Friday, July 17, 2015

Is belief in miracles irrational?

I will comment on this:

Take Jesus’ resurrection. Given how nature works, dead people stay that way. 

Absent the intervention of a rational, omnipotent agent. 

It didn’t have to be that way. Just as the freezing temperature of water might have been 34º F rather than 32º F, maybe one in ten dead could have “naturally” come back to life. 

i) That's a bit too facile. In principle, the freezing point for water could be different. However, that's not a discrete variable. To change that would impact other things. To make everything balance out, there'd have to be corresponding changes. You can't just alter the freezing point of water and leave everything else unaffected. Other adjustments must be made to accommodate that particular change. And maybe there's not that much give in the system. 

ii) Under what scenario does he think one in ten dead could "naturally" come back to life? How much necrosis has the body undergone? 

But, water does freeze at 32º F, and dead people stay dead (barring unforeseen medical advances that certainly were not available 2000 years ago). That’s why, if Jesus really did return to life, something must have intervened to block the otherwise inevitable march of natural laws.

That's roughly true.

Back to miracles. Even granting the tremendous reliability of the witnesses to Jesus’ resurrection, the case for accepting their account is very weak. How many people return from the dead? It must be very low, far less than the number of people who have the serious disease in our analogy. Suppose, for the sake of argument, that God resurrects one in a billion people. This means that even if the witnesses to the resurrection were incredibly reliable (perhaps they misidentify non-miraculous events as miraculous only one in a million times), the chance that they were correct about Jesus’ resurrection would be only one in a thousand.

That frames the issue as if it's a roll of the dice. The natural odds. But if it happened, the Resurrection was the result of divine intervention. Not letting nature taking its course, but reversing nature. Circumventing nature. 

It's like asking what are the odds of throwing sixes ten times in a row? Well, that depends. Are they fair dice or loaded dice? 

Natural processes involve unintelligent causes–like a computer that's programmed to perform a task. It always does the same thing. Only does what it was programmed to do.

But the odds for what a computer will do–given the status quo–are very different from what a computer programmer will do. He can change the program. He can make the computer perform a different task. 


  1. In addition to Steve's stellar points:

    1. Human life is only supported by the narrowest margin on Earth. Just changing a single variable like the freezing point of water would dramatically change water itself, which in turn could arguably lead to water no longer existing, or not existing in a form conducive to human life on Earth. To say nothing about other life on our planet.

    Exobiologists have alternative theories for life surviving in the absence of water, but my understanding is this is almost always in reference to life on a micro scale. And even if what exobiologists theorize were relevant to higher forms of life, life as we know it on Earth would nevertheless be radically different to say the least. You and I wouldn't be here.

    2. A human body could arguably survive if all the necrosis were contained to non-vital areas of the body. However, if there's enough necrosis in vital areas, then all bets are off. And this would certainly be the case in raising the dead to life.

    3. If the mind is ultimately reducible to the physical brain, then even if future science could raise the dead to life, would the person who had been raised to life be the same person if the entire brain itself had undergone complete necrosis?

    4. Why does Shapiro frame the issue of detecting miracles in terms of diagnostic medical tests? How is this an appropriate let alone ideal methodology for detecting a miracle like raising the dead to life? A metal detector is useful for detecting metal. But it'd be far less useful for detecting wood. Is Shapiro's methodology appropriate to the claim?

    5. Imagine a society where people only died of natural causes i.e. old age and its complications. Imagine this is what has always happened. Imagine this is what people have always known and naught else. Imagine one day a police officer or detective discovers a dead body. Imagine the only theory under consideration is the body died of natural causes. That's how it's always been. Why think anything different this time? But imagine this was the first case in which a person was in fact murdered.

    a. Sure, it would go against known knowledge and experience that a person didn't die of natural causes. But it would in fact be false in this case.

    b. Also, I suppose it's "rational" to a certain degree to believe the person died of natural causes. But that's only if we restrict or limit ourselves. If we refuse to "think outside the box" as it were. If we refuse to investigate other lines of inquiry and evidence. Such as the possible involvement of human agency. Hence, if we persistent in our stubborn close-mindededness, then what may have been "rational" becomes "irrational."

  2. Larry Shapiro

    "To summarize, the extreme rarity of divine interventions works against the rationality of believing in them. What about other very rare events, such as winning lottery tickets or perfect bridge hands? Are we never justified in believing in these things? Of course we are, but this is because, in the case of the lottery, we know independently that one ticket or another must win; in the case of the perfect bridge hand, we know that it is no less likely than any other bridge hand. And, in both cases, we might ask for corroborating evidence. Miracles differ from other rare events in each of these respects."

    To summarize, the extreme rarity of the universe works against the rationality of believing in the universe.

    To summarize, the extreme rarity of life works against the rationality of believing in life.

    To summarize, the extreme rarity of humans works against the rationality of believing in humans.

    And so on.

    How do we know the universe, life, or humans must exist independently from the existence of the universe, life, or humans? How do we know there must have been something rather than nothing?

    How do we know the universe, life, or humans are no less likely than any other universe, life, or humans?

  3. "Suppose, for the sake of argument, that God resurrects one in a billion people. This means that even if the witnesses to the resurrection were incredibly reliable (perhaps they misidentify non-miraculous events as miraculous only one in a million times), the chance that they were correct about Jesus’ resurrection would be only one in a thousand."

    On the contrary, isn't it more plausible to believe a miracle is random i.e. one miracle is statistically independent of another miracle? Or to look at it another way: if we had a string of miracles every single day for the past 1000 days, it would not necessarily increase the probability that we will have a miracle tomorrow.

  4. This is embarrasingly poor stuff from Shapiro. His notion of witness reliability is bizarre. Is this how we evaluate anything in real life, "Hmmm, how often does my wife misidentify something as a car accident when it was really not a car accident?"

    A little decent understanding of Bayes factors could have cured all that nonsense. As it stands, it's pretty much just Hume dusted off a little.

    Tim and I have an article in this volume on this very question:

    The short version of the story is that in terms of witness testimony this concept of "witness reliability" is so unwieldy that it creaks at the joints. Not only do we not have actual statistical data on witness reliability in respect to particulars like that, but "reliability" is not the whole story. For example, the power of the evidence is often a result of factors in the surrounding situation, such as what the witnesses have to gain or lose from their testimony. Those factors are not well-described as "reliability." For example, the fact that a witness has everything to lose and nothing to gain increases the evidential power of his testimony but isn't what one would normally call "reliability." Similarly, the fact that a witness has better access to some types of purported miracles than to others is not well-picked-out by a broad question like, "How often does this person mis-identify non-miraculous events as miraculous?" Perhaps if the non-miraculous event were, say, a healing, he would have more difficulty telling if it were really miraculous, whereas telling that you are seeing a close friend again and talking to him up close (when he was previously dead) is much harder to be mistaken about.

    And so forth. This sort of faux probabilistic calculation he is doing is ludicrously crude and is not either a realistic or an epistemically responsible way to evaluate witness testimony to particular events.