Richard Carrier

No worries. Your new explanation was excellent and I completely get your point now. You are right to make this distinction. I am not sure how effective it is against Craig's argument, but you are still making a good point about how his argument lacks certain elements of clarity.

This reminds me of a similar complaint I have made against Craig (and others like him) that distinguishes between evidence and explanation, e.g. often they keep referring to "the empty tomb" as evidence, when in actual fact that is an explanation of the evidence. Craig is really saying an actual empty tomb is the cause of a later story about an empty tomb, whereas skeptics say there are other possible and plausible causes of this story besides there having been an actual empty tomb. This confusion between fact and theory is related to (but not identical to) your point about identifying a confusion between asserting what was the case and asserting why that was the case. And you now make this point much better than I just did.

But you still need to keep this point distinct from Craig's argument about theoretical simplicity, which derives from the methodology of McCullagh. Craig could sort his mess out and thus take into account the distinctions you rightly note he must, and then still come up with a simpler theory accounting for all the facts than any naturalist can present (though you are right he does face a harder time against rival supernaturalists, like Muslims or Orthodox Jews, who can also come up with simpler theories, like your demon-trickery hypothesis).



No, you may be right after all, since I am not up on anything Craig has specifically argued against rival supernaturalists--I only had in mind naturalists, but obviously you are taking into account supernatural theories alternative to the Christian theory, and that does change the playing field a bit (I forgot I had made a similar point in Why I Don't Buy the Resurrection Story).



I can't predict how Craig will respond. Because this would require an entirely different track of argument--in particular, this runs into the debate over the inspiration and prophetic vindication of biblical prophecy, and then into the deeper meta-debate over whether the OT has anything in it that was inspired by the Devil (as Marcion argued) or serves the specific aim of tricking people (as can be inferred from the theory I propose in The End of Pascal's Wager: Only Nontheists Go to Heaven).

In other words, I expect Craig would have to argue that the resurrection was not a divine trick because it corresponds too accurately with the promises of God in the OT and the OT contains only the honest promises of God. This means he would put forward these two "facts" as further "evidence" that his theory explains and yours does not (i.e. his theory now explains three things while yours only explains two). The problem he will encounter is in making the case that these two propositions are true and therefore "facts" in need of explaining. I don't know where your debate with him then would end up.

And that's not even counting the inevitable: that Craig will end up appealing to his own personal communication with the Holy Spirit (the very argument he went straight for when he debated me on TV). This would then enter an entirely new debate over whether the Holy Spirit is sincere or tricking him (assuming you both posit that there is such a thing and that he knows when he is really experiencing it rather than mistakenly thinking so).

As you can see, this is a can of worms, and not one I've ever opened before.



Formally, I'm referring to Craig's own methodology as borrowed explicitly from McCullagh, where there are several criteria for establishing historicity and a theory must meet all of them to be truly secure, and must at least meet more of them than any other theory before you can proclaim it more probable than other theories. Simplicity is only one of those criteria. For "everything else" to be "equal" his theory must meet all the other criteria exactly as well as all other theories. In actual practice, that just isn't so.

To just raise one obvious example, without getting too technical, his theory depends upon the presumption that a particular undiscovered agent exists (a particular god with particular powers and interests--indeed a god who must be so extraordinarily complex as to explain why he would work out his will in such an exasperatingly complicated and roundabout way as the whole Christian theology requires). But no naturalist theory requires the presumption of any agency or cause that we have not already demonstrated the existence of. This difference affects prior probability--the prior probability of any natural explanation is always significantly positive (no matter how low) because all the required factors are known to exist and could have existed in the time and place required, but the prior probability of an undemonstrated agency is always near zero, and therefore his theory fails miserably on the criterion of prior probability in a way that no natural theory does (even ones that do very poorly on that same criterion).

