Another Counterexample to Plantinga's Proper Functionalism

This one is from leading epistemologist Richard Feldman (“Proper Functionalism,” Nous 27 (1993), pp. 34-50). His is a modification of one of Plantinga's own counterexamples to the sufficiency of pure reliabilist accounts of knowledge.[1] Plantinga's counterexample to reliabilism goes as follows:

"Suppose I am struck by a burst of cosmic rays, resulting in the following unfortunate malfunction. Whenever I hear the word 'prime' in any context, I form a belief, with respect to a randomly chosen natural number less than 100,000, that it is not prime. So you say "Pacific Palisades is prime residential area" or "Prime ribs is my favorite"...; I form a belief, with respect to a randomly selected natural number between 1 and 100,000 that it is not prime. The process or mechanism in question is indeed reliable (given the vast preponderance of non-primes...) but my belief -- that, say, 41 is not prime -- has little or no positive epistemic status. The problem isn't simply that the belief is false; the same goes for my (true) belief that 631 is not prime, if it is formed in this fashion. So reliable belief formation is not sufficient for positive epistemic status." (Warrant: The Current Debate (OUP, 1993), p. 210.)

Feldman then points out that one need merely change the counterexample so that the cause of the beliefs about primes is not a burst of cosmic rays, but rather a cognitive faculty formed by an intelligent designer who designs the person to naturally and spontaneously form such beliefs about prime numbers whenever they hear the word 'prime'. In such a case, all of Plantinga's conditions of warrant are satisfied: according to the thought experiment, we have a cognitive faculty that, when functioning properly, reliably produces sufficiently firmly-held true beliefs when in epistemically congenial environments. However, such beliefs have little by way of warrant. Therefore, the conditions laid out in Plantinga's proper functionalism are not sufficient for warrant.
-------------------------------
[1] Very roughly, a pure reliabilist account states that a true belief counts as knowledge if, and only if, it is produced by a reliable belief-producing process or mechanism. So, for example, consider my belief that I see a tree. That belief was formed by my visual perceptual system. If that system or mechanism or process is sufficiently reliable (i.e., most of the perceptual beliefs it produces are true in the appropriate actual and counterfactual circumstances), then (according to pure reliabilism), that belief counts as knowledge; if it's not reliable, then it doesn't count as knowledge.

23 comments:

Mike Almeida said...

I'm having a hard time seeing this as a counterexample. The designer knows that your chances of thinking of a non-prime are extremely high if you arbitrarily think of a number in that range. So he designs you in such a way that the number arbitrarily selected from that range occurs to you. This is part of a design to have you form true beliefs about these numbers. How do you lack warrant?

Look, God designs me in such a way that I form the belief 'that's a live tree' when in the presence of live trees and dead trees. Given the preponderance of live trees, the mechanism, though fallible, provides warrant to my belief that that's a tree.

exapologist said...

Hi Mike,

I'm afraid that strikes me as pretty hard bullet-biting. If my intuitions line up with one of the greatest living epistemologists, that's fine with me!

In any case there are clearly deep disanalogies between the tree case and the non-primes case, as there is a deep affinity between tree-experiences and trees; not so between (say) hearing, "I love prime ribs" and forming a belief about some random number below 1,000 that it is prime.

I'm guessing the reliability of the (c'mon: utterly bizarre) non-primescase is what's driving your intuitions about the belief having warrant. But if reliability is doing all the work, then the "proper function" clause isn't doing any substantive work.

Plantinga says the motivation for the proper function condition is to screen off the weird cases (e.g. his non-primes case) as counting as warranted -- a big problem, he thinks, for straight reliabilist theories. But Feldman pointed out that such weird cases aren't screened out by adding the proper function requirement.

Mike Almeida said...

I'm guessing the reliability of the (c'mon: utterly bizarre) non-primescase is what's driving your intuitions about the belief having warrant.

