Why there probably wasn’t a guard at Jesus’ tomb

Christian apologist Professor Tim McGrew recently defended the historicity of Matthew’s account of the guard at the tomb, in a post put up by his wife, Dr. Lydia McGrew. Professor McGrew’s post was written in response to a challenge he issued to me, in response to my (generally positive) review of Michael Alter’s book, The Resurrection: A Critical Inquiry (2015), which was published at The Skeptical Zone last year. Not wishing to address the bulk of Alter’s arguments, which he considered unconvincing, Professor McGrew challenged me to narrow the focus of our discussion, by listing three of Alter’s arguments which I had found particularly convincing. The first topic on my list which Professor McGrew chose to address was the question: was there a guard at Jesus’ tomb? However, it turns out that McGrew’s argument for the historicity of Matthew’s story of the guard is based on faulty math – a surprising flaw, coming from a man who has written extensively on the subject of Bayes’ Theorem and its role in Christian apologetics. Before we have a look at the math, though, I have a special announcement: Michael Alter himself has decided to weigh in on the controversy, and I have included his remarks in this post.

Curious, that!

Before I continue, let me begin by pointing out four curious facts, which may be of interest to readers.

First, I actually wrote a comment in reply to Professor McGrew’s post, which was never published. Funny, that. Now, I don’t wish to complain about this – after all, the people in charge of What’s Wrong With the World (the blog on which Professor McGrew’s post can be found) have every right to make their own rules, as it is their blog. However, I will note for the record that since the tone of my reply was entirely civil and (I believe) in keeping with the blog’s stated posting rules, I had a reasonable expectation that it would be published.

Second, in his post, Professor McGrew only quoted the first half of my argument, omitting the second half, which I consider to be by far the stronger half. Let me be clear that I am not accusing Professor McGrew of making a deliberate omission, as he subsequently attempted to rectify his omission with a comment attempting to address the second half of my argument. Nevertheless, the original omission betokens a certain carelessness on Professor McGrew’s part: a dangerous failing, when one is engaging in controversy.

Third, the list of three “test cases” displayed in Professor McGrew’s post which we agreed to debate is not the same as the list which I originally provided him. Here is his list of the three topics to be covered:

Since I am unimpressed by Alter’s arguments, I asked Torley to pick three particular arguments as test cases… Torley chose the three following points for this test:

1. Was there a guard at Jesus’ tomb?

2. Did Jesus’ mother and the beloved disciple stand at the foot of the cross?

3. Was Jesus buried in a new rock tomb? (specifically, a tomb owned by Joseph of Arimathea)

However, in actual fact, I offered Professor McGrew a choice of two topics for the first test case: (i) Were Jewish saints raised at Jesus’ death? or (ii) Was there a Guard at Jesus’ tomb?

Professor McGrew then chose the former option, in an early response to my review of Michael Alter’s book. I am therefore a little puzzled as to why he has suddenly changed course, and chosen the latter option instead. Now, Professor McGrew has a perfect right to change his mind, and I presume he has done so for tactical reasons. That is his privilege. Nevertheless, a public acknowledgment of this change of tack would have been courteous to his readers.

Finally, Professor McGrew made no attempt to notify either me or Michael Alter about the post which he had authored. I stumbled on it by accident a short time after he had written it, as I occasionally peruse the What’s Wrong With the World blog for its articles of topical interest. Michael Alter heard about the post from Professor Joshua Swamidass, who thought he might wish to respond – which he did. I hereby invite readers to carefully examine Michael Alter’s response, which I have included in this post. Here it is, in full. Below Michael Alter’s response, I have written my own detailed, mathematical response to Professor McGrew’s defense of the historicity of Matthew’s story of the guard.

=======================================================

CORRECTED and UPDATED RESPONSE TO TIM MCGREW’S “Was there a guard at Jesus’ tomb?” by Michael Alter

Recently, Dr. S. Joshua Swamidass kindly sent me an e-mail providing information that Dr. Tim McGrew had published a guest post response to Vincent Torley’s review of my text, The Resurrection: A Critical Inquiry (2015) in his wife’s group blog: What’s Wrong With the World. First, I am honored that a respected, knowledgeable, and published authority has taken time out of his busy schedule to respond to Vincent Torley’s review of my text. His time is respected.

Now then, let me respond to several points addressed by Tim:

  1. Tim wrote: Since I am unimpressed by Alter’s arguments…

RESPONSE: Tim is unimpressed of Vincent’s review / summary of my text. His opinion is absolutely respected. However, it is significant that presumably, Tim has not examined my text. At least this is my gut feeling after having read his introduction. In my opinion, if Tim has read my text, he should have made this point clear to his readership. If I am in error, please correct me and excuse me. Therefore, it must be repeated for emphasis that he is only responding to Vincent’s review of my text. Perhaps, it might be harsh, but imagine a movie critic critiqued a movie without seeing it. Or, imagine that a music critic published a review of the Cleveland Orchestra’s performance of Beethoven’s 9th symphony without having heard and seen the actual performance. Finally, imagine that a chess expert analyzed a chess match without having witnessed or seen the chess notations of that match. In the three examples just identified, the evaluation was merely based on an earlier reviewer’s published review. Question: Do you think that the evaluation is fair?

2. It [the guard at the tomb] is mentioned only in Matthew’s Gospel, not in the other three… the argument from silence in such cases is generally terribly weak… As Torley has not attempted to argue that the silence of the other evangelists meets the probabilistic challenge laid out there, I will not belabor the point.

RESPONSE: To the contrary, numerous bible commentators (on both sides of the religious aisle) doubt or question the historicity of Matthew’s account of the guard at the tomb (Dale Allison, C H Dodd, Raymond Brown, R H Gundry, The New Catholic Encyclopedia, etc. On pages 297-299, I offer several “SPECULATIONS” for Matthew’s rationale for inventing the Guard Episode. Although various explanations are possible, I elected to focus on, and explore the rationale discussed by Elaine Pagels. Unfortunately, in my opinion, you did not evaluate her writing.

Most significant, the historicity / veracity of the tomb itself is a point of contention and discussion. On pages 337-339, I identified at least fifteen scholars, theologians, and historians. Numerous times, I include their commentary. Yet, you chose not to belabor the point. Your decision is respected, however, it is NOT fair to my text.

3. Torley objects that the account does not explain why the body could not have been stolen on Friday night.

RESPONSE: On pages 340-343, I specifically presented William Lane Craig’s discussion on this specific topic. After presenting his apologetic, I offer my rebuttal. Your essay responds to Vincent’s rationale, it does NOT respond to my rationale. Once again, this is not fair, and disingenuous to your readers.

4. You wrote: “The third objection is that Matthew’s narrative does not tell us why Pilate would acquiesce in the request of the Jewish leaders.”

RESPONSE: Pilate’s rationale is subject only to scholarly speculation. And yes, you offer several thoughtful ideas on this topic. Thank you! But, these too, are just scholarly speculations. However, here too, these speculations are depended upon the historicity / veracity of a tomb burial. Furthermore, there is scholarly speculation as to the meaning of Pilate’s words: “Take a guard,” or “You have a guard.” That topic, too, is discussed on page 294.

5. You wrote: The fourth objection is that the Jewish leaders would not have asked Pilate to set a guard at the tomb, since it was the Sabbath day, and Jewish law would have forbidden them to hire a gentile to do such work on the Sabbath.