An analogy would be a "space aliens" theory of the resurrection. That requires positing an agency not yet demonstrated to exist--yet I'd say the evidence for alien visitors is actually better than the evidence for Craig's particular and peculiar God (he would debate that, but then that's the point), so I'd even predict that an aliens theory would, in the final analysis, have a higher prior probability than his theory (though one so low as to rule it out against most other naturalist theories). Note that "prior probability" is the probability of something being true before taking into account the specific evidence. So winning on just this one criterion does not count as winning--e.g. if the space aliens theory failed on more criteria than Craig's theory or failed on them more miserably than Craig's theory, it would still be less probable in the final analysis than his theory even if it had a higher prior probability. As you can see, this gets complicated--and I'm making this much simpler than it really is.

But the point is, even if Craig's theory wins on the simplicity criterion, if it then loses on the prior probability criterion, we are back to square one: not knowing whether his theory is the correct one (assuming the win and loss each balance each other in degree). Then if Craig's theory loses on another criterion, but one alternative theory does not, Craig's theory becomes less probable than that one--even if it is simpler. Thus, he must assume all else is equal before he can argue from simplicity.



This is a good example to pick on: as you observe, he is inventing an additional theoretical element to explain the actual evidence, and this does make his theory more complex than he thinks. Whether the natural explanations of the same discrepancy are simpler or more probable and so on is another matter, but you get the picture I hope.

spin

I can see that these are attempts at analysis, but they already assume too much to be useful.

1) If there was no Jesus then there is an empty tomb because there was no Jesus to put in it.

If there is no Jesus, why must there be a tomb?

2) Most likely he did live and did bring the message attributed to him.

This is not based on what came before it. "Most likely" attempts to refer back to a prior situation that needs comment, but doesn't attach to anything. So, this sentence stands as an unsupported opinion.

3) The message attributed to Jesus did not come out of thin air, if he didn't bring it, then who did?...

There are as many messages as there are tellers. That provides an answer which you don't contemplate for "then who did?"

4) and why jump thru hoops to deny Jesus when the message attributed to him is here.?

I personally am not denying him. I'm waiting for someone to make a serious effort to demonstrate why anyone should accept this literary figure as anything other than a literary figure.

You need to get a better hold of logical connections.


spin

Richard Carrier

Hallucination is a construct of the brain's imagination, and as such is extraordinarily sensitive to suggestion and to both cultural and environmental cues--by contrast, ordinary sensory experience is much less vulnerable to such influences. For example, what a person hallucinates will be highly determined by their cultural expectations (e.g. no one hallucinated a robot in antiquity, but always hallucinated what they expected their god to look or sound like; likewise, hypnagogic hallucinations of being choked in the middle ages manifested as winged demons whereas today they manifest as alien abductors), and by their subconscious needs (someone who needs to see something, and then hallucinates, will see what they most needed to see, usually to quell anxiety or guilt or satisfy a great longing), and by their circumstances (lifeboat survivors typically hallucinate ships on the horizon, not elephants in the sky, or when at a dinner table someone will more likely hallucinate food than car parts, in each case their brain taking visual cues from the environment).

But above all is the phenomenon of anchoring: a trusted authority figure can guide a person's hallucination, and consequently can guide the hallucinations of many people in the same room who all equally trust and look up to him. They tend to edit their hallucinations, as well as their memories, to "agree" with what their leader says he saw or heard. People are so susceptible to this effect that it happens even to ordinary sensory memory in otherwise mundane circumstances (e.g. it has been proven that eyewitnesses allowed to discuss what they saw with each other will end up misremembering what they saw to agree with one coherent story, usually closest to the version told by the most charismatic or trusted member of the group). Hallucination is many times more vulnerable to this, because there is no real sensory input or memory to counter anchor suggestive influences--since the brain is inventing the experience out of whole cloth, it will latch onto every assisting source of advice, and thus will readily follow someone else.

We have already documented mass hallucination in both senses: mass groups in which everyone hallucinates something different (usually in an undirected event where no one was expecting to see or hear anything and there is no common cultural or social structure of expectation), and mass groups in which everyone claims to have seen the same thing (one famous case involves lifeboat survivors, but religious visions are the most common examples, precisely because these are usually the only occasions where there is a socially anchored expectation, i.e. the hallucinators all expect to see a certain thing and are receptive to charismatic suggestion by trusted leaders within a context of shared wishes, expectations, and beliefs).

