Mirza, EAAN, and the TuQuoQue Objection
June 14, 2008 — 21:12

Author: Andrew Moon  Category: Atheism & Agnosticism  Comments: 49

Omar Mirza has a forthcoming paper in Phil. Studies (available in SpringLink in the ‘OnlineFirst’ section) where he (among other things) examines three standard objections to EAAN, shows that Plantinga’s responses are faulty, and then provides his own responses. (For other Prosblogion discussion of EAAN, see here, here, and here. For Plantinga’s most important paper on it, see here.)
I want to examine his response to the tu quo que objection. Now it’s dangerous to reach into the middle of a complicated dialectic and pull out relevant little bits for discussion, but that’s what I’ll try to do! (There is a possibility that I will make hermeneutical errors; I take full responsibility and am open to correction!)


EAAN is as follows:
Probability thesis: P(R/N&E) is low or inscrutable
Defeater thesis: The naturalist who accepts N, E, and the probability thesis has a defeater for R.
It then follows that the naturalist has a defeater for all his beliefs, including his belief in naturalism. So naturalism is self-defeating.
The tu quo que objector attacks the defeater thesis. Where austere theism is the view that God exists (but does not include any content about God creating us as reliable knowers or in his image or whatever), he points out the following:
Probability thesis*: P(R/A) is low or inscrutable
Defeater thesis*: The theist who accepts A and probability thesis* has a defeater for R.
So the theist has a defeater for all his beliefs, including his belief in theism. So theism is self-defeating.
The naturalist can therefore push that he has as much reason to accept the defeater thesis as the theist has to accept the defeater thesis*. The arguments fall or stand together.
The defender of EAAN must give a good reason why the naturalist should accept the defeater thesis that is not a good reason for the theist to accept probability thesis* [TYPO, I SHOULD HAVE SAID ‘defeater thesis*’ HERE. THIS EDIT WAS MADE ON 10:36pm, 6/15/08]. Plantinga has his own strategy for doing this, but I’m interested in Mirza’s right now. The sort of defeater that Mirza thinks is relevant for EAAN is an undercutting defeater, which he defines as a defeater that “undermines the trustworthiness of [a] belief’s source”. How does naturalism do this? Given that Plantinga’s reasons for accepting the probability thesis are good (that’s a whole other discussion), it is the case that if naturalism is true, then “during the operation of the process that resulted in the creation of human cognitive faculties, there were no conditions that could serve to filter out unreliable cognitive faculties. Or, as we might put it, the process did not involve any filter of unreliable cognitive faculties” (section 2 of his paper).
Now what’s the difference between the theist and the naturalist? He writes, “the theist has not been given any grounds for doubting that the process she believes created her cognitive faculties involved a filter of unreliable cognitive faculties…” (section 4.2.5). So while the naturalist, via the probability thesis (and the reasoning for the probability thesis) has been given positive grounds for doubting that the process which created her faculties involved a filter of unreliable cognitive faculties, the theist does not have these positive grounds.
This response seems to me promising. Any thoughts?

Comments:
  • the naturalist … has been given positive grounds for doubting that the process which created her faculties involved a filter of unreliable cognitive faculties, the theist does not have these positive grounds.
    What do you mean by ‘given positive grounds for doubting?’ Do you mean positive grounds for thinking it unlikely that there is a filter for unreliable faculties?

    June 15, 2008 — 13:51
  • Clayton,
    I think that that’s part of it. More of Mirza’s own words: “the relevance of the Probability Thesis in EAAN is that the naturalist who accepts it either has good reason to think that the process that created human beings did not involve a filter of unreliable cognitive faculties, or at least good reason to doubt that the process that created human beings involved a filter of unreliable cognitive faculties;”
    So it’s not only reason to think that there was no filter, but also a reason to doubt (or withhold belief?) that there was a filter. I wonder if, in addition to having reasons for believing that p and believing that ~p, there can be reasons to doubt that p? It seems intuitively plausible; anyway, that’s what Mirza seems to need, and I can’t see anything wrong with it.

    June 15, 2008 — 16:11
  • Maybe it’s just conversational implicature or something, but this sounds really odd to me:
    So it’s not only reason to think that there was no filter, but also a reason to doubt (or withhold belief?) that there was a filter.
    This, however, seems slightly less odd:
    It’s not only reason to doubt or withhold belief that there was a filter, it’s reason to think that there was no filter.
    Now that this terribly important point has been taken care of, if I’m reading you right you think something like this:
    (*) While we should say that on both austere theism and naturalism there is reason to doubt that there was a filter for reliable cognitive faculties, only on naturalism is there reason to think there was no such filter.
    If that’s what you are suggesting, doesn’t (*) nevertheless entail:
    Probability thesis*: P(R/A) is low or inscrutable.
    (*) entails:
    (**) On austere theism there is reason to doubt that there was a filter for reliable cognitive faculties.
    And (**) seems to entail that P(R/A) is low or inscrutable.

    June 15, 2008 — 20:45
  • Andrew Moon

    Clayton,
    Just to be clear, that sentence of mine that you quoted was an attempt to capture the disjunction in the Mirza quotation (in my last comment). My ‘but’ may have been a bit misleading, but what is important is that both disjuncts are understood.
    So it is either the case that there is reason to think that there is no filter or there is reason to doubt that there is a filter. Either of these options is the case for the naturalist (if Plantinga’s arguments for the probability thesis are correct), but neither is the case for the austere theist (even if the tuquoque objector is right). So we have a reason to accept the defeater thesis that is not a reason for accepting defeater thesis*. This makes them disanalogous.
    So Mirza would not accept (*) because (and this was the main point) he doesn’t think the austere theist has been given reason to either doubt that there is a filter or believe that there is no filter. The naturalist has to accept one of these disjuncts (if Plantinga’s earlier arguments are correct). Also, the issue is not about whether the austere theist has a reason to accept probability thesis*. Sorry about this, but I made a typo in my opening post that may have mislead you. I will flag it. I hope this clears things up.
    (Also, the distinction between “having reason to think there is no filter” and “having reason to doubt that there is a filter” may need some more illuminating. When the factory manager tells me that those widgets have red lights shining on them, it’s not the case that “I have reason to think that they are not red”, but it is the case that “I have reason to doubt that they are red. I hope that helps clarify.)

    June 15, 2008 — 23:53
  • Why doesn’t the naturalist reply in roughly this way, supposing the probability thesis asserts this,
    Probability thesis: P(R/N&E) is low or inscrutable
    Let T be the probability thesis, and assume the probability thesis is true. In that case,
    P(T/N&E) is low or inscrutable.
    That is, if naturalism leaves me cognitively unreliable, then it leaves me cognitively unreliable relative to EAAN too, and so too with respect to T. So, I would have no reason to trust the fact that I find EAAN persuasive unless I already believed that naturalism were false. But then EAAN is something close to circular. I’m guessing some naturalist has already given this sort of argument.

    June 16, 2008 — 10:19
  • Andrew Moon

    Mike,
    I think I must’ve been unclear in the opening post, so I’ll give it another shot. If the naturalist accepts the probability thesis for the reasons that Plantinga gives, and if he accepts N&E, then either he has been given reason to doubt that the process which created his cognitive faculties had a filter for reliable faculties or he has been given reason to think that there was no such filter. This gives him an undercutting defeater, and so a reason to accept the defeater thesis. This situation is not the same for the austere theist. This is also not the case for the person who accepts T; there is no undercutting defeater; there is little similarity to the naturalist’s predicament.
    (Btw, your objection sounds like the ‘perspiration objection’, which basically points out that it’s a bad strategy to go from ‘P(A/B) is low or inscrutable’ to ‘B is a defeater for A’. Mirza thinks his general strategy works against the perspiration objection as well, as should be clear from what I said in the last paragraph. Plantinga has his own response to the perspiration objection.)

