Friday, August 14, 2020

Minimal facts vs. Maximal Data


Minimal facts vs. Maximal Data

(Originally published at What's Wrong With the World. Link to original post at 'permalink' below.) 

Recently I had some correspondence with someone who found my interview from several years ago with atheist Luke Muehlhauser and asked me to look around Muehlhauser's (now-archived) web site and respond to some posts there. In the course of doing so I came upon this part of Muehlhauser's deconversion story:

What I learned, even when reading Christian scholars, shocked me. The gospels were written decades after Jesus’ death, by non-eyewitnesses. They are riddled with contradictions, legends, and known lies.

That the gospels were written decades after Jesus' death is apologetically fairly unimportant. From an historical perspective, there is no reason to distrust a document because it was written "decades" after the events it tells about. This statement is even compatible with a document's being written by a careful and truthful eyewitness of the events recounted! Moreover, "decades" could mean as little as twenty years.

But what about the rest: "Riddled with contradictions, legends, and known lies"? That would be problematic, if it were true. And "written by non-eyewitnesses" definitely implies that we know that John wasn't written by John and Matthew wasn't written by Matthew. Then there are Luke and Mark, which on the traditional view were written by people who had access to and conversations with eyewitnesses, but Luke M. obviously thinks that he "learned" that no such thing is the case.

There is an approach to arguing for the resurrection of Jesus Christ known as the minimal facts approach. Versions of this argument have been made by William Lane Craig and Gary Habermas, for both of whom I have the greatest respect. But there is one problematic aspect of minimal facts arguments: They tend to be presented in such a way as to imply that it doesn't matter to the strength of this case for the resurrection if the gospels are historically unreliable. Proponents of the minimal facts approach do not rely heavily upon the details of the resurrection accounts in the gospels themselves, except for a few general aspects (e.g., that Jesus was believed to have appeared to a variety of people), preferring to put more weight upon Paul's creedal statement about the resurrection in I Corinthians 15. The very strong impression given is that we can get a very strong case for the resurrection of Jesus even if the gospels are historically unreliable and even if the resurrection stories in the gospels are beefed-up, legendary accretions.

I submit that this is highly problematic. A minimal facts argument would be fine as a first statement of some of the issues, as a first approach, but it becomes positively misleading if those who learn this method think that they can lightly toss the gospels to the likes of Bart Ehrman and that we can be fully justified in believing in Jesus' resurrection even if the gospels are, in the words of deconvert Muehlhauser, "riddled with contradictions [and] legends."

First of all, some evidence that I am not misrepresenting this loose approach to the reliability of the gospels. Here, from William Lane Craig:

The Christian apologist seeking to establish, for example, the historicity of Jesus’ empty tomb need not and should not be saddled with the task of first showing that the Gospels are, in general, historically reliable documents. You may be wondering how it can be shown that the Gospel accounts of the discovery of Jesus empty tomb can be shown to be, in their core, historically reliable without first showing that the Gospels are, in general, historically trustworthy. Read chapter 8 to find out. Reasonable Faith, Preface to the Third Edition, pp. 11-12.

We shouldn't "saddle" ourselves with the task of showing that the gospels are historically reliable. We can get what we need without that.

Even stranger and more surprising is a passage here, where Craig (who is among the most careful, brilliant, and worthy analytic philosophers of religion now living) seems to move confusingly from Biblical inerrancy to historical reliability and back again to inerrancy without noting the shift of emphasis:

[W]e have a very strong case for the resurrection of Jesus. That case in no way depends on the Bible’s being inerrant. This became very clear to me during my doctoral studies in Munich with Wolfhart Pannenberg. Pannenberg had rocked German theology by maintaining that a sound historical case can be made for the resurrection of Jesus. Yet he also believed that the Gospel resurrection appearances stories are so legendary that they have scarcely a historical kernel in them! He did not even trust the Markan account of the discovery of the empty tomb. Rather his argument was founded on the early pre-Pauline tradition about the appearances in I Corinthians 15.3-5 and on the consideration that a movement based on the resurrection of dead man would have been impossible in Jerusalem in the face of a tomb containing his corpse.

Evangelicals sometimes give lip service to the claim that the Gospels are historically reliable, even when examined by the canons of ordinary historical research; but I wonder if they really believe this. It really is true that a solid, persuasive case for Jesus’ resurrection can be made without any assumption of the Gospels’ inerrancy.