An additional problem is the limitations of reporting. If you and I both see someone, we can never really be sure we saw the same person--no matter how many details we try to compare, there could always remain details that we actually experienced differently. And rarely do we even engage such a comparison of details (even less did they do so in antiquity, when there was little popular conception of hallucination as a benign malfunction of consciousness). If a bunch of people say they saw Jesus, they will all assume every one of them saw the same Jesus--even if they didn't. Indeed, this corrupts testimony, too, as reporters who assume everyone saw the same Jesus will ignore incongruencies and only report shared details, thus making the record seem as if there was greater agreement than there really was.

This is all the more so when the actual hallucinations are highly amorphous and ambiguously described. It is most likely, for example, that the original visions were like Paul's: amorphous lights and a voice. No body or clothes or facial details were present to compare, much less wounds, and it is extremely difficult to compare notes on whether a disembodied voice sounded the same to everyone. Finally, given the effects of anchoring, what people "hear" a voice say (or remember it to have said) will depend largely on what the most respected leader present tells them they heard. And, of course, we will only hear what one party heard--those who heard something else and thus broke off into their own sect (e.g. Apollos and those others whom Paul says in Galatians heard new gospels from angels) we will not hear from them, because the winning sect did not preserve any of their writings--once again making the record seem as if there was greater agreement than there really was.

Of course, the stories of bodily encounters are much more likely just straight fabrications, not hallucinations, though hallucination is still possible--especially considering how much disagreement there was among those who "saw" him (many not recognizing him or doubting it was him or seeing him in a different form and so on) and how little we have in the way of precise detail regarding what they saw.

Celsus

Hallucinations are not the only way an accepted story can come about that had little basis in reality. "Collective memory" can also be understood as a way of consolidating ethnic, national, or other group histories into a coherent whole. These are later accepted, even by witnesses, despite any discrepancies, because the witnesses accept the moral truths that they illustrate. Liisa Malkki's study on memory of the Burundian massacre in Purity & Exile is very instructive (Richard, I recommend it highly for certain other approaches that I rarely see in your debates). There are a lot of instances where one can draw parallels with the Judahite Exile, and plenty of the theories about how "memory" functions in a different social context are explored. "Hallucination" is an unfair term to use, IMO, because the people in question did not necessarily share our rationalistic valorisation for the past as it really happened, and so remembered in it specific ways, some clearly didactic, and others clearly aetiological.

Joel

Hobbs

This reminds me of something I read about the early Mormons when they moved to Utah, but I can't find the reference now. Apparently, when they were debating what to do after Joseph Smith died, Brigham Young made a persuasive speech to the community to move to Utah, and most people agreed and others went along. This was according to the diaries and letters that were written at the time. But diaries and letters written a decade or so later, even by witnesses who had been at the meeting, described it as Young standing up but having Smith's voice miraculously and unmistakably speaking through Young's mouth, and a unanimous decision to agree with what was clearly a message from God. So, in their retellings of the story, legendary aspects quickly arose.

But I may be misremembering what I think I read because it so neatly illustrates something I think is, or at least should be, true. Does this sound familiar to anyone else, and do you remember a reference?

Richard Carrier

I completely agree with everything you say here, and it is a point carried forward in a different way by Evan Fales in his contribution to The Empty Tomb. But my discussion of hallucination here was started by someone else--I'm only clarifying that issue, and thus my focus on that is in the context of this thread.

In terms of public debate, however, your point is much harder to explain to an audience who has no training at all in the differences between our culture and those more like the culture early Christianity found itself in, and no training in the relevant psychology. They generally become immovably hostile to such arguments and assume this is just more liberal academic cultural relativism claptrap, and consequently stop hearing anything else you say. So one must take baby steps, and hallucination is the baby step, because everyone accepts that people hallucinate the divine, because they have to believe this, for nothing else can explain why so many non-Christians see and speak to God and see and hear things completely at odds with Christianity.