    June 16, 2008 — 12:59
  • Andrew, the post is clear enough. My worry wasn’t directly related to the tu quoque response (or its value). I made an independent observation on behalf of the naturalist. It goes this way,
    1. P(R /N&E) is low
    2. P(Cogency of EAAN/N&E) is low
    Suppose it’s true that, if naturalism is true, I should not put much credence in the reliability of my cognitive apparatus. That’s (1). If that’s so, then if naturalism is true then I shouldn’t put much credence in the cogency of EAAN. After all, if I can’t trust my cognitive faculties assuming naturalism, then I couldn’t trust them in relation to the cogency of EAAN assuming naturalism. But this shows that no naturalist should find EAAN cogent.

    June 16, 2008 — 13:22
  • Mike,
    Ah, I got you; so you weren’t after the perspiration objection. The objection you’re thinking of was considered by Plantinga in his original presentation of EAAN (WPF, pp. 234-236). (Your objection and his are different, but I think that they are basically the same.) Here’s his statement and part of the response:
    “When the devotee of N&E notes that he has a defeater for R, then at that stage he also notes(if apprised of the present argument) that he has a defeater for N&E; indeed, he notes that he has a defeater for anything he believes. Since, however, his having a defeater for N&E depends upon some of his beliefs, what he now notes is that he has a defeater for his defeater of R and N&E; so now he no longer has that defeater for R and N&E. So then his original condition of believing R and assuming N&E reasserts itself: at which point he again has a defeater for R and N&E. But then he notes that that defeater is also a defeater of the defeater of R and N&E; hence… So goes the paralyzing dialectic. After a few trips around this loop, we may be excused for throwing up our hands in despair, or disgust, and joining Hume in a game of backgammon.”
    I think that what is important is that the line of reasoning you suggest the naturalist should take won’t ultimately get rid of the defeater. It doesn’t stop where you stop it; if the naturalist keeps thinking, the defeater will come back (and then be defeated and then reappear and then defeated…) It will end in a vicious circle, which is also not a happy place for the naturalist to be in. As I understand him, Plantinga ends by basically saying: the defeater will always still be there!

    June 16, 2008 — 14:04
  • As I read this, the argument goes,
    1. P(R / N & E) is low
    2. P(N & E/ N & E) is low
    So, if I believe N&E, then I should not believe N&E. But then I no longer have my defeater for R or N&E, since I no longer find credible my defeater, viz., N&E. Since I no longer find credible N&E, I have lost my reason not to believe both R & N&E. And we start the cycle again.
    But that’s not my argument. My version does stop, as far as I can see. The naturalist says this to the theist:
    1. Either your argument EAAN is sound or it isn’t.
    2. If it is unsound, then I have no reason to believe it’s conclusion.
    3. If it is sound, then I have no reason to believe it’s conclusion.
    4. :. I have no reason to believe the conclusion of EAAN.
    The premise that needs justification is (3). Why would a naturalist make that assertion? Well, if EAAN is sound, then a naturalist would be in no position to trust his judgment concerning the soundness of EAAN. He would be in no position to trust his judgment concerning the soundness of EAAN, since EAAN entails that no one can believe that naturalism is true and also trust his cognitive ability to assess complex arguments such as EAAN. So whether or not EAAN is a good argument, it should not affect the beliefs of a naturalist. We cannot continue the argument without assuming EAAN is good. But that’s something no naturalist has reason to believe.

    June 16, 2008 — 15:11
  • Not this in Plantinga’s argument (which is of course false),
    1. P(N & E/ N & E) is low
    But something like this,
    1′. P(S has reason to believe N & E/N&E) is low

    June 16, 2008 — 15:51
  • Mike,
    Thanks for persevering on this point. Quick question. Wouldn’t you want to put the argument in terms of cogency (as in the earlier post)? On (2), we might have reason to believe in an unsound argument (just as we can sometimes have reason to believe in false premises).

    June 16, 2008 — 21:38
  • Andrew,
    Right, I’m sure it needs to be tightened. I’m not sure what you mean by we might have reason to believe in an unsound argument. Maybe you have in mind that some otherwise reliable person tells me that an argument is sound when in fact it is unsound? Maybe we’d have to say that EAAN itself, if unsound, gives me no reason to believe it’s conclusion. Anyway, as I said, given the amount of literature on this argument, I’m sure some naturalist has given an argument that is close to this.

    June 17, 2008 — 8:11
  • Kyle

    Mike,
    I’m not sure I properly understand your objection.
    But shouldn’t your no. 3 read:
    3′. If EAAN is sound (and N&E is true), then I have no reason to believe it’s conclusion
    This seems trivially true if EAAN is sound because you have no reason to believe that anything is true if N&E is true. However, why believe N&E?

    June 17, 2008 — 9:18
  • Andrew Moon

    Mike,
    Sometimes, we have reason to believe false things. It seems to me that the widgets are red. Unbeknownst to me, the widgets are really white with red lights shining on them. I have reason to believe the widgets are red even though it is false that they are red. I would be justified in using my belief that the widgets are red in a premise in an argument which would ultimately be unsound but which I would be justified in believing. The reason I have trouble with the 4-lined argument you give is because the actual soundness of EAAN seems irrelevant to whether one has reason to accept the argument.
    I’ve read most of the EAAN literature, and nobody’s quite formulated this objection as you have (as far as I know), but it still seems to me to be pretty close to the objection Plantinga dealt with above. The person who (significantly) doubts R, yes, has reason to doubt not only N, but the Probability thesis. I grant that. This is because he has a defeater for all of his beliefs, as Plantinga points out. But then, you say, he now should doubt his belief in EAAN, so he should conclude that he doesn’t have a defeater for any of those beliefs he thought were defeated just a second ago. So he is back to square one; he is reasonable in thinking that all of his beliefs are justified again (or at least not defeated by the reasoning presented by EAAN). But if he thinks those beliefs are justified again, then he must think this of his belief in the Probability thesis and the Defeater thesis, and now we have reason to doubt R again… and so on. I don’t see how your argument is significantly different from the one Plantinga deals with, and it doesn’t seem to avoid the vicious Humean circularity.

    June 17, 2008 — 11:40
  • . . . if EAAN is sound because you have no reason to believe that anything is true if N&E is true.
    EAAN is sound only if naturalists have no reason to believe it is sound. So if you are a naturalist, you should remain unmoved by EAAN.
    However, why believe N&E?
    EAAN is directed to those who already believe naturalism is true. I incidentally think there are lots of reasons in favor of naturalism. It’s not some silly position to hold. My point is that EAAN cannot persuade a naturalist to abandon naturalism.

    June 17, 2008 — 13:27
  • Sometimes, we have reason to believe false things. It seems to me that the widgets are red. Unbeknownst to me, the widgets are really white with red lights shining on them.
    Yes, clearly, but what does this have to do with “believing in” an argument? I was asking what that phrase “believing in” EAAN might mean. I don’t see how the fact that we might be justified in believing something that is false is relevant here. Arguments aren’t false.
    But if he thinks those beliefs are justified again, then he must think this of his belief in the Probability thesis and the Defeater thesis, and now we have reason to doubt R again… and so on.
    I just can’t see how this is relevant to the argument I provided. My argument has this form.
    1. P v Q
    2. P -> R
    3. Q -> R
    4. :. R
    Add whatever premises you like (yours, mine or Plantinga’s . . . or Jove’s or Thor’s) to the argument, it doesn’t matter a whit. Add contradictions or circular assumptions, it still does not matter. The argument is valid and remains valid under the addition of any premises you like. I’ve tried to show that the premises are true, and concluded (tentatively) that it’s sound. Since the argument is deductive, it’s monotonic. Add what you want, if it is valid (as it is) it remains so. If is it sound (as I think it is) it remains so. If you have an objection, it must be that one of the premises is false or that it is invalid. But if the argument is sound, then no naturalist has reason to believe the conclusion of EAAN. So which premise is false?

