buy this book! Maybe I’ll have to forego some ramen noodles for a few weeks, lol.

    Minor nitpick: Noticed a couple spots where ‘carrier’ is not capitalized.

    P.S. Your review shows up third in Google for the unquoted search ‘proving history’ (for me, anyway; I know Google tweaks its results based on one’s past search history, so my result might be anomalous).

  10. Second Thaumas’ “please keep updating”.

    I’ll be interested to see what approach Carrier takes with Bayes’ Theorem. Speaking as a maths grad and skeptic-with-a-k, I get a bit uncomfortable when people try to feed qualitative arguments into BT – it seems like cherry-picking would be a major issue here. For example people have tried to use Bayesian reasoning to prove God’s existence – argh.

    There are undoubtedly good, rigorous uses of BT. For example it’s used in insurance to adjust industry-wide life tables to particular schemes’ mortality experience. But it’s also quite easy to abuse as a tool for confirming our prejudices.

This blog is no longer in use; NO comments will post.

Please log in using one of these methods to post your comment: Logo

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

Google photo

You are commenting using your Google 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 )

Connecting to %s

%d bloggers like this: