Monday, November 29, 2021

On the minimal facts case for the resurrection, Part 2: Confusing sociology and epistemology

 I have frequently said that those who advocate the minimal facts argument for the resurrection (MFA) confuse sociology with epistemology. Here I'm thinking first of all the statements you hear such as, "Nothing else will work" or "We have to do this," which are ambiguous as between, "This is objectively the strongest way to argue" and "This is sociologically the way to argue that is most likely to get a hearing in today's world." I think that if people are honestly and astutely observing the evangelical apologetic landscape they will recognize this ambiguity in many statements. 

This is related to what I call the "Yugo" model of apologetics--the idea that the best (in some ambiguous sense of "best") vehicle for our endeavors is the one that is the most stripped down. Of course, no reasonable person ever thought that the Yugo was an especially powerful car. 

As I'll note in the next post in this series, on p. 161 in The Historical Jesus, Gary Habermas actually states that using only four minimal facts (rather than twelve) "will strengthen the earlier apologetic." In a footnote he explains that what he means by this is two-fold: Using a smaller number of facts will gain recognition by an even larger number of critical scholars, and doing so will reveal how strong the case is. The latter is sort of a "half my brain tied behind my back, just to make it fair" motive. But in point of objective fact, reducing the evidence base does not strengthen the argument at all, even if it allows you to say that more scholars are on board. Interestingly, both of these explications of "strengthen" are sociological. Yet the phrase "strengthen the earlier apologetic" sounds like an epistemic evaluation. Such a usage is a good example of this kind of ambiguous language. 

Apparently my saying that MFA advocates conflate sociology and epistemology is considered by some people to be offensive or even a misrepresentation. I'm not likely to let that stop me, since if something like this is going on, it's better for it to be stated and faced than to be avoided on the grounds of not causing offense. And I explain carefully what I mean by saying this and document it, so you can decide for yourself whether it is a misrepresentation. But I will admit that the conflation takes various forms. I've just illustrated one popular form--the ambiguous use of words like "have to," "better," "strengthen," etc.

We find some rather striking illustrations of similarly ambiguous statements in the book Reasonable Faith by William Lane Craig.

Language in Reasonable Faith that conflates epistemology and sociology

In Reasonable Faith, William Lane Craig lays out the history of the argument for the resurrection. He discusses the way that William Paley did it, but the purpose of this history is to say that we can't do that anymore. Since the rise of higher criticism, Craig believes, we must argue for the resurrection (and also for Jesus' self-understanding of his own deity) in a very different way. We must use various "criteria of authenticity" to show which facts about Jesus, sayings of Jesus, and facts about the events after his death can be supported, preferably by two or more such criteria, and use those to make our arguments. It's a minimalist approach, though Craig prefers that it not be confused with Habermas's MFA, since Habermas's most limited set of facts is smaller in number and since Craig relies more on the criteria of authenticity. 

In The Eye of the Beholder, I discuss the interesting way in which Craig first warns against a negative use of the criteria of authenticity but then, apparently unconsciously, seems to agree tacitly that a saying or event that cannot be thus verified is open to special historical doubt, at least if we're only using objective historical evidence. I argue there that this is a kind of negative use of the criteria, though not the one that Craig discusses and rejects. (Craig says that we should not conclude that something didn't happen if it cannot be specifically supported by "the criteria." But he doesn't reject the idea that historical Jesus scholars qua historians should be agnostic about something if it cannot be thus supported, and in fact, in practice this seems to be what he does.)

In these chapters, Craig has several very odd statements about the older form of argument for the resurrection. These statements sit precisely in that in-between zone--neither entirely epistemological nor entirely sociological. One of the strangest of these is the place where he refers to the trilemma argument for Jesus' deity, relying upon "prooftexts" for Jesus' claims to deity, as "forever obsolete."

Often one hears people say, “I don’t understand all those philosophical arguments for God’s existence and so forth. I prefer historical apologetics.” I suspect that those who say this think that historical apologetics is easy and will enable them to avoid the hard thinking involved in the philosophical arguments. But this section ought to teach us clearly that this is not so. It is naïve and outdated simply to trot out the dilemma “Liar, Lunatic, or Lord” and adduce several proof texts where Jesus claims to be the Son of God, the Messiah, and so forth. The publicity generated by the Jesus Seminar and The DaVinci Code has rendered that approach forever obsolete. Rather, if an apologetic based on the claims of Christ is to work, we must do the requisite spadework of sorting out those claims of Jesus that can be established as authentic, and then drawing out their implications. This will involve not only mastering Greek but also the methods of modern criticism and the criteria of authenticity. Far from being easy, historical apologetics, if done right, is every bit as difficult as philosophical apologetics. The only reason most people think historical apologetics to be easier is because they do it superficially. (Reasonable Faith, p. 328)f

There are several highly problematic phrases here. One is "naive and outdated." Really? Did Jesus or didn't he assert his own deity in verses like John 8:58 and John 10:30? If he did, what is wrong with "trotting out" the trilemma and adducing these prooftexts? I note here that Craig does not use either of these texts in his chapter on Jesus' self-understanding, and this lecture on how hard it is to do historical apologetics, especially with respect to the sayings of Jesus, raises the troubling question of why those verses are not adduced. 

Perhaps the phrase "do the requisite spadework of sorting out those claims of Jesus that can be established as authentic" provides a clue, since the obvious implication is that there are things ostensibly recorded in the Gospels that cannot be established as authentic. This seems to involve internalizing, as epistemically binding, the notion that each saying reported must be separately "authenticated," rather than arguing for the reliability of a Gospel (say, the Gospel of John!) and considering that that "establishes as authentic" the sayings of Jesus recorded therein. The notion of "sorting out" those claims of Jesus that "can be established as authentic" can hardly be understood in anything other than an epistemological way. But that is a restrictive, passage-by-passage method. I've argued that this is not a good method and that we should not give it normative force.

The phrase "forever obsolete" is also highly problematic. Obviously neither the Da Vinci Code (a reference to which sounds rather outdated in 2021!) nor higher criticism renders anything "forever obsolete." If all that Craig means is "out of fashion," why say "forever"? Fashion is changeable as the wind. Nor should we care very much about mere fashion.

This is not the only language in Reasonable Faith conflating epistemology and sociology. In the chapter on the resurrection there is this:

The historical apolgetic for the resurrection played a central role in the case of the Christian apologists during the Deist controversy....Too often today Christians employ an apologetic for the resurrection that was suitable for use against eighteenth-century opponents but is today ineffective in dealing with the objections raised by modern biblical criticism. p. 333

Speaking as someone who is self-consciously reviving a Paley-style argument, I find this a problematic lead-in to a chapter on the history of resurrection apologetics. The fact is that Paley and his 19th-century followers (some of whom, living later, had even heard of D. F. Strauss!) would have been quite unimpressed by higher criticism. They would have had no real trouble "dealing with the objections raised by modern biblical criticism." It's not like such objections are all that strong. 

I cannot resist pointing out that the minimalist approach that Craig favors doesn't "deal with" these objections but instead concedes things to them in handfuls for the sake of the argument. Elsewhere Craig has said that his approach is to concede for the sake of the argument virtually all that the skeptic wants to allege in the way of contradictions and discrepancies in the Gospels and then to argue that we still have enough evidence left to make our case. In his introduction to this very edition of Reasonable Faith, Craig emphasizes that it is not actually necessary to argue that the Gospels are reliable, because true propositions can be mined out of even unreliable documents. So apparently the new, updated, non-obsolete way of "dealing with the objections of modern biblical criticism" involves a large amount of pre-emptive concession for the sake of the argument, which hardly sounds like an "effective" way of "dealing with" the objections that became popular in the 19th century! I think I would rather take my place with people who, confronted with the poorly-supported objections of the higher critics and "modern biblical criticism," would actually have "dealt with" them in the sense of answering them, followed by continuing to use the traditional historical apologetic that Craig is telling us is "ineffective."

But notice, again, the conflation of the purely social fact that a Paley-style argument will be considered outdated (by modern critical scholars) with an epistemological claim (or what looks like one) that it is actually not effective. This is even stranger when one reads on in the chapter and finds Craig merely laying out the bare fact that Strauss & co. claimed that the Gospels are riddled with legend and religious imagination. It is not as though Craig claims to show that these assertions have merit to them or strong arguments undergirding them. Presumably he thinks that they don't. But in that case, why not stand up to them? 

One more quote will illustrate the conflation of sociology and epistemology:

William Paley's View of the Evidences (1794) constituted the high-water mark of the historical apologetic for the resurrection. During the nineteenth century this approach dramatically receded. Indeed, it would be difficult to find a significant and influential thinker defending the Christian faith on the basis of the evidence for the resurrection. It seems to me that there were two factors that served to undermine the traditional apologetic. p. 342

The "two factors" are the rise of higher criticism and what Craig calls "the tide of subjectivism." The phrase "served to undermine" is another of those ambiguous phrases. Nothing that Craig says over the next few pages shows (nor even seems intended to show) that the "traditional apologetic" was objectively "undermined" by argument. Rather, what Craig chronicles is a series of beliefs that took over much of European thought, so that many people came to think of the Gospels as highly unreliable. But once again, why not point out that these trends were unsupported by anything  like good arguments? Why not point out that they haven't actually "undermined" anything? Why introduce the historical survey by saying that the historical apologetic of previous centuries was "undermined"?

I also feel the need to make an historical point. I take that statement about how it would be difficult to find a significant and influential thinker, etc., to mean "in the nineteenth century," or perhaps "after Paley," and it sounded dubious to me, especially since Tim and I have revived various 19th-century thinkers like T. R. Birks and J. J. Blunt. These gentlemen apparently didn't get the memo about how the Paleyan approach was obsolete and undermined. And, though he spent most of his career focused on topics like authorship and patristics, I'd be rather surprised if J. B. Lightfoot would agree that the traditional apologetic was obsolete. Esteemed Husband sent me the following list (and this is only a partial list) of thinkers in the 19th and early 20th century who did defend Christianity on the basis of the evidence for the resurrection. (One of these was a Unitarian, and is noted as such.) You can argue about how "significant and influential" they were, but it's not like they were nobodies. Chalmers, for one, was undeniably influential.

Thomas Chalmers, Scottish minister, theologian, social reformer, and leader of the Free Church of Scotland. See his Evidence and Authority of the Christian Revelation, 4th ed., (1822) and his Lectures on Paley's Evidences.

Andrews Norton, Harvard professor and unitarian theologian. See his Evidences of the Genuineness of the Gospels (1846).  

Henry Rogers, English nonconformist minister, man of letters, and president of Lancashire Independent College, notably in his Eclipse of Faith (1852) and A Defence of the Eclipse of Faith (1854).

Richard Whately, Oxford professor, logician, and Anglican Archbishop of Dublin. See his annotations to his edition of Paley's Evidences (1859).

Charles Pettit McIlvaine, Episcopal Bishop, professor, and chaplain to the United States Senate. See his Evidences of Christianity in their External, or Historical Division (1859).

Charles Aiken, Princeton Seminary professor, in his lectures on Christian Apologetics (1879).

James Orr, Scottish minister, theologian, and professor. See The Resurrection of Jesus (1908)

The above examples of language from Reasonable Faith show an important part of what I mean by confusing sociology and epistemology, and I think just bringing out such language can serve to warn people. When someone tells you that an approach or an idea is obsolete or outdated, ask him if it's false, and what his argument is that it's false. If the Paleyan approach to arguing for the resurrection is said "not to work" or to be "ineffective," ask him to define what he means by that. If all we are talking about is sociology, we can go there, and we can discuss that, because there are actually sociological reasons why I would argue that the MFA itself "doesn't work" socially. But it's more important to know whether something is a good argument. We shouldn't make it sound like something is "ineffective" in the sense of not being a cogent argument, merely on the grounds that other people (even influential ones) have made silly ideas highly popular! D. F. Strauss and all his followers from his own time to the present have unfortunately made silly ideas popular, but that doesn't objectively undermine anything.

But some may think that all this time spent on WLC is unfortunate, since Habermas and Licona are the bigger advocates and architects of the real MFA. So now I turn my attention to them.

"Historical bedrock"--using sociological considerations too restrictively 

I want to say right at the outset that I am well aware that Habermas and Licona have said that they require both good evidence and a certain type of varied, broad scholarly consensus as requirements for placing a fact in the category of "historical bedrock" or "minimal fact." It is an odd thing, but it seems that when one states that MFA advocates confuse sociology and epistemology, one is likely to be told indignantly that Habermas requires good arguments as well as consensus, so there, end of discussion, such a criticism is a misrepresentation, and there's nothing more to say. Well, no, there's still a lot more to be said. For one thing, the kind of rhetoric quoted from Dr. Craig in the last section is fully compatible with a method in which one requires good arguments for the premises of one's own MFA argument. One could still say of the Paley-style method that it is outdated, was undermined by higher criticism, is "ineffective," and other such ambiguous phrases.

But there is more still to say about the epistemological problems with the historical bedrock approach of Habermas and Licona. So I want to request that if you read this post and if you want to reply to it from a pro-MFA position, please do not bring forward quotations (there are several) in which Licona or Habermas says that their method requires both good evidence and broad consensus for inclusion in historical bedrock. I am fully aware of those and acknowledge those. But I'm raising problems nonetheless, problems that are not answered by that. Read on if you're interested.

In the final section of this post (see below) I will have something rather controversial to say about whether, in Licona's case, these two requirements are collapsing into one another in practice. Perhaps I should provide a "trigger warning" about that. But even there, it is not that I am unaware of the claim that "good evidence" is a separate requirement for inclusion in "historical bedrock."

In this section, I'm going to consider the mixed status of the "historical bedrock" category and even the somewhat larger category that Habermas refers to as "known historical facts." That mixed status is admitted by Habermas and Licona and is a central part of their method. What I mean by "mixed status" is that these categories are defined both by the arguments for the propositions in question and by the presence of a consensus of living scholars (or recently living scholars?) in favor of them. This approach unproductively mixes epistemology and sociology.

A term like "known facts" or "historical bedrock" appears to be a term of epistemic approval. It is natural to think of these propositions as those that have, on the basis of publicly available evidence, a very high probability. They are rock solid. There is so much evidence for them that they should be in no doubt to anyone in possession of the publicly available data. But in that case, why not make that the definition? Even if one wanted to have various levels, one could do that in terms of what the data support. The "bedrock" could be those propositions that, on the basis of presently-known historical information, are very, very probable, while the "next-level" facts (however one wants to label those) could be those that are at least "quite" probable. And so forth.

Habermas and Licona do not do this. For any of the categories that they treat as data to be explained, they require a certain consensus of scholarship--either extremely high, such as 90-95%, or at least a majority, perhaps 2/3 or 75%. This consensus, moreover, must be among scholars within a particular time period (a point I will return to below). They also want this consensus to consist of scholars across the scholarly spectrum--Jews, Christians, liberals, conservatives, skeptics, etc. 

As Bob Stewart has pointed out in a recent article on the MFA (his contribution to Raised on the Third Day, a festschrift for Gary Habermas), there are various problems with restricting the facts to be explained in this way. For one thing, the most significant facts (historically speaking) aren't always those that happen to command the highest scholarly consensus. Stewart also rightly points out that propositions are or aren't justified by a certain set of data even if no one believes them--a salutary and important reminder. For another thing, Stewart warns, consensus could change. It's at least in principle possible that consensus would swing so far "to the left" (that's my terminology, not Stewart's) that there would be no propositions anymore that would meet the consensus cut-off that Habermas and Licona are looking for. I would note that this could, in principle, happen even with a lower consensus cut-off. What then? 

And I would add "what then" epistemologically: Would we, as individuals, not be able to be justifiably extremely confident that, say, Jesus was buried if a time came within our own lifetime that this fact no longer commanded a large and ideologically broad scholarly consensus?  Just what sort of influence upon our own epistemic confidence is it supposed to have if some proposition doesn't command this sort of consensus? This is left unclear, to say the least, in the historical bedrock methodology.

This question is especially urgent when we consider the stated reason for the inclusion of the consensus requirement in the MFA methodology--namely, to guard against personal bias. Licona spends a great deal of space in The Resurrection of Jesus talking about "horizons" and the need for consensus to guard against individual "horizons"--i.e., personal biases. In Habermas's review of Licona's book, here, Habermas singles out this rationale and praises it:

When establishing a consensus of views, it is important to show that such a near-unanimity is “composed of scholars from all interested camps” (p. 64). We are not guessing about where researchers stand, and neither are we basing the case on a small, sectarian element within the academic community. Rather, the scholars should hold a variety of religious and philosophical positions (p. 65). Later, Licona reported that:  

These scholars span a very wide range of theological and philosophical convictions and include atheists, agnostics, Jews and Christians who make their abode at both ends of the theological spectrum and everywhere in between. We therefore have the heterogeneity we desire in a consensus, and this gives us confidence that our horizons will not lead us completely astray (p. 280).