No, it's not the reliability. It's the fact that I'm designed in the way Feldman assumes. I actually don't see the disanalogy you're alluding to. Is it that I've noticed some correlation between tree observations and the presence of trees? But haven't done so in the case of prime numbers? If so, then stipulate that I have noticed the correlation between utterances of 'prime' and true beliefs about prime numbers. Take seriously the idea that I was so designed, odd as the design is. Tell me why the belief lacks warrant? I honestly couldn't tell you why I don't have warrant assuming God has made me this way.

I've incidentally posted in the comments over at Prosblogion, if you'd like to reply there. Maybe other epistemologists would chime in.

exapologist said...

Hi Mike,

No, it's not mere correlation. Rather, it's that the phenomenology matches the referent in a way that's lacking in the "prime ribs"/non-primes case. Of course, it's notoriously difficulty to pin down this "matching" relation (as Wittgenstein and many others have pointed out), but that doesn't change the fact that we know it when we see. it.

Thanks for calling my attention to your reply at Prosblogion. As it happens, I commented there right after I posted here, and have since commented again. The comments haven't showed up yet, for some reason. I'm chalking it up to people being busy with Father's Day festivities -- like me. ;-)

Mike Almeida said...

Rather, it's that the phenomenology matches the referent in a way that's lacking in the "prime ribs"/non-primes case.

How could that matter? There is no principled reason to think that one particular phenomenology is crtical epistemologically. It's perfectly possible that God created me so that I detect trees auditorily. He didn't do that, but he might have. If my hearing were sufficiently acute, I could hear the sap moving. Or, he could have created me in other ways that would be epistemologically significant. He might have created me in a way that, when in the presence of trees, my feet swell up. All of these are possible. As t happens, they are not actual. But here's what is actual: God created me in a way that (by hypothesis) I form a true belief about non-prime numbers when someone mentions 'prime'. Perfectly possible, I concede. And there is no reason to deny warrant on the basis of the fact that the hypothesis is unusual. We are assuming that it is actual, and so not unusual.

exapologist said...

Hi Mike,

I'm a bit taken aback by your remarks, as they sound -- contrary to fact -- that the sort of view I'm advancing does not have a long and distinguished philosophical pedigree, and is still put forth by many philosophers -- christian and non-Christian alike. Indeed, it's built into, e.g., Alston's theory of appearing.

Also, I'm afraid I still find your endorsement of the "primes" case as baffling. It's just utterly bizarre to me -- beyond the pale, even. But let me see if I can push the intuition a bit. Suppose some scientists "truetemp" you by putting a device in your brain that relays reliable information about the position of one of the moons of Jupiter. And suppose they design the faculty's triggering-conditions so that you form a firm and unwavering, reliable belief about the position of that moon whenever someone says "Jupiter" (or, say, "Elvis Presley"). Then all the conditions of Plantinga's account are met, and yet the belief has little by way of warrant.

exapologist said...

Also, once you add the observed correlation requirement, there are problems:

(i) epistemic circularity issues loom.

(ii) You've granted the inadequacy of Plantinga's analysis of warrant as he has developed it, which is the point I'm bringing up with this post.

Mike Almeida said...

I'm a bit taken aback by your remarks, as they sound -- contrary to fact -- that the sort of view I'm advancing does not have a long and distinguished philosophical pedigree

I'm really just asking some questions. I'm not aiming to offend against tradition or take you aback or doubt recieved epistemic truth. I was simply pointing out what looks true to me, viz., that in worlds where God designs me so that I form true beliefs in ways that seem odd to us here, I see no reason to deny that, there in that world, I have warrant for my beliefs. When I'm asked to entertain a hypothesis about a mechanism for the formation of true beliefs that God causes me to have, I'm asked ot entertain such a world. If I were to discover that this world is indeed that hypothetical world, would I say that my beliefs about non-primes were warranted? It seems like it (God whispers to me, "you are designed to detect truths about non-primes in this way, I have my reasons"). Not knowing that this world is such a world, do I think they are warranted? Seems not. But those are just my intuitions, so don't take them too seriously.

exapologist said...