(I will have much more to say about Pannenberg below.) I am particularly disturbed by the implications of this sentence: "Evangelicals sometimes give lip service to the claim that the Gospels are historically reliable, even when examined by the canons of ordinary historical research; but I wonder if they really believe this." Prima facie, this sentence is expressing a lot of dubiousness about the claim that the gospels are historically reliable when examined by the canons of ordinary historical research. In fact, a natural reading of the sentence would be that Craig is so dubious about this claim that he doubts that even many evangelicals who give "lip service" to it really believe it! Well, I'm not sure whether I count as an evangelical (being that I'm a sacramentalist of sorts), and I'm even less sure that I count as an inerrantist according to the most natural interpretation of the Chicago Statement, but let it be noted here, now, and for the future that I really do believe that the the gospels are historically reliable when examined by the canons of ordinary historical research. More: I think that's pretty important to apologetics, as I will argue here. And if a lot of evangelicals, especially leaders and pastors, are just giving "lip service" to that proposition, then they need to do some digging and learn more. Not so that they can sign off on a statement of faith that includes inerrancy (which wouldn't be guaranteed by historical reliability in any case), but so that they can present the arguments of historical apologetics with confidence that they aren't dealing with a handful of legend-riddled documents.

Here is a fairly typical statement of the minimal facts case for the resurrection of Jesus. The point I intend to focus on is Fact #3, stated here as:

On multiple occasions and under various circumstances, different individuals and groups of people experienced appearances of Jesus alive from the dead.

In this interesting discussion of different versions of the minimal facts approach, Gary Habermas quotes from Mike Licona a slightly different wording of this same general minimal fact:

Very soon afterwards [i.e., after Jesus' death], Jesus’ disciples had experiences that they believed were appearances of the resurrected Jesus.

One might think, reading these statements, that they mean to take as a "minimal fact" to be explained that the disciples and others (such as the women at the tomb) had experiences of the kind that are described in the gospel resurrection stories--that it seemed to them and that they claimed that they talked with Jesus (experiences both auditory and visual), that Jesus ate food in their presence, that they spent periods of time interacting with Jesus in groups on repeated occasions, that Jesus invited them to touch his scars, and the like.

It's important to emphasize, however, that that is not what is meant by this minimal fact. The reason is that "consensus of New Testament scholars" is very important to the minimal facts approach. Here is how Habermas puts the consensus issue:

From the outset of my studies, I argued that there were at least two major prerequisites for an occurrence to be designated as a Minimal Fact. Each event had to be established by more than adequate scholarly evidence, and usually by several critically-ascertained, independent lines of argumentation. Additionally, the vast majority of contemporary scholars in relevant fields had to acknowledge the historicity of the occurrence.


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.

It is a little surprising that Habermas should appear to agree with Licona about the importance of this scholarly agreement between "conservative" and non-conservative scholars, given that Habermas also says,

Of the two criteria, I have always held that the first is by far the most crucial, especially since this initial requirement [being supported by more than adequate scholarly evidence] is the one that actually establishes the historicity of the event. Besides, the acclamation of scholarly opinion may be mistaken or it could change.


I have always doubted, given what I do know about New Testament scholarship, that the vast majority of New Testament scholars agree that the disciples had experiences of the kind recorded in the gospels that gave them the idea that Jesus was risen from the dead.

In researching this post I found confirmation of this suspicion directly from William Lane Craig here. Craig makes statements that one might at first (incorrectly) take to mean that the consensus of scholarship supports the disciples' experiences as recounted in the gospels:

First, the resurrection appearances. Undoubtedly the major impetus for the reassessment of the appearance tradition was the demonstration by Joachim Jeremias that in 1 Corinthians 15: 3-5 Paul is quoting an old Christian formula which he received and in turn passed on to his converts According to Galatians 1:18 Paul was in Jerusalem three years after his conversion on a fact-finding mission, during which he conferred with Peter and James over a two week period, and he probably received the formula at this time, if not before. Since Paul was converted in AD 33, this means that the list of witnesses goes back to within the first five years after Jesus' death. Thus, it is idle to dismiss these appearances as legendary. We can try to explain them away as hallucinations if we wish, but we cannot deny they occurred. Paul's information makes it certain that on separate occasions various individuals and groups saw Jesus alive from the dead. According to Norman Perrin, the late NT critic of the University of Chicago: "The more we study the tradition with regard to the appearances, the firmer the rock begins to appear upon which they are based." This conclusion is virtually indisputable.

