Saturday, February 20, 2010

Mackie on ad hoc hypotheses

Bill Vallicella objects to the Scholastic notion of suppositum because he takes it to be entirely ad hoc, “a mere invention pulled out of thin air to render coherent an otherwise incoherent, or not obviously coherent, theological doctrine.” As we have seen, Dale Tuggy objects to the characterization of the Trinity as a “mystery” for similar reasons. In both cases, the underlying assumption is that the introduction of a concept or hypotheses for the specific purpose of answering a certain objection is per se philosophically questionable.

But as logicians know, not every appeal to authority is a fallacious appeal to authority; not every ad hominem criticism is a fallacious ad hominem; not every inference from part to whole involves a fallacy of composition. And by the same token, not every appeal to ad hoc considerations is fallacious or otherwise suspect. When is such an appeal legitimate? In his article “Fallacies” in The Encyclopedia of Philosophy, J. L. Mackie – no friend of theology – tells us that we must distinguish fallacious ad hoc hypothesis formation from “the respectable procedure of interpreting new observations in light of an established theory,” and that “in the respectable procedure, we are working with a hypothesis that is already well confirmed, [while] it is a fallacy to ‘save’ a hypothesis for which there is no strong independent support.”

As I have said in earlier posts, the Trinitarian theologian does not pull his doctrine out of thin air. The claim is rather that the doctrine of the Trinity follows from divine revelation – that is to say, from infallible testimony – where the fact that such a revelation has occurred is itself something that can be established by solid arguments. If this is correct, then the doctrine is indeed “already well confirmed” and has “strong independent support.” The appeal to notions like suppositum or mystery is therefore not objectionably ad hoc.

Obviously the skeptic is going to deny that the doctrine really is independently well confirmed, but that is a different issue. The point is that the objection to the notions in question on the grounds that they are ad hoc has, by itself, no force.

16 comments:

  1. where the fact that such a revelation has occurred is itself something that can be established by solid arguments.

    Such as?

    You don't have to give me the arguments themselves; just tell me what kind of arguments they would be. Theological? Metaphysical? Empirical?

    ReplyDelete
  2. It's all the less ad hoc given that the concept is not exclusively found in the context of the Trinity; scholastics use the concept of a suppositum when talking about the Incarnation, singular reference, whether the soul is uniform or pluriform, the relation between form and actual being, etc. This is what puzzles me about Vallicella's argument; you can't prove that something is ad hoc except by showing that there is no reason independent of the mere resolution of the problem to make use of it. But that cleary requires looking at more than how it functions in the context of Trinitarian doctrine. And when we do we find that the notion was demonstrably not just invented merely for laying out the Trinity; in fact, the use of the term in the context of the Trinity seems clearly to have originally been borrowed from discussions of what today we would call reference. Even if you think it was a philosophical dead end, the ad hoc charge just won't stick.

    ReplyDelete
  3. I've been going around and around on this with Dr Vallicella over there, and it's come to the point where I'm beginning to doubt whether he's right and I'm an idiot who doesn't understand what's going on in the discussion, or I'm right and he's approaching the subject totally backwards.

    ReplyDelete
  4. Hello George,

    I sketched out the line of argument I had in mind in my first post on this subject some months back:

    http://edwardfeser.blogspot.com/2009/06/trinity-sunday.html

    ReplyDelete
  5. Hello Brandon,

    Good point. The same can be said about the appeal to mystery, since (as I've emphasized) it has also been put to use by naturalists, in a way that is defensible even if (in my view) not successful. In both cases, we have concepts which, whatever their origins, show themselves to be applicable and philosophically interesting outside the Trinitarian context.

    There is also the fact that it wouldn't really be relevant even if the suppositum concept had been introduced merely to defend Trinitarianism. That would merely be a fact about the history of ideas and the psychology of the thinkers who introduced it. It has no logical significance -- as if an argument or concept is philosophically legitimate only if you get it in by a certain deadline. And the use to which it has been put in other contexts only underlines the irrelevance of historical origins, the motives behind its introduction, etc.