Nevertheless, in my own (and largest) chapter in The Empty Tomb I argue that Mark was probably writing pretty much what you are talking about: a symbolic representation of the meaning of the movement and the ideas of the community, and not a historical account in the sense we are familiar with.

Matthew and John may have been doing this, too, but Luke explicitly, and John "indirectly" (via the final paragraph), assert something closer to historical veracity as a goal. Luke in particular consciously follows the genre of historians of his day who did valorize the past as it really happened and made a particular point of attacking those who didn't, and Luke would not have done that had he not meant to imply the same (though what he himself really believed is another matter--and as to his methods, that is yet another matter: e.g. his reliance on Mark as a historical document suggests Luke was not a good historian in any critical sense).

But ultimately, the issue comes to Paul's letters, and the way Paul writes about his many "communications" with God is clearly evocative of audio hallucination. However, as Malina (IMO) persuasively argues, this same language has an equally strong explanation in terms of social expectation (see my discussion here: http://www.columbia.edu/~rcc20/christianity/crisis.html), i.e. Paul would have good reason to claim such experiences in order to get his program accepted and adopted by others. The reason such experiences carried such authority probably stems from a very ancient feature of human consciousness that employs hallucination to serve the community (in the role of the shaman originally, but a role later usurped by so-called "prophets" and priests and priestesses and other "holy men"). This is a point also well argued by Malina, and his books on this are required reading on the origin of Christianity (for this and many other reasons).

There were no doubt hallucinators in antiquity who were taken seriously as having seen or spoken to God (good examples are in Robin Lane Fox's book and elsewhere). For that very reason, these forms of contact with God became the respected cultural norm. Therefore, to get a message accepted as divine required either actually having one of these experiences or claiming to have had one. And there is no way to tell the difference from the written record--there is no way to tell if Paul actually saw and heard what he claimed, or only claimed to have, because what gets written in either case would be exactly the same.

Indeed, human psychology blurs the line here: needing to have such an experience can often actually cause such an experience, so that there is no clear line between having one and inventing one. The same needs and circumstances can produce both. Therefore, we can talk about Paul and other early Christians as hallucinating, without slighting the possibility that all their talk isn't just a convenient copying of the language of others who really had hallucinated an encounter with gods or spirits.

Matthew_Green

I am not sure either but at least this seems to me to put the burden of explanatory power on Craig to explain why skeptics of the resurrection can't devise a theory of causation that explains both the empty tomb and the post-mortem "appearances". I have been thinking a lot about your response over this past weekend since I read it and I realized something. When I noted the distinction between the nature of an historical event and its cause, I realized that if skeptics develop a complex theory to explain both the cause and nature of the post-mortem appearances that Craig cannot claim that his theory is much simpler for reasons that I will try to elaborate.

Let him try. I realized that if Craig buys into the anachronistic distinction between resuscitation and resurrection (i.e. why the dead rose before Christ and Christ is supposedly the first fruits of those who slept is a discrepency "explained" by such a distinction) then one can argue that a simpler idea than the resurrection is a "resuscitation" theory. One can argue, for instance, that Jesus is not the Son of God, or the Jewish Messiah, nor was he part of the holy trinity, but just a Jewish prophet of sorts who was wrongly punished. One can argue that Yahweh felt sorry for the poor guy and decided to resuscitate him to the body he had before the crucifixion (not any glorified body of flesh or "transphysical" as N.T. Wright might call it) as a way of taking pity on Jesus and vindicating him. Craig's "resurrection" theory becomes more complex because he has to argue that not only did Jesus come back to life but that Jesus was given a glorified body of flesh rather than just an earthly pre-mortem body of flesh that was still subject to decay and death. Such a "resuscitation" hypothesis is simpler than Craig's "resurrection" hypothesis.