Licona makes an insightful comment here regarding guarding against our own horizons. We must beware of our own imported biases, as well. When discussing the Minimal Facts, I have always purposely included notes at each juncture that list representative numbers of skeptics of various stripes who still affirm the data in question. This is a significant methodological procedure that serves more than one purpose. Among others, it assures the readers that they are not being asked to accept something that only conservatives believe, or that is only recognized by those who believe in the veracity of the New Testament text, and so on. After all, this sort of widespread recognition and approval is the very thing that our stated method requires.

Apparently the answer to the question of why we don't bestow such terms on the basis of the strength of the evidence alone is this: "Doing so isn't guarding enough against our own biases." If that is a correct way to approach things, shouldn't you then be afraid that, if you think the arguments for something are very strong, but the "heterogenous consensus" doesn't support it, you are being too moved by your own biases and should therefore correct your "too high" confidence in this proposition? This is one of the things I mean by conflating sociology and epistemology. I am referring to giving what I consider to be gravely undue weight to purely sociological considerations. The bestowal of a term of apparent epistemic approbation (such as "known historical facts" or "historical bedrock") is being held hostage to a purely contingent sociological fact--whether or not a certain varied consensus obtains concerning that fact. And that's being done on the grounds that otherwise we'd be too likely to be personally biased. 

In a recent video on the topic of historical bedrock, Licona says that a proposition does not need to have the very high degree of consensus required for "historical bedrock" status in order for him to use it in arguing for the resurrection. For example, he says that he sometimes uses the proposition that the disciples believed and proclaimed that Jesus was physically raised from the dead or the proposition that the disciples had group experiences even though these are (he thinks) affirmed only by 75 or 80 percent (or thereabouts) of scholars rather than by a consensus of nearly 100%. This might be brought up to counter my claim here that the use of these concepts is not too restrictive. 

I would answer: 1) Even these additional propositions allegedly have a fairly high consensus of scholarship. It remains a question whether, on this methodology, an individual would be considered reckless in some way if he had a very high probability for a proposition when it lacked this degree of majority consensus within the discipline. 2) The concern about "horizons" and the worry about guarding against "horizons" would definitely seem to caution against believing strongly against consensus. I will return to this point in the next section. While Licona has occasionally (in The Resurrection of Jesus) made a comment about how it can sometimes be okay to buck consensus, he far more often warns about the dangers of doing so and how dubious research is that goes against contemporary scholarly consensus. So the "horizons" concern does indicate, at least in Licona's case, a reluctance to go against consensus, and Habermas appears (see above) to endorse this concern as insightful. 3) I have repeatedly referred here both to the term "historical bedrock" and to the term "known historical facts" because it doesn't really matter where you put the cutoff. The problem lies in withholding terms of apparent epistemic approbation on the grounds of the absence of social approval of those propositions. This point lies at the heart of the "too restrictive" concern that I am raising here and at the heart of my saying that this is a conflation of sociology with epistemology. At the barest minimum, this use of terminology is creating a "mixed" category that combines an epistemic criterion (being well-evidenced) with a sociological criterion (being accepted by some level of consensus or other) and then bestows what sounds like an epistemic compliment. This is just not a good idea at all, even if one has various levels of such terms.

Making things even stranger and more concerning, the consensus required is (as far as I can tell) supposed to be a consensus of modern scholars, even very modern scholars, not of the "democracy of the dead." Yet C.S. Lewis noted explicitly long ago that reading old books can serve exactly the supposedly desired purpose of correcting for our own bias, since living at a certain time period can induce bias. That sort of time-bound bias is especially likely within a given scholarly guild, where careers are at stake and even people from across the "ideological spectrum" influence one another. 

This issue arose recently in a relevant social media exchange. Someone asked Dr. Licona for his opinion on the "democracy of the dead" in the field of New Testament studies. The questioner was interested particularly in the differences between my work and that of Licona and Craig Evans, where I'm agreeing more with older scholars. The question about the democracy of the dead is quite relevant to the issue of consensus. For example, the consensus landscape would look very different on the question of whether or not the Gospel authors deliberately made changes to the facts if we were to include those who intensively studied the Gospels in the past than if we were only to consider modern New Testament scholars. Licona's answer was revealing:

I don't think modern scholarship is "much superior" to previous scholarship. However, our knowledge base builds. Method is pondered on further and fine-tuned. And I'm of the opinion that these things assist us in furthering our knowledge. For example, in the discipline of astronomy, Hugh Ross may not be a smarter and more careful scientist than Galileo. However, he understands a lot more about our universe because his knowledge base is vastly greater than that which Galileo had.

So here we have an argument that we don't (apparently) need to include these older scholars in our consensus-seeking. Why not? Because of some sort of science-like discoveries that have been made in the meanwhile, analogous to discoveries in astronomy, that add to our "knowledge base" and render those scholars simply less-informed than those living in our own time. Such a claim leaves me shaking my head. I really doubt that the older scholars listed above would agree, upon being informed of the work of Strauss or of the supposed "discovery" of fact-changing literary devices in the Gospels, that this is on a par with the discovery of a new planet or scientific law. Suffice it to say that, despite Licona's oft-repeated concerns about "guarding against horizons" and seeking consensus "across the ideological spectrum," apparently he isn't at all seeking to guard against horizons by looking seriously at older, conservative scholars writing before the era of biblical higher criticism or at dead scholars who, during that area, bucked its trends. 

(Also indicative of the use of modern consensus in a way that is too restrictive is something from this video, starting at about minute 9. I will be discussing this video more below. Licona says here, as I have heard him say elsewhere, that "classicists" don't establish the reliability of whole documents. He bases this on the authority of something that allegedly classicist John Ramsey said to him in personal communication. Licona says that Ramsey said that "classicists" don't establish holistic document reliability but rather just ask if certain specific events or passages are correct. Regardless of what Ramsey said, it's certainly easily possible to find classicists talking about the reliability of a given ancient author. Which would, of course, translate into expectations--more positive or more negative or mixed--about the historical reliability of their writings. I expect, however, that this means that Licona will now be citing an alleged consensus of modern "classicists" against the holistic method of supporting whole-Gospel reliability as opposed to the passage-by-passage approach.)

The modern limitation in the consensus sought by MFA architects has the potential both to withhold consensus from propositions which would otherwise receive it (based on a less time-bound measure of consensus) and to bestow apparent consensus upon propositions that otherwise (in the larger context of the generations) would not receive it. This is only one concern, but it is one worth bearing in mind.

So, to begin with, Licona and Habermas are (I argue) conflating sociology and epistemology because, by their own account, they are restricting epistemic accolades by requiring a certain kind of scholarly consensus for their bestowal.

But that's not all.

"Historical bedrock" as a category that is too loose

Right here I want to say that it isn't really true that agnosticism is cautious. I emphasis this in The Eye of the Beholder. There is nothing particularly "responsible," "careful," or "cautious" about agnosticism. If the evidence available to a reasoner justifies a high probability for proposition P and he gives P only a mediocre probability and hence remains agnostic about P, he's not thereby being "cautious." Instead, he's being too optimistic about not-P! 

But it can be hard for some people to shake the idea that, by being hesitant and agnostic about more and more things, we're being more and more careful. To these people, it might seem that the too-restrictive definition of "historical bedrock" and "known historical facts" discussed in the last section is no biggie, because at the worst it would mean that a person was made hesitant about something or didn't use something in his argument when it really was highly justified by the evidence. Now, I actually think that is a pretty big deal. Once again, what if that kind of consensus is withheld from something really important that is really well-supported by the facts? What if that proposition affects some important practical matter? In that case, being overly diffident about it, due to lack of consensus, could have significant negative effects. Not to mention the fact that someone who is epistemically diffident in that case would be misevaluating the actual force of the evidence, which some of us think is a big deal in itself.

