Geivett’s Exercise in Hyperbole (1999, 2005)
Richard Carrier
[Part 4B of a larger Review of In Defense of Miracles.]
Going Too Far
Douglas Geivett’s basic conclusion that it is possible to use miracles as evidence for the existence of God is formally correct (as is his reversed version of the argument, which I address in my conclusion below). I have acknowledged and expanded on this in my review of Corduan and Purtill. But it simply does not work when given the actual evidence that we have. Geivett tries to defeat this problem with an unusual rhetorical tactic: constant repetition of ridiculous exaggerations. Since I will be addressing the actual evidence in my reviews of Newman and of Craig and Habermas, and have already covered the relevant historiographical issues in my review of Beckwith, and since I already agree with Geivett that the argument from miracles is at least logically sound, I will use this space here to exhibit his strange penchant for hyperbole. For there is no better sign of a man’s desperate situation than a resort to a blatant overstating of his own case.
The Argument from Miracles
First, I will describe where Geivett is basically correct. Given that an event is accepted as having no explanation in natural causes, he describes the logical dilemma that arises: either there is no cause at all, or the cause is nonnatural (i.e. “supernatural”). The problem of defining “natural” is something I address in my review of Nash, but the dilemma here is linguistically correct, whatever the definition of “natural” may be. Of course, if it means what many scientists actually use the term to mean (i.e. anything that can be observed is a part of this universe and hence “natural”), then the dilemma collapses into a single option (the absence of any cause). But what Geivett has in mind is what would be better stated as follows: Given that an event is accepted as having no unintentional cause, either there is no cause at all, or the cause is intentional. This would divide all events that “merely” follow the laws of physics from all events which proceed from an intentional agent, namely humans and, in his argument, God.
In other words, if it becomes reasonable to believe that a certain event (e.g. the resurrection account) has no explanation in terms of unintentional causes, then we must believe that the event was not caused at all (that it just happened out of the blue), or that it was caused deliberately (either by human or divine methods)–unless of course we reserve all judgment entirely. Since regarding an event as complex as this as totally “uncaused” requires believing something about the universe that we are least justified in believing (having never seen such a random occurrence), we are most justified in believing in an intentional cause. This is where the argument must go from here: given that the event is accepted as having no explanation in human causes, either there is an alien cause, or there is a divine cause. If we subsume all possible alternatives to God under “alien” (such as demons, or other gods, or people with superhuman powers, etc.), then this dilemma is as linguistically valid as the first. If there is no evidence of any particular “alien” origin (as there is not in the case of the resurrection), but some evidence of a divine origin (as there is in the case of the resurrection, however weak), then it is reasonable to believe the event was of divine origin.
So far so good. But notice the two catches: first, the evidence must be such as to lead us to reasonably accept that a particular event has no explanation in either unintentional or human causes. This is essential for the actual success of the “argument from miracles.” In other words, to use the argument from miracles so as to convince someone to believe in God, it is necessary to demonstrate that there can be no other explanation. As long as there are plausible explanations in the laws of biology and physics or in human design, which cannot be eliminated, the argument from miracles cannot succeed. It is not enough that “natural” explanations be improbable, because improbable things happen all the time. Rather, it is necessary that there be positive evidence against these natural explanations sufficient to compel us to reject them as impossible in the specific case at hand.
Now, this is the argument as Geivett presents it, and it can be summed up simply as follows: if an event is such that only a god could have caused it, then the event is evidence for the existence of a god. The trick is in establishing the “only” part (which equates to a similar difficulty in Purtill’s Definition). But there is an even better argument: if we saw God regularly acting today, it would automatically be reasonable to believe that he has acted before. This argument is even more difficult to launch, however, since it requires what isn’t the case: namely, that God be acting regularly today. And this brings up an important defeater to Geivett’s approach: why isn’t God more visible and active today? One can speculate at whim, but no one will be able to offer any proof to support one speculation over another. And insofar as the most reasonable answer is that there is no god to act in the first place (since there is a collection of evidencing reasons to support that theory, as even theists must acknowledge–for example, see my book Sense and Goodness without God 2005), the attempt to appeal to the divine is undermined even if there is a good evidential case for a miracle, since the second dilemma (alien or divine) will no longer be so easy to resolve–unless the divinity theory can be so revised as to explain all the other facts about the universe which seem to contradict it, and these revisions have some evidential support other than the sole fact that they rescue the theory. But as it is, no real case ever gets that far.