    June 17, 2008 — 14:18
  • Mike,
    Okay, I’ll attack your argument more directly. First, let me state EAAN a little more precisely:
    A) The naturalist who believes N&E, “P(R/N&E) is low or inscrutable”, and Plantinga’s reasons for believing the probability thesis has a defeater for his belief in R (the defeater thesis)
    B) If someone has a defeater for his belief in R and he reflects appropriately about this fact, then he has a defeater for all his beliefs.
    C) The naturalist who meets the description in (A) and who reflects on the fact that he has a defeater for his belief in R has a defeater for all his beliefs.
    D) The naturalist who meets the description in (A) and who reflects on the fact that he has a defeater for his belief in R has a defeater for his belief in naturalism.
    Now let’s look at your two crucial premises:
    2. If it is unsound, then I [the naturalist] have no reason to believe its conclusion.
    3. If it is sound, then I [the naturalist] have no reason to believe its conclusion.
    But I think that both premises are false. (2) is false because he may still have reason to believe in an argument which has justified false premises. (See my widget example.)
    More importantly, (3) is false because it is talking about a naturalist simpliciter, not the reflective naturalist described in (D). The naturalist has no reason to believe its conclusion only if he meets the description of the naturalist described in (D) and has considered how his defeater for all his beliefs affects his belief in EAAN. But at that same point of time (if he is considering (A)-(C)), he has also has a defeater for naturalism. He has a defeater for both EAAN and N at the same time.
    Suppose he reflects a little further and realizes (as you point out) that since he has a defeater for EAAN, he can give it up. Now he can hold his beliefs again and the skepticism no longer threatens. He can conclude: “I have no reason to believe in EAAN, so I can hold my beliefs again!” But with the return of his beliefs comes the return of the belief in premises (A)-(C). And we are back in the Humean circle.
    In short: at any time the naturalist is in a state that he has a defeater for his belief in EAAN, he also has a defeater for his belief in N. And at any time the naturalist stops believing in EAAN (because of EAAN), he is able to hold his beliefs again, but then (if he is reflective) he will start believing the premises of EAAN again until he has a defeater for N (and EAAN). And then the cycle continues.

    June 17, 2008 — 14:34
  • Mike,
    I was working on that last comment before I got to see your most recent comment (so it may not seem like I’m responding appropriately).
    Now that I’ve gotten a chance to read it, to clarify, one believes in an argument if they believe all the premises and the conclusion of the argument.
    My criticism of (2) (from the last comment) should’ve said that one may have good reason to believe in the conclusion of an argument even if it is false (or the argument it is based upon is unsound). So I was attacking (2) in the earlier comment, but I didn’t make that clear.
    You’re right that I needed to just attack your argument; hope the last comment helps.

    June 17, 2008 — 14:46
  • 2. If it is unsound, then I [the naturalist] have no reason to believe its conclusion.
    3. If it is sound, then I [the naturalist] have no reason to believe its conclusion.
    But I think that both premises are false. (2) is false because he may still have reason to believe in an argument which has justified false premises. (See my widget example.)
    More importantly, (3) is false because it is talking about a naturalist simpliciter, not the reflective naturalist described in (D). The naturalist has no reason to believe its conclusion only if he meets the description of the naturalist described in (D) and has considered how his defeater for all his beliefs affects his belief in EAAN. But at that same point of time (if he is considering (A)-(C)), he has also has a defeater for naturalism. He has a defeater for both EAAN and N at the same time.

    Let’s try to save some ink. Modify (2) in the way suggested (above) with the naturalism part fussed a bit (henceforth, assume I mean by ‘naturalism’ what you have in mind by it),
    2′. If EAAN is unsound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    3′. If EAAN is sound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    About this,
    But at that same point of time (if he is considering (A)-(C)), he has also has a defeater for naturalism. He has a defeater for both EAAN and N at the same time.
    If I’m a naturalist [as described in D] then if EAAN is sound, then EAAN does not give me reason to believe that my belief in naturalism is unreliable. EAAN can give me a reason to believe that my belief in naturalism is unreliable only if I am not a naturalist. That’s the point of the argument. So EAAN is not a defeater for EAAN and N at the same time.

    June 17, 2008 — 15:41
  • Mike,
    Well, I guess I still don’t understand, and we may be talking past one another. (Such conversations are easier in person.) I still don’t see the justification for (2) (or 2′) since I don’t see how actual unsoundness of an argument makes a direct difference to whether one has a reason to accept it (or its conclusion). What matters for whether one has reason to accept an argument is whether one is justified in believing the premises, not whether the premises are true or false.
    And on (3′), as it is now formulated, when you assume the antecedent (that EAAN is sound), it follows that the reflective naturalist has a defeater for his belief in N (since EAAN is sound only if the conclusion is true, and the conclusion is true only if the reflective naturalist has a defeater for N). So it may be that the antecedent of (3′) implies the consequent you mention (that EAAN does not itself give the naturalist a reason to accept its conclusion), but it also implies that the naturalist has a defeater for his belief in N. That’s pretty weird, and I’m sure (as you mentioned) there’s probably something in the formulation to be fixed up.
    Well, I think this is interesting, and I want to repeat that I haven’t seen this specific form of the argument in the literature and I’m familiar with the literature, so if I’m wrong that this is getting at the same argument as Plantinga’s argument (above), then you may be on to something worth pursuing. I’ve gotta spend a little less time blogging now and go work more on my dissertation proposal more (though I will make sure to read your response if you have one, but I may not respond)!

    June 18, 2008 — 13:21
  • Paul

    Hi Mike,
    I am confused as well.
    First, let’s look at a paradigm case of defeat:
    Say there is a drug, call it XX, that cures those who take it of some malady but has a side effect of causing massive hallucinations and dementia or temporary insanity to occur in 90% of those who ingest XX. Say S comes to believe she has ingested XX. It would appear that S now has a defeater for (almost all, all?) of her beliefs. Say S comes to believe this due to her fathers testimony. It appears that, on your analysis, S now has a defeater for her belief that she ingested XX. So would it be right for her to reason that she does not have a defeater for her beliefs? We can switch subjects and talk about one who believes he is the victim of a Cartesian demon bent on causing epistemic havoc. Or perhaps one believes she is victim of Alpha Centaurion gamma rays shot down to earth causes unreliability on ones cognitive faculties. On all these cases I could switch the terms of your argument but keep the form the same. But now it seems like *any* case of defeat (analogous to EAAN) can de deflected or defeated by your argument. But, it is wrong to suppose that the people in the above paradigm cases should be able to defeat or deflect defeat using your argument. Your argument allows *paradigm* cases of defeat to not be defeat. Any argument that allows a paradigm case of defeat to not be defeat has got the wrong end of the stick.
    Secondly, the situation looks like this for the naturalist.
    N&E have *built within it* the defeater for N&E (for the reflective naturalist). So, the naturalist reflects on her cognitive situation. Sees that P(R|N&E) = low or inscrutable. She then has a defeater D for all her beliefs. This includes belief in the *conjunction* of N&E. She could give up N, but doesn’t. She’s read her Dawkins. The arguments are just too good to pass by! 🙂 So, she obtains her defeater. But since she has a defeater for all her beliefs, she has a defeater for D. So all is well. She holds N&E. But this, for the reflective naturalist, entails EAAN. So D comes back. She has a defeater for all her beliefs. So Plantinga,

    “This is, of course, extraordinary: ordinarily, if one acquires a defeater-defeater for a belief B, i.e., a defeater for a defeater of B, one no longer has that defeater for B–or else its defeating power is neutralized. But not so here. The difference is that here the original defeatee shows up at every subsequent level. When that happens–when, roughly speaking, every defeater in the series is really the defeatee plus a bit, the defeater-defeater doesn’t nullify the defeater. The defeater gets defeated, all right, but the defeatee remains defeated too.

