Dispatches from the Creation Wars

Evangelical Outpost Joins the Fun

Joe Carter at Evangelical Outpost has jumped into the Leiter/VanDyke fray, in a post filled with misconceptions and illogical statements. He begins:

For a legal scholar and professor of philosophy, Brian Leiter has a remarkably poor grasp of basic logic. For the past week Leiter has been bashing a defender of Intelligent Design theory using his typical rhetorical style of bullying and bluster. Instead of thinking up creative new ad hominem attacks, though, he should be paying closer attention to his reasoning.

At the risk of being pedantic, I have to point out this very common mistake in claiming the ad hominem logical fallacy. An ad hominem, contrary to how seemingly everyone conceives of it, is not merely an insult. Calling someone a jerk is not an ad hominem. An ad hominem is a logical fallacy, so there must be a mistake in reasoning in the formulation. The logical fallacy in an ad hominem attack is in responding to a substantive claim by referring to an irrelevant personal trait of the person making the argument. For example, if I said, “Joe Carter shouldn’t be listened to when he talks about ad hominems, look at the way he dresses”, that would be an ad hominem. I would be rejecting his arguments based upon an irrelevant personal trait. While Leiter is often rude and harsh in his attacks on people, those are not ad hominems. They may be insulting, but that doesn’t make it ad hominem.

Joe quotes this passage from Brian:

The difficulty, however, is that science did not “a priori pick a naturalistic methodology”; it adopted, based on evidence and experience (i.e., a posteriori), the methods that worked: it turns out that if you make predictions, test the predictions against experience, refine the hypotheses on which the predictions are based, test them again, and so on, you figure out how to predict and control the world around you. This is what the Scientific Revolution, the Enlightenment, and a few other ancient events apparently not covered in Mr. VanDyke’s education, were about: the a posteriori discovery of the most effective ways to predict and control the world. This, of course, distinguishes the naturalistic worldview of science from the supernatural view of religion, which is genuinely a priori.

And begins his response:

There are numerous problems with Leiter’s reasoning but I will point out just three. The first is that his methodology would lead to conclusions that Leiter himself woudl presumably reject. Take for example the “anthropic principle.” We could predict, post hoc, what type of universe would be required to produce human life, but we’d be unable to test the theory (we aren’t able to repeat the Big Bang).

Does Joe really think that if we can’t repeat an event, we can’t test explanations for that event? This would rule out whole fields of science, including the one he mentioned. Big bang cosmology is entirely testable, and has been tested, without having to repeat the big bang itself. Testability requires making predictions about the nature of new evidence, not repeating the event itself. It would also, by the way, rule out the entire field of forensic science, which is used to convict people and even put them to death on a daily basis in this country. By Joe’s reasoning, you would have to actually recreate the murder in order to test forensic explanations for the murder. But that’s not how it’s done, of course. You test the forensic explanation by making predictions. If the bullet came from gun X, then we make predictions Y and Z. If Y and Z are confirmed, the explanation is validated.

He continues:

We could, however, determine the likelihood that the event could have occurred by pure chance. Since the probability of such a series of events occurring by coincidence would be close to zero, we would be lead, by evidence and experience, to the conclusion that the universe was “designed.” (To conclude otherwise would require taking an a priori prejudice against supernaturalism.)

I’ll take issue with Joe’s claim that we can determine the likelihood of the big bang, or the so-called anthropic coincidences, occuring by “pure chance”, and I’ll challenge him to produce such a calculation. We hear this argument over and over again, but it’s never accompanied by an actual probability equation. If you think we can calculate the probability of either of those two things, let’s see the probability equation.

It should also be noted here that even if such a probability equation were possible, it wouldn’t tell us anything meaningful about whether the event could have occured naturally or supernaturally. The perfect illustration of this is Marshall Berman’s example of the rock in the backyard:

Go outside and pick up a small rock. The probability of that rock being on that spot on the earth *by chance alone* is roughly the area of the stone divided by the surface area of the earth, or about one chance in 10 to the 18th power (one followed by 18 zeros). If picking up the stone took one second, the probability of such an event occurring at this precise moment over the lifetime of the universe is now even smaller by another factor 10 to the 18th power! This simple event is so incredibly unlikely (essentially zero probability) that one wonders how it could be accomplished!

Joe continues:

The second reason is that the “what works” approach gives us no reason to believe that our conclusions are true. I may believe, for example, that my dryer contains a black hole that causes socks to disappear. Every time I put a load of clothes in the machine I find that I’m missing a sock. The more I repeat this experiment the more socks I lose, thereby providing sufficient evidence to confirm my theory.