In the process of all this, Geivett makes an argument that is not only correct, but absolutely precious, and should be paraphrased in every relevant debate: “Failure to eliminate the possibility of a naturalistic explanation does not entail that a naturalistic explanation currently exists or is close at hand” (184). Substitute “theistic” for “naturalistic” in this sentence and you will have something that atheists have been trying to drill into the heads of theists for centuries. He is quite right, and naturalists should take heed, to avoid committing the “naturalist fallacy” which I address in my review of Corduan. But theists should take heed, too: concocting an unfalsifiable definition of God does not make belief in God rational. It is perfectly reasonable to disbelieve in the existence of a God that you cannot prove doesn’t exist, so long as you have no other good reason to believe it (see my article Proving a Negative for more on this point).
Slow Down, Pal!
But when it comes time to apply this argument to reality, Geivett goes overboard. For instance, in defending the possibility that it would be reasonable to accept that there is no unintentional or human cause for an event, he brings up Quine’s notion of a “recalcitrant” experience which is “one that stubbornly resists explanation within the framework of a given paradigm or web of beliefs” (182). All well and good. But then he doesn’t show even a hint of sarcasm or qualification when he calls the resurrection of Jesus “a recalcitrant experience of the highest order,” so “bodacious” that “if [it] doesn’t count as a violation of presumed natural law, then nothing does” (183). This is an astonishingly absurd statement. Certainly even he can think of something that is more obviously a violation of natural law than a mere resurrection, which is accomplished on a regular basis today using CPR and electric defibrillators, and which is also known to happen naturally, however rarely (see How Do We Know He was Dead? from my article on the Resurrection of Jesus). And something cannot be a recalcitrant experience of the “highest” order when I can think of a dozen experiences that would be even more recalcitrant still (and that’s just off the top of my head), and when Geivett’s colleagues and I myself have described just such events (from my talking “fish” example to Corduan’s “regrown hand” and modern “holy man” examples–see my review of Corduan and Purtill). Hence, Geivett is far overselling the resurrection here, and that reflects badly on either his objectivity or his honesty. He sounds more like a preacher or a sophist than a scholar.
This is no isolated slip. Nor is it trivial, for his “evidence” for this magnificent recalcitrance is cleverly couched in a false dilemma which excludes a crucial state of affairs. Geivett says that “given a choice between treating the…resurrection as an anomaly that will eventually be explained by science and repudiating the historicity” of the account, “today’s naturalist will generally opt for the latter” and “this is presumably because of the recalcitrant character of the event” (183). It is curious that he would think such a dilemma valid, when I know of no naturalist today who actually thinks it likely that the resurrection “will eventually be explained by science,” simply because there is no good prospect that we will invent the needed time machine to investigate the matter “scientifically” in the first place. Rather, most naturalists think there are already good, plausible, natural explanations which are consistent with their total worldview, and that if we could go back in time to test them, we would find one of them correct.
Moreover, naturalists, such as myself, who favor the “unhistorical features” explanation as a more likely account of the record as we have it, base this on the obvious fact that historical error or deception, or both, is already so very likely, having such a vast amount of repeated examples to prove its frequency, that there is hardly any need to resort to other explanations, even though there are many for the taking (see my article on the Probability of Survival vs. Miracle and my two chapters, “The Plausibility of Theft” and “The Burial of Jesus,” in the book The Empty Tomb: Jesus Beyond The Grave). This is all the more so because we actually have positive evidence of error or fabrication in the accounts themselves and in the history of the church, which we can use to build hypotheses from, whereas we are unable test any natural explanations of the event itself, either to prove or refute them, since all physical evidence is lost, and what little we do know comes to us from late, biased, and uncritical sources. Geivett is thus not only vastly overplaying the recalcitrant nature of the resurrection, but he is using his own hyperbole as if it were “proof” that naturalists actually believe the account to be recalcitrant. Does Geivett understand skeptics so poorly? Or is he simply eager to invent a straw man?
Julius Caesar Crossed the Rubicon,
But Was Jesus Resurrected from the Dead?
Geivett tries to press this tactic further with the assertion that “if one takes the historian’s own criteria for assessing the historicity of ancient events, the resurrection passes muster as a historically well-attested event of the ancient world” (185). Of course, as is typical of Christian apologetics, he never cites a single historian or explains what these “criteria” are, or who actually uses them and why (I give a good accounting of what historical method really looks like in my review of Beckwith). He then issues a comparison, in the voice of a mock critic, asserting that the resurrection of Jesus is as historically evidenced as Julius Caesar’s crossing of the Rubicon in 49 B.C. But let’s examine that claim:
- First of all, we have Caesar’s own word on the subject. Indeed, The Civil War has been a Latin classic for two thousand years. On the other hand, not only do we not have anything written by Jesus, but we don’t even have anything written by anyone who actually knew him–unless we accept the questionable authenticity of some of the non-Pauline epistles, but they don’t describe the resurrection and thus present no direct evidence of that event anyway.
- Second, we have many of Caesar’s enemies, including Cicero, reporting the event, whereas we have no hostile or even neutral records of the resurrection until long after the Christian’s own claims had been written down and widely spread across the whole Empire.