    So I think, for at least these two reasons, the out you’ve postulated for the naturalist won’t work.
    Best,
    Paul

    June 18, 2008 — 21:41
  • N&E have *built within it* the defeater for N&E (for the reflective naturalist). So, the naturalist reflects on her cognitive situation. Sees that P(R|N&E) = low or inscrutable. She then has a defeater D for all her beliefs.
    Thanks Paul. This (above) is exactly what I am denying. Let N be the reflective naturalist. You offer to N this argument EAAN. His response ought to be that he cannot be perusaded away from his naturalism by such an argument. If EAAN is sound, then the naturalist cannot trust his own assessment of EAAN, and so cannot reasonably be perusaded by it. I cannot reasonably be persuaded by any argument whose cogency I believe I am in no position to assess. On the other hand, if EAAN is unsound, then I cannot reasonably be persuaded by it.
    Now, suppose we assume (1) is false,
    1. If EAAN is sound, then the naturalist cannot trust his own assessment of EAAN, and so cannot reasonably be perusaded by it.
    So, we are assuming (a) EAAN is sound and (b) I can reliably come to believe that EAAN is sound and (c) I can be persuaded away from my naturalism. But if (b) is true, then
    (i) my cognitive faculties are reliable in the assessment of the cogency of EAAN.
    But wait! If EAAN is sound, then
    (ii) my cognitive faculties are not reliable in the assessment of EAAN (or otherwise),
    Contradiction from (i) and (ii). So, by reductio, (1) is true.

    June 19, 2008 — 9:52
  • We can switch subjects and talk about one who believes he is the victim of a Cartesian demon bent on causing epistemic havoc. Or perhaps one believes she is victim of Alpha Centaurion gamma rays shot down to earth causes unreliability on ones cognitive faculties. On all these cases I could switch the terms of your argument but keep the form the same. But now it seems like *any* case of defeat (analogous to EAAN) can de deflected or defeated by your argument.
    No, these classic arguments for defeat claim that you do not know P unless you know that you are not being decieved by a demon/alpha centurian, etc. My argument does nothing against these. But that’s not how Plantinga argues. Instead, he argues that because EAAN is sound, the naturalists beliefs are defeated. But as I say just above, this is not a conclusion that a naturalist can reach.

    June 19, 2008 — 13:32
  • Paul

    Hi Mike,
    I also posted another response besides this one. Perhaps it hasn’t made it “through” the screening process yet.
    You wrote:

    No, these classic arguments for defeat claim that you do not know P unless you know that you are not being decieved by a demon/alpha centurian, etc. My argument does nothing against these. But that’s not how Plantinga argues. Instead, he argues that because EAAN is sound, the naturalists beliefs are defeated. But as I say just above, this is not a conclusion that a naturalist can reach.

    That’s not how I meant it. Rather, I meant it that if a cognizer S finds himself believing that he has, say, ingested XX, and that it causes hallucinations or temporary insanity is 90% of those who inject it, S has obtained a defeater for S’s beliefs. Same if S finds herself believing that she is a victim of an evil demon. It’s not that she can’t know that P unless she knows she’s being deceived, it is merely that she believes the probability is low (or inscrutable) that she isn’t being deceived, and so obtains a defeater.
    Likewise, I take it that Plantinga’s argument isn’t exactly how you say. Indeed, Plantinga states that his argument depends on the naturalist “seeing” that the probability that she would have reliable cognitive faculties given N&E is either low or inscrutable. That’s it.
    So, the naturalist can either reflect on the argument or not. I addressed various naturalist responses in my other response. So rather than repeat myself I’ll hope it is posted.
    Also:
    2′. If EAAN is unsound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    3′. If EAAN is sound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    Take the argument: If you ingest XX you have a defeater for R. Call this argument XXAAR
    Now:
    2”. If XXAR is unsound, then XXAR (itself) does not give me [the ingester as described in D] reason to believe its conclusion.
    3”. If XXAR is sound, then XAR (itself) does not give me [the ingester as described in D] reason to believe its conclusion.
    But if anything gives a cognizer reason to doubt R, XXAR does. Thus your argument can’t be sound.
    Best,
    Paul

    June 19, 2008 — 17:29
  • I meant it that if a cognizer S finds himself believing that he has, say, ingested XX, and that it causes hallucinations or temporary insanity is 90% of those who inject it, S has obtained a defeater for S’s beliefs.
    We need to consider parallel arguments. This does not so far look parallel to mine. So, let me run mine, and then yours. Assume for reductio that (1) is false.
    1. If EAAN is sound, then the naturalist cannot trust his own assessment of EAAN, and so cannot reasonably be perusaded by it.
    So, we are assuming (a) EAAN is sound and (b) I can reliably come to believe that EAAN is sound and (c) I can be persuaded away from my naturalism. But if (b) is true, then
    (i) my cognitive faculties are reliable in the assessment of the cogency of EAAN.
    But wait! If EAAN is sound, then
    (ii) my cognitive faculties are not reliable in the assessment of EAAN (or otherwise),
    Contradiction from (i) and (ii). So, by reductio, (1) is true.
    Now, the parallel argument. Assume for reductio that (2) is false.
    2. If it is true that S have ingested XX, then S cannot trust the testimony from his father that he has ingested XX, and so cannot reasonably be perusaded by such testimony that he has a defeater for all of his beliefs.
    So, we are assuming (a) S has ingested XX and (b) S can reliably come to believe the testimony that he has ingested XX and (c) S can thereby be persuaded that he has a defeater for all of his beliefs. But if (b) is true, then
    (i) S’s cognitive faculties are reliable in the assessment of the testimony of S’s father.
    But wait! If S has ingested XX, then
    (ii) S’s cognitive faculties are not reliable in the assessment of the testimony of his father,
    Contradiction from (i) and (ii). So, by reductio, (2) is true.
    So no one who has ingested XX can reasonably come to believe on the basis of testimony concerning his ingestion of XX that he has a defeater for his beliefs. Notice that I do not deny that S can unreasonably come to believe he has a defeater. He can rely on his own judgment and come to such a belief. But his judgment is by hypothesis unreliable after the ingestion.

    June 20, 2008 — 10:13
  • Paul

    “1. If EAAN is sound, then the naturalist cannot trust his own assessment of EAAN, and so cannot reasonably be perusaded by it.”
    But this only happens *after* the naturalist has “went through” the argument. As Plantinga has stated, the argument only “works” for those who *grant* the low probability.
    After they “see” the problem, *then* they might refuse to go through the argument again. But in this case the cognizer is not a *reflective* naturalist. I mean, if this is what the naturalist is reduced to, so much the worse for naturalism. Any theist who refused to re-listen to an argument that seemingly refuted theism would be chided.
    Furthermore, the answer might be that because the naturalist *can* be persuaded by it, naturalism can’t be true. The one who *gives* the argument assumes 1b, and that’s because that person believes N is false. The one to whom the argument is given, after she go through it, grants the low/inscrutable probability, *then* loses the EAAN defeater.
    “But wait! If EAAN is sound, then
    (ii) my cognitive faculties are not reliable in the assessment of EAAN (or otherwise),”
    Not true. As I said above, the cognitive faculties *are* reliable. EAAn does not entail the *fact* that the naturalist has unreliable cognitive faculties.
    As far as my counter example: I only assume S forms the belief that she has ingested XX. Reliably or not, that doesn’t matter. If she believes herself an ingester of XX, she has a defeater for her beliefs.
    Same with your 2(ii). Since it *may* be that S’s faculties *are* reliable (since 10% *do not* suffer the temporary insanity), then I *do not* assume that the faculties are *in fact* unreliable. I merely assume that there is a *defeater* for the *belief* that they are.

    June 20, 2008 — 23:28
  • “1. If EAAN is sound, then the naturalist cannot trust his own assessment of EAAN, and so cannot reasonably be perusaded by it.”
    But this only happens *after* the naturalist has “went through” the argument. As Plantinga has stated, the argument only “works” for those who *grant* the low probability.