But if you still think it's no biggie, I want to raise yet another question: What do the MFA architects propose as the means of knowing how strong the evidence is, independently of scholarly consensus? Here's why I'm asking that: I've heard repeatedly that it would be a terrible misunderstanding of Habermas and Licona to say that they are placing too much weight on consensus because, after all, they insist that things that belong in these favored categories must have both consensus and good arguments. Well, the previous section has shown that even then, even granting that, they can be placing too much weight on sociology. But here's another problem: This way of stating matters assumes that "the evidence" or "the strength of the argument" is significantly accessible in some way that is quite independent of the mere presence or absence of consensus. In that way, the hope is, we won't include something in "historical bedrock" or some other favored category merely on the basis of heterogenous consensus when the arguments and evidence for it are poor. So at least we're supposedly being somewhat cautious about our acceptance of things on the basis of consensus, right? This is the ground for the indignation: "Haven't you noticed? Habermas and Licona are very clear that evidence and consensus are both required."

Suppose, then, that we lacked independent access to the strength of the argument. Suppose that our very access to the strength of the argument was to say, "Oh, look, a large, diverse consensus thinks this is true." Or suppose that we were so diffident about our own ability to evaluate the strength of the argument while going against consensus or working without consensus that we were scarcely ever willing to buck consensus or to go ahead and have very high confidence in something lacking that consensus. In that case, the two requirements would functionally collapse or nearly collapse into one. Hence, it wouldn't really matter if one kept insisting, "I'm taking this consensus to indicate that the arguments are really strong." If you aren't separately evaluating the arguments for yourself, you might as well just say that you're going to take the proposition to be probably true if it has this kind of diverse, modern consensus, and you're highly reluctant to go against that. The supposed requirement that it also be "well-evidenced" would not then be playing a significant, independent, epistemic role. Such a collapse would make it entirely possible for propositions to be treated as highly probable solely on the basis of the appearance of diverse consensus (and the fact that the scholars themselves say that the evidence is good), even if, to put it bluntly, the arguments in question really stink.

Even in The Resurrection of Jesus, there is a passage that disturbingly hints at this sort of collapse in Licona's own approach. (I do not know of anywhere that Habermas has endorsed this apparent collapse.)

When we investigate matters such as the resurrection of Jesus, historians in every camp operate with their own biases, agendas and hopes, all of which serve as unseen advisors. By requiring hypotheses to account for the historical bedrock, a check is placed on the explanatory narratives that are constructed. Any narrative unable to account for the historical bedrock should be returned to the drawing board or be relegated to the trash bin. Of course, this is a guideline rather than a law, since the majority of scholars have been mistaken on numerous occasions in the past. Accordingly there is a risk involved in requiring hypotheses to account  for the historical bedrock before their serious consideration by other historians, since this may result in excluding a hypothesis that denies one or more of the facts belonging to the bedrock but [that?] may later turn out to be mistaken in light of new information. This risk notwithstanding, minimizing the impact of biases and agendas is a serious matter, and historians must weigh the possibility of a mistaken consensus on strongly evidenced facts against the certain presence of horizons. Guidelines are not to be enforced in a wooden manner. The Resurrection of Jesus, p. 58

If the evaluation of strong evidence is really operating as a separate and independent requirement for the bestowal of an accolade like "historical bedrock," this passage is very odd indeed. Licona is considering the concern that a false proposition might make its way into the category dubbed "historical bedrock" and might, from that position of influence, cause historians to exclude (wrongly) some hypothesis that involved denying or "failing to account for" that proposition. A very interesting and real possibility. Licona's response is weak sauce, especially given the resources his own position ought to give him for a better answer. He says that we really need to worry more about being overly influenced by individual agendas but that we must vaguely try not to be "too wooden" in trashing maverick hypotheses that swim upstream against scholarly consensus. Well, that's reassuring. That sounds like a great way to guard against including things in "historical bedrock" that are falsely believed by a majority of our contemporaries!

I myself could use Licona's and Habermas's own statements to construct a much more reassuring-sounding response. Supposedly, or so we are repeatedly assured, no proposition makes it into the coveted category of historical bedrock solely on the basis of any consensus, however broad or diverse. No, no, we are told, it must also have strong evidence for it. Well, if that "also" is to have any practical meaning whatsoever, then it must mean that the individual investigator looks at the evidence for himself, independent (as much as he can bring himself to be) of the knowledge of what all the other people think, and decides whether or not it is strong. Why doesn't Licona say that, then? In response to this concern, he could have said something like this: 

But it must be remembered that our method requires very strong evidence as well as heterogenous consensus. If a historian is confronted with a theory that denies something he has previously regarded as bedrock, he can re-evaluate the evidence for himself, taking into account whatever arguments the maverick theorist brings, and consider whether the evidence is as strong as he previously regarded it to be. If he changes his mind, he should no longer regard that proposition as part of historical bedrock, leaving him more open to the new theory. In this way, the requirement of good evidence, and the possibility of evaluating that evidence for ourselves, can act as a check upon scholarly consensus, just as we hope that seeking heterogenous consensus can act as a check upon individual bias.

See? That would have been a lot more helpful than vague talk about "not being wooden" combined with further fretting about the "certainty of [individual] horizons." It is difficult to read Licona's statement of this problem, and his answer to it, in any way other than as a tacit admission (though it probably did not occur to him that it was any sort of an admission) that he has no really robust notion of independently evaluating the arguments for a position that runs contrary to consensus. Rather, the idea seems to be that a major way that we decide that something has good arguments going for it is by noticing that it has a large, heterogenous, and (at least in the field of biblical studies) modern consensus going for it. But practically speaking, this comes close to collapsing the requirement for strong evidence into the requirement for heterogenous contemporary consensus!

Some comments that he has made in a recent video support this concern. Here (starting around minute 3) he insists, per script, that it would be a terrible misunderstanding to think that he and/or Habermas is saying that something is true merely due to consensus. No, no, it's that it has such great arguments for it, and this is why it commands such a high consensus. But just a few moments later he endorses the proposition that if the Republicans and Democrats agree on the origin of the Covid virus, since they hardly agree on anything, why then, that proposition is "probably true"!

Some people have misunderstood Gary on this about the minimal facts, thinking that we should accept these facts because the majority of scholars grant them, and that's never what Gary said. …It's just a matter of here are 12 facts for which the supporting data is so strong that a majority of scholars, including sceptical ones, grant them as facts. So the importance of that, of course, is if you have a sceptic, a non-believer who grants those facts, you know, they may have biases, but they're not the same bias as a Christian would have. So it's kind of like, look, if both Republicans and Democrats were in agreement let's say that the Covid virus came from the Wuhan lab in China, well, then we could have a pretty good degree of confidence that's the case, because…they don't get along with one another, the Republicans and Democrats, they agree on hardly anything. If they agree on this, well, then it's probably true. 

[Notice that there is no mention here of diving in and evaluating for oneself the evidence about the origin of the Covid virus. Rather, the agreement between the Republicans and Democrats is being treated here as sufficient grounds for believing that the conclusion is “probably true,” due to their usual disagreement. LM] 

So if you're looking at both sceptic and believer alike who are willing to grant certain things based on the data because the data is strong, they think the data strong, well, that gives you some more confidence that that probably is correct. (emphasis added)

Hmmm. So how are we accessing, "There is strong evidence for the claim that the Covid virus came from a lab in China"? Apparently by noticing the consensus between people who often otherwise disagree and noticing that the people we're counting in the consensus think that the data for their position is strong. That's just not very encouraging as far as the permissiveness of the dependence on consensus. It's not very encouraging about the power of the "strongly evidenced" criterion to do any real epistemic work in correcting for a wrong consensus. 

Elsewhere in the video (around minute 9:50) Licona does refer to doing both: evaluating the arguments for the claim that Jesus died by crucifixion, finding them to be strong, and also noting that the consensus agrees with us. But what if we evaluate the arguments for ourselves and find that our independent evaluation of the evidence for some claim strongly differs from a "heterogenous consensus"? What happens then? And is it really necessary to do an independent evaluation if, as in the Wuhan virus example, the proposition is "probably true" if a heterogenous majority of people agree on it?