- Third, we have coins and inscriptions produced in the very same years of the Republican Civil War, and shortly after, which serve to corroborate the event. But we have absolutely no physical evidence of any kind in the case of the resurrection.
- Fourth, we have the story of the Rubicon crossing from several historians of the period, including the most prominent scholars of the age: Tacitus, Suetonius, Appian, Cassius Dio, and Plutarch. Moreover, these scholars have at least some measure of proven reliability, since a great many of their reports on other matters have been confirmed in material evidence and in other sources. In addition, they all quote and name many different sources, showing a wide reading of the witnesses and documents, and they show a regular desire to critically examine claims for which there is any dispute. If that wasn’t enough, all of them cite or quote sources which were written by witnesses, hostile and friendly, of the Rubicon crossing and its repercussions. Compare this with the resurrection: we have not even a single prominent historian mentioning the event, and of those few people who do bother to mention it, none of them show any wide reading, never cite any other sources, show no sign of a skilled or critical examination of conflicting claims, have no other literature or scholarship to their credit (which could in turn be tested for accuracy by comparison with other evidence), and have an overtly declared bias towards persuasion and conversion.
- Fifth, the history of Rome could not have proceeded as it did had Caesar not physically moved an army into Italy. Even if Caesar could have somehow cultivated the mere belief that he had done this, he could not have captured Rome or conscripted Italian men against Pompey’s forces in Greece. On the other hand, all that is needed to explain the rise of Christianity is a belief that the resurrection happened. There is nothing that an actual resurrection would have caused that could not have been caused by a mere belief in that resurrection. Thus, an actual resurrection is not necessary to explain all subsequent history, unlike Caesar’s crossing of the Rubicon.
It should be clear that we have a huge number of reasons to believe that Caesar crossed the Rubicon, all of which are lacking in the case of the resurrection. In fact, when we compare all five points, we see that in four of the five evidences of an event’s historicity, the resurrection has no evidence at all, and for the one kind of evidence it does have, it has not the best, but the very worst kind of evidence–a handful of biased, uncritical, unscholarly, unknown, second-hand witnesses.
Indeed, you really have to look hard to find another event that is in a worse condition than this as far as evidence goes. And this is the case even before we consider the relative probability of the event. In other words, I have not even mentioned the fact that an “actual” resurrection–in light of the incredible paucity of evidence that any miracle has ever really happened–is very, very improbable. It is so improbable, in fact, that even a very unlikely natural explanation would still be more probable, as I explain in Adding Things Up. For more on the relevant issues of historical method in general, see my review of Beckwith. But I discuss the Resurrection issue at great length in my collection on Why I Don’t Buy the Resurrection Story, especially the Main Argument thereof, where I improve and expand on the Rubicon comparison, as again in my book Sense and Goodness without God.
If a Lie is Big Enough, Suckers will Believe It
Even though this is the real status of the evidence, Geivett pretends that the exact opposite is the case, and continues with a stream of hyperbolic assertions that are so blatantly false even the Pope would be ashamed to repeat them. For instance, in Geivett’s world, “it is unlikely that…the resurrection would otherwise enjoy such historical authentication according to the highest standards of historical inquiry” [emphasis added] “if the tomb was not in fact empty and Jesus was not observed alive again” (186). Geivett calls this a “modest” suggestion! How on earth can he claim that the New Testament, or any of the post-biblical literary evidence, meets the “highest standards” of historical inquiry? Am I to believe that he thinks the unnamed author of the Gospel of Matthew is to be ranked with Thucydides, Polybius and Tacitus? That having no physical or contemporary evidence of any kind is somehow the epitome of “historical authentication”? Does he smell what he’s shoveling?
Just in case you think he just slipped once into a bit of an exaggerative mood, he does not let up: there are “numerous observations” of the “relatively innocent facts” of the empty tomb and the post mortem appearances of Jesus, making them “supremely well attested historically” (186). Indeed, he even dares to say that “there is no compelling reason to dispute their historicity.” Yet we have no eyewitnesses, no mention of an empty tomb or physical appearances in any of the epistles–the earliest literature–and of the four gospels that were chosen as orthodox, two of them unmistakably copied from the first, and the fourth is the last of them all to appear.
Maybe Geivett does not understand the difference between numerous sources and what those sources (and only those sources) claim to have been numerous observers, although no historian would make such a mistake. But that is not all that is wrong here: remember that skeptics do not regard natural explanations as implausible. Even if we were to accept, on such feeble evidence (as we sometimes may, for less incredible things), that there may have been an empty tomb and post-mortem appearances, this would still not demonstrate a miracle, since these are indeed “relatively innocent facts” which do not necessarily follow an actual death, for if the appearances are genuine, we have no real proof that Jesus died. Moreover, these things still have a natural explanation even if he did die: for there are several reasons that his tomb might end up or be mistaken as empty, no matter how odd they may be, and there are good reasons to believe that the earliest appearance traditions could have been of a spiritual, not a physical nature (for example, see my collection on Why I Don’t Buy the Resurrection Story and my contributions to The Empty Tomb: Jesus Beyond The Grave).
Geivett even makes the absurd claim that the empty tomb and physical appearances are “as well attested historically as the crucifixion itself” (187). Does he really think that the crucifixion is not more widely attested than the resurrection? After all, the crucifixion story is repeated in at least one early-second-century non-Christian source (Tacitus), whereas no details of the resurrection get such a publication, and although both the crucifixion and the resurrection appear in the pre-gospel epistles, none of those letters ever refers to an empty tomb, and none of them make any clear reference to physical post-mortem appearances (this I will discuss at greater length in my review of Craig). Thus, even Geivett must concede that the crucifixion is at least somewhat better attested than the empty tomb and the post mortem appearances.
Geivett also thinks that a miracle “may be the best explanation for first-century belief in the resurrection.” But if we regard as “best” that explanation which is most consistent with what we are all (believer and unbeliever) certain to be true about the universe and the period and place in question, then credulity and fanaticism is clearly the “best” explanation for that belief (see my review of Beckwith). Geivett is correct, however, in noting that a proper theory must say more than merely “they were mistaken” or some such dismissal, but he never cites or addresses or even mentions the existence of any of the developed theories which have been set out by various historians (such as Hermann Reimarus, Ernest Renan, David Strauss, Bruno Bauer, Maurice Goguel, A.E. Harvey, H.C. Kee, W.H.C. Frend, and W.S. Kissinger, just to name a few). Thus, his argument gives the appearance that no such theories exist.
The Top-Down Approach
Geivett concludes with what is definitely a correct argument: that if we had sufficient reason to believe that a God exists who is capable of working miracles, then “miracle” would become a more accessible explanation for unusual events, especially if said events could be shown to have some definite connection with God. But this is of little use when there are insufficient reasons to believe there is a God, for on this approach miracles cannot be appealed to as evidence. Rather, miracles only serve to identify the ways and purposes of the actions of a God already accepted as real. This of course explains why believers are quick to attribute anything orthodox and Christian as genuinely miraculous (while disparaging all other miracles as false), but it does little to justify their belief in God in the first place. But the idea is that once you believe in God, miracles become the stamps of approval of God, confirming which revelations or deeds He advocates (hence one can, Geivett suggests, argue from miracles that a theist of any other creed ought to be a Christian in particular).
But a new and decisive problem arises even with this argument: why the heck would a superbeing only communicate his revelations by using historically isolated miracles? Can’t he just talk, like any normal sentient being? Or at least establish a permanent miracle which continually confirms his approval? For instance, if Catholic churches were mysteriously impregnable to all forms of harm, that would be a useful proof of God’s approval for that creed. So why would a God, who is capable of this, ask us to trust (let’s imagine) that a single, miraculous resistance to one harmful event by a single Catholic church centuries ago is just as good? Likewise, if all true Bibles, and all faithful translations, were indestructible–unharmed by fire or water or pencil or ink–or instantly reconstructed whenever injured or destroyed or soiled or marked, that would be a useful proof that it contained God’s word. So why would a God, who is capable of this, ask us to trust in his approval of a compendium that we can’t even get straight–owing to thousands of variant readings and hundreds of translations, each carrying a different nuance of meaning, and whose books’ canonicity is questioned and debated even among Christians–all on the weight of certain isolated, ancient, and questionable events? Geivett must think his God the most dimwitted divinity in all the pantheons ever envisioned by man, or at least the most impotent. His is a God who chooses the most feeble and ambiguous means of reassuring us of what he wants. One wonders why it would ever be advisable to follow such a deity. I couldn’t trust the sanity of such an engineer, much less his ability to save me.
A second, and related problem arises: if it really is so reasonable to accept miracles as proof of divine approval, doesn’t that mean that fraud will be far more rampant in the case of miracles than in any other event? In other words, since it is accepted that miracles confer approval, anyone who wants their ideas to be adopted will naturally want to demonstrate divine approval by inventing miracles, either physically or in persuasive rhetoric. This presents an awful problem for the theist: how is a fraud to be distinguished from a genuine miracle? If in most cases it cannot be distinguished, owing to lack of evidence, yet we know fraud must inevitably be most common in the case of miracles, it follows that identifying genuine miracles will be even more difficult than might otherwise be the case, and even more difficult than identifying genuine historical events of any other kind. In other words, Geivett’s argument actually provides a strong reason for treating miracles with greater skepticism than other historical events.
Return to this review’s Table of Contents to read more detailed critiques of specific chapters in In Defense of Miracles.