    That helps, since no doubt details have been lost. We can reassert these and tighten things up. The naturalist can reach his conclusion without ever being persuaded by EAAN. He can reason entirely a priori. That is, he can be hypothetically persuaded. Here’s the argument, supposing I’m the naturalist.
    1. Either EAAN is sound or it isn’t, I don’t know which.
    2. I believe firmly that N & E are true.
    3. I recognize that if EAAN is sound, then N & E together render R low.
    4. So, I recognize that if EAAN is sound, then someone such as I–someone committed to the truth of N & E–is also committed to believing that he is in no position to assess it’s soundness and so cannot reasonably be persuaded by it.
    5. So if EAAN is sound, I (i.e., a person committed to N&E) cannot reasonably be persuaded by it.
    6. If EAAN is unsound, it’s not relevant to what I believe.
    7. So whether EAAN is sound or unsound, I cannot reasonably be persuaded by it.

    June 21, 2008 — 10:30
  • Paul

    Mike,
    The arguments begins with a question:
    What is the probability of R given N&E (P(R|N&E)=?).
    That’s it. Can I not ask that of a naturalist? Is it blasphemous! Would all naturalists have to say in response: “I can’t even try to figure it out because I run the risk of finding out that P(R|N&E) = low or inscrutable. And if I found that out, then I couldn’t figure out the probability in the first place, so I won’t even try.” That position just seems odd to me. The above seems a perfectly legitimate question to ask.
    Furthermore, as Plantinga explicates the argument he makes this move: “Now, let’s not ask this of ourselves, rather, let’s begin by looking at another group of agents similar to ourselves that arose via the popular evolutionary story, and did so in the context of naturalism. What would P(R|N&E) be with respect to them? Are you saying that naturalist shouldn’t try to answer that question?
    I also take issue with your 4:
    “4. So, I recognize that if EAAN is sound, then someone such as I–someone committed to the truth of N & E–is also committed to believing that he is in no position to assess it’s soundness and so cannot reasonably be persuaded by it.”
    One response is that this begs the question because if, per the grant, EAAN is sound, then they “he is in no position to assess” that “he is in no position to assess it’s soundness and so cannot reasonably be persuaded by it.”! The naturalist can’t have his cake and eat it too. So if he’s going to reject his ability to assess EAAN given that EAAN is sound, then he would also have to reject his assessment of that. In fact, he would have to reject his entire reductio if he grants EAAN is sound.
    So it would seem to me that once he grants the first disjunct, he can’t then go about assuming that everything else he’s doing isn’t likewise defeated.
    That is, he can’t say, “Well if EAAN is sound then I couldn’t assess it but I can asse4ss that I couldn’t asses it.” He’s cheating.
    Second, if EAAN were sound, then it is *irrational* to accept N&E. So if he grants that EAAN is sound, he can’t say, “but I’m committed to N&E anyway,” unless he wants to add, “And I’m irrational for doing so.”
    I also reject your 5
    “5. So if EAAN is sound, I (i.e., a person committed to N&E) cannot reasonably be persuaded by it.”
    Not only does he also lose the right to be reasonably persuaded by this answer if he grants EAAN is sound (see above), but:
    I pointed out above that this is a false inference. N&E may be *false*, the naturalist is made in God’s image, say, able to reason, cognitive faculties designed to produce most true beliefs, etc., and so he can *in fact* be persuaded by it. That he is committed to N&E doesn’t imply that N&E is in fact the case.
    Lastly, you may have developed a new theodicy!
    Take a theist like me. I don’t think P(R|N&E) is high. I think I need something like Christian theism for that. Thus I think, currently, that if there were no God, and so evolution the best alternative to all the diversity and apparent design, and also naturalism would have to be the case, I would have no reason to assign a high number to P(R|N&E).
    Now, this is the view of many Christians.
    Enter the PoE:
    (One version of the) PoE states that given all the evil that exists, God’s existence is less probable and N&E is more probable. So Dawkins,

    “In a universe of blind physical forces and genetic replication some people are going to get hurt, other people are going to get lucky, and you won’t find any rhyme or reason in it, nor any justice. The universe we observe has precisely the properties we should expect if there is, at bottom, no design, no purpose, no evil and no good, nothing but blind, pitiless indifference.”

    Evil is not a defeater to this position. It’s what we would in fact expect. But not so on theism.
    Okay.
    If (this) PoE were sound, then my theism would be defeated.
    But, if that happened, I wouldn’t be able to assess this PoE.
    If it is unsound I do not need to worry.
    And so here’s my theodicy:
    Let T = Theism.
    Let PoE = Problem of Evil.
    1. Either PoE is sound or it isn’t, I don’t know which.
    2. I believe firmly that T is true.
    3. I recognize that if PoE is sound, then N & E together render R low.
    4. So, I recognize that if PoE is sound, then someone such as I–someone committed to the truth of T–is also committed to believing that he would be in no position to assess it’s soundness and so cannot reasonably be persuaded by it.
    5. So if PoE is sound, I (i.e., a person committed to T) cannot reasonably be persuaded by it.
    6. If PoE is unsound, it’s not relevant to what I believe.
    7. So whether PoE is sound or unsound, I cannot reasonably be persuaded by it.
    And thus your argument against EAAN suggests that all a theist needs to do to defeat almost any PoE is form the believe that R is very likely on T and not very likely on N&E.
    Theists (like me and many others) do not even need to think about and debate the PoE. “He can reason entirely a priori.” And what’s worse, the atheist cannot debate me on this without hearing my reasons for EAAN, which he can’t do if he wants to remain insulated as you try to do for him!
    So, not only do I not accept your argument against EAAN, if I did accept it, then it helps (most) theists out by giving them an easy defeater to one of the greatest atheological weapons, the PoE.

    June 21, 2008 — 18:03
  • I’ll try to keep it brief, since we might have exhausted the dialectic. So, here are some short replies (though I appreciate the pressing questions).
    I pointed out above that this is a false inference. N&E may be *false*, the naturalist is made in God’s image, say, able to reason, cognitive faculties designed to produce most true beliefs, etc., and so he can *in fact* be persuaded by it. That he is committed to N&E doesn’t imply that N&E is in fact the case.
    No, this is mistaken. You cannot be reasonably persuaded by any argument that you firmly believe you cannot assess. In short, it is incoherent to say “I believe firmly that I cannot assess the argument EAAN, but it is reasonable for me to believe the conclusion of EAAN based on the premises of EAAN.” Similarly, I cannot believe that I am being decieved by a demon and also believe that I can rely on my senses. Given that belief, what I perceive will simply not be convincing or persuasive to me (even if there are no demons!).
    There’s probably a simpler way of thinking about this argument. The believer in N&E has no idea whether EAAN is sound, and has no belief about it’s actual soundness. He does not need one. He knows this: in every world in which he is committed to N&E, he is committed to believing that either EAAN is unsound or he cannot assess it’s soundness. Either way, he must remain unpersuaded by it.

    June 21, 2008 — 21:17
  • nacisse

    Mike Almeida, wouldn’t the committed E&N person have to be more committed to E&N then to R? is that even possible?
    William Alston wrote an essay called ‘Plantinga, Naturalism and Defeat’ (i think that was the name) where he questioned Plantinga’s way of arguing that N&E defeats R… he said that R has basic warrant and so however committed we may be to N&E R wouldn’t be defeated by it – rather R would defeat N&E. so if someone committed to N&E found EAAN sound they would have a defeater for N&E not R — i think that is how it went anyway…

    June 22, 2008 — 1:45
  • test….