But Craig goes on to make his meaning quite clear:

At the same time that biblical scholarship has come to a new appreciation of the historical credibility of Paul's information, however, it must be admitted that skepticism concerning the appearance traditions in the gospels persists. This lingering skepticism seems to me to be entirely unjustified. It is based on a presuppositional antipathy toward the physicalism of the gospel appearance stories.

In other words, when the architects of the minimal facts approach speak of a vast scholarly consensus on the "appearances" experienced by the disciples, they do not mean a consensus on the physical-type experiences recounted in the gospels. To those aspects of the stories, Craig acknowledges that many scholars still actually have an antipathy, though he (correctly) suspects that this is based on an ideological rather than a scholarly objection.

The weakness of what the minimal facts approach is claiming about the disciples' experiences is further confirmed by Craig's discussion of Wolfhart Pannenberg. Here is more on Pannenberg's view from Craig:

I don’t think that Pannenberg’s objection to the appearances is based on naturalism or a bias against miracles. He is already committed to miracles in affirming the empty tomb. Rather, I think it would be exegetical, frankly. He is convinced by Grass’ exegesis of 1 Corinthians 15 that when Paul talks of a spiritual body, what he is talking about is an immaterial, invisible, unextended body. Therefore, the Gospel appearance stories are late legendary developments that represent a kind of materializing of the original, primitive, spiritual experiences. The original experiences were just these visions of Jesus. It would be similar to Stephen’s vision of Jesus in Acts 73. When Stephen is being stoned, he sees the heavens open and he says, “I see the Son of Man in the heavens.” Nobody else saw anything, but Stephen saw this vision of Jesus. And I think that Pannenberg would say that that is similar to what the original resurrection appearances were. They were these visionary events and then they got corrupted and materialized and turned into the Gospel appearance stories, which are very, very physicalistic.

That is quite clear. The "experiences" that Pannenberg accepts that the disciples had were some kind of "visionary events," others who were present would not have seen anything, and the gospel accounts are physicalistic "corruptions."

The extremely minimal nature of the scholarly consensus on the disciples' experiences is also confirmed by the careful wording used in describing what is agreed upon by New Testament scholars. Habermas has done extensive research in this area. Indeed, much of the information we have about the scholarly consensus comes from his meticulous research. Here are some of the wordings he uses in discussing the results of his research on scholarly opinion:

The nearly unanimous consent of critical scholars is that, in some sense, the early followers of Jesus thought that they had seen the risen Jesus. "Resurrection Research from 1975 to the Present: What Are Critical Scholars Saying?" Journal for the Study of the Historical Jesus, 3.2, p. 151.

The phrase "in some sense" is worth noting.

The vast majority of scholars agree that these persons certainly thought that they had visual experiences of the risen Jesus. "Resurrection Research From 1975 to the Present," p. 152.

The reference is to visual experience, not to tactile or even auditory experiences, much less to the length of the experiences nor to conversational interaction with Jesus. Habermas does imply that there is a high degree of scholarly consensus that these visual experiences (whatever they were like), were reported to have been "witnessed both by individuals and groups" (p. 152).

I think it is necessary to be blunt: If all that we are going to assert and seek to explain is the claim that Jesus' disciples had some kind of visual experiences soon after his death that they took to be appearances of the risen Jesus, and if we are allowing that these experiences could, for all we know, have been fleeting, unclear, intersubjectively inaccessible (that is, invisible to anyone other than the disciples), and involving no senses other than sight, then the case for the resurrection is gravely weakened.

If one hangs onto the idea that these experiences (whatever their precise nature) came "both to individuals and groups," and if one includes James, Jesus' brother, among those who had an individual experience (Habermas discusses the question of whether an appearance experience on the part of James should be included as a "minimal fact"), then this will provide an interesting coincidence, and naturalistic explanations will be somewhat strained. I admit that. Why should these various people, including a former skeptic of Jesus' ministry (his brother) have had these experiences shortly after his death, even if they may (for all we know) have been somewhat vague and visionary in nature?

But let's be clear: The conclusion we thought we could support was that Jesus was risen from the dead. Vague, fleeting, or visionary experiences provide a weak case for that conclusion. In fact, if the minimal fact of the appearance experiences is compatible with minimal experiences, then paranormal explanations become an interesting option, which I gather is what New Testament scholar Dale Allison is exploring. Maybe there's just "something weird" in this world that we don't know much about that isn't a miracle, and isn't a resurrection, but that causes people to have brief experiences "of" a person after his death.