Licona's tendency to place far too much evidential weight upon consensus is also evident in his treatment of my own work, when he has referred again and again to an alleged majority of scholars "including evangelicals" that (he claims) agree with his views and disagree with mine. Repeatedly in his response videos in 2020, he would make a fairly blatant scholarly bandwagon argument. Sometimes he would  undeniably exaggerate the alleged consensus (though no doubt sincerely), giving listeners the impression that my views are so bizarre that virtually nobody else agrees with them and that virtually all even among living conservatives agree with his literary device views. This is wildly untrue, and I must resist the temptation to take the time to point out the various occasions in that series when Licona simply gave a factually false impression (one he no doubt believes himself, but should know better about) concerning the scope and nature of the alleged scholarly consensus. But that is not my point here. My point here is that he could barely restrain himself from making an argument from consensus repeatedly. Then he would pull himself up, make some brief gesture in the direction of saying, "Well, that consensus doesn't automatically mean McGrew is wrong" followed by, well, but, man, wow, she is so out of touch with the consensus of scholars, and the consensus shouldn't be dismissed too lightly. Lather, rinse, repeat. 

Here are just a couple of examples. Licona preemptively sneered at The Eye of the Beholder, which had not yet come out at that time and has now been released to scholarly critical acclaim .(This at least refutes the implication that my work is so crazy and fringy that no real scholar in the field takes it seriously.) 

The point I’m making is there is so much more going on behind John’s Gospel that McGrew fails to appreciate. Despite the fact that Johannine specialists find John’s Gospel to be a challenging conundrum, including evangelical scholars who have spent years focused on John and have published commentaries on it, McGrew apparently thinks the matter is grossly overblown and has announced that she is presently working on her book on John’s Gospel. One wonders what she will find that has gone totally unrecognized by those who’ve spent lifetimes studying the Fourth Gospel. 

Please remember that the supposedly radical, fringe idea that I'm promoting, that Licona is dismissing here, is that John's Gospel is robust reportage and that John never made anything up or changed any facts. It is not even inerrancy! It is not the idea that John recorded every word of Jesus verbatim, like a tape recorder. "So much more going on behind John's Gospel that McGrew fails to appreciate" is code for the idea that John invisibly changed things, reporting things in ways that made it look like they were true (in his narrative) when he knew that they weren't, for theological or literary motives. And that is being treated here as something so strongly supported by scholarly consensus that I'm obviously out to lunch for rejecting it. Of course, the consensus is wildly exaggerated as well. My views on John's Gospel are very close to those of living scholars like Peter J. Williams and D.A. Carson, recently dead scholars like Leon Morris, and older scholars like J. B. Lightfoot. What I'm noting here though is the extremely heavy-handed use of alleged consensus against a view deemed too conservative, even though it is a view that would have been deemed "the" evangelical view up until very recently.

Here's another:

Now McGrew does not think the Gospels belong to the genre of Greco-Roman biography and asks the following question: “Do the Gospels belong to or resemble the genre of GrecoRoman Βίοs in the informative sense that the authors were probably influenced by the conventions of this genre and chose to write their Gospels according to the conventions of this genre?” She answers “no” for 2 reasons: First, she says “Burridge’s arguments are utterly unconvincing.” However, a very large and heterogenous majority of New Testament scholars have found the arguments of Burridge and others quite convincing. Of course, this does not make them right. But such a large and heterogenous majority should not be dismissed too quickly. So, let’s look at McGrew’s second reason.

Here's that back and forth movement. I don't simply say that Burridge's arguments are utterly convincing. I go into why they are unconvincing in some detail. Licona apparently thinks he can move on from my "first reason" (that Burridge's arguments are too weak to support so strong a substantive position) merely by stating that "a very large and heterogenous majority of New Testament scholars have found the arguments convincing"! Then he stops, says that that doesn't make Burridge right, then he goes back to saying the consensus "should not be dismissed too quickly." Then he doesn't examine why I say that Burridge's arguments are unconvincing and moves on to a different argument I give against Burridge's thesis! This is typical of the use of "large and heterogenous consensus" as a way to brush off doubts about the sufficiency of argument. How dare someone say that the positive arguments offered for the position are weak? A large, heterogenous majority of scholars thinks they are strong. They can't be all that weak! Next! 

(And again, Licona himself takes this genre identification to mean that it's quite probable that the Gospel authors used a certain type of "literary devices" that involved changing what happened. By no means do all of the scholars who have adopted the idea of "Greco-Roman bioi" agree with that conclusion. Indeed, one often sees the "Greco-Roman bioi" idea touted as supporting Gospel historicity. So even the invocation of the large consensus is being stated in soft focus and then pushed for more content than it really supports.)

You can search the transcript to see how often Licona refers to a majority or a large majority consensus to imply that I'm wrong. While he will try to say that he isn't assuming that this majority is correct, his continual use of the claim of majority as grounds for dismissing a more conservative view says otherwise.

The mention of a "heterogenous" consensus and phrases like "including evangelical scholars" is particularly noteworthy. 1) "Heterogenous consensus" is what Licona has given so much weight to in the definition of "historical bedrock." So are we now supposed to act like it's something at least akin to "historical bedrock" that the Gospels are specifically "Greco-Roman bioi" and that they make use of Licona's suggested "compositional devices"?!  2) As already mentioned, this includes only contemporary scholars, due to a faulty, pseudo-scientific notion of modern progress in biblical studies. This means that modern biases influencing a large number of scholars in various camps are getting a free pass. 3) The views in question, which he is trying to bolster in this way, are the sort of things that just a very few decades ago would have been considered "liberal" scholarly views by definition.

This last point deserves emphasis and will be the final point I make in this post: When Licona, Habermas, and anyone else talks about a "heterogenous majority" of scholars "across the ideological spectrum," they will often list labels that are taken to indicate real diversity of underlying approaches. These include terms like "skeptic," "liberal," "conservative," "evangelical," "Jewish," or "Muslim." The only assumption that could make such a consensus even a weak indicator of truth is that these terms indicate stable, independent, definitional commitments by such scholars that really would tend to lead to diverging opinions in the area of interest. That, in turn, is what makes it seem like a remarkable thing to find that they all agree. By golly, if an evangelical and a liberal both agree with this proposition about the Gospels, then it must have something going for it. Again and again, in a number of things I have read, Licona will make use of this idea: If (some) evangelical scholars agree that John moved the Temple cleansing, moved the day of Jesus' crucifixion, elaborated Jesus' discourses, and more, then who are you, you non-expert, to disagree with them? The arguments must be good, right?

But it was just approximately yesterday (in the larger sweep of history) that accepting the idea that the Gospel authors changed such facts would have been considered to exclude one, by definition, from being an "evangelical" at all! So the very concept of what counts as an "evangelical" has shifted, and the "scholarly spectrum" has, effectively, narrowed. That is part of what I mean by the biases inherent in all living at the same time. While the "evangelical" label has remained, the commitments involved in that label have not remained stable. We certainly should not assume that just because someone whom someone else labels "evangelical" holds a certain conclusion, when that conclusion might sound "liberal" to the ears of the uninitiated, it must be well-evidenced. And it has been a major part of my mission in writing my last two books to show in detail that such propositions are not well-evidenced. The arguments leading some "evangelicals" to embrace these conclusions are the same tired, old, poor arguments that cause those previously labeled "liberals" to embrace them. Indeed, on occasion the "evangelicals" have even out-liberaled the "liberals" in coming up with a farfetched view supported by Byzantine literary arguments. (I'm thinking here of Daniel Wallace's theory that John changed "My God, why have you forsaken me?" into "I thirst.") The labels therefore mean nothing as far as how strong the argument is for the conclusion. 

It turns out that it just isn't true that people labeled "evangelicals" are moved in a more "liberal" direction only because of the sheer force of overwhelming evidence. Far from it. Indeed, witnessing the way that Licona and others have used the notion of scholarly consensus, one can see that younger evangelical scholars, seminarians, apologists, pastors, etc., are more likely to be bludgeoned into reluctantly accepting more liberal views by the sheer force of peer pressure. Thus the claim of "heterogenous consensus" becomes self-perpetuating. In the contemporary milieu, the very concept of "heterogenous majority," applied to people who are all part of the guild and seeking a place at the table during our own time, is often a misleading concept. It is an illusion of diversity, in a field where truly original thought in a too-conservative direction is laughed off the stage by the charge of hyper-conservatism or even (gasp) fundamentalism.

Taken together, what all this means is clear: We really need to stop worrying about consensus in biblical studies. We need to stop assigning it any significant, independent epistemic force. We need to make our investigation all about the evidence and arguments, as best we can evaluate them. We need to stop pretending that the consensus of modern scholarship, diversity of labels notwithstanding, indicates real, interesting diversity of viewpoints.