    June 22, 2008 — 9:42
  • Paul

    Mike,
    “No, this is mistaken. You cannot be reasonably persuaded by any argument that you firmly believe you cannot assess. In short, it is incoherent to say “I believe firmly that I cannot assess the argument EAAN, but it is reasonable for me to believe the conclusion of EAAN based on the premises of EAAN.”
    I see your point. But,
    1. I was speaking to the claim that his faculties were *in fact* not reliable. It seemed at times this is what you were hinting at. You did so with AXXR, and in other places.
    2. I also pointed out that this is ‘having your cake and eating it too.’ As soon as he grants soundness, and then claims that “if sound/no assess,” but (!), I could asses that, then he’s cheating.
    So, “if sound” then he can’t be persuaded by his claim that he can’t assess EAAN either. Now we’re back at square one.
    “There’s probably a simpler way of thinking about this argument. The believer in N&E has no idea whether EAAN is sound, and has no belief about it’s actual soundness. He does not need one. He knows this: in every world in which he is committed to N&E, he is committed to believing that either EAAN is unsound or he cannot assess it’s soundness. Either way, he must remain unpersuaded by it.”
    3. He has no idea because he won’t reflect on the argument. That’s not epistemically virtuous. I for one do not think naturalists have a blank check to ignore questions like: “What is the probability of R given N&E?”
    4. So basically you have our naturalist saying: “Well N&E means so much to me that I flat out refuse to even think about someone questioning the probability that our cognitive faculties have the purpose of producing mostly true beliefs.”
    Well, it seems then he must remain agnostic as to R.
    So, to the question: Hi, Mr. Naturalist, what do you think P(R|N&E)=?
    He must say: “I’m agnostic.”
    But this, of course, gives him a defeater for R.
    5. It just seems odd to me that this kind of dialogue would be considered acceptable:
    S = Supernaturalist
    N = Naturalist
    S: I think naturalism has serious problems.
    N: What’s that?
    S: Well, for starters, there’s EAAN.
    (Now, note that the naturalist can’t say, “What’s that, because then he gets drawn into a discussion, and this fails to insulate him, as you indicate above. So, he says something like:)
    N: Give me the bottom line.
    S: Well, the bottom line of EAAN is that if naturalism + evolutionary theory of our cognitive development were true, the probability that our cognitive faculties have the purpose of producing mainly true beliefs is low or inscrutable, which gives the reflective naturalist a defeater for all his beliefs.
    N: Well, N&E is true. And so if your argument were sound, I couldn’t assess it. Thus I deflect defeat.
    S: And if it were sound you have a defeater for your defeater-deflector. So if it is sound, what you just said can’t count, for you, as a reason to deny EAAN. So if EAAN is sound, then you can’t assess its soundness and you can’t assess that you can’t assess its soundness.
    6. I’m also unsure if you’ve avoided the XX counter example. AXXR did not depend on (a) S reliably believing she took XX, (b) and it does not presumes that R is not reliable for S since 10% do not suffer the side effects. So if I went up to someone and said that they took XX, they could automatically ignore me and refuse to reflect on their cognitive situation. That seems to me a disaster in the making. I wouldn’t want that person on the road. But you give her the epistemic right.
    7. I was thinking about my argument for defeating PoE’s above. It’s actually worse.
    Given that I (and many other Christians) hold to something like EAAN, AFR, etc., then if one shows us that God does not exist, or his existence is unlikely, then as I (and most, perhaps all Christians) also believe, N&E would be the best account of things. But then I would not be able to assess the defeaters for God belief. So if your counter-argument to EAAN were cogent you would have to say that I can reason thus about any apologetic challenge:
    Let T = theism
    Let A = any Apologetic argument
    He knows this: in every world in which he is committed to T, he is committed to believing that either A is unsound or he cannot assess it’s soundness. Either way, he must remain unpersuaded by it.
    I assume you do not want to write apologists everywhere a blank check for outright ignoring and dismissing any atheological argument whatever, but you would have to given your argument against EAAN.
    8. Lastly, the naturalist atheologian would not even be able to question this strategy of mine since then they’d have to hear my reasons for EAAN, but this loses the insulation you want to provide them.
    At the end of the day, isn’t this what you prescribe for the naturalist:
    “Just dig in your heels and dogmatically declair N&E to be the case and refuse to look into the argument that R would be low or inscrutable on N&E.”
    ?
    Say our naturalist is an epiphenominalist. So her beliefs to not enter into the causal nexus (either totally of semantically). It seems they’d be invisible to natural selection and it seems R would be low given this. But the naturalist can just whistle Dixie and not even bother to think about the implication of epiphenomenalism as it relates to R given N&E. I find this highly problematic.

    June 22, 2008 — 23:50
  • 2. I also pointed out that this is ‘having your cake and eating it too.’ As soon as he grants soundness, and then claims that “if sound/no assess,” but (!), I could asses that, then he’s cheating.
    Paul, I confess to not reading the entire post, since these really should be getting much shorter at this point in the discussion. In any case, mine will be shorter.
    Concerning the above, I think I’ve said several times now that he does not grant the soundness of of the argument. He concedes that it might be sound and considers worlds in which it is sound. So his argument is entirely hypothetical.

    June 23, 2008 — 7:56
  • …wouldn’t the committed E&N person have to be more committed to E&N then to R? is that even possible?
    nacisse,
    I’m not sure I follow you. He is committed to his N & E, and that places him where he cannot evaluate EAAN, if EAAN is sound. It leads him to conclude that worlds in which EAAN is sound (if there are any) are ones in which EAAN is not a defeater for R (for him).

    June 23, 2008 — 8:02
  • Paul

    Mike (a shorter post!),
    As Plantinga states, the argument is only for the one who accepts N&E and the probability thesis. It is not designed for the dogmatic naturalist who says N&E is true and so he refuses to look at arguments purporting to show R has a low or inscrutable probability given N&E. In other words, your response to EAAN for the naturalist is to tell them to not even think about the argument.
    So, what do they say when I ask them what would the probability of R be on N&E? What do they say? Nothing? That’s odd. Do they say, “I don’t know?” But then they have the defeater? They certainly can’t say “low.” And they can’t say “High,” because then we’d get into the debate. I suppose they can say, “High!”, cross there arms, and say, “Period. End of story. Look at the time. I’m late. Got to go.”
    Now, in answer to your response:
    “I think I’ve said several times now that he does not grant the soundness of the argument. He concedes that it might be sound and considers worlds in which it is sound. So his argument is entirely hypothetical.”
    Right. That’s fine. And my response is that in worlds where it is be sound are worlds where he can’t assess that he can’t assess it….hypothetically.
    In worlds where it might be sound he’s saying: “Well, if it’s sound, then I couldn’t assess it.” I got that. But what I want to add is that he’s cheating. The full sentence would be: “If it’s sound, I can’t assess it and I couldn’t assess that I couldn’t assess it.”
    He has no way out, hypothetically or not.
    If it might be sound then he might not be able to assess it and he also might not be able to assess that he can assess it.
    So it seems like he’s in this situation:
    Either EAAN is sound or not. If it is, then I can’t dismiss it because I’d have a defeater for any belief that dismisses it. If it isn’t, then I need not worry about it. So, I guess I better look into the argument.”

    June 23, 2008 — 10:15
  • Paul

    Mike,
    Another way to respond, which I alluded to above, is that the argument proceeds by having us look at a hypothetical population of creatures like us.
    The naturalist can engage here, right? If R for these creatures is low on something like N&E, then they have a defeater for R, but not our actual, earthly naturalist….yet 🙂
    So, at this stage your ‘insulator’ does not come into play.
    Plantinga:

    Is this impossible or unlikely? That depends upon the relation between belief and behavior. What would or could that relation be? To try to guard against interspecific chauvinism, I suggested that we think, not about ourselves and our behavior, but about a population of creatures a lot like us on a planet a lot like earth (Darwin suggested we think about monkeys in this connection). These creatures are rational: that is, they form beliefs, reason, change beliefs, and the like. We imagine furthermore that they and their cognitive systems have evolved by way of the mechanisms to which contemporary evolutionary theory direct our attention, unguided by the hand of God or anyone else. Now what is P(R/N&E), specified, not to us, but to them?