Hi Mike,

Ok, but let's go back to my question about becoming aware of a correlation between the phenomenology (or a phenomenology-free belief, if that even makes sense. Plantinga of course thinks it does, but it's not clear he's right about that) and the referent: do you think it's required or not? So, for example, consider my "Jupiter" case: Suppose you were designed to form true beliefs about the position of one of Jupiter's moons whenever you heard the word "Jupiter" (or, say, "Elvis Presley"). Suppose further that you were unusual in this regard. Finally, suppose you had no idea whether such beliefs are reliable (though they in fact are). What do your intuitions tell you? Warranted or no?

Unknown said...

Hi exapologist,

Just wandered over from Proslogion for this interesting subject. I should look at the Feldman piece myself... but here I am, lazy and counting on you to save me the time searching the phil index.

It's been awhile since I've looked carefully at the Plantinga proper function story, but isn't there a clause of some sort about the properly functioning cognitive mechanism being "aimed at truth"? Might we wonder if the design plan of Feldman's designer meets this condition?

--Dan Speak

exapologist said...

Hi Dan,

Yep, you're remembering right (and if this the Dan I think it is, you probably remember first-hand from Plantinga himself ;-)). So the cognitive faculty at issue has to be aimed at producing true beliefs (as opposed to, say, comfort, or survival, or...).

Now as you know, Plantinga realizes that there is a distinction between proper function and good function. An incompetent designer might make your faculties so that they are "aimed" at producing many more true beliefs than false beliefs, and yet fail miserably. So to handle this sort of problem, he talks of cognitive faculties properly functioning according to a design plan successfully aimed at truth.

Well, how does he spell out the notion of success? Answer: high objective probability that the beliefs so produced are true. In other words, we have something like Goldmanian reliabilism + a proper function requirement to rule out cases of accidental reliability.

Now go back to Feldman's variation on Plantinga's "prime" case: we have a design plan that's aimed at truth (as opposed to comfort, say). Furthermore, the objective probability of forming true beliefs via this faculty is high (given the predominance of non-primes between 1 and 1,000). Still, wouldn't we say that such beliefs have little by way of warrant?

Mike Almeida said...

Suppose you were designed to form true beliefs about the position of one of Jupiter's moons whenever you heard the word "Jupiter" (or, say, "Elvis Presley"). Suppose further that you were unusual in this regard. Finally, suppose you had no idea whether such beliefs are reliable (though they in fact are). What do your intuitions tell you? Warranted or no?


I'll begin by saying that I agree with Plantinga that I might be designed in such a way that beliefs are evoked in me in unexpected ways. I may come to believe it God or grow spiritually wiser by a phenomenological experience that appears to have nothng to do with God. I guess that's not consistent with the phenomenological claim.

It is difficult to answer your questions about being designed in suhc a way that I have true beliefs about Jupiter. In the case you describe, the designers are not God. So I would say that, if anything, they are interfering with what I'm designed for epistemologically. God designs me.

Concerning the uniqueness claim, it would depend on whether God's designs admit of idiosyncratic design. Plantinga doesn't think so, at least on what I've seen. But maybe this is a contingent fact about design. Maybe there are worlds in which God does idiosyncratically design agents. So, suppose it is possible that the holy spirit gives me this gift of knowing that others don't have.

Suppose all of that is true. Then given this view about proper function--one that allows for individuals to have specialized epistemic abilities--then I think I would have warrant. This is of course different from claiming that I would come to believe that I have warrant in such cases. I probably wouldn't.

Here's why. Suppose I die, and the holy spirit says to me. "You know, I gave you this special ability to know the location of Jupiter on occasions when you heard 'Jupiter'. I might reply: "I didn't really notice" or "I noticed, but didn't think much of it". Here's what I wouldn't say: "You gave me this special ability, but I still didn;t have warrant for the beliefs I formed this way". I wouldn't say it, because I would then have learned that it wasn't true that I did not have warrant.

Unknown said...