In terms of the argument for the resurrection, this also means that we shouldn't really make a big deal about sheer scholarly support for any premise we want to use, even if it is a conservative-sounding premise that does have overwhelming evidence going for it. If you want to mention scholarly consensus on, say, the fact that Jesus existed, be sure to make it clear that you've investigated the evidence for yourself and that that is where your real emphasis lies.

Next up, in Part 3: A discussion of confusion in the statements of the MFA about what is really granted by a large consensus of scholarship.

Monday, November 22, 2021

On the Minimal Facts Case for the Resurrection: Part I--A statement of my central objections

 In light of some recent events, I've decided that it would be a good idea for me to do a new, three-part, written series on my objections to the minimal facts argument (hereafter the MFA) and related cases like the "core facts" case of William Lane Craig. I have written on this before. Here are several of my older posts. (Here, here, here.) Here is a longer treatment of the issues in the form of a webinar. Here is my Youtube playlist. If interested in more details beyond those in this post, please see these other resources.

The three parts currently planned will be as follows.. In this post, Part I, I will state what I take to be the most salient epistemological problems with minimalist approaches to arguing for the resurrection. These are approaches that attempt to rely only on a small number of facts widely granted by scholars and to eschew reliance on the proposition below labeled DT. I will also reply to some attempted responses to my criticisms that I've encountered over the last several years. 

In Part II, I plan to talk about another problem with minimalism, which is what I call the confusion between epistemology and sociology. I am deferring this to the second post, because the objections in this post stand on their own even if someone finds the concern in Part II to be offensive and/or thinks that the scholars in question are neither confused themselves about this nor fostering such confusion in others. 

In Part III, I plan to provide and discuss quotations from Dr. Gary Habermas and possibly from others and to explain how these caused confusion initially for Tim and me concerning what is or is not granted by the wide consensus of scholars. That confusion led to some incorrect statements in our 2009 Blackwell article about the scope of scholarly consensus (statements that we have since publicly repudiated). Some have now tried to use these earlier quotations against us, with the unsubtle implication that what we were doing there was at least quite close to or a lot like an MFA (in fact, it was not, despite our mistaken idea about the scope of consensus) and also that we should have known better than to be at all confused about how much scholars do and don't grant. I would contend instead that such a mix-up is at least somewhat understandable on the part of those who don't expect confusion on the part of prominent, admired apologists (which might seem uncharitable). I think that this is why so many other laymen and even philosophers and apologists have also been confused about the appearance experiences claim in the MFA. I'll be discussing this more in Part III.

Since there has been some surprising confusion in a recent social media exchange about the terms "use" and "rely on" in the context of argument, I want to state for the record at the outset that I intend those terms to be interchangeable in this series, and the same for "depend on."

Throughout this post I will be talking about a proposition that I will label DT (for "details of testimony").

DT: The Gospel accounts and the account in Acts 1 of Jesus' resurrection appearances and of the finding of the empty tomb reliably represent what the disciples/alleged witnesses (both male and female) claimed about their experiences at that time. This includes such matters as that Jesus ate with them more than once, that they were able to touch him, that he appeared to them multiple times and to varied groups, that he had lengthy conversations with them, and so forth.

Readers of our 2009 article on the resurrection will hopefully recognize that DT is included as an aspect of both what we call W there (the testimony of the women) and what we call D (the testimony of the [male] disciples). Our 2009 article crucially relied on DT in order to set the Bayes factor (a measure of the force of an item of evidence) for these facts, something we emphasize repeatedly in the article. I will not argue here for the claim that our 2009 article crucially relied on DT, though I could bring multiple quotations to show it. I believe that the article speaks for itself and that most people who read it can see this, despite some very surprising statements that I have recently encountered to the contrary. I also note that DT is not question-begging, since it is a statement about what the alleged witnesses claimed; their testimony, with this content, must then be explained. We argued that the resurrection was the best explanation.

When I speak of minimalist approaches to arguing for the resurrection in this post, I particularly have in mind those that try to avoid using DT, relying instead on more minimal claims about appearance experiences that are granted by a wide variety of scholars. Such approaches sometimes include the statement that the Gospel accounts "present Jesus' resurrection as physical," but this is different from relying on DT, since the premise instead is merely the very broad fact that the accounts present the resurrection as physical. This is meant to support the conclusion that the apostles claimed that Jesus' resurrection was physical. It does not amount to relying on the claim that the disciples actually testified to the specific experiences that are told in the Gospels. (This is an important distinction.)

A classic and fairly typical statement of the "minimal fact" about the appearance experience claims is found in Michael Licona's book, The Resurrection of Jesus: A New Historiographical Approach. Licona, along with Gary Habermas, is one of the major proponents of the MFA. Here (with some surrounding context) is how Licona words that.

In a similar manner, historians may conclude that, subsequent to Jesus’ death by crucifixion, a number of his followers had experiences in individual and group settings that convinced them Jesus had risen from the dead and had appeared to them. We may affirm with great confidence that Peter had such an experience in an individual setting, and we will see that the same may be said of an adversary of the church named Paul. We may likewise affirm that there was at least one occasion when a group of Jesus’ followers including “the Twelve” had such an experience. Did other experiences reported by the Gospels occur as well, such as the appearances to the women, Thomas, the Emmaus disciples, and the multiple group appearances reported by the tradition in I Corinthians 15:3-7 and John? Where did these experiences occur? Historians may be going beyond what the data warrants in assigning a verdict with much confidence to these questions. I reiterate that historicans may conclude that subsequent to Jesus’ execution, a number of his followers had experiences, in individual and group settings, that convinced them Jesus had risen from the dead and had appeared to them in some manner. This conclusion is granted by a nearly unaniumous consensus of modern scholars and may therefore be added to our “historical bedrock.” The Resurrection of Jesus: A New Historiographical Approach, p. 372

Similar quotations about the disciples having had experiences that "in some sense" seemed to them to be experiences of the risen Jesus can be documented from Habermas. This appearance fact definitely does not include DT. Indeed, I have recently been told that it is so obvious that the MFA doesn't rely on the details of the Gospel accounts, not even on taking them to be what the original witnesses claimed to have experienced (the premise we used in the 2009 article), that everyone should know this and that Tim and I should never have been confused about whether or not a wide consensus of scholars grants DT. (In other contexts, laymen who advocate the MFA have tried to tell me that the MFA and/or William Lane Craig's cousin "core facts" approach does include DT among its premises and that it is uncharitable for me to think otherwise. This understandably makes me feel that I'm in a no-win situation. When I try to clarify what the MFA doesn't include, I'm told by one set of MFA defenders that it does include that. When I admit openly that we were originally confused ourselves about what Habermas's research shows and the fact that DT is not included, I'm told that we should have known better all along.) I will not take the time here to argue further that the MFA and other minimalist approaches do not rely on DT. While I could bring multiple quotations to support this, I will take it that the prominent, scholarly advocates of those approaches are well aware that they are not relying on DT and are quite willing to admit it, so hopefully I won't be accused of misrepresentation for simply stating this: The MFA doesn't rely on DT as a premise. DT is not meant to be included in the MFA premise stating that the disciples had "experiences" that they believed were appearances of the risen Jesus. Moreover, the original advocate of the MFA (Habermas) apparently did not intend to say that a large consensus of scholars grants DT, even though DT merely states that this is what the alleged witnesses claimed.

Therein, I believe, lies a crucial problem with the MFA and other minimalist approaches. It is my firm contention that it is not possible to provide a strong argument for the bodily resurrection of Jesus without being prepared to defend the proposition that the Gospel accounts of Jesus’ resurrection are unembellished, reliable accounts of at least what the initial putative witnesses claimed. By “strong argument” I mean (roughly) an argument that would require a good bit of evidence to defeat it, an argument that is sufficient to overcome even a moderately low prior probability for the event and to leave one with a high posterior. (To put this in the terms of our Blackwell article, a "strong argument" here refers to an argument that provides a significantly top-heavy Bayes factor.) I'm reluctant to quantify these statements further, but for reasons of concreteness I will tentatively say that I do not think that evidential arguments for the resurrection that do not rely on DT can take a prior probability of, say, .1 or lower to a posterior probability of, say, .9 or higher.

A major reason for my contention about the need for DT is this: A strong argument for the resurrection needs to contain good evidence that the disciples were rational in believing that Jesus was physically raised (not merely that they did believe this). Many people irrationally believe religious propositions, and the skeptic is not going to grant (and perhaps shouldn’t grant) that, since the disciples believed Jesus was physically raised, they must have had very good reasons for believing this. Any skeptic quite reasonably wants to know more specifically what they said they experienced and whether or not that would have been good reason to accept that Jesus was physically alive. 