    {Insert various arguments and grant Plantinga shows P(R|N&E) = low or inscutable}
    Now the next move is:

    Now return to Darwin’s doubt, and observe that if this is the sensible attitude to take to P(R/N&E) specified to that hypothetical population, then it will also be the sensible attitude towards P(R/N&E) specified to us. We are relevantly like them in that our cognitive faculties have the same kind of origin and provenance as theirs are hypothesized to have.

    If the naturalist grants the Probaility Thesis with respect to those creatures, he should do so with respect to himself since there are only (apparently) accidental differences. That is, if he admits that R is defeated for creatures in a situation just like his, then he should admit that for himself.
    Since we are asking him to reason about other creatures than himself, your ‘insulator’ is irrelevant in this regard. If he grants that creatures that arose via evolutionary processes, just like he did, in a non-supernatural environment (say no God guides the development of our CFs to ensure they are reliable), just like he did, have defeater D for R, then he owes us an explanation for why he would not have D for R.
    Since we spoke of other creatures, the ‘insulator’ is not needed, and when it is needed, it’s too late.

    June 23, 2008 — 13:37
  • nacisse

    He is committed to his N & E, and that places him where he cannot evaluate EAAN, if EAAN is sound.
    Mike Almeida,
    I was thinking that it would place him where either he would accept that he can’t evaluate EAAN (or anything else), if EAAN is sound, or he rejects E & N, if EAAN is sound. so I thought he’d have to be more committed to his N & E than to his own Reason – since giving up on R is not a real possibility he should give up on N & E.

    June 23, 2008 — 13:42
  • In worlds where it might be sound he’s saying: “Well, if it’s sound, then I couldn’t assess it.” I got that. But what I want to add is that he’s cheating. The full sentence would be: “If it’s sound, I can’t assess it and I couldn’t assess that I couldn’t assess it.”
    You’ve got a fair gripe here, since the argument should be phrased in terms of “being reasonably persuaded by EAAN” rather than in “being in a position to assess EAAN”. Let @ be our world and let w be the hypothetical world where, we are imagining, EAAN is sound. I make the argument here in @ (@ is, as they say, the world of utterance), I’m not making the argument in w (w is, as they say, the world of evaluation). I say, here in @, about this world w: If I WERE in w, and I WERE committed to N&E, and EAAN WERE sound, then I would not be reasonably persuaded by EAAN.
    Why do I say that “I would not be reasonably persuaded by EAAN in w”? It is not because N&E are true in w, but (as I noted above) because I believe it is true. Suppose for reductio that there is a world w in which I believe N&E are true and I am reasonably persuaded by EAAN. If am reasonably persuaded by EAAN, then I believe that I have reasonably reached the conclusion that EAAN is sound. Since in w I believe that N&E are true, and that I have reasonably reached the conclusion that EAAN is sound, I believe in w that I have the cognitive resources that allow me to reliably reach the conclusion that I do not have reliable cognitive resources. To reach the conclusion that EAAN is sound, I would have to believe a contradiction. Therefore, naturalists don’t reasonably reach the conclusion that EAAN is sound in worlds where it is sound.

    June 23, 2008 — 13:45
  • If the naturalist grants the Probaility Thesis with respect to those creatures, he should do so with respect to himself since there are only (apparently) accidental differences.
    Paul, the naturalist would conclude that those creatures would not be reasonably persuaded by EAAN, supposing it is sound. So EAAN would not provide a defeater for them.
    The entire worry I have can be summed up fairly quickly. The naturalist has to have already abandoned his commitment to naturalism in order to be reasonably persuaded by EAAN that naturalism has unwelcome implications!

    June 23, 2008 — 14:05
  • Paul

    Mike,
    Here’s your defeater-deflector De-Def. Call De-Def (*):
    (*) If N&E were true, which I believe it is, and if EAAN were sound, which is what I would believe in w, then I would not be able to be reasonably persuaded that EAAN is sound, and I could not be reasonably persuaded of that, I have no reason to put any stock into EAAN.
    Our naturalist N takes (*) to be a defeater deflector. But I claim it is cheating. If EAAN were sound in W then N could not use (*) to deflect potential defeat since (*) would be also defeated. That is, N could not be reasonably persuaded of (*). So, in w N deflects EAAN with (*) and also defeats belief in (*).
    “The entire worry I have can be summed up fairly quickly. The naturalist has to have already abandoned his commitment to naturalism in order to be reasonably persuaded by EAAN that naturalism has unwelcome implications!”
    Part of the draw-back to our convo is that we can’t get to in-depth due to time or whatever. Fpr example, what do you mean by “EAAN.” Is it just this:
    P(R|N&E) = low or inscrutable?
    But then what you say is false since many naturalists have, in response to Plantinga, argued that they grant the probability thesis (are reasonably persuaded by it) but that it still does not give them a defeater for R. For example, they might claim, and have claimed, that they “hit the evolutionary jackpot.”
    “Paul, the naturalist would conclude that those creatures would not be reasonably persuaded by EAAN, supposing it is sound. So EAAN would not provide a defeater for them.”
    That doesn’t matter. We’re not talking, right now at least, about those creatures believing R is low (or inscrutable) conditioned on N&E. We’re talking about how we would judge R with respect to their situation. If we judge R as low (or inscrutable) with respect to them, then why don’t we judge R as low or inscrutable with respect to us? We just admitted that the probability is low that creatures that arose in situations very similar to our own had reliable cognitive faculties, but when we now ask that about ourselves we, what, say nothing? We say that we believe R is low for them but not us? Based on what? That we don’t want to lose our precious N&E? I’m confused.
    Lastly, my Apologetic Blank Check Argument (see above) has this consequence:
    If Mike’s counter to EAAN argument were cogent, he’d have to grant the Apologetic Blank Check Argument. Mike will not so grant. Therefore, Mike’s counter to EAAN is not cogent.

    June 23, 2008 — 19:23
  • Our naturalist N takes (*) to be a defeater deflector. But I claim it is cheating. If EAAN were sound in W then N could not use (*) to deflect potential defeat since (*) would be also defeated. That is, N could not be reasonably persuaded of (*). So, in w N deflects EAAN with (*) and also defeats belief in (*).
    But (*) is not something he believes in w. It is something he believes in @, where he is reasoning about w (the hypothetical world in which EAAN is sound). He has no defeater for (*) in @.

    June 24, 2008 — 14:36
  • Paul

    Mike, here’s what you said:

    Suppose for reductio that there is a world w in which I believe N&E are true and I am reasonably persuaded by EAAN. If am reasonably persuaded by EAAN, then I believe that I have reasonably reached the conclusion that EAAN is sound. Since in w I believe that N&E are true, and that I have reasonably reached the conclusion that EAAN is sound, I believe in w that I have the cognitive resources that allow me to reliably reach the conclusion that I do not have reliable cognitive resources. To reach the conclusion that EAAN is sound, I would have to believe a contradiction. Therefore, naturalists don’t reasonably reach the conclusion that EAAN is sound in worlds where it is sound.

    You clearly indicate (to me) that the bolded portion (at least) is something you believe in w. This allows defeat, in w, to be deflected – ’cause in w you’d realize you were being asked to believe a contradiction, and to so ask is an automatic defeater-deflector. So it appears to me that you have claimed that in w (where you believe EAAN sound) you are forced to believe A and ~A, and so in w you have no reason to believe that which forces you to believe A and ~A.
    My response has been, but all that is defeated too.
    If he doesn’t believe that he can’t reliably conceed EAAN without admitting his faculties unreliable (since he still wants to hold to N&E), then what’s the problem?
    I’d also point out that your argument could also be given against something like Reppert’s AFR, Lewis’ arguments, Haskers, &c.