The example of Pannenberg, so far from showing that the gospel narratives are unnecessary to a defense of the resurrection, shows just the opposite. Pannenberg believes that God miraculously made Jesus' dead body disappear from the tomb, raised Jesus in an invisible "spiritual body," and took him to heaven, from which he sent visions to the disciples. This (with variants on it) is generally known as the "objective vision" theory of the resurrection.

To say as Craig does that Pannenberg believes in "the resurrection of Jesus" is to say something that ought to be rather theologically controversial. Pannenberg does not believe in the physical resurrection of Jesus, and that is precisely why he ditches the gospel accounts! He realizes that they are the strongest evidence for the physical resurrection he rejects. There is something oddly backwards about implying that we do not need to support the reliability of the gospels in order to support a belief in the resurrection of Jesus and illustrating this claim by reference to a theologian who doesn't even believe that Jesus walked about visibly on the earth after his "resurrection."

In fact, it is worth asking just how strongly any supernatural explanation is supported by an attenuated body of evidence that has been deliberately weakened by a willingness to waive the whole question of whether the "physicalist" gospel resurrection accounts are legendary late additions.

A case in point here is Gerd Ludemann, a New Testament scholar whom Craig expressly cites to support the contention that there is scholarly consensus on the resurrection "appearances." Ludemann says, "It may be taken as historically certain that Peter and the disciples had experiences after Jesus's death in which Jesus appeared to them as the risen Christ."

I hope that by this point in my post, however, readers have learned to question dodgy, New-Testament-scholar-like expressions such as "had experiences after Jesus' death in which Jesus appeared to them as the risen Christ." Such statements are by no means as straightforward as they might appear to the unsuspicious eye. And so it is with Ludemann, who believes that the disciples had hallucinations, starting with Peter, who felt guilty after Jesus' death because he had denied that he knew Jesus. The other disciples then got fired up by Peter and sort of "caught" the tendency to have hallucinations in which Jesus appeared to them.

Now, don't misunderstand me: This is a very foolish hypothesis, and Craig and Habermas are right to critique it. But we are critiquing it with a huge handicap if we insist on doing so without using the gospel resurrection accounts as reliable indicators of what the disciples claimed. One can, perhaps, just barely wrap one's mind around a sort of "hallucination virus" that spreads among grief-stricken followers if the phenomena alleged are a lot different from what the gospels recount--a lot less like actual meetings and conversations with a real, tangible, physical person and a lot more like spooky ghost stories. That is precisely where we need to press in answering Ludemann.

Interestingly enough, Gary Habermas goes a bit in that direction in answering Ludemann, but in doing so he has to go beyond what he has supported as being part of the consensus of the vast majority of New Testament scholars. Habermas asks,

What about the natural human tendency to touch? Would no one ever discover, even in a single instance, that their best friend, seemingly standing perhaps just a few feet away, was not really there?

That's very interesting. As far as I know, Habermas's own careful wording elsewhere shows that he does not have scholarly consensus on the idea that the disciples had long enough and detailed enough "appearance experiences" that this question would have arisen. Do the vast majority of scholars concede that Jesus appeared accessible to touch, that he appeared to be standing just a few feet away, that the disciples had the opportunity to recognize him by distinct visual inspection as their dearest friend? As far as I know, no such consensus has been ascertained, and the minimal facts approach does not depend on asserting those types of appearances.

Of course, the gospels say that Jesus invited his disciples to touch him, but the minimal facts approach does not take that data into account even as part of what the disciples claimed.

Similarly, Habermas argues that the disciples were not in a frame of mind like that of pilgrims to a holy site, a mindset of expectation, that might give rise to ecstatic hallucinations. To some degree he argues this on the basis of what they "would have felt like"--the trauma, sadness, and certainty of Jesus' death following upon his crucifixion. But of course such a case would be much strengthened by the gospel accounts of what frame of mind the disciples were in.