RESPONSE: You added: “First, even supposing the objection to be fairly stated, there is no guarantee that the Jewish authorities would be particularly scrupulous in the matter of hiring a Roman guard to do their work, as they had already shown their willingness to hold a trial by night in prima facie violation of their own rules.” To be one hundred percent honest, this statement makes me cringe. Is it possible that this invented episode (and the trials) is/are, in fact, an argumentum ad hominem against the Jewish leadership (Jewish people)? If you excuse me, you continue the myth of the degradation of the Jewish leadership. I will be the first to admit that not everyone is “wonderful”… And, the Tenakh is clear that numerous times the Jewish people have fallen short of the mark / not been Torah faithful. However, numerous commentators, on both sides of the religious aisle frankly discuss plentiful examples of anti-Semitism recorded in Matthew, and elsewhere in the Christian Bible. Unfortunately, since you presumably (again, I could be in error) have not read my text, you did not comment on pages 343-344. Not to hit a dead horse, but this failure on your part is not fair and it is disingenuous to your readers.

In closing you write: “I conclude that on the first point, Alter’s argument, as summarized by Torley”… This reminds me of a famous quote by the Jewish poet Haim Nachman Bialik. Hopefully, the analogy will be self-evident. “ Reading the Bible in translation is like kissing your new bride through a veil.” Those who understand, will understand…

Take care.

Mike

===========================================================

I’d like to thank Michael Alter for his well-written response to Professor McGrew. I’d now like to address the substance of Professor McGrew’s argument for the historicity of Matthew’s story of the guard at Jesus’ tomb.

A preliminary observation

First of all, I’d like to draw Professor McGrew’s attention to a remark I made, in the same comment where I posed my challenge to him:

In what follows, the question I’d like to address is not whether these claims are true, but whether or not they are probable, when judged on purely historical grounds, by a fair-minded historian with no anti-supernaturalistic bias. [The emphases have been added by me – VJT.]

In his post, Professor McGrew concludes that “Alter’s argument, as summarized by Torley, completely fails to undermine the credibility [of] Matthew’s account of the setting of a guard at the tomb where Jesus had just been buried.” But all that goes to show is that Matthew’s story of the guard is possible, not that it is historically probable. More is needed.

I should also note that Professor McGrew’s post contains no less than four instances of “might have” or “might still have.” Once again, this is not the sort of language one employs when attempting to build a probabilistic case.

So where’s the beef?

Surprisingly, the real substance of Professor McGrew’s argument for the historicity of Matthew’s story of the guard at Jesus’ tomb is not contained within his post, but in a comment he wrote four days later, in response to a critic. Here’s the relevant excerpt:


It is a matter of balancing probabilities and inclining to the most likely. There are three independent variables here: the prior probability that a guard was set, P(G), the probability of our having the Matthean account, given that a guard was set, P(M|G), and the probability of our having that account, given that a guard was not set, P(M|~G). I contend that, on the basis of such information as we actually possess, P(G) is not particularly low, and therefore the ratio P(G)/P(~G) is not significantly less than 1. I have disposed of Alter’s attempt to argue to the contrary. P(M|G) is not itself wildly low; if that is what happened, this is more or less the sort of account we might hope to have of it. P(M|~G), however, is very low; I cannot see why anyone would think it is even on the same order of magnitude as P(M|G). Therefore, P(G)/P(~G) ≈ 1, and P(M|G)/P(M|~G) >> 1; therefore, P(G|M)/P(~G|M) >> 1; therefore, P(G|M) is easily more likely than not.

That’s all.

Professor McGrew’s mathematical reasoning is sound, given his three premises, which are that:

(i) the prior probability P(G) of a guard being set over Jesus’ tomb is not particularly low;

(ii) the probability P(M|G) of our having Matthew’s account, given that a guard was set, is not wildly low – indeed, M is something we might expect, given G;

(iii) the probability P(M|~G) of our having Matthew’s account, given that a guard was not set, is very low – in fact, orders of magnitude lower than P(M|G).

In a nutshell: the premises which I contest are premises (ii) and (iii). I think P(M|G) is very low, while P(M|~G) is low, but nonetheless greater than P(M|G) – even on the assumption of an empty tomb.

Why Matthew’s story actually makes more sense if there were no guard

Let’s address Professor McGrew’s premises in reverse order, and start with premise (iii). Why do I contest McGrew’s claim that P(M|G) is massively greater than P(M|~G), while conceding that P(M|~G) is low? Briefly, there are two probabilities we need to consider when looking at the story of the guard in Matthew’s gospel: the probability of such a story being created in the first place, and the probability of its being circulated. The latter is of vital importance, because even a true story will never end up in anyone’s gospel, unless it circulates well enough to reach the ears of the gospel’s author. Now, if there actually were a guard at Jesus’ tomb, then the creation of the story poses no problem: Matthew is simply narrating what actually happened. However, the circulation of a story like the one we find in Matthew’s gospel is extremely unlikely, since it would have been counter-productive for the guards to spread it, as they would have been putting their lives in danger by doing so, for reasons we’ll discuss below. And even if we throw in the additional fact of the empty tomb, and grant that the tomb of Jesus was mysteriously opened while the guards were on duty, one would still not expect them to circulate the story which Matthew claims they did: “His disciples stole the body while we were all asleep.” To quote Alter, “it would have made much more sense for the Sanhedrin to have told the soldiers to say nothing at all or to report that everything was in order and that they had left at the proper time (Schleiermacher 1975, 430).”

On the other hand, if there were no guard at Jesus’ tomb, then there would have been no obvious motive to invent the story of the guard, in the first place. In his book, Alter suggests possible motives, but they are simply that: possibilities. (Incidentally, while we’re on the subject of possibilities, I should mention another interesting hypothesis for the origin of the guard story in Matthew, suggested by Alter in his book (2015, p. 342): “Matthew creatively and skillfully weaves a legendary account incorporating passages from Joshua 10 and Daniel 6 that are supposedly fulfilled by Jesus.” Make of it what you will. All I will say is that the account in Joshua 10:16-27 resembles Matthew’s account in several respects: it features a cave whose mouth is covered up with large rocks, with bodies inside [live ones, in the book of Joshua], and several guards posted outside.) However, once the story of a guard at Jesus’ tomb had been invented (for whatever reason), there would have been no powerful reason not to allow it to circulate freely, as it would have jeopardized no-one, meaning that no-one had any motive to suppress it.

In short: the main hurdle for P(M|~G) to overcome is its creation: why invent such a story in the first place? Once created, however, it could freely circulate. P(M|G), on the other hand, faces no creation hurdle, but a massive circulation hurdle, for two reasons: (a) circulating the story posed a real danger to the guards, who were allegedly the first people to propagate it, since (as we’ll see below) spreading the story put their lives at risk, and (b) the guards would have been better served by spreading another story, instead: “Nothing happened.” Had this lie been too difficult to sustain in the face of contrary evidence, then how about this one: “The earthquake [which Matthew 28:2 tells us was a violent one] broke the seal of the tomb and also caused the body of Jesus to disappear down a crevice.” And even supposing there had been a public clamor for Jesus’ dead body, in order to quell the rumors of its having been resurrected, the guards could have stolen the body of another executed criminal from somewhere (say, a burial pit), and placed it in Jesus’ tomb. Would this have been difficult to carry out? Not at all. We need to bear in mind that according to Jewish religious law, corpses were deemed to be no longer legally identifiable with any certainty if they were more than three days old (see here). The apostles didn’t start publicly preaching Jesus’ Resurrection until seven weeks after the Crucifixion – which means the guards would have had plenty of time to organize their fraudulent scheme, before word of Jesus’ Resurrection got around. There are lots of other stories, then, which the guards could have told, to spare themselves public embarrassment and punishment at the hands of Pilate.

To sum up: the flaw in Professor McGrew’s claim that P(M|G) is massively greater than P(M|~G) stems from the fact that he considers only the probability of the account we find in Matthew’s gospel being created in the first place, while ignoring the much greater problem of its being circulated, initially by the guards themselves, and subsequently, throughout the wider community, over a period of decades.