Thanks ex. That helps a great deal... and it is bringing back some memories of the proper function story. I'll say, though, that I don't remember getting any (or much) of this stuff from Al himself while I was visiting ND. I just remember him darting past me up the the stairs as if our ages were reversed. What a freak. I'd punch him if I didn't think he'd kick my ass in response.

Further confession: it looks to me like some chisholming is going to allow the proper functionalist to deny that the intelligent designer has installed a mechanism that is successfully aimed at truth. But your (and feldman's) point is absolutely well taken given Plantinga's appeal to objective probabilities.

I'm wondering now (and this is somewhat unfair, so feel free to ignore my wonderment here) if you have reason to believe that the proper functionalist story reflects a flaw or flaws that can't be patched up with tweaks. Is there a problem that suggests to you that the strategy is just on the wrong track?

exapologist said...

Hi Mike,

Given your intuitions, I guess there's not much else I can say. Well, except this: I'd be happy if Plantinga would slap a sticker on all copies of Warranted Christian Belief stating something like your reply, and that accepting Plantinga's account of warranted Christian belief might well commit one to it.

Mike Almeida said...

Given your intuitions, I guess there's not much else I can say.

I'll try harder not to merit ad hominems. How about this way to proceed? I say that in the following case I would have learned that I was warranted.

Suppose I die, and the holy spirit says to me. "You know, I gave you this special ability to know the location of Jupiter on occasions when you heard 'Jupiter'. I might reply: "I didn't really notice" or "I noticed, but didn't think much of it". Here's what I wouldn't say: "You gave me this special ability, but I still didn;t have warrant for the beliefs I formed this way". I wouldn't say it, because I would then have learned that it wasn't true that I did not have warrant.

Consider a world (maybe just an epistemically possible world) in which this happens to you. In that world, you die and God informs you of this ability that he placed in you without you knowing it. You would then learn that (unlike the Feldman case) it is not coincidental that when you hear 'Jupiter' you correctly believe that Jupiter is at spatial location L. Would your response be that, despite the fact that it is not coincidental, and despite the fact that it is highly accurate, and despite the fact that you were deliberately designed to detect the location of Jupiter in just this way, you still had no warrant? My guess is that you would not say that. But if you did say that, I wonder what you would add to make the belief warranted.

exapologist said...

I'm sorry if my tone came off as harsh, Mike! I certainly respect you, and of course your smarts and philosophical abilities are off the charts. And as I've said on several occasions, I think you're one of the best, fairest, and most creative philosophers of religion alive. Just take what I'm saying as giving the point I'm making a fighting chance to see how it fares.

Yes, like leading epistemologists of the likes of Bonjour, Chisholm, Lehrer, Feldman, Conee, Huemer, et al., I would say that such beliefs about the position of one of Jupiter's moons is not warranted. The agent has no inner phenomenology or appearances. They have no corroboration that such a belief-forming mechanism or process is reliable. They're in the situation of (at least one version of) Bonjour's clairvoyant. And beyond that, most people don't share the agent's ability. None of this makes a dent, though, in the degree of warrant for your belief in the position of one of Jupiter's moons? Not even Plantinga accepts that, and that's in print in his reply to Lehrer's variation of his Truetemp case against Plantinga's analysis of warrant. Plantinga thinks it's enough to defeat the warrant-basic status of such a belief by realizing that you're like other people, and most other people lack such an ability. Indeed, he even thinks that the fact that others smile condescendingly or sneer at the agent (for thinking they have such an ability) would deflate the belief's warrant. So if Plantinga won't side with you on this, perhaps it comes as no surprise that I won't.

exapologist said...

Hey Dan, (We know each other, by the way!)

Nah, I'm inclined to think Plantinga's account of warrant is going in the wrong direction at a fairly fundamental level. It seems to me to be an account of a Sosa-style version of "animal knowledge", but without even the internal phenomenology of seemings (Plantinga of course thinks there are often no seemings to be had for very many basic beliefs -- unless one just reduces seemings to beliefs, or something like the felt conviction that a belief seems apt, or something in that neighborhood. I think Huemer and others make a good case that seemings/appearances are distinct from believings and convictions, though).