Habermas also states, "Men and women, hard-headed and soft-hearted alike, all believing that they saw Jesus, both indoors and outdoors, by itself provides an insurmountable barrier for hallucinations." But is that supported by the vast consensus of scholarship--that Jesus was seen both outdoors and indoors, by hard-headed and soft-hearted alike, and by men and women? Is that clearly asserted as a "minimal fact"? Habermas himself has acknowledged a difficulty in deciding what to include in a list of minimal facts agreed upon by a large array of New Testament scholars:

[S]ince I have surveyed this material for decades, I can report that most contemporary critical scholars actually concede far more facts than those included even in the long list, let alone just the few Minimal Facts alone. But the problem is that, as the numbers of events expand, fewer scholars agree on each one. So there could be more give and take on “whose facts” ought to be utilized. Obviously then, longer lists would not fulfill especially the second strict criterion of the Minimal Facts method.

As far as I can tell, this difficulty would plague any attempt to attribute to the "great consensus of scholarship" a claim about the great variety of circumstances (soft-hearted and hard-headed, men and women, indoors and outdoors) in which Jesus allegedly appeared to his followers after his resurrection. For example, Licona points out that the number of scholars who even address the conversion of James and its cause is small. Furthermore, how easily such a variety of appearances can be explained naturalistically will depend greatly upon the surrounding circumstances and the nature of the alleged appearance experiences.

Let me try to put this in broad probabilistic terms: The data constrain the explanations we bring to bear. Whether an explanation is far-fetched and ad hoc or not, and how far-fetched or ad hoc it is, depends on what data we are explaining. If we are unwilling to get at all nitty gritty about the nature of the "experiences" the disciples claimed to have had of Jesus risen, then it is easier to argue that naturalistic or paranormal explanations will do the trick. In Bayesian terms (if you happen to be interested in those terms), the Bayes factor for an "appearance" is considerably weaker in favor of the conclusion that that appearance is veridical if one leaves vague the question of what the appearance was like. When the assertion that the disciples had appearance experiences is so weak that it is consistent with purely visionary experiences of an intangible Jesus, inaccessible to any but his followers, experiences that, for all that is stated to the contrary, might have been fairly brief, involving sight and no other senses, then it becomes a much, much harder task to argue that there must have been a supernatural explanation for what happened. It becomes harder still to argue that the correct explanation is that Jesus really was physically risen from the dead. I won't go so far as to say that a minimal facts case thus construed provides no evidence for Jesus' literal resurrection, but it is a much weaker case than a case that includes, as data indicating what the disciples claimed, the types of experiences actually recounted in the gospels.

To argue that this was indeed what the disciples claimed it was like for them to see the risen Jesus, one needs to argue that the gospels are not riddled with legends. To some degree, one needs to be willing to buck the trends of overly literary New Testament scholarship and to argue that the gospels are reliable historical memoirs of Jesus, written by those close to the events (some of them eyewitnesses).

But this argument can be made! Both external and internal evidence support this forward position on the memoir nature of the gospels. My husband and I have spent the last several years mustering such evidence from older authors (see some more sample links here), and I am happy to say that various other apologists are doing the same. To name just two: Jay Warner Wallace has discussed undesigned coincidences; Peter J. Williams has independently begun discussing external evidences for the historical veracity and eyewitness sourcing of the gospels.

In no way do I mean to disparage the work of Dr. Craig and Dr. Habermas. Their immense work is a gift of God to the church. But I think the ambiguity of the minimal facts approach on the matter of the appearances is a subject that has not been discussed enough in apologetics circles. This means that the weakness I am concerned about has gone largely unnoticed, because it is easy to assume that the "appearance experiences" referred to are the types of experiences we read about in the gospels.

I have become concerned over the past few months as I have realized that there are young apologists in training who believe that even major concessions to skeptical New Testament scholars like Bart Ehrman are not all that damaging to the case for the resurrection so long as we can "get some historical information" out of the gospels. I fear that Craig's own writings, perhaps influenced by his interest in Pannenberg and by a desire to induce evangelicals not to lean too heavily upon inerrancy, have encouraged this conclusion. So I have decided that it is important to step up and disagree with any such implication concerning the relative apologetic unimportance of the gospels' reliability.

When it comes to arguing for miracles, both God and the Devil are in the details. That is why it is such a dicey business in the area of miracle claims to set aside pertinent information and to waive contentious questions just because they are contentious. To see a strong probabilistic case aright, we should base our conclusion on all the available evidence. The existence of the gospel accounts and the evidence for their closeness to the facts are highly pertinent data to be taken into account. I therefore propose that, instead of arguing from Minimal Facts, we should argue from Maximal Data.

No comments: