Do Extraordinary Claims Require Extraordinary Evidence? Part 1

Post Author: Bill Pratt

“Extraordinary claims require extraordinary evidence.”  I cannot count how many times skeptics of Christianity have trotted out this statement when conversing with me, usually in the context of Jesus’s miracles and resurrection.

There are many possible responses to this statement, but in this four-part series of posts, I want to present responses from William Lane Craig and Mike Licona.  First, William Lane Craig.

This sounds so commonsensical, doesn’t it?  But in fact it is demonstrably false.  Probability theorists studying what sort of evidence it would take to establish a highly improbable event came to realize that if you just weigh the improbability of the event against the reliability of the testimony, we’d have to be sceptical of many commonly accepted claims.

Rather what’s crucial is the probability that we should have the evidence we do if the extraordinary event had not occurred.  This can easily offset any improbability of the event itself.  In the case of the resurrection of Jesus, for example, this means that we must also ask, “What is the probability of the facts of the empty tomb, the post-mortem appearances, and the origin of the disciples’ belief in Jesus’ resurrection, if the resurrection had not occurred?”  It is highly, highly, highly, improbable that we should have that evidence if the resurrection had not occurred.

Let me restate Craig’s claim here.  He is basically saying that if we actually employed this criteria across the board, we would have to rule out all kinds of claims that we all believe are true.  Extraordinary and improbable events happen all the time, and we usually do not have extraordinary evidence for these events.

A simple example is Alexander the Great’s extraordinary conquests.  Everyone agrees it happened, but we have no documentation of it until centuries after it occurred.  Is documentation hundreds of years afterward enough to know what happened?  Most people seem to think so.  Is this level of documentation extraordinary?  It doesn’t seem particularly extraordinary.  So should we say Alexander’s conquests never happened?  Obviously not, so that means this criteria is far too restrictive, and is, therefore, not useful.

The better question to ask is this: “Given the evidence for an extraordinary event, what is the probability we would have that evidence if the event had not occurred?”

We will next look at Mike Licona’s detailed analysis of this skeptical maxim.  Stay here!

  • tildeb

    Everyone agrees it happened, but we have no documentation of it until centuries after it occurred.

    Umm… I think there was lots of ongoing documentation as well as a great deal of physical evidence that would directly support Alexander’s conquests. The reliance on only documentation by testimonials would be a very poor way to determine whether or not Alexander did in fact conquest to a remarkable degree of success. In comparison, Jesus’ miracles has only second and third hand testimonials without any accompanying physical evidence. And the testimonials do not always agree in detail, making their reliability suspect.

    Understandably, miracles are hard to prove. Yet testimonials of miracles are quite common but seem impossible to verify in any other way. In every modern case where conditions can be applied to isolate the miraculous part, no miracles are forth-coming. Physical laws always seem to remain unbroken. So why do we undertake this process rather than simply believe those who make these claims?

    It seems to me that the best way to attest to miracles is to use the scientific method to show it is utterly unable to account for the phenomena that breaks these natural physical processes. (This is the underlying reason why the Catholic Church, for example, sends out a team of investigators to look into these claims.) The important point here is to realize that the key must be the inability of scientific inquiry to explain what might have happened. And, unfortunately for the believers, there are reasonable explanations for the claims made about Jesus. These reasonable explanations have been pointed out to Craig time and again to no avail.

  • Mark

    We do have contemporary references to Alexander. Have you heard of the Decree of Philippi?

  • While it is true that the extant biographies of Alexander date from long after his death, those biographies all cite biographies that were written by Alexander’s contemporaries including the official biographer that Alexander chose.

  • Boz

    so Bill Pratt, are you agreeing or disagreeing with Bayes’ theorem?
    I can’t tell from the OP.

  • But none of the contemporary biographies of Alexander exist today, so we only have extant copies of biographies written hundreds of years later. Is this extraordinary evidence? Hardly. It is ordinary evidence of the kind one would expect for events that occurred 2500 years ago.

  • The blog post mentions Alexander’s conquests, not his mere existence. The Decree of Philippi is an arbitration on a land dispute, so it has little to do with describing the extraordinary extent of his conquests.

  • I agree with Craig’s point that this is a better question:
    “Given the evidence for an extraordinary event, what is the probability we would have that evidence if the event had not occurred?”

  • I’ve never thought that “extraordinary claims require extraordinary evidence” is all that helpful for precisely the reason that “extraordinary” is such an imprecise term. The scope of Alexander’s accomplishments were anything but ordinary, however, we have seen other great military conquerors like Napoleon and Genghis Khan.

  • Boz

    ‘extraordinary claim’ is best defined as the prior probability of that claim.

    ‘extraordinary evidence’ is best defined using bayes’ theorem.

  • Pingback: Do Extraordinary Claims Require Extraordinary Evidence? Part 2 | Tough Questions Answered()

  • Boz,

    “Extraordinary claims require extraordinary evidence” is a useful principle if the terms “extraordinary claims” and “extraordinary evidence” are sufficiently defined. Unfortunately, arguments based on the principle often seem to me to spend much more time on definition than application.

  • tildeb

    If we replace the term ‘evidence’ with what we actually have – ‘testimonials’ – and ‘extraordinary event’ with ‘miracle’ then let’s look at Craig’s point: Given the testimonials for a miracle, what is the probability we would have those testimonials if the miracle had not occurred?

    The veracity of testimonials are well known to be affected by many factors, all of which mitigate rather than enhance their reliability. The probability of those testimonials being true (based only on testimonials) is reasonably low in comparison to the probability that physical processes we understand and utilize successfully as if they were consistent over time and place remained consistent in spite of contrary claims. We also know that the reliability of these testimonials is further diminished and not improved by significant differences between them. So Craig’s question is best answered (given the nature of the totality of the ‘evidence’ we have – testimonials) by assigning a very low probability, which does not advance but reduces the probability case that the miracles really happened.

  • If testimonials are suspect, then I take it you also reject the veracity of Alexander’s conquests and the veracity of all ancient history that is known through testimonials?

  • tildeb

    Suspect doesn’t mean wrong when we’re talking about probabilities; it means of a lower comparative value to evidence that includes more than just testimonials. Alexander’s conquests, by comparison, have nothing but supportive evidence from many mutually compatible avenues of inquiry when it doesn’t have to be this way, which significantly raises the probability by comparative value that these conquests by this historical figure really did happen.

    Remember, Bill, you’re the one who preferred Craig’s question of assessing probabilities based on evidence. What he doesn’t go on to explore with his audience is why the probability that miracles historically happened is so low: because of the paucity of evidence and it’s poor quality.

  • Boz

    That question is included in the denominator of Bayes’ Theorem. See the hair/train example here: http://en.wikipedia.org/wiki/Bayes'_theorem


    so Bill Pratt, are you agreeing or disagreeing with use of Bayes’ theorem?

  • Jeff Lowder