    ReplyDelete
  6. Hello Michael,

    It seems to me that the trouble with Bill's argument is that he wants to assess what he calls the "rational acceptibility" of the doctrine in the abstract, just considering the doctrine by itself apart from the reasons why Trinitarians accept it. And as I've emphasized, that's a mistake. Sure, if we treat the doctrine as if it just popped into our heads out of nowhere, the ad hoc objection to the notions of suppositum or mystery has force. But that's not the situation we're actually in. The Trinitarian, unless he is a fideist of some sort, thinks he has solid rational grounds based on divine revelation confirmed by miracles verified against the background of a detailed system of natural theology (as in the approach I sketched in the post I referred George to above). All of that must be factored in as well in determining the doctrine's rational acceptibility, and when it is we see that the ad hoc objection has no force by itself. (As I've said, someone might reject the grounds for accepting that such a revelation really has occurred, but that is a separate issue.)

    Maybe I've misunderstood Bill, but that's the way it strikes me.

    ReplyDelete
  7. Dr. Feser: It seems to me that the trouble with Bill's argument is that he wants to assess what he calls the "rational acceptibility" of the doctrine in the abstract, just considering the doctrine by itself apart from the reasons why Trinitarians accept it.

    I just looked at the most recent exchange on Dr. Vallicella's blog and I agree with your assessment.

    I think a comparison with physics will help, since every reasonable person accepts the results of physics (even in an anti-realist way, as I am inclined to do): the model behind quantum electrodynamics is bizarre -- the idea that a photon could "virtually" travel every possible path on its way from point A to point B is fraught with, well, philosophical problems. But since it is prima facie not irrational to adopt a realist interpretation of QED, the philosopher is not really in a position to be demand that the physicist make QED acceptable to him before he starts using it.

    Now with regard to Christian trinitarian theology no non-Christian will grant that the doctrine of the Trinity is in quite the same bargaining position as modern physics. Still I think the situation is analogous: if we are convinced on other grounds that the Trinity is true (i.e. because we believe that God has revealed it to be true), then how we go about making sense of it is by no means necessarily ad hoc. True, there may be a more dynamic relationship between our grounds for accepting the Trinity and making logical sense of it than our grounds for accepting QED and making logical sense of it. But the former is not something, as you say, that "just popped into our heads," and to consider it in that manner is an intellectual exercise with less probative force vis-à-vis our acceptance of the dogma of the Trinity than Dr. Vallicella seems to believe.

    Apologies for the inelegant prose.

    ReplyDelete
  8. Obviously the skeptic is going to deny that the doctrine really is independently well confirmed, but that is a different issue

    Perhaps a skeptic could also object that Mackie's view on fallacious ad hoc hypothesis formation refers mainly to science, not to theology or metaphysics, which deal mainly with demostrations, not with empirical hypotheses.

    So perhaps an argument for the relevance of Mackie's distinction, as applicable to theology or philosophy, would be required to avoid misunderstaindings.

    Regarding ad hoc hypothesis, a philosopher of science that I like posed this classification:

    1-Bona fide ad hoc hypothesis: They're hypothesis intented to save an hypothesis from falsification, by positing an ad hoc entity or explanation.

    But what's essential to bona fide ad hoc hypothesis is that they're by themselves TESTABLE by independent means.

    An example: let's to assume the hypothesis "Heart attacks ocurrs only in men over 30 years old". Let's call this hypothesis the "main hypothesis"

    If there is evidence that a guy of 25 years old suffered a heart attack, then the "main hypothesis" could be saved with this ad hoc hypothesis: "There are congenital diseases that can cause a condition similar to a heart attack in persons of any age."

    This hypothesis is testable, because in principle it's possible to know if the guy of 25 years old suffered from a congenital diseases related to the heart.

    If he didn't, then the main hypothesis is refuted. If he did, then the ad hoc hypthesis saved the main hypothesis.