    June 25, 2008 — 9:38
  • Yes, complicated. I am reasoning here in @ about what I would be committed to in worlds w where, by hypothesis, I believe N&E are true and I am persuaded by EAAN. I am saying that someone who believed both of those things would be committed to a contradiction. So, no one could believe both. Or rather, no rational person could believe both. I don’t think I need to say that I would be able to trace out the contradiction in w. I trace out the contradiction here.

    June 25, 2008 — 14:06
  • Paul

    Mike,
    You’re viewing it from the *outside*. But EAAN is applied to *those people* in the specific epistemic situation EAAN puts them in. *Those people* have a defeater for *all* their beliefs. And, yes, no rational person could hold both, that’s the point of EAAN. If E&N together give you EAAN, then one needs to be dropped (I actually argue that in all contexts N needs to be dropped and in *most* E needs to be dropped, so in most situations E&N need to be dropped, but that argument takes us beyond the scope of our comments). So, if EAAN is *believed* in @, then this defeater-deflector won’t work because *it* is *itself* defeated in @ or any world w where EAAN is believed.
    If the Mike in w believed EAAN, the Mike in w would have a defeater for all his beliefs. The Mike in @ doesn’t believe EAAN the case. The Mike in @ must admit that if the Mike in w believed EAAN then the Mike in w has a defeater for all his beliefs. By parity, then, the Mike in @ should believe that if the Mike in @ believes EAAN, then the Mike in @ has a defeater for all *his* beliefs just like the mike in w did.

    June 26, 2008 — 10:47
  • You’re viewing it from the *outside*. But EAAN is applied to *those people* in the specific epistemic situation EAAN puts them in. *Those people* have a defeater for *all* their beliefs. And, yes, no rational person could hold both, that’s the point of EAAN.
    Paul,
    We might have accidentally agreed about something. Let’s try to be more careful…
    What I’ve been doing is reasoning “from the outside”, if you like, as a naturalist would. The naturalist does not know whether EAAN is sound or not. But he does know that in worlds where it is sound, he cannot be persuaded by EAAN without first abandoning his naturalism. It is not that EAAN persuades him to abandon his naturalism. It is rather that he must abandon his naturalism before EAAN can be persuasive at all.
    I guess you’d agree that the point of EAAN is, inter alia, to persuade the naturalist of the bad epistemic implications of his naturalism. But EAAN cannot persuade a naturalist of anything, for the reasons I’ve given. EAAN is persuasive only to those who have already abandoned naturalism.

    June 27, 2008 — 10:41
  • Paul

    Mike, this will constitute my last reply. Thanks for taking the time to discuss this with me.
    No argument is persuasive to anyone who does not want to give up their position. Anyone can say, “I believe P no matter what. Your argument is that ~P. So if I believed your argument too I would hold P and ~P. Since P is true, I’d have to give it up to accept ~P. But I won’t do that since P is true.”
    EAAN asks the question what the probability of R would be on N&E.
    That’s it.
    I see no reason why a naturalist can’t answer that question. They don’t have a problem asking what the probability of a life-sustaining universe is on naturalism, do they? It sure seems to me that R would be low given naturalism & evolution.
    Indeed, many naturalists have no problem answering that question. Some say it is high. Some say that it is low but that it does not matter.
    The argument seeks to show that if it is low then it does matter.
    At any rate, I showed:
    i) Your argument, if cogent, provides theists committed to EAAN with a defeater-deflector for any atheological argument whatever. Either you accept this or you do not. If you do not, then you can’t accept your argument against EAAN. If you do, hey, thanks for the apologetic blank check! 🙂
    ii) Your argument applies to AFR and other arguments in this vein.
    iii) Our naturalist can surely admit that R is defeated for someone other than him who arose via N&E. He then sees that he has a defeater too since he is in the same position as the person he admits R is defeated for.
    iv) “But he does know that in worlds where it is sound, he cannot be persuaded by EAAN without first abandoning his naturalism.”
    Soundness doesn’t even matter. So long as he believes R has a low probability on N&E, that’s enough.
    The only way for your naturalist to avoid EAAN is to stick his fingers in his ears, wag his tongue back and forth, and declare “Nee ner nee ner nee ner” at the top of his lungs. The argument isn’t meant for those kinds of people.
    The minute he avoids the “playground rebuttal” and actually reflects on the argument, he has an alethic rationality defeater.
    This is like Hume’s problem. Hume believed that we were irrational in believing our inductive inferences. He nevertheless agreed that to think inductively was a “habit of the mind.” Thus when S sees a truck coming at her, S does not doubt her inductive inferences; she jumps out of the way. But, later that night when S reflects upon her belief in inductive inferences, she finds that she has no reason to believe them, and thus S has a defeater for her inductive inferences.
    v) I still think what I said above was cogent:
    If the Mike in w believed EAAN, the Mike in w would have a defeater for all his beliefs. The Mike in @ doesn’t believe EAAN the case. The Mike in @ must admit that if the Mike in w believed EAAN then the Mike in w has a defeater for all his beliefs. By parity, then, the Mike in @ should believe that if the Mike in @ believes EAAN, then the Mike in @ has a defeater for all *his* beliefs just like the mike in w did.
    Best,
    Paul

    June 27, 2008 — 12:47
  • Pau

    I beg you let me add one more point:
    Here’s your argument, Mike:

    Let’s try to save some ink. Modify (2) in the way suggested (above) with the naturalism part fussed a bit (henceforth, assume I mean by ‘naturalism’ what you have in mind by it),
    2′. If EAAN is unsound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    3′. If EAAN is sound, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.

    I claim that “sound” and “unsound” is improper.
    A defeater can be sound or unsound. One can reason poorly, have false premises, and invalid structure, and still obtain defeaters. I’ll assume I don’t need to give examples.
    So, I suggest we remove the “sound/unsound” terms in your argument and replace it, in fact, let’s make a couple other adjustments too:
    2”. If the probability of R on N&E is not believed low, then EAAN (itself) does not give me [the naturalist as described in D] reason to believe its conclusion.
    3”. If the probability of R on N&E is believed low, then EAAN (itself) does give me [the naturalist as described in D] reason to believe its conclusion.
    These changes are necessary for many reasons. Surely Mike would admit that even if EAAN were unsound and a naturalist N happened to believe that R was low given N&E, then N has a defeater regardless of whether EAAN is sound or not. I also replaced “EAAN” with another term that makes the case of defeat in 3” explicit.
    Okay, now I’m out.
    Regards,
    Paul

    June 27, 2008 — 13:15
  • No argument is persuasive to anyone who does not want to give up their position. Anyone can say, “I believe P no matter what. Your argument is that ~P. So if I believed your argument too I would hold P and ~P. Since P is true, I’d have to give it up to accept ~P. But I won’t do that since P is true.
    This really misconstrues my argument entirely. You can’t ask a naturalist to give up his position BEFORE he hears the alleged problem for it from EAAN. My argument shows that, if he does not give up his naturalism BEFORE reviewing EAAN, he cannot be persuaded by EAAN. It has nothing to do with stubbornly holding on to naturalism.
    On the other hand, if I’ve ever been certain of anything, it’s that this point is not getting through. So it likely makes little sense for us to pursue it further.
    Thanks for the discussion.

    June 27, 2008 — 16:01
  • A defeater can be sound or unsound. One can reason poorly, have false premises, and invalid structure, and still obtain defeaters. I’ll assume I don’t need to give examples.
    This point is just irrelevant to my argument, as far as I can see. I’m talking about (and have been talking about) the argument that Plantinga offers against the naturalist, which purports to be sound. I exhaust the possibilities: either it is sound or not. In neither case does it persuade the naturalist. I have no idea what it means to call a defeater sound or not. You are of course free to develop your own argument where you talk about defeaters being sound and unsound or whatnot. But let’s not confuse that with the argument I have offered.

    June 29, 2008 — 14:44