Joe seems to be confusing facts with explanations. Socks disappearing from your dryer would be a fact; the “black hole hypothesis” would be a potential explanation for that fact to be tested. The continued occurence of the fact does not test the potential explanation. Doing laundry would not constitute “repeating the experiment”, since doing laundry does not test the explanation at all. In other words, this is an absolutely absurd analogy for how science tests a hypothesis.

The black hole hypothesis could of course be tested in other ways, as black holes have predictable effects. If there was a black hole in your dryer, it would have a quite noticable effect on gravitational pull around the dryer. It would also not be able to distinguish between socks and other types of clothing, since black holes are not conscious entities and would simply obey the laws of physics. Now, Joe might invoke a violation of the laws of physics here and say that this is a magic black hole that only makes socks disappear and doesn’t have any other predictable effects. But in doing so, Joe would demostrate for us exactly why science rules out supernatural explanations, because once you allow them, all bets are off – there is absolutely no way to discern true supernatural explanations from false ones. Can you propose a means of distinguishing the “magic black hole” hypothesis from the “malevolent demons” hypothesis or the “mischievous leprauchans” hypothesis? Of course not. In other words, Joe’s analogy is a really good argument against his thesis.

Leiter, of course, would claim that we should use Occam’s razor and exclude the necessity of the black hole to explain the missing socks. But this would require us to take an a priori position in favor of the principle of parsimony in order to preserve methodological naturalism. My theory would work well enough that I would have no reason to test it further and while it might not be “true”, the a posteriori examination of the evidence makes it a plausible explanation. After all, naturalistic methodology doesn’t require us to take a priori assumptions about truth.

Leiter would not have to invoke Occam’s Razor to distinguish between two explanations here, because Joe’s hypothetical explanation hasn’t been tested at all, and if it was tested by making predictions about the effects of a black hole, it would be falsified. Joe is pretending that he has two equally plausible explanations that explain the exact same things equally well, when in fact he doesn’t have such an explanation at all. He has one very bad analogy that, if made more analogous, would fail miserably as a theory.

The third reason Leiter’s argument fails is that he has no justification for excluding other theories or methods that don’t rely on methodological naturalism. Just because a method works doesn’t mean it is infallible. The method provided us with Newtonian physics, a hypothesis that “worked” well enough…until it didn’t. Do we regard the theory as having always been an implausible scientific hypothesis just because it was replaced by another? Of course not. The same applies to methods. Just because methodological naturalism “works” (at least sometimes) does not mean that it is the only valid method or that it cannot be replaced. Besides, you can’t (without resorting to an a priori assumption) exclude other methods as invalid without allowing them to be tested.

This would be a serious objection if, and only if, there was some means of testing those “other methods”, in this case the ID explanation. And if Joe can come up with a testable hypothesis that flows from ID, or a way to falsify ID, he’ll be the first to do it.

Leiter’s reasoning shows that his bias against intelligent design theory is not rooted in science but in prejudice. By acknowledging that science does not require an a priori submission to naturalism he inadvertenly undercuts his own argument. He can’t claim that methodological naturalism is the “most effective ways to predict and control the world” while refusing to allow other methods to be tested.

Again, Leiter is not ruling out ID without allowing it to be tested. He’s challenging the ID advocates to put forth a real model with testable hypotheses that flow logically from it and propose a means of testing those hypotheses. But they haven’t done that, and I don’t think they can. It’s not by accident that all of their publishing efforts to this point have been trying to poke holes in evolution. The entire ID argument up to this point comes down to one big God of the Gaps argument – “Evolution can’t explain X, therefore God (sorry, the unnamed – wink, wink – intelligent designer that we know nothing about) must have done it”. There are no testable hypotheses that flow from that. So it simply isn’t a question of anyone “ruling out” ID without testing it, it’s a question of there not being any means of testing it. And if the ID advocates think that’s false, all they have to do is actually publish some means of doing so, as we have been challenging them to do since at least 1997′s NTSE conference. That deafening silence you’ve been hearing in that regard is quite telling, don’t you think?

Comments

  1. #1 Joe Carter
    April 7, 2004

    At the risk of being pedantic, I have to point out this very common mistake in claiming the ad hominem logical fallacy.

    While I agree that Leiter may not have technically committed such a fallacy, that is the basic purpose of his insults. He attempts to cast doubts on his opponents arguments by demeaning their intellectual abilities. Leiter may be a smart guy but you couldn’t tell if from the way he argues. Strip away the bluster and insults and he rarely has anything substantive to add.