    2-Mala fide ad hoc hypothesis: they're posited to save another hypothesis from falsification, but the ad hoc hypothesis is NOT testable by independent means.

    An example: The main hypothesis is "Heart attacks ocurrs only in men over 30 years old"

    When evidence is presented that a 25 years old guy suffered of a heart attack, the defender of the main hypothesis replies with this mala fide ad hoc hypothesis: "An immaterial and invisible demon, by mysterious means, caused the heart attack in that guy"

    Given that you cannot test this ad hoc hypothesis by independent means, it's a mala fide hypothesis. And should be unacceptable.

    Note that a mala fide hypothesis could be true; but we cannot know it nor (what's important to science) test it.

    ReplyDelete
  9. This comment has been removed by a blog administrator.

    ReplyDelete
  10. Shamdon: Boldy asserted and without a shred of supporting argument. Bravo. Of course, once I read your comment that the notion of substance was "pre biological/chemistry" you removed all doubt about whether you knew what you were talking about. Perhaps your concept of pre-modern philosophy is "pre-bookreading/classtaking"?

    ReplyDelete
  11. It's just Perezoso yet again, doing his angry young man dance.

    ReplyDelete
  12. I'm afraid I must disagree with everything that's been stated so far, because this entire notion of "ad hoc" is built upon an inherent flaw necessarily within reason; this correction alone is the present and future philosophical and logical reality and its law. Here it is, and learn to know this ever more deeply: from now on every single instance of "ad hoc" demands decision, and every implementation responsibility. But every logical statement and legitimization itself moves within a certain relation to the current mental manifold of the logical system. Therefore, branding something "ad hoc" is itself not a thing, thus nothing which is, and yet it remains constant in its implementation without being something concrete like the beings in time, and hence, we cannot legitimately label a judgment as "ad hoc," since "ad hoc" as commonly understood, in the sense of the distance measured between two standards of opposing logic, is the negative collapse of a TIME calculation, which is in essence nothing. In spite of this, stubborn logicians still are the house of the truth of unforgivable judgmental lapses. They should click their heels three times and go back to their mothers' basement.

    ReplyDelete
  13. Well, put. I think I entirely agree with this, Ed.

    I don't think one gets very far in objecting to the Trinity on purely procedural grounds. One will have to examine the foundations on which it is allegedly built. I think that exploring how these appeals to mystery are supposed to work highlights the importance of examining the epistemic credentials of the theory. This is why in the historical supplement to my SEP piece I start to analyze arguments from the scriptures to the "full deity" of Christ and the distinct and "full deity" of the Holy Spirit.

    ReplyDelete
  14. Anon 4:22

    Your point seems very insightful, but personally I can't figure out exactly what the hell you're talking about. Could you clarify?

    Or, could it be? Could we have a real, live continental philosopher (or continental philosopher wannabe,) complete with impenetrable, obscure prose and an utter lack of argumentation, relying instead on Zen-like statements of paradoxical nonsense?

    Either way I'm fascinated to know personally, so please explain.

    ReplyDelete
  15. I don't think that the ad hoc objection to the concept of suppositum is a good one, even if the doctrine of the Trinity did "just pop into our heads". The suppositum is invoked to reply to the objection that the Trinity is unintelligible. The issue isn't whether the doctrine of the Trinity is true. There is thus no reason to avoid introducing a concept invented just for the occasion; the ad hoc problem arises only when confirmation is at issue. Of course, if the suppositum is itself unintelligible, invoking it won't help, but the ad hoc objection is irrelevant.

    ReplyDelete
  16. I too agree with David. A distincion or use of concepts cannot legitimately be dismissed as "ad hoc". Regardless of where the concepts and distinctions have come from to us, they can be legitimately used, unless an argument is produced by the other party to show that something is wrong with them.

    There can be "ad hoc" assumptions, but not distinctions or concepts. Using a concept or making a distinction does not constitute an assumption. What IS an assumption, on the other hand, is the claim that certain distinction is not valid (that is, the two concepts are equivalent).

    ReplyDelete