Also, I think there are decent counter-examples against the necessity of an actual, intended design plan for warrant. James Taylor and (I believe Feldman and Sosa) give some good cases here. Here's one I'm toying with (though it's admittedly rough at the moment): Suppose there is a multiverse with infinitely many universes, each one with different values assigned to the fundamental constants. As one might expect, there is at least one universe within it roughly qualitatively identical to ours, including creatures like us with cognitive faculties like ours. However, unlike us, their cognitive faculties arose by unguided evolutionary processes. The only difference is that divine intentions lie behind our faculties, while theirs don't. Now it seems to me that their beliefs have warrant just in case ours do. But surely ours do; therefore, so do theirs. So, proper function according to an actual, intended design plan isn't necessary for warrant.

Third, as DeRose, Chignell, and (most recently) Beilby have argued, Plantinga's account of warrant requires a very high degree of firmness of belief for warrant. But if so, then since very many Christians have something along the lines of weak and wavering belief ("I do believe; help thou mine unbelief"), then if Plantinga's analysis of warranted Christian belief is correct, then the belief of many Christians will have little by way of warrant (an ironic result!). I think this is a deep problem that goes way back to Plantinga's pre-warrant phase, and was ably brought out in the Plantinga-Quinn exchanges.

Finally, I think Kvanvig is probably right that Plantinga's account is a piece of patchwork with deep internal tensions. See his remarks on this here.

Best,
EA

Mike Almeida said...

(1) The agent has no inner phenomenology or appearances. (2) They have no corroboration that such a belief-forming mechanism or process is reliable. (3) They're in the situation of (at least one version of) Bonjour's clairvoyant. And beyond that, (4) most people don't share the agent's ability. None of this makes a dent, though, in the degree of warrant for your belief in the position of one of Jupiter's moons?


Exap, it might be inevitable, but the case you describe here is not the case I describe. I don't take (1) to be an intrinsic feature of cases where we have warrant. (1) matters when the correlation between hearing 'Jupiter' and believing something about Jupiter is coincidental (unlike the correlation between seeing a tree and believing there is a tree present). The cosmic ray case makes this correlation coincidental.

But it is really hard to see how (1) matters if God is deliberately designing us to detect the location of Jupiter in this way. Why would God have to observe such a restriction on epistemic design? In this case, the correlation between hearing 'Jupiter' and forming the belief is not at all coincidental. In this case, unless I'm missing something important, (1) seems to lose its epistemic force.

Whether (2) or (3) is true depends on whether you believe that you are designed by God. Plantinga talks about this extensively, as you know. If you do believe that, then you do have reason to believe that your belief-forming mechanisms are in general reliable. Justification doesn't depend on such non-epistemic assumptions, on Plantinga's view, but warrant does.

(4) has important implications for Feldman's counterexample. Feldman assumes that S has this special epsitemic ability and this ability is not widely shared. I agree that, so far forth, Plantinga cannot consistently call this a case of warrant. But there are two things he should say. The first is that the case Feldman advances is not possibe, since God does not create us with special epistemic abilities in the way Feldman describes. Second, if we assume that God does create us with special epistemic abilities, then (4) presents no problem. What Feldman can't do is build into the counterexample the fact that God creates beings with epistemic abilites that are not widely shared and call it a counterexample to Plantinga. Plantinga denies that we are (or might be) created in that way. And if Plantinga were to concede that God might do that, then he would be changing his mind about (4). It would not present a problem.

I appreciate the clear discussion of this issue, exap. I suspect that you can't take seriously the cenrtal role in epistemology that Plantinga gives to God and God's design. If you see epistemology as an autonomous enterprise, independent of theology, then none of this is going to sound even moderately plausible.

exapologist said...

Hi Mike,

I guess I'm still not convinced (along with virtually all epistemologists, including Christian epistemologists). I'll start with this:

I don't take (1) to be an intrinsic feature of cases where we have warrant. (1) matters when the correlation between hearing 'Jupiter' and believing something about Jupiter is coincidental (unlike the correlation between seeing a tree and believing there is a tree present). The cosmic ray case makes this correlation coincidental.