    Does Joe really think that if we can’t repeat an event, we can’t test explanations for that event?

    No, I don’t. But then, that wasn’t my claim. We may be able to test individual explanations but we cannot test unrepeatable events. That is the problem with your forensic science analogy. While we may be able to confirm whether the bullet came from a particular gun, we are unable to test all the elements that could have been involved in the murder.

    We hear this argument over and over again, but it’s never accompanied by an actual probability equation. If you think we can calculate the probability of either of those two things, let’s see the probability equation.

    U= The known universe
    N= Occurrence of an event necessary for human life to exist

    Prob(U)= prob(N) + prob(N1) + prob (N2) + prob (N3)…

    The perfect illustration of this is Marshall Berman’s example of the rock in the backyard.

    Actually, Berman’s is a bad example since it commits a form of the inverse gambler’s fallacy. Since there is nothing special about each of the events that are being used to calculate the probability, there is no reason to assume the event was unlikely.

    The continued occurrence of the fact does not test the potential explanation. Doing laundry would not constitute “repeating the experiment”, since doing laundry does not test the explanation at all. In other words, this is an absolutely absurd analogy for how science tests a hypothesis.

    While it may be an absurd analogy and a bad hypothesis, there is nothing in it that violates Leiter’s explanation for how science works.

    Now, Joe might invoke a violation of the laws of physics here and say that this is a magic black hole that only makes socks disappear and doesn’t have any other predictable effects. But in doing so, Joe would demostrate for us exactly why science rules out supernatural explanations, because once you allow them, all bets are off – there is absolutely no way to discern true supernatural explanations from false ones.

    But in order to get to that conclusion you would have to make an a priori assumption about science, which is precisely my point about Leiter’s silly misunderstanding of how science works.

    In other words, Joe’s analogy is a really good argument against his thesis.

    No, it is a silly analogy that provides a good argument against Leiter’s thesis.

    And if Joe can come up with a testable hypothesis that flows from ID, or a way to falsify ID, he’ll be the first to do it.

    That’s hardly the case. Dembski provides a rather good one that uses the bacterial flagellum: “To falsify such a claim [that it is irreducibly complex], a scientist could go into the laboratory, place a bacterial species lacking a flagellum under some selective pressure (for mobility, say), grow it for ten thousand generations, and see if a flagellum–or any equally complex system–was produced. If that happened, my claims would be neatly disproven.”

  2. #2 Ed Brayton
    April 7, 2004

    No, I don’t. But then, that wasn’t my claim. We may be able to test individual explanations but we cannot test unrepeatable events.

    This makes no sense whatsoever. Only explanations need to be tested. You don’t need to “test” an event, you need to test the explanation for that event.

    That is the problem with your forensic science analogy. While we may be able to confirm whether the bullet came from a particular gun, we are unable to test all the elements that could have been involved in the murder.

    And therefore….what? The logical conclusion of your argument is that we therefore cannot test the explanation fully. That would mean that we have to let everyone out of prison because we were “unable to test all the elements that could have been involved in the murder”.

    On the question of a probability equation, I challenged Joe to provide a probability equation for the anthropic coincidences. He responded:

    U= The known universe
    N= Occurrence of an event necessary for human life to exist

    Prob(U)= prob(N) + prob(N1) + prob (N2) + prob (N3)…

    But this is just silly. How many Ns would you like to add to it? No matter how many it is, I can show you lots and lots of events that happen all the time with an equal probability using this same reasoning. Take a deck of cards and shuffle them randomly, then using this kind of simple probability formula, figure out the probability of them being in the exact order they ended up in. You’d end up with an astronomically high number. Yet they were just found in exactly that order.

    The real problem with such probability equations is that you simply don’t have enough data to reach a conclusion. You don’t know how many trials were possible, you don’t know that the conditions were actually random (not being intentionally controlled is not the equivalent of “random”), and you don’t know that those conditions were actually necessary in order to have life, as there are other possible forms of life.

    Actually, Berman’s is a bad example since it commits a form of the inverse gambler’s fallacy. Since there is nothing special about each of the events that are being used to calculate the probability, there is no reason to assume the event was unlikely.

    Quite the opposite is true. Remember that this is, by your own admission, post hoc reasoning. That is, you’re looking at what DOES exist and saying, “imagine all of the things that could have gone wrong in all of the contingent events leading up to this point that could have screwed up what exists now.” It’s like tracing back your specific lineage. In every generation, you are able to imagine a million possible things that could have happened that would have resulted in you not being here – a single ancestor who gets trampled by an oxen (an essentially random event) and you’re not here now. You’re doing the same thing for life, by comparing what is with the millions of possible might have beens and concluding that what is must be staggeringly unlikely. But again, you don’t know how many trials took place, you don’t know if those alternates really were possible because you don’t know that the physical features of the universe were truly random (as opposed to not being “willed”, which is not the same thing), and you are presuming true randomness without having any idea if that was the case. And none of this really has anything to do with the question of testability, which is the real issue.