Is Matthew’s account what we would expect, if there were a guard at Jesus’ tomb?

The Resurrection of Jesus Christ by Hans Multscher (1437). Wurzacher Passionsaltar, left panel inside: bottom right .

I’d now like to explain why I contest premise (ii) of Professor McGrew’s argument, that
the probability P(M|G) of our having Matthew’s account, given that a guard was set, is not wildly low: indeed, McGrew thinks P(M|G) might appear reasonably high, if we only had the first part of Matthew’s guard story in front of us (Matthew 27: 62-66), in which the chief priests ask Pilate to give the order for Jesus’ tomb to be made secure. If there actually were a guard, then that’s the kind of historical record we might expect. But there’s more to the story than that. An angel comes down from heaven, rolling away the stone and causing the guards to faint; the guards subsequently report what has happened to the chief priests, who bribe them to spread the story that they fell asleep, and that the disciples stole the body while they were asleep. Once we include this part of Matthew’s guard story, P(M|G) drops dramatically, as it rests on a massive psychological implausibility, which has nothing to do with miracles. Put simply: nobody in first-century Palestine would believe the story peddled by the chief priests, that all of the guards fell asleep at the same time, and none of them woke up while the disciples broke the seal of the tomb, rolled back the stone, and removed the body of Jesus, despite the fact that the penalty for guards falling asleep was crucifixion upside down! That story just wouldn’t wash. Even if there were a guard at Jesus’ tomb, one would not expect an account of the guards behaving in such a silly fashion: first, accepting a bribe, and then spreading a story which would put them all in mortal danger.

At this point, P(M|G) appears to be very low indeed. Let’s have a look at the apologetic attempts to reinflate it.

At the very end of Matthew’s account of the guard, the chief priests try to allay the guards’ fears of being executed on a charge of falling asleep at their posts, by promising to persuade Pilate not to punish them. One commenter on Professor McGrew’s post proposed that the chief priests thought they could convince Pilate to (at least publicly) go along with the false story that they were peddling for public consumption, even though he would have known perfectly well that it was false. This commenter was fair enough to acknowledge the inherent unlikelihood of Pilate listening to the Jewish leaders’ advice and agreeing to “hush up” the issue and let the soldiers live, but then argued that he would have reluctantly agreed to do so, in order to avoid an even greater evil: public insurrection. On the scenario proposed by the commenter, the Jewish leaders may have said to Pilate: “A story that Christ’s followers stole his body will help quell his faction, and a story that some supernatural power overwhelmed the soldiers will foment unrest, so it is in your interest as well as ours to back the version we put out.” Professor McGrew then contributed a clarifying remark of his own, in response to the commenter’s suggestion (emphases mine – VJT):


We are not told whether the move to shield the soldiers worked; we are told only that this is how they were induced to acquiesce in the tale.

Many real events seem far less probable on their face than this. The career of Julius Caesar is an instance — or far more incredibly, that of Napoleon Bonaparte. If we were allowed to use uncalibrated personal incredulity as a principle of inference, it would send a wrecking ball through the discipline of history, ancient and modern. Donald Trump, anyone?

So where are we now? Has Professor McGrew succeeded in restoring P(M|G) to the level of reasonable probability? Not at all: the examples he cites merely demonstrate that highly improbable events sometimes happen – like a dead whale ending up in a mangrove forest. But that does not render these events any less improbable.

Professor McGrew also objects that arguments from personal incredulity would undermine the study of history. But the difference here is that Matthew’s account of Jesus’ Passion and Resurrection is not just any old historical account: it is avowedly biased (being written from a Christian perspective), heavily supernatural in its subject matter, and contains what appear to be numerous dramatic embellishments which the other Gospel accounts lack – such as Jewish saints coming to life at Jesus’ death and subsequently appearing to people, or an angel rolling back the stone. I put it to my readers that a responsible historian would be grossly remiss in accepting these accounts at face value. They deserve to be subjected to a more skeptical kind of scrutiny than most other historical accounts. Allow me to quote from the words of a real historian: Dr. David Miano, Lecturer in History at UC San Diego. In his blog article, How Historians Determine the Historicity of People and Events (June 10, 2018), he writes:


Some general rules of thumb that can be used in evaluating a source would be: (1) the closer in time and place that the source is to the historical event, the better. This makes sense, because the longer the gap in time between the event and the writing, the more time there is for the story to be embellished, confused, doctored, misunderstood, or exaggerated. This is not a hard-and-fast rule, because it is possible for a later source to be more reliable than an earlier source, but generally speaking, time allows a story to pass out of memory and to be transformed into legend. (2) separate the core testimony from the biased presentation. Oftentimes the wording that a writer may use to tell a story employs loaded language designed to sway the reader in a certain direction. A historian is wise to wade through the rhetoric to get to the basic witness of the document.

When we examine the testimony given in any written source, we first try to ascertain plausibility, and then probability. A claim that is plausible might not be probable, after all.

“A claim that is plausible might not be probable, after all.” Words well worth remembering. And let us add: the scenario we are considering here isn’t even a plausible one: the most one could say is that it’s possible, despite its massive implausibility.

Is there any other plausible way of boosting the probability P(M|G)? The commenter whom I quoted above nominates the publicly known fact of the empty tomb: in his opinion, “the probability that given an empty tomb fact, they [the chief priests] could convince Pilate to allow the theft account to go out for public consumption… is indeed quite reasonable.” But as I’ve argued above, even if the fact of the empty tomb became publicly known, the earthquake would have served as a convenient excuse for the body’s absence. Additionally, it would not have been difficult for the guards to substitute the body of an executed criminal for the missing body of Jesus, had they wished to: they had seven weeks to do it.

One last possible way of boosting P(M|G) is by supposing additionally that Jesus actually rose from the dead, and that an angel rolled back the stone. The flaw in this assumption should be readily apparent to everyone: it assumes the very thing which it sets out to prove: the Resurrection. I am forced to conclude, then, that the attempts to render P(M|G) reasonably probable are all failures.

Too incredible to invent: Professor McGrew appeals to Aristotle

Roman copy of a bust of Aristotle, after a Greek bronze original by Lysippos.

In a comment attached to his post, Professor McGrew also chides me for expressing incredulity at an attempt by a Christian apologist (Wenham) who argues from the improbabilities in the story (conceived as a story) that the best explanation for why it is told is that it was notoriously true. As I put it in my review: “Wenham is inclined to credit the story of the guard, precisely because it’s so full of obvious holes that he thinks no-one would have made it up in the first place.” McGrew contends, quoting Aristotle, that I am being grossly unfair to Wenham. But even if I am, that, in and of itself, does nothing to establish that the story of the guard is probably true – a story which Wenham himself concedes “bristles with improbabilities.” Surely a fair-minded historian would take note of these improbabilities, and evaluate accordingly. Once again, I ask: are we to believe Matthew’s claim that the chief priests and the guards (Mt. 28:12-13) deliberately circulated the story that all of the guards fell asleep, which would leave them liable to a capital charge? Is that historically probable?

But let us examine Aristotle’s argument. Does he say what Professor McGrew claims he says in his Rhetoric 2.23.21 (1400a)? Aristotle writes: “For the things which men believe are either facts or probabilities: if, therefore, a thing that is believed is improbable and even incredible, it must be true, since it is certainly not believed because it is at all probable or credible.” This argument deployed here is very similar to Tertullian’s “certum est, quia impossibile” (which is often misquoted by skeptics as credo quia impossibile).

Professor McGrew seems to be interpreting Aristotle as arguing that a claim is more credibly true if it is prima facie incredible, on the grounds that its very incredibility militates against its having been made up: no-one would be dumb enough to make up a story full of holes. With the greatest respect, I don’t think that’s what Aristotle is arguing in the passage quoted above. Instead, I think that the point he is making is that a claim is more credibly true if it is widely accepted (“believed”), despite its prima facie incredibility, as such an incredible-sounding claim is unlikely to be widely believed by men unless it has strong independent support, which nobody can gainsay. As I read him, Aristotle is putting forward something like Nachmanides’ kuzari argument, which features heavily in Jewish apologetics.

Assuming my interpretation is correct, the question we need to answer is: was the story of the guard at the tomb ever widely believed by the Jews? Aside from Matthew’s Gospel, we have absolutely no grounds for thinking that it was. All we know, from later Jewish polemics against Jesus, is that the Jews believed the disciples had stolen his body. For instance, Justin Martyr, in chapter 108 of his Dialogue with Trypho, mentions the Jewish claim that Jesus’ disciples “stole him by night from the tomb, where he was laid when unfastened from the cross, and now deceive men by asserting that he has risen from the dead and ascended to heaven.” But that polemic proves absolutely nothing about the existence of a guard.

So it seems that Aristotle’s weighty authority does not support Wenham’s argument, after all.

Where Professor McGrew and I more or less agree

Professor McGrew will be delighted to learn that I am prepared to concede the first premise of his argument. (I would be happy to set P(G) at 0.1, or 10%, for the sake of argument. Certainly I would put it at more than 1%, after carefully weighing the arguments which Professor McGrew marshals in his post.) Premise (i) is of course predicated on the assumption that Jesus was actually buried in a tomb. I had previously maintained that an independent historian would conclude Jesus’ body was most likely thrown in a common burial pit for criminals, having been influenced by Professor Bart Ehrman’s spirited defense of this view. Since writing my review of Michael Alter’s book, I have examined the literature on the subject more carefully, and I now think the matter is far from settled. Jesus’ burial in a tomb remains a strong possibility, although I continue to vigorously maintain that it was not a new tomb owned by Joseph of Arimathea, as three of the Gospels claim. (Mark’s Gospel doesn’t say if it was new or not.) I would also maintain that John’s account of Jesus’ burial by Joseph of Arimathea (and Nicodemus) is heavily embellished. However, I am now prepared to grant that the prior probability of a guard being posted at Jesus’ tomb, as Matthew narrates, is not as low as I had previously thought.

What prompted my change of mind? Let’s examine some of the key arguments I brought forward in my review of Alter’s book. Professor McGrew has helpfully summarized these arguments under four points, which I have listed below (with very slight modifications in the interests of clarity), along with his responses and my counter-responses. (Of these four points, C and D relate directly to the prior probability of a guard being set over Jesus’ tomb – in other words, P(G).)

A. The guard at the tomb is mentioned only in Matthew’s Gospel, not in the other three. [McGrew’s response: “the argument from silence in such cases is generally terribly weak.”] [My reply: fair point.]

B. Matthew’s account fails to explain why the body could not have been stolen on Friday night. [McGrew’s first response: some scholars have argued that when properly interpreted, Matthew’s account actually implies that the chief priests could have made their request for a guard on Friday evening, which means that if Pilate promptly granted the request, Jesus’ body could not have been stolen on Friday night, after all.] [My reply: the scholars McGrew cites (Doddridge, Paulus, Kuinoel, Thorburn) all wrote in the eighteenth and nineteenth centuries. Is this the best he can do? In his book, Alter mentions another source dating from 1893, making the same claim, but goes on to note: “Most commentators presume that the visit was held on Saturday morning” (2015, p. 288).] [McGrew’s second response: even if the Jewish leaders had to wait until Saturday to obtain a guard from Pilate, they “might have left someone of their own to keep an eye on the tomb overnight. Failing that, they might still have thought that it would be better than nothing to have a guard set for the remainder of the time period specified” (italics mine – VJT.)] [My reply: McGrew is clearly reaching here. “Might have” establishes mere possibility, not probability.]

C. We are not told why Pilate would agree to the Jewish leaders’ request for a guard over Jesus’ tomb. In particular:

1. The request concerned a purely religious matter, and we would not expect Pilate to care much about such things. [McGrew’s response: “An imposture {on the part of Jesus’ disciples} might well raise civil trouble in Jerusalem… Preventing civil unrest lay squarely within Pilate’s sphere of responsibility. On this count, the matter is exactly the sort of thing we would expect the Jewish rulers to request of Pilate.”] [My reply: if we assume for the sake of argument that the Jewish leaders were actually aware of a rumor that Jesus had claimed he would return to life after his death, then McGrew has a valid point here. However, the assumption McGrew is making here does not enjoy a scholarly consensus: it is possible, but has not been shown to be probable.]

2. Pilate had just been pressured into ordering Jesus’ crucifixion, and therefore any further request would be unlikely to meet with a favorable reception. [McGrew’s response: “The theft of a body and proclamation that the individual in question was alive was the sort of scenario a Roman governor under Tiberius could not safely ignore.”] [My reply: the example McGrew cites to support his case relates to a conspiracy which the Roman Emperor Tiberius feared, against his own life. Jesus posed no such threat, although McGrew could perhaps urge in reply that Jesus was crucified as the “King of the Jews,” making him a pretender in the eyes of Rome, and hence someone whose resurrection would be bad news for Pilate.]

D. The Jewish rulers would not have made such a request of Pilate, since a gentile employed by a Jew would not be allowed to work on the Sabbath. [McGrew’s response: “there is no guarantee that the Jewish authorities would be particularly scrupulous in the matter of hiring a Roman guard to do their work, as they had already shown their willingness to hold a trial by night in prima facie violation of their own rules.” In any case, nothing in Jewish law prevented them from “making a request to Pilate, as the civil governor, that he would secure the tomb with a guard.”] [My reply: the first part of the response naively assumes that the negative portrait of the Jewish authorities in Matthew’s gospel is historically accurate. The second part of the response is more substantial, and makes a valid point.]

When all is said and done, I’m prepared to concede that McGrew’s responses to my foregoing arguments at least show that the prior probability P(G) of a guard being set over Jesus’ tomb is not as low as I had imagined. However, nothing in his responses suggests that this prior probability would be especially high, either. For argument’s sake, I’m prepared to accept that P(G) = 0.1.

Summing up

As we saw above, Professor McGrew’s reasoning was as follows:
“P(G)/P(~G) ≈ 1, and P(M|G)/P(M|~G) >> 1; therefore, P(G|M)/P(~G|M) >> 1; therefore, P(G|M) is easily more likely than not.” I, on the other hand, would argue that
P(G)/P(~G) ≈ 0.1, and P(M|G)/P(M|~G) is considerably less than 1; therefore, P(G|M) is quite unlikely, after all. I leave it to readers to decide who has the better of the argument.

I would like to conclude by thanking Professor McGrew for this exchange of opinions. He is welcome to comment on this post. What do readers think? Over to you.

APPENDIX

The following is an excerpt from Michael Alter’s book, The Resurrection (2015, pp. 340-342), in which Alter proposes a scenario as to how Matthew’s story of the guard might have originated. After presenting this scenario, I’ll briefly examine Professor McGrew’s criticisms of it. I hope this information will help readers form a better evaluation of the probability P(M|~G), discussed by Professor McGrew above – i.e. the probability of Matthew’s account being composed if there were no guard at Jesus’ tomb. Without further ado, here’s Alter’s scenario (emphases are mine – VJT):

…An obvious argument by doubters is that anyone could have removed the body before the tomb is discovered early Sunday morning by the several women. To circumvent this objection, Matthew is forced to invent a guard at the tomb. However, the presence of a guard will require a rational explanation. Consequently, Matthew is forced to invent the account of the Jewish leadership going to Pilate. But when is it possible for this visitation to have occurred? The earliest possible day would have been the Jewish Sabbath. However, a visit by the Jewish leadership on the Jewish Sabbath will have seemed unlikely to most knowledgeable readers or listeners to the text. Consequently, Matthew 27:62 obscures from its readers and listeners that this visitation occurs on the Sabbath: “Now the next day, that followed the day of the preparation, the chief priests and Pharisees came together unto Pilate.” This lie is necessitated because of Mark’s chronology (Mk 15:47; cf. Lk 23:54-56). That is, there is not enough time for the Jewish leadership to return to Pilate before the Sabbath and request a guard.

But why did the Jewish leadership need to see Pilate? There has to be a reason. Consequently, the previous lie necessitates Matthew inventing the idea that the Jewish leadership knew about Jesus’ prophecy that he will rise again after three days, “Saying, Sir, we remember that the deceiver said, while he was yet alive, After three days I will rise again” (Mt 27:63). How the Jewish leadership knew about this prophecy is not provided by Matthew.

Up to now, Matthew has explained why a guard is at the tomb, and he also provides information for his readers and listening audience that the guard stayed there for an undetermined length of time until the women arrive. This scenario now creates an even bigger problem. How can the women examine the tomb and verify that the tomb is empty if it is guarded by a Roman watch? Somehow these Roman soldiers must be eliminated from the scene. To resolve this problem, figuratively speaking, the angel descending from heaven and removing the stone, this terrifying the guard into a state of paralysis, kills two birds with one stone. Matthew has now explained how the tomb is open for the women to verify that Jesus’ body is missing and how the guard became immobilized to permit the women’s investigations at the tomb.

However, Matthew has now dug an even deeper hole for himself. Given that there is a guard at the tomb, why is there no record of what they saw? That is, why is there no record that the guard saw both the angel descending from heaven and the removal of the stone? To take care of this problem, Matthew invents the bribe: “And when they were assembled with the elders, and had taken counsel, they gave large money unto the soldiers” (Mt 28:12). However, this bribe creates yet another loophole. Would all the guards accept such a bribe, knowing that, if they were found out, it would mean their certain execution? Consequently, Matthew needs to invent another lie to protect his narrative. Thus, Matthew 28:14 states that the Jewish leadership will come to their assistance: “And if this comes to the governor’s ears, we will persuade him, and secure you.” [Note: Cassels (1902, 828) posits: “The large bribe seems to have been very ineffectual, since the Christian historian is able to report precisely what the chief priests and elders instruct them to say.”]

In brief, Matthew could have created a better lie. Every time he tells a lie it requires another and bigger lie to cover up the problem created by the previous lie. All these lies are ingeniously interwoven.

Professor McGrew read only my abbreviated version of the foregoing scenario, which did not impress him greatly. Here’s how I summarized it in a single paragraph (see also here):


…Alter suggests (2015, pp. 340-342) that the story was originally created in order to forestall an anti-Christian explanation for the empty tomb: maybe the reason why it was found empty is that Jesus’ body was stolen. To forestall that possibility, someone concocted a fictitious account of the Jewish priests going to Pilate and requesting a guard, in order to quell popular rumors that Jesus would rise from the dead on the third day. But that created a problem: if there were a guard at the tomb, then the women wouldn’t have been able to enter and find it empty. So in the story, the guard had to be gotten out of the way. This was done by inserting a terrifying apparition of an angel just before the women arrived at the tomb, causing the guards to fall into a dead faint, and conveniently providing the women with the opportunity to enter the tomb. And in order to explain why there was no public record of the guard seeing the angel remove the stone, the story of the guards being bribed into silence by the Jewish chief priests was invented. In short: the lameness of the guard story cannot be used to establish its authenticity. The story is an ad hoc creation, designed to forestall a common objection to the empty tomb accounts.

Professor McGrew is having none of it. He writes (emphases mine – VJT):


There is certainly something ad hoc going on in Alter’s treatment of the matter, but the problem lies in the methodology Alter employs here rather than in the story as told in Matthew’s Gospel. Start with a surmise — “Maybe it didn’t really happen.” Faced with the fact that there isn’t much reason to doubt it, make up a purely hypothetical motivation that someone might have had for inventing such a story: “Maybe Jesus’ body really was stolen, and they had to create a cover story for that fact.” Faced with the further problem that this particular cover story is hardly what one would invent to answer to that hypothetical state of affairs and could easily be contradicted by people on the ground in Jerusalem who knew the guards, ignore the problem and instead double down on creating hypothetical rationales for other parts of the story. “The guards have to be gotten out of the way so the women can enter …” Okay, why not just have Jesus’ resurrection itself knock them out instead of resorting to the awkward fabrication of their falling asleep? Simple questions like this suffice to show how specious such reasoning is. What historical narrative, however faithful, could not be dissolved (at least in the imagination of the critic) by the application of such methods? 

Professor McGrew asks a fair question: why did Matthew feel the need to introduce an angel, rather than have Jesus himself roll back the stone and in so doing, knock out the guards? The reason, I would suggest, is that the story of the angel at the tomb was already part of the Christian tradition, as it is found in all four Gospels, in some form or other. Even in Mark’s Gospel, the young man dressed in a white robe, whose presence alarms the women, is meant to be an angel (compare with Luke’s description of two men in dazzling clothing). So the logical thing for Matthew to do would be to give the angel a more dramatic role – that of rolling back the stone – and an intimidating appearance (“like lightning”), so that the terrified guards fall into a faint and become “like dead men.” (Matthew does not say that they fell asleep; that was the lie they supposedly agreed to circulate.) So to my mind, Alter’s scenario is not as fanciful as Professor McGrew evidently thinks. It sounds plausible to me. The only caveat I would add is that we don’t know who invented the story: it could have been Matthew, or it could have been some members of an early Christian community, who composed it in to counter objections by hecklers.

Finally, in response to Professor McGrew’s objection that any historical narrative, even a reliable one, could be dissolved by the universal acid of Alter’s fanciful speculations, I would remind him of the point I made above: Matthew’s account of Jesus’ Passion and Resurrection is not just any old historical account: it is biased, heavily supernatural, and contains what appear to be numerous dramatic embellishments. Such an account therefore invites a special kind of scrutiny, which we would not normally subject other historical accounts to.

I shall lay down my pen here, as I have written enough. Over to you, readers.

310 thoughts on “Why there probably wasn’t a guard at Jesus’ tomb

  1. Alan:

    As I pointed out at the time, keiths’s comment was not really relevant.

    Sure it was. It corrected your misconception regarding the applicability of probabilities to past events.

    You had written:

    Not sure how probabilities work with past events.

    I responded:

    They’re epistemic probabilities, so they work just fine with past events.

    Consider a simple example: You put a fair coin in a box, cover the box, and shake it vigorously. The coin comes to rest on the bottom of the box with either heads or tails facing up. Before you remove the cover, someone asks you “What is the probability that the coin landed heads up?”

    If you say the probability is 1/2, you are correct, even though you are talking about a past event. Why? Because the probability in this case is an epistemic probability, not a metaphysical one.

  2. keiths: That’s understandable, because it would be embarrassing for you.

    I’m certainly not embarrassed by the Bible as most any conversation here will attest.

    keiths: It would require you to defend goofiness like the mass resurrection story in the gospel of Matthew.

    Arguing with you about an obscure single sentence in a vast ancient work by multiple authors that is categorically rejected in it’s entirety by you would be probably the biggest waste of time that I could possibly imagine.

    It would be worse than arguing with a YEC that Darwin was correct on the first sentence of the third paragraph of page 124 of the Origin of Species. Instead of talking about common descent or carbon dating.

    peace

  3. newton: Is this an example of devolution?

    “The gene that codes for CCR5 is situated on human chromosome 3. Various mutations of the CCR5 gene are known that result in damage to the expressed receptor. One of the mutant forms of the gene is CCR5-delta32, which results from deletion of a particular sequence of 32 base-pairs. This mutant form of the gene results in a receptor so damaged that it no longer functions. But surprisingly, this does not appear to be harmful:”

    Loss of function?

    Surprisingly, neither does the broken mousetrap…😉

    But how does one evolve a 5 pound land walking mammal into a 50 ton whale, if Darwin mainly Devolves functions whether they are helpful, benign or damaging?
    Let’s try to evolve a 5 pound toddlers bicycle into a 50 ton plane by mainly breaking the bicycle’s parts…
    I hope you understand my problem?

  4. Corneel:

    So now I had to google the terms [‘epistemic probability’ and ‘metaphysical probability’] and regret having done so. Statistics was complicated enough without bringing philosophy into it, thank you very much.

    Heh. It gets even better when you bring in the Bayesian-vs-frequentist debate and the principle of indifference.

    I sense an OP in the offing.

    Welcome back keiths.

    Thanks, Corneel.

  5. keiths,

    I was hoping you were going to justify the claim that one can assign math probabilities to the likelihood that an event (any event which is unknown) occurred, based on a written text.

    I think if one can claim they are using math, and yet anyone can claim their numbers are the correct one (and not be more wrong or less wrong), that is not really math. That is just using a symbol, to mean whatever you want it to mean. Math is a pointless concept if the symbols relate to nothing.

  6. Mung: He’s not flipping out over it. He’s simply pretending the conversation never took place.

    https://discourse.peacefulscience.org/t/the-inevitability-of-improbability/4910/200

    I had a look at PS. I often complain about the moderation at TSZ, but looking at this:

    Michael_Callen
    Herculean Skeptic
    Old Earth Moderator
    7h
    Can we cease this aspect of the discussion. People are so invested now, that wars may start over this. Clearly T_Aquaticus intended that the outcome was known in his example, whether or not Mung read it that way. Just leave it be and move on.

    But having some schlep who happens to have an internet connection (and who calls himself a Herculean skeptic no less) telling you what you can discuss? Screw that, man.

    No sure why anyone would participate in that.

  7. phoodoo,

    I was hoping you were going to justify the claim that one can assign math probabilities to the likelihood that an event (any event which is unknown) occurred, based on a written text.

    Why ask me to justify a claim I haven’t made?

  8. keiths,

    Oh ok, fair enough. I thought you were agreeing with VJ that a mathematical probability is rational.

    I see no reason to accept his assumption on that. In fact its preposterous.

  9. But regarding the discussion on biological probabilities at PS, I think the lottery example is completely misguided.

    I think for a fair analogy, one needs an example where there are , let’s say, 100 trillion possibilities for failure, and 12 possibilities for success.

    That is of course being facetious. The real numbers are more like 10*2772 possibilities for failure, and 2 for success. As such should we expect success, without somehow tipping the scale?

    That is nothing like a lottery example.

  10. keiths:

    Why ask me to justify a claim I haven’t made?

    phoodoo:

    Oh ok, fair enough. I thought you were agreeing with VJ that a mathematical probability is rational.

    I see no reason to accept his assumption on that. In fact its preposterous.

    You aren’t being fair to either of us. You wrote:

    I was hoping you were going to justify the claim that one can assign math probabilities to the likelihood that an event (any event which is unknown) occurred, based on a written text.

    Vincent isn’t making that claim, and neither am I.

  11. Seems to me there are several sets of issues that we’re muddling.

    A list:

    1. Can we use statistics to help decide historical claims, in particular whether Jesus’ tomb was guarded?

    2. Is statistics a valid and useful mathematical tool.

    3. Can statistics be usefully applied to past events?

    4. Is it correct to say that past biological events can be both improbable and inevitable?

    5. Other?

  12. My answer is no to number 1 and yes to 2, 3 and 4.

    Nobody other than Vincent has yet agreed with him about 1 so it would be good to hear from him.

    Who disagrees on the other points?

  13. fifth:

    I’m just not all that interested in doing a bible study with atheists.

    keiths:

    That’s understandable, because it would be embarrassing for you. It would require you to defend goofiness like the mass resurrection story in the gospel of Matthew.

    fifth:

    I’m certainly not embarrassed by the Bible as most any conversation here will attest.

    I didn’t say you were, though you should be. I said it would be embarrassing for you if you tried to defend the goofy parts of the Bible.

    Here’s the story in question:

    51 At that moment the curtain of the temple was torn in two from top to bottom. The earth shook, the rocks split 52 and the tombs broke open. The bodies of many holy people who had died were raised to life. 53 They came out of the tombs after Jesus’ resurrection and went into the holy city and appeared to many people.

    Matthew 27:51-53, NIV

    Do you actually believe that ridiculous story? (I asked colewd the same question, but he bailed.)

    Vincent has the sense to question it, and I give him credit for that.

  14. Alan Fox,

    Adding:

    Can we apply statistics to the question: does the event under consideration violate the laws of physics such as gravity or the second law of thermodynamics?

  15. Alan,

    Your questions are really more about probability than statistics, so I’ve revised them accordingly.

    1. Can we use statistics probability to help decide historical claims, in particular whether Jesus’ tomb was guarded?

    Yes, and in fact I would argue that people do that sort of thing all the time, whether they realize it or not. Note that this can be done without calculating explicitly numerical probabilities.

    For example, most of us reject the story of King Arthur pulling Excalibur from the stone. Why? We judge it highly improbable based on what we know about the world and how it works. That is a probabilistic analysis, whether or not numbers are explicitly invoked.

    Another example is the use of DNA evidence to exonerate falsely convicted prisoners, though the mathematics is more explicit in those cases. A belief about a past event — that the prisoner committed the crime — is changed on the basis of new information. The prior probability is revised downward.

    2. Is statistics probability a valid and useful mathematical tool?

    Yes, of course.

    3. Can statistics probability be usefully applied to past events?

    Yes, as I explained to you here:

    They’re epistemic probabilities, so they work just fine with past events.

    Consider a simple example: You put a fair coin in a box, cover the box, and shake it vigorously. The coin comes to rest on the bottom of the box with either heads or tails facing up. Before you remove the cover, someone asks you “What is the probability that the coin landed heads up?”

    If you say the probability is 1/2, you are correct, even though you are talking about a past event. Why? Because the probability in this case is an epistemic probability, not a metaphysical one.

    Alan:

    4. Is it correct to say that past biological events can be both improbable and inevitable?

    That’s ambiguously worded. Assuming you’re trying to ask…

    4. Can a particular biological event be both improbable and inevitable?

    …my answer is no, because an unambiguously defined event cannot be both improbable and inevitable. But that isn’t what T. aquaticus was trying to say. As I explained to Mung:

    T. aquaticus phrased it ambiguously. He/she isn’t saying that every improbable event is inevitable. Instead, the idea is that it’s inevitable that you’ll get one out of the many improbable outcomes.

    In other words, if the probability is spread more or less evenly over a very large number of possible outcomes, then the probability of any specific outcome is low. That’s the ‘improbable’ part. Yet there is a high probability that the actual outcome, whatever it turns out to be, will be one of those low-probability outcomes — that’s the ‘inevitable’ part.

    A concrete example: You roll ten fair six-sided dice, each of a different color. Each specific result has a tiny probability, less than 1/60,000,000. That’s the improbable part. Yet the probability that you’ll get one of those improbable results is 1. That’s the inevitable part.