I'm not following this bit. You just said that experience or appearances or phenomenology matters when the connection between belief and referent is coincidental, correct? Now objective reliability is certainly an important kind of non-coincidental connection here, but so is evidence of reliability that is available to the subject. And when it comes to non-inferential beliefs about the contingent world, this amounts to experiences. But unfortunately, this is precisely what is lacking in the Jupiter case. How is this different from Bonjour's clairvoyance case? In that case, the agent's beliefs have a reliable, non-coincidental connection to the referents, and yet, intuitively, the agent's true beliefs are not warranted. However, if we stipulate that the clairvoyant could "see" the events she has beliefs about, or if she could get evidence of reliability, then our intuitions tell us she has knowledge.

Mike Almeida said...

ex,

The Bonjour case involves someone who is a reliable clairvoyent, right, but not designed to be such. It is just chance that Norman's clairvoyent perceptions are accurate ones, whereas my perceptions of hte future are not. So, he has no reason to believe he would be reliable in this respect; indeed, if he thought about it, he'd have reason to believe he is not.
This is not the sort of case I'm describing. In my case, you do have reason to believe that your special epistemic abilities are reliable ones. It is not chance that they are reliable.

The argument I offered was that Feldman's counterexample describes a case that is either impossible (from Plantnga's point of view) or untroubling. It's impossible if God does not bestow specialized epistemic abilities. It is untroubling (for Plantinga) if God does bestow such abilities. Take the latter case. I find myself with a remarkable ability to predict future events. If God exists, that ability is reliable and it's reliability is non-chancy. I was designed that way; it doesn't just happen that I'm reliable. So this is not the Bonjour case where I just happen to be reliable. So, if theism is true, then I have warrant for my beliefs abut the future. Of course, I may not know that I have warrant for those beliefs. But that's a different issue. That gets us to worries for internalists that I don't have.

exapologist said...

Hi Mike,

I find myself with a remarkable ability to predict future events. If God exists, that ability is reliable and it's reliability is non-chancy. I was designed that way; it doesn't just happen that I'm reliable. So this is not the Bonjour case where I just happen to be reliable. So, if theism is true, then I have warrant for my beliefs abut the future. Of course, I may not know that I have warrant for those beliefs. But that's a different issue.

This isn't doing anything to change my fundamental intuitions: no appearances or experiences = not properly basic.

In any case, I guess I'm unclear on this "just chance"/"not just chance"distinction. In what sense is the reliability of Norman's clairvoyance chancy? Suppose it's metaphysically necessary that anyone in a world with laws of nature like ours and a Clairvoyance mechanism like Norman's has reliable clairvoyant beliefs. How would that be inferior to the kind of reliability in the God case?

And how reliable is the mechanism in the God case? Suppose there are possible worlds at which God makes the mechanism at least fairly unreliable in order to achieve some greater good. That doesn't seem implausible to me. If so, then the reliability isn't metaphysically necessary even in the God case.

The argument I offered was that Feldman's counterexample describes a case that is either impossible (from Plantnga's point of view) or untroubling. It's impossible if God does not bestow specialized epistemic abilities.

It's not clear to me that appeal to God is a part of Plantinga's analysis of warrant. I thought it was (roughly) that a belief is warranted if and only if it's formed by a properly functioning, (successfully) truth-aimed cognitive faculty that is operating in an epistemically congenial (mini- and maxi-) environment (where there degree of warrant then corresponds to the degree of firmness or conviction with which it is held). In any case, are we now conceding that Plantinga's account is inadequate as stated, and thus requires an extra "no specialized epistemic abilities" clause? (at any rate, might not the sensus divinitatis and the internal instigation of the Holy Spirit be abilities or processes or faculties of just this sort?).