The MFA states that the disciples had experiences shortly after Jesus’ death that convinced them that Jesus had risen. But the “appearance experiences” referred to in the MFA have to be left vaguely specified in order to obtain the high consensus of scholarship required by the MFA method. Vaguely-defined appearance experiences are insufficient to show with a high degree of confidence that the disciples were rational in believing that Jesus was physically raised. The appearance experiences allowable could include everything from the kinds of empirical, polymodal, sustained, intersubjective group experiences described in the Gospels to some sort of experience of light and sound, a vision at a distance, or a spiritual ecstasy. 

Consider: Atheist scholar Gerd Ludemann is included in the "scholarly consensus" surrounding the experiences claim in the MFA. Habermas and William Lane Craig have quoted the following statement from Ludemann as showing how much support this minimal fact has, and one will often see lay apologists quote it on social media as though it's a big deal. (I just saw someone do this within the past week.)

It may be taken as historically certain that Peter and the disciples had experiences after Jesus' death in which Jesus appeared to them as the risen Christ. (What Really Happened to Jesus? p. 80)

But Habermas and Licona have elsewhere made it clear that Ludemann's concession here really doesn't amount to much. Indeed, in the same book, he says that the "objectification" of the Gospel resurrection accounts is the result of later embellishments. By objectifying he means the physical details such as Jesus' eating and being tangible. He insists that "the original seeing...was a seeing in the spirit" (p. 69)--some sort of religious ecstasy and, in some cases, an individual hallucination. One wonders very much in that case why we should quote Ludemann's statement with such enthusiasm, unless perhaps merely to argue (from Ludemann's authority) that that the disciples were not the merest fraudsters, making the whole thing up from literally nothing.

What is often not well-understood is that the details of evidence matter. It is a crucial principle of probabilistic reasoning that we must always use the most specific version of a piece of evidence that is relevant to the conclusion in question. Otherwise, merely by cherry-picking a  degree of "focus" for the evidence, we can make it sound like it supports a conclusion that, if fully understood, it either doesn't support at all, supports far less than we think, or even disconfirms!

Here is a well-known example in epistemology: Suppose that you are arguing that John is Roman Catholic. Suppose that you have as a piece of evidence the specific proposition that John is from a city in Ireland that is almost entirely Protestant. But instead of describing the evidence in that way, you describe it as, "John is from Ireland." You then argue that, since most Irishmen are Roman Catholic, probably John is Roman Catholic. This is obviously illicit. By deliberately giving the evidence in a "fuzzy" form and leaving something out, you make it sound like you have a strong premise for the desired conclusion, when in fact stating the evidence in the specific form in which you really have it would show that the evidence points in exactly the opposite direction.

Similarly, to say that Ludemann and others grant that the disciples had some sort of appearance experiences that led them to think that Jesus was risen, while including in that "consensus" widely different views about what kind of experiences we are talking about, we are basing our conclusion on supposed expert opinions that might or might not support the desired conclusion--that Jesus rose bodily from the dead. In fact, some such opinions might even tend to support the opposite conclusion. For example, if Ludemann is right that the disciples merely had a "seeing in the spirit" experience, it isn't at all clear that this proposition supports the proposition that Jesus was literally risen from the dead. Or suppose that a scholar thinks that the disciples had an experience as of Jesus up in the sky, seen at a distance--something that Bart Ehrman has suggested might be the most that a group of disciples experienced. (Ehrman doesn't grant even this, but he thinks at most this would have been the type of "group experience" that they had.) An experience like that might be well explained by the theory that Jesus did not physically rise from the dead but instead was dead while his spirit was in heaven, communicating with his disciples from there.

Here is another example of the "specificity of evidence" problem: Suppose that I for some reason want to argue that a number in a box is the number 99. I know ahead of time that it must be some number between 1 and 1000. Suppose that I have never looked in the box and that ten people come and look into the box briefly and then make statements to me about the number in the box. Two of them tell me that the number that they saw was 200. One of them tells me that the number was 99. Four of them tell me that the number was 42. And the remaining three state that the number was 800. It would be completely illegitimate for me to combine the four votes for 42, the one vote for 99, and the two votes for 200 and state, "70% of those surveyed agree that the number in the box is less than 250. Thus this survey supports the proposition that the number is 99, by reducing thee possibilities to numbers less than 250." In point of fact, 9 out of 10 of those surveyed stated a number other than 99! There is nothing about being "less than 250" that unifies all of these claims so as to make them all support the conclusion that the number is 99. It is obviously false to say that the survey supports the conclusion that the number is 99. The "consensus" is gerrymandered by combining specific claims, most of which point away from the desired conclusion, describing them in a vague fashion, and thus making it sound like seven of the speakers asserted something that supported the conclusion that the number is 99.

Similarly, Ludemann's belief that the disciples experienced a "seeing in the spirit" and Ehrman's suggestion that (at most) perhaps they experienced a "seeing at a distance" really have nothing much in common with the belief of an evangelical scholar included in Habermas's survey who thinks that the disciples really had experiences like those contained in the Gospel accounts. These do not form an epistemic "natural kind" supporting a literal resurrection, except in the very tenuous sense that they all involve some notion that the disciples were not mere religious frauds.

The issue of the disciples' rationality and trying to do without DT is exacerbated when one is arguing (as of course a Christian should be) for Jesus' true bodily resurrection and bodily appearances rather than a vision by the disciples. For it is the bodily details that are included in DT as elements of the disciples' testimony. These elements were crucial to Tim's and my argument in 2009 because they helped to rule out various religious, secular, or paranormal explanations of the disciples' testimony other than physical resurrection. As I often put it in presentations, whether or not you and your friends spent a lot of time in conversation with another mutual, well-known friend is not the kind of thing you can be easily mistaken about. All the less so if all of you agree that you had meals and long conversations with him in varying groups over a period of weeks. These are the kinds of robust details of testimony included in DT that make honest mistake, grief hallucination, ghostly apparition, and religious ecstasy non-starters as explanations of the testimony. And this is why trying to do without DT is such a problem.

I would say that a skeptic confronted with the MFA would have reason to think that something-or-other happened after Jesus' death that was odd and intriguing, but he need not have too much concern (on the basis of the MFA alone) that it was something supernatural, much less that it was specifically the resurrection of Jesus of Nazareth. If there is a lot of evidence for "the paranormal" (I'm not convinced that there is, but I bring this up for the sake of completeness), this would provide even more categories for some sort of odd "spiritual" event at that time, after the death of Jesus, sufficient to account for the minimal facts, without strongly supporting the Christian explanation. A non-Christian like Dale Allison will be inclined to give a lot of weight to this latter possibility.

Several responses have emerged over the years as I have laid out these problems. I'll deal first with the most recent--a new one on me, and quickly answered. In a recent presentation at the ETS, the audio of which I possess, one presenter attempted to say that Habermas has already addressed the kind of criticism I am leveling here. Here is the quote from Habermas that supposedly addresses it:

[I]t has been concluded that the resurrection can be historically demonstrated even when the minimum of historical facts are utilized. Neither can it be concluded merely that “something” occurred which is indescribable due to naturalistic premises or to the character of history or because of the “cloudiness” or legendary character of the New Testament texts. Neither can it be said that Jesus rose spiritually but not literally. These and other such views are refuted in that the facts admitted by virtually all scholars as knowable history are adequate to historically demonstrate the literal resurrection of Jesus according to probability. pp. 165-166 The Historical Jesus

I am rather astonished that anyone would think that that quote addresses my concern about the appearances. This is Habermas's assertion of what he believes his argument supports. Of course he believes that his argument supports a stronger conclusion than I think it supports! It is not surprising at all that Habermas asserts that we can indeed conclude something stronger than merely that "something" occurred or that a spiritual resurrection occurred. This is not addressing my arguments that the MFA doesn't provide enough grounds for thinking that the disciples were rational in their belief. I am well aware that Habermas and Licona and other MFA advocates think that they have a good argument for the bodily resurrection of Jesus. That is not news! I never claimed that those who use the MFA do not believe that Jesus rose bodily from the dead. The question is whether they do have a strong argument for that conclusion, using only premises granted by a wide consensus of living scholars. I've argued that they don't, as summarized here. It doesn't "address" that argument to quote Habermas asserting something about how strong his own argument is.