  16. Alan:

    Adding:

    Can we apply statistics to the question: does the event under consideration violate the laws of physics such as gravity or the second law of thermodynamics?

    The SLoT is itself a statistical law, so you can apply probability and statistics to it.

    The probability of SLoT violations is extremely low (but not zero) for most systems. However, SLoT violations have been observed experimentally in small systems over short periods of time.

  17. keiths:

    For example, most of us reject the story of King Arthur pulling Excalibur from the stone.Why? We judge it highly improbable based on what we know about the world and how it works.That is a probabilistic analysis, whether or not numbers are explicitly invoked.

    This is almost correct, but I’d rather say:

    “We judge it highly improbable based on what we believe about the world and how it works.”

    We are unconsciously doing Bayesian statistics most of the time, and we assign our a-priori likelihoods on the basis of our already existing beliefs. This is how we end up with what we consider a consistent notion of what is possible and what is likely, what is not possible and what is unlikely. Hence when a believer reads the Bible they will accept far more of what is described as actually having happened then when a non-believer reads it. We interpret everything through the lense of our own personal beliefs. This makes these discussions so fruitless – how can you agree on the probabilities in a Bayesian analysis when you assign such different values to the priors?

    Lizzie’s aspirations for this site are admirable, but doomed to fail. None of us can leave our priors at the door. We wouldn’t be able to talk about very much at all if we did.

  18. keiths: Here’s the story in question:

    LOL

    We all have our favorite Bible passages, here is one of mine

    The stupid man cannot know; the fool cannot understand this:
    (Psa 92:6)

    😉

    peace

  19. keiths:

    For example, most of us reject the story of King Arthur pulling Excalibur from the stone.Why? We judge it highly improbable based on what we know about the world and how it works.That is a probabilistic analysis, whether or not numbers are explicitly invoked.

    faded_Glory:

    This is almost correct, but I’d rather say:

    “We judge it highly improbable based on what we believe about the world and how it works.”

    Fair enough. Some people might reject the Excalibur story on the basis of bogus beliefs, although I would bet that in your case and mine, the relevant background beliefs are justified well enough to qualify as knowledge.

    We interpret everything through the lense of our own personal beliefs. This makes these discussions so fruitless – how can you agree on the probabilities in a Bayesian analysis when you assign such different values to the priors?

    That points to one of the benefits of making the analysis, including the prior probabilities, explicit. If you disagree (sufficiently) on the priors, then it may make sense to focus the discussion there.

    Something similar can happen in logical arguments, when the discussants work backwards from a contested conclusion and end up identifying a difference in premises.

    Lizzie’s aspiration for this site are admirable, but doomed to fail. None of us can leave our priors at the door. We wouldn’t be able to talk about very much at all if we did.

    True. A better admonition would be “Bring your priors along, but be prepared (and willing) to update them.”

  20. keiths: Do you actually believe that ridiculous story? (I asked colewd the same question, but he bailed.)

    I have no way of knowing what sort of odd twisted interpretation you have imposed on that text and I just don’t care.

    I won’t share my interpretation with you because I have been warned not to cast pearls before swine.

    I would be happy to discuss your interpretation with you further once you tell me why I should trust that any interpretation you could have about any text whatsoever could possibly be the correct one.

    before you do that you need to post a listing of the tenets of your new faith.

    peace

  21. keiths:

    A better admonition would be “Bring your priors along, but be prepared (and willing) to update them.”

    When I see the difference in how believers look at the Gospels and the Iliad I have to conclude that this is mainly because of the different a priori likelihoods they assign to these texts being more or less factual.

    When you consider the origins of the texts, how they have been transmitted, and when they were eventually codified in writing, there is not much between them. Granted, the time lines for the Gospels are shorter, but does that really impact the probability of a largely fictional origin and/or embellishment to a significant degree? Hard to tell.

    Even so, believers rate the credibility of the Gospels far higher than that of the Iliad, especially when it comes to the finer details and in particular so when it comes to miracles and Acts of God(s).

    The question then is, how do people arrive at such different a priori’s? I’d venture that this is largely caused by the three E’s of Environment, Education and Emotion. Rational analysis doesn’t seem to play a huge role in it. I am not optimistic that entrenched positions taken up via ‘EEE’ can easily be modified by mere rational analysis.

  22. keiths: Fair enough. Some people might reject the Excalibur story on the basis of bogus beliefs, although I would bet that in your case and mine, the relevant background beliefs are justified well enough to qualify as knowledge.

    I think so too.

  23. keiths: I was hoping you were going to justify the claim that one can assign math probabilities to the likelihood that an event (any event which is unknown) occurred, based on a written text.

    Vincent isn’t making that claim, and neither am I.

    Well, he did say this:

    I have written my own detailed ,mathematical response to Professor McGrew’s defense of the historicity of Matthew’s story of the guard.

    And this:

    Professor McGrew will be delighted to learn that I am prepared to concede the first premise of his argument. (I would be happy to set P(G) at 0.1, or 10%, for the sake of argument. Certainly I would put it at more than 1%, after carefully weighing the arguments which Professor McGrew marshals in his post

    Gee, that sure sounds a lot like assigning math probabilities to the likelihood of an historical event.

    I have no idea why you feel otherwise.

  24. fifth:

    I have no way of knowing what sort of odd twisted interpretation you have imposed on that text and I just don’t care.

    I interpret it in a straightforward, non-twisted fashion. When the author writes

    At that moment the curtain of the temple was torn in two from top to bottom.

    I interpret that as meaning that the curtain of the temple was torin in two, from top to bottom, at that moment. What is your interpretation?

    When he writes…

    The earth shook, the rocks split and the tombs broke open.

    …I interpret it as meaning that the earth shook, the rocks split, and the tombs broke open. Do you know what ‘earth’, ‘rock’, and ‘tomb’ mean? How about ‘shook’, ‘split’, and ‘broke open’? This is not rocket science, fifth.

    The bodies of many holy people who had died were raised to life.

    I interpret that to mean that the bodies of many dead holy people were raised back to life. Do you know what ‘body’, ‘dead’, ‘people’, and ‘raised to life’ mean?

    They came out of the tombs after Jesus’ resurrection and went into the holy city and appeared to many people.

    I interpret that to mean that they (the aforementioned resurrected bodies) came out of the tombs — you know, the tombs that broke open when the earth shook. Then they went into the city and appeared to many people.

    That interpretation is not twisted at all. But let’s hear your interpretation — I have a feeling it’s twisted like a pretzel.

  25. J-Mac: But how does one evolve a 5 pound land walking mammal into a 50 ton whale, if Darwin mainly Devolves functions whether they are helpful, benign or damaging?
    Let’s try to evolve a 5 pound toddlers bicycle into a 50 ton plane by mainly breaking the bicycle’s parts…
    I hope you understand my problem?

    Yes, I understand the problem. You are stuck with a position that is devolving.

  26. faded_Glory: I’d venture that this is largely caused by the three E’s of Environment, Education and Emotion.

    And the greatest of these is emotion.

  27. A modified list:

    1. Can we use probability theory to help decide historical claims, in particular whether Jesus’ tomb was guarded?

    2. Is statistics a valid and useful mathematical tool.

    3. Can probability theory be usefully applied to past events?

    4. Is it correct to say that past biological events can be both improbable and inevitable?

    5. Can we usefully apply probability theory to the question: does the event under consideration violate the laws of physics such as gravity or the second law of thermodynamics?

  28. So, I say 1. No, 2. Yes, 3. Yes, 4. No, 5. No.

    Keiths says 1. Yes, 2. Yes, 3 Yes, 4 No, 5. Yes.

    Just trying to clarify if there is consensus on some point so it would be more profitable to focus on what folks are disagreeing about and their reasoning. Borrowing a dog and going for a walk now but looking forward to responses.