(Even waiving all this, I'm not sure about the "It's impossible if God does not bestow such faculties" move. I tend to think Kit Fine et al. are right that essence claims and analyses aren't properly spelled out in terms of strict biconditionals; I think they're too coarse grained. Now of course it's notoriously hard to spell out the semantics, but I think (e.g.) Tony Roy's "Making Sense of Relevant Semantics" is on the right track (ms. available at his department website. I'm not sure if it's published yet. In any case, there are several other accounts.). On such an account of the semantics, appeals to impossibility won't help, I don't think. I have to think about that a bit more, though.)

Mike Almeida said...

I guess I'm unclear on this "just chance"/"not just chance"distinction. In what sense is the reliability of Norman's clairvoyance chancy?

It is chance that Norman is in possession of such a mechanism, and that I am not. There is no reason to believe that there are such mechanisms, that he would be in possession of such a mechanism or that anyone would. But if God is designing me with such a mechanism, then there is excellent reason to believe that there are such mechanisms and excellent reason to believe that I'd be in possession of one.


Suppose it's metaphysically necessary that anyone in a world with laws of nature like ours and a Clairvoyance mechanism like Norman's has reliable clairvoyant beliefs.

It is chance that he has M or that anyone does. There's no reason to believe that there would be such a mechanism. I don't mean that it is chance that it works well. I'm granting it's reliability, once you've got it.

How would that be inferior to the kind of reliability in the God case?

Norman is not designed with M, but just happens to find himself with it. I am designed in such a way that I am functioning properly when I know things about the future. Nothing chancy about it: it's not chancy that there are such mechanisms, and not chancy that I have one.

It's not clear to me that appeal to God is a part of Plantinga's analysis of warrant.

Surely my functioning properly depends on my doing what I was designed to do. Plantinga, if I recall right, doesn't think the naturalist can appeal to proper function. So God has to be in the picture.

In any case, are we now conceding that Plantinga's account is inadequate as stated, and thus requires an extra "no specialized epistemic abilities" clause?

Actually, it is Feldman that needs the clause. Suppose we deny it. In that case Feldman's example does not describe something possible on Plantinga's view. I'm happy with that.

This isn't doing anything to change my fundamental intuitions: no appearances or experiences = not properly basic.

I'm not sure I follow this. How different from the claim that belief in God is properly basic is the claim that belief that some number is non-prime is properly basic? Both beliefs are evoked by odd experiences. I come to believe God exists by such unrelated observations like viewing the grand canyon.

exapologist said...

Hi Mike,

Re: whether Plantinga thinks proper function requires design and a supernatural designer: I'm not convinced by Plantinga's reasons for this, and for roughly the reasons Swinburne, Wunder, and Bardon offer: it's not at all clear that Plantinga has shown that proper function requires intelligent design (Swinburne, Wunder, Bardon), and even if it did, it appears that the artifactual notions of proper function and related, inter definable terms are at best analogically applicable to organic beings and their parts (where the latter are best referred to by means of the organic notions of health, sickness and related, inter-definable terms) (Wunder).

In any case, if warrant requires proper function, and proper function requires intelligent design, then one might worry that self-referential incoherence looms for Plantinga's account. It should be noted that Plantinga has anticipated the latter problem, and has attempted to address it (in a footnote(!) in Warrant and Proper Function, if I recall correctly). But at the very least, it's not clear his reply is successful.

Re: Feldman's counter-example: I mentioned this parenthetically in a previous comment, but even if we waive problems with motivating appeal to God and design in Plantinga's account, I don't think the appeal to impossibility ("God wouldn't allow cases envisioned by Feldman to be actualized") wouldn't save Plantinga's account. For I think possible-worlds accounts of analysis and of essence are too coarse-grained. I agree with Daniel Nolan, Berit Brogaard, Joe Salerno and others that we need to go hyper-intensional here.

In any case, thanks for your stimulating comments!

Best,
EA

Ricki Bliss's Cambridge Element on Grounding, Fundamentality, and Ultimate Explanations

 ...is now out , and available for free download for a limited time. Required reading.