You want to know of an interesting irony? In trying to be a Christian, I was always troubled by the so-called Virgin Mary "appearances" at Fatima and elsewhere. I thought that if these were some kind of collective "vision", "hallucination", or "delusion" than why couldn't a skeptic argue similarly about the post-mortrem "appearances" of Jesus? Some apologists (Norman Geisler) would argue that some appearances of Mary were demonic in origin. But then, I would reason, the post-mortem "appearances" of Jesus couldn't be used as evidence of his resurrection because other faith groups like the Zoroastrians and Muslims could argue the same thing about the "appearances" of Jesus. Apologists assumed a very difficult task of having to defend the "appearances" of Jesus against theories of delusion or demonic trickery. This always puzzled me and sowed seeds of strong doubts for years.

Craig would play into my hands. If he appearls to the OT for correlation, I can appeal to the works of Jewish scholars like Rabbi Toviah Singer and other Jewish scholars who would call Craig's bluff and reject this kind of reasoning.

I would then respond that his existential "experience" with any such being is special pleading because numerous other religions and faiths offer glowing testimonials and stories of deep religious experience. I would then note that Craig has had his historical arguments ripped to shreds and is resorting to emotional experience and has jumped the border between objective and subjective.

Yesterday, I re-read a paper by Jeff Lowder in which he responds to Craig's arguments about the empty tomb. I also read part of Craig's essay in his debate with Gerd Ludemann. Both used appears to McCullagh's methodology. Craig didn't mention the criterion of "prior probability" and neither did Lowder. Perhaps both discussions weren't meant to be an exhausive list.

Craig's theory would be in worst shape than you make it out to be. I realized that not only does Craig's theory presume that a particular god exists but that this particular god also decided to give Jesus a glorified body of flesh for his resurrection and not a mortal and mundane body that he would've had prior to his death. Craig would have to justify his preference of resurrection over resuscitation. (This is the resurrection/resuscitation note that I made earlier). The "resusciation" option is actually much simpler and although it fails the prior probability criterion as well, Craig's theory fails it even more because Craig's argument is carrying more metaphysical baggage in this case. Craig's begging way too many questions.

I see your point. But I have one question left over from the above discussion. In terms of simplicity- is this simplicity in terms of explaining well-known facts or is this in terms of fulfilling all historical criteria? In other words is simplicity related to history or epistemology?

Matthew

Richard Carrier

"Prior probability" is a term from the Bayesian formulation of the McCullagh method (also in McCullagh's book, but Swinburne is the one who runs with that, while Craig prefers the more "colloquial" version because it plays well with audiences, even though the Bayesian formula is the formally correct version--not that Swinburne uses it correctly).

In McCullagh's colloquial list, the criterion of "plausibility" corresponds to the formal term of "prior probability." McCullagh's criterion of "ad hocness" also attenuates prior probability (since an increase in ad hocness entails a decrease in prior probability, but usually an increase in evidential probability at the same time--and when Craig accuses a theory of being less simple, he usually means more ad hoc).



Simplicity relates to elements of a theory that are not directly in evidence. A simpler theory is one in which fewer independent elements are needed to explain the same evidence. Simplicity is measured only as the quantity of independent elements, because dependent elements by definition follow necessarily from other elements, so the probability of those elements appearing and acting together is 100% or very nearly so. Only when that probability is substantially < 100% do you have independent elements, and the more you have to "propose" to explain the evidence you have, the more complicated your theory is.

The issue is complicated by the fact that the set of evidence to be explained is somewhat arbitrarily defined. For example, Doherty's theory explains a much wider evidence set than any other theory (it thus has greater "explanatory scope" in McCullagh's terminology, since it explains all evidence outside the NT better), and that makes it a simpler theory with respect to that evidence set, but not necessarily with respect to a smaller evidence set (e.g. the content of the New Testament). His theory might even be more complex relative to explaining just the NT than Craig's theory (I'm not saying it is--just that it could be...I haven't analyzed this enough to say one way or the other), yet it could at the same time be less complex than Craig's theory relative to explaining the larger evidence set (for example, Craig's theory entails things about God's plans and powers that cannot itself explain why Christianity broke so quickly and so easily into so many opposing factions, especially those particular factions which were more dependent upon direct revelations from the holy spirit than orthodox Christianity--so Craig must propose additional theoretical elements to explain this, whereas Doherty does not).

Again, in terms of method and analysis, there is nothing simple about any of this.