  29. Hi everyone,

    Here’s a short but fascinating article I stumbled across a few days ago, which I’d like you all to have a look at:

    The concept of probability is not as simple as you think by Assistant Professor Nevin Climenhaga (Aeon, February 26, 2019). In place of the three prevailing interpretations of probability (viz. that probabilities express (a) frequencies, (b) built-in propensities or (c) levels of confidence) Climenhaga proposes a new interpretation, called the “degrees of support” interpretation:

    Here, probabilities are understood as relations of evidential support between propositions. ‘The probability of X given Y’ is the degree to which Y supports the truth of X. When we speak of ‘the probability of X’ on its own, this is shorthand for the probability of X conditional on any background information we have….

    Because they turn probabilities into different kinds of entities, our four theories offer divergent advice on how to figure out the values of probabilities. The first three interpretations (frequency, propensity and confidence) try to make probabilities things we can observe – through counting, experimentation or introspection. By contrast, degrees of support seem to be what philosophers call ‘abstract entities’ – neither in the world nor in our minds. While we know that a coin is symmetrical by observation, we know that the proposition ‘this coin is symmetrical’ supports the propositions ‘this coin lands heads’ and ‘this coin lands tails’ to equal degrees in the same way we know that ‘this coin lands heads’ entails ‘this coin lands heads or tails’: by thinking.

    But a skeptic might point out that coin tosses are easy. Suppose we’re on a jury. How are we supposed to figure out the probability that the defendant committed the murder, so as to see whether there can be reasonable doubt about his guilt?

    There’s more, and I leave it to readers to peruse the rest of the article. I think it answers the questions and skeptical concerns which some readers have been raising, about the legitimacy and possibility of calculating probabilities for past events. I’m tossing it out for people’s consideration; for my part, I thought it made good sense.

  30. Hi Alan Fox,

    For the record, my answers to your questions would be pretty much the same as keiths’s.

    Hi phoodoo,

    Gee, that sure sounds a lot like assigning math probabilities to the likelihood of an historical event.

    As keiths pointed out above, the probabilities I’m concerned with are epistemic probabilities rather than metaphysical propensities, so there is no problem in applying them to past events.

    Also, my calculation of the likelihood of the guard story is not based merely on the text itself, but primarily on considerations relating to (i) what kind of stories people will and won’t believe, and (ii) what people will and won’t do, when their lives are at risk. Briefly, I argued that it’s extremely unlikely that anyone in first-century Palestine would have believed the false story Matthew says the chief priests instructed the guards to peddle, and it’s extremely unlikely that the guards would have agreed to spread such a story, since (a) it would have put their lives at risk and (b) there were other stories they could have spread, which would have better suited their purposes without exposing them to the same risk.

    Hi colewd,

    What is the criteria for declaring a historical record false?

    I suggest you have a look at the article by historian Dr. David Miano, whom I referred to in my OP. Cheers.

  31. I think the problem is the informal use of the concept of probability.

    Equations are simply misapplied to historical events (unless we are discussing physical events, such as the movements of planets and asteroids).

    No one on a jury (I’ve been on one) does calculations.

  32. There probably wasn’t a guard at Jesus’ tomb. So what. Improbable things happen all the time. Improbable events are actually quite inevitable.

  33. keiths: …my answer is no, because an unambiguously defined event cannot be both improbable and inevitable.

    And that’s exactly what I was saying at PS.

    But that isn’t what T. aquaticus was trying to say. As I explained to Mung:

    And T_aquaticus avoided my attempts to clarify.

  34. It is probable that guards were placed throughout the region. It is entirely reasonable to believe that one just happened to be at the site of the tomb by pure serendipity alone. Unless it can be shown that it could not possibly be the case … yada yada yada.

    Why are skeptics so selective in their skepticism?

  35. Has anyone noticed that the OP is semantically equivalent to, “Have you stopped beating your wife?”

  36. In my job I continuously had to deal with uncertainty. We had to try and quantify the probability of success or failure when drilling an oil or gas well, to be able to evaluate how sound it would be to spend money on it. Setting aside the technical intricacies of how we did that, the spread of our understanding would be threefold: some estimate of success, some estimate of failure, and in between a (usually large) region of ‘not enough data to say one way or another’. When this region was large we would consider the prospect insufficiently mature to make an estimate of succes, and therefore we would not (yet) run the economics for the well.

    With more and more data gathering and technical work this central region usually got smaller and smaller over time, until we felt that we had a good enough grasp on the variables to be reasonably confident in our chance-of-success estimates. At that point we would run risk-based economics, and if these came out positive we would propose the well to be drilled.

    In case of this guard question I feel that the central region of ‘not enough data to say one way or another’ is vast, and dwarfs the end regions of ‘likelihood of there being a guard’ and ‘likelihood of there not being a guard’.

    In other words, absent more data, I would shove this question back into the drawer where it came from.

  37. faded_Glory: In other words, absent more data, I would shove this question back into the drawer where it came from.

    But this is important. This is in the top three of reasons to doubt the resurrection.

  38. phoodoo:
    But regarding the discussion on biological probabilities at PS, I think the lottery example is completely misguided.

    I think for a fair analogy, one needs an example where there are , let’s say, 100 trillion possibilities for failure, and 12 possibilities for success.

    That is of course being facetious.The real numbers are more like 10*2772 possibilities for failure, and 2 for success.As such should we expect success, without somehow tipping the scale?

    That is nothing like a lottery example.

    Those numbers are entirely made up.

  39. Mung: And that’s exactly what I was saying at PS.

    And T_aquaticus avoided my attempts to clarify.

    You attempted to change the analogy so you could say something different.

  40. vjtorley: For the record, my answers to your questions would be pretty much the same as keiths’s.

    Can you assign a probability to that, Vincent? 😋

  41. petrushka:
    Has anyone noticed that the OP is semantically equivalent to, “Have you stopped beating your wife?

    Well, it is, in my view, rather a loaded question. Is it sensible to argue over whether Edmund’s Turkish delight was coated in icing sugar.

  42. OT can thoroughly recommend dog borrowing. There’s apparently also an upside to dog lending.

  43. Alan Fox: Well, it is, in my view, rather a loaded question. Is it sensible to argue over whether Edmund’s Turkishdelight was coated in icing sugar.

    The icing question is about a book marketed as fiction. The guard question is about a book that millions of people regard as the truth.

    Moving on to historical fiction, some might consider it entertaining to discuss whether this detail or that are historically accurate, and wheter a character resembles an actual person.

    The Bible is somewhere in between. Marketed as fact, it has many passages that have no external supporting evidence. And many passages that seem to be based on fact, but seem inaccurate or fanciful.

  44. faded_Glory: In other words, absent more data, I would shove this question back into the drawer where it came from.

    But it’s not like the data shortage regarding the probability of say, whether there is currently a vaccine for Alzheimer’s that some drug company won’t tell us about until after the 2020 election,,

    Even though we don’t have sufficient data to estimate that probability. It doesn’t make it all the way into “The Silly Drawer” with this Guard question, which is much more like the Turkish Delight query or the question of whether any of Siddhartha’s siblings could sing as well as he could.

  45. vjtorley: As keiths pointed out above, the probabilities I’m concerned with are epistemic probabilities rather than metaphysical propensities, so there is no problem in applying them to past events.

    It doesn’t matter, what are you basing your numbers on, belief? Math isn’t supposed to be about belief.

    If someone says the number should be .02 and you say, no, no, the number should be .03, then some else can just as well claim the number should be .479. That’s not math. Its a bunch of numbers, sure, but that doesn’t make it math. There are no calculations involved. There is no right or wrong numbers. Its just..I like 4. No, I like 7…

Leave a Reply