    Me: The continued occurrence of the fact does not test the potential explanation. Doing laundry would not constitute “repeating the experiment”, since doing laundry does not test the explanation at all. In other words, this is an absolutely absurd analogy for how science tests a hypothesis.

    Joe:While it may be an absurd analogy and a bad hypothesis, there is nothing in it that violates Leiter’s explanation for how science works.

    Huh? You’re making very little sense here. Leiter’s explanation dealt only with whether methodological naturalism is a priori or a posteriori. You responded with not only an absurd analogy and a bad hypothesis, but an obviously false means of testing it. You claimed that socks disappearing was a test for the “black hole” hypothesis, but that’s ridiculous. Socks disappearing is a fact; the black hole hypothesis is a potential explanation for that fact. The continuation of the observational pattern is not a test of whether the hypothesis is true or not. So you claimed that science as practiced “gives us no reason to believe that our conclusions are true”, and then built a perfect little straw man that had nothing whatsoever to do with how hypotheses are tested in science.

    Me: Now, Joe might invoke a violation of the laws of physics here and say that this is a magic black hole that only makes socks disappear and doesn’t have any other predictable effects. But in doing so, Joe would demostrate for us exactly why science rules out supernatural explanations, because once you allow them, all bets are off – there is absolutely no way to discern true supernatural explanations from false ones.

    Joe: But in order to get to that conclusion you would have to make an a priori assumption about science, which is precisely my point about Leiter’s silly misunderstanding of how science works.

    Do you have a way of testing supernatural explanations and discerning true ones from false ones? We can come up with lots of supernatural explanations for why socks disappear in the dryer. Do you have a means of discerning which supernatural explanations for that phenomenon are likely true and which are likely false? If you do then, by all means, share it with us. If you don’t have such a means, then you are proving that what you are referring to as an “a priori assumption” is entirely justified. Science is about testing explanations; therefore, explanations that can’t be tested aren’t a part of science. That doesn’t mean that all non-testable explanations are automatically false, it means that there is no way of knowing whether they are or not. They are simply useless as explanations. This is not an a priori assumption, it’s simply the nature of reality. We have no choice but to not waste our time testing that which is not testable. If you have an alternative to this, please offer it.

    Me: And if Joe can come up with a testable hypothesis that flows from ID, or a way to falsify ID, he’ll be the first to do it.

    Joe: That’s hardly the case. Dembski provides a rather good one that uses the bacterial flagellum: “To falsify such a claim [that it is irreducibly complex], a scientist could go into the laboratory, place a bacterial species lacking a flagellum under some selective pressure (for mobility, say), grow it for ten thousand generations, and see if a flagellum–or any equally complex system–was produced. If that happened, my claims would be neatly disproven.”

    No Joe, this is not a test for the truth of ID, it’s a test for one of the many ancillary hypotheses of evolution (not a good one, but that’s an entirely different question). The implication is, as I said above, a God of the Gaps conclusion – if evolution can’t explain it, God must have done it. Worse than that, the argument is “if evolution can’t repeat it, God must have done it. That is not a testable hypothesis that flows from ID, it’s a purely negative argument. In addition, Dembski has backed off from this claim, for the obvious reason that even if scientists could prove beyond all reasonable doubt how the flagellum evolved, Dembski would simply move on to other unanswered questions and repeat the same argument – “they haven’t explained exactly how this one happened, therefore God must have done it”. You’ve made my point for me perfectly, Joe. I said, “He’s challenging the ID advocates to put forth a real model with testable hypotheses that flow logically from it and propose a means of testing those hypotheses. But they haven’t done that, and I don’t think they can. It’s not by accident that all of their publishing efforts to this point have been trying to poke holes in evolution”, and you ducked into the punch by giving a perfect example of how the only thing they have to offer is to poke holes in evolution with the implication that if evolution can’t (yet) explain , God must have done it.

  3. #3 Jim Flannery
    April 8, 2004

    Joe: “He attempts to cast doubts on his opponents arguments by demeaning their intellectual abilities.”

    No, he’s making an inference about their intellectual abilities based on the quality of their arguments. That a priori/a posteriori thing is a tough one, isn’t it?