In fact, if anything, providing this quotation saves me the trouble of searching to document the fact that Habermas has claimed that he has a really strong argument for the resurrection! Suppose that someone wants to retreat to saying merely that the MFA provides some reason (not a strong reason) for the resurrection, or that it's better than nothing or "might get someone thinking," etc. The above quotation from Habermas shows that he thinks he can "historically demonstrate the literal resurrection of Jesus by probability," using only propositions granted by "virtually all" scholars! So any such retrenchment by MFA advocates would amount to a very significant retreat from the founder's ambitions.

A more popular attempted answer to my concerns has been to supplement the barest minimal facts with a slightly larger set of facts, including those designated by Habermas as "known facts" and by Licona as "second-order facts." The most salient of these for purposes of trying to answer my concerns is the proposition that the disciples themselves believed that Jesus was raised bodily. Note that this is somewhat stronger than the fact that they had some sort of experiences that convinced them that Jesus was risen, since it includes more specifically that they really believed that he was risen bodily. (By the way, DT doesn't even make it into this larger list of facts used by Habermas.)

This response to my criticisms--slightly supplementing the barest MFA with the disciples' belief in bodily resurrection--usually involves saying that they would not have believed so firmly in bodily resurrection if they did not have experiences like those contained in the Gospels. It thus reverses the order of dependence from that which I am suggesting. Instead of arguing that the disciples did claim the things found in the Gospels, and hence that (if they were not lying) their belief that Jesus was raised from the dead in bodily form was rational (since such experiences would provide good reason to think Jesus was raised bodily), this proposal is to argue that they probably experienced something-or-other like what is found in the Gospels, since they probably believed rationally.

This, however, is a gravely inferior way to argue. For it requires us to assume without the support of DT that the disciples were rational and then to argue that something akin to DT is true as a conclusion. To use DT both as a premise and as a conclusion would be circular. So the apologist trying to do without DT as a premise is forced to use only evidence that the disciples really believed in Jesus' bodily resurrection, hoping that the skeptic will grant him that they were probably rational in doing so. But that "something akin to DT" is not independently supported by documentary evidence (namely, the Gospel accounts). Rather, it is simply supported by the reasoning that they must have experienced something like that or else they wouldn't have believed and therefore, probably, that their corporate belief was not only rational but true. This is much weaker than arguing for the bodily resurrection of Jesus as an inference to the best explanation, using the records in the Gospels as unembellished, varied testimony from different individuals and groups who were in a position to know. (See this post on the conspiracy hypothesis, which also is relevant to hypotheses of hallucination. Note, too, that the approach I am criticizing here does not allow us to multiply Bayes factors at all, since it treats "the disciples" merely as a group with a particular bare belief, without looking at even partially independent, varied, unembellished, specific accounts.)

Now, I am not going to say that the disciples' sincere belief that Jesus was raised bodily would not be some evidence that they had good reason for believing this, and thereby some evidence that Jesus was really raised, but if I really know very little independently about why they believed this, the argument is quite weak.

Compare: Suppose that I don't know Alice very well. She doesn't seem wildly irrational to me, but I don't have much to go on. I then find out that Alice believes that she has seen a ghost, due to some experience or other that she had last week. But I'm unable to find out about the details of her claims in any way that I am confident is not embellished. So I don't really know in detail what she claims happened to her. Does the bare fact that a person who isn't certifiably insane thinks that she saw a ghost provide some reason for believing 1) that she believed this for good reason and hence 2) that there are really ghosts? Yes, some. But not much. It is not a strong argument.

It makes a big difference which of the following I am saying:

1) Alice does firmly believe that she has seen a ghost. Therefore, she probably experienced detailed interactions with a person she knows to be dead (though I have no access to her actual accounts which I can trust to be unembellished by others). Therefore, probably, she really did see a ghost.

2) Alice has described in detail a series of interactions with a person she knows to be dead, and I have access to these descriptions and reason to think that they are her own, actual claims, unembellished by others. I can see for myself that these would be good reason to believe in ghosts. Since I have other reasons to think that Alice is not a liar, it is probable that she has really seen a ghost.

I can't help thinking that people who make the argument, "They sincerely believed it, so they must have had experiences like those in the Gospels, so Jesus was probably really physically risen" do not appreciate the magnitude of the difference between these. But structurally they are completely different.

This is also why it is not enough simply to argue that the Gospels portray Jesus' resurrection as physical, while carefully saying that we don't know if those portrayals are embellished, not (specifically) what the putative witnesses claimed. There one is using the general fact of the portrayal only to argue that the disciples preached a message of Jesus' physical resurrection, while not using the claim that these stories are what the witnesses claimed. (Licona, The Resurrection of Jesus, pp. 329-330, 338-339)

Arguing from the disciples' belief in Jesus' bodily resurrection (while eschewing reliance on DT) only postpones the need to press the skeptic to accept something he is highly unlikely to accept--i.e., that if the disciples sincerely thought Jesus was physically raised, they were probably rational in thinking so. Worse, the deliberate attempt to do without DT as independently supported (due to a desire not to confront squarely skeptical claims that the Gospel resurrection accounts are embellished) leaves the claim that they believed rationally as a bare posit. That is, it leaves one without the resources to say why we should think that they believed rationally. Surely we cannot rely on a generalization to the effect that most people who firmly believe that they have witnessed supernatural or paranormal events are rational in drawing that conclusion!

Here we are up against what I have called elsewhere the problem of the bottleneck. The capacity of evidence for "the disciples firmly believed that Jesus rose bodily" to support "Jesus rose bodily" has a hard upper limit: That upper limit is the probability of "Jesus rose bodily" given that "The disciples firmly believed that Jesus rose bodily"--given it, that is, at probability 1. I am saying here that even if the skeptic fully grants that the disciples firmly believed this, that has fairly limited force to support the conclusion that it really happened, absent more information about why they believed it. You cannot make up the deficiency there simply by providing more and more and more evidence that they believed it. For some reason it is difficult to get people to understand this, but it's very important. (Erik Manning of the Youtube channel Testify has done a nice job here of restating this issue.)

Bringing the Apostle Paul into the picture really does nothing to alleviate the bottleneck problem about whether or not the disciples were rational. Suppose that we argue that Paul believed and taught that Jesus was raised bodily. This is certainly true, and arguments to that effect (e.g., from N.T. Wright) serve to rebut the rather silly (but previously popular) skeptical attempt to pit Paul against the Gospels' physical accounts, based on his use of the phrase "spiritual body" in I Cor. 15. It's good to rebut that skeptical move, but in itself that is not positive evidence that the disciples rationally believed that Jesus was raised. 

All the talk one hears about the disciples offering Paul the right hand of fellowship (Gal. 2:9), hence he was teaching the same Gospel they were teaching, hence the disciples were teaching Jesus' bodily resurrection, runs smack into the same bottleneck problem that I just discussed. When all is said and done, what does this support? Once again, that the disciples sincerely believed and taught that Jesus was physically raised! And how, again, does more and more support just for that proposition independently support the proposition that they believed this reasonably, on the basis of good and sufficient evidence? It doesn't, at all.

Finally, it is of very limited value to argue (again, using Wright) that Jewish expectations of resurrection pointed to a large, apocalyptic resurrection of the righteous rather than individual resurrections in the disciples' own time. While this is no doubt true, it provides only a little evidence that the disciples "would have" required some overwhelming evidence before accepting that Jesus was individually risen. We should remember that this Jewish theology did not prevent the superstitious belief both by Herod and by some unspecified number of Jewish people in the surrounding population that Jesus was John the Baptist risen from the dead (Mark 6:14-16). So generally accepted theology did not preclude human superstition in particular cases, causing a false belief in an individual resurrection.

Compare: Suppose that we argued that members of Alice's religious group tend not to believe in ghosts. Just how far would that go in the absence of more specific evidence about Alice and Alice's claims to convince us that her belief in ghosts must have been rational? Not very far. All the less so if we have a specific record of other people who are part of Alice's religious group, in the same time period, who did believe (falsely) in a ghost.

So none of the popular attempts to use a minimalist account addresses the problem that is induced by trying to do without DT. I suggest instead that we should rely on DT and be prepared to defend it, if and when it is challenged, using the plethora of evidence for Gospel reliability.

I hope that this post is useful as a go-to for my current statement of my central objections to the MFA. Next up: Confusing epistemology with sociology.