Sunday, February 21, 2010

Lukeprog's Excellent Podcast

Hi gang,

If you didn't know already: lukeprog over at Common Sense Atheism has an excellent podcast called Conversations from the Pale Blue Dot, on which he regularly has discussions with excellent philosophers of religion (e.g., Graham Oppy, Wes Morriston, Richard Otte, Nick Trakakis). You can subscribe to his podcast on iTunes. Required listening!

Thursday, February 18, 2010

Intrinsic Defeaters and the Plantinga-Quinn Debate

I've worked up some notes on a strand of the debate between Alvin Plantinga and Phillip Quinn in the 80s and 90s. My aim is to get clearer on their discussion of Plantinga's notion of an intrinsic defeater. As such, I focus on that strand of their exchanges, and not so much on their exchanges on modern and classical foundationalism.

Plantinga[1]: Belief in God is properly basic, i.e., it's rational or justified or warranted wholly apart from supporting arguments. For there is parity between universally accepted properly basic beliefs (e.g., perceptual beliefs and memory beliefs) on the one hand, and basic theistic beliefs on the other. Both sorts of beliefs are naturally and spontaneously "triggered" in a wide variety of circumstances. So, for example, one will naturally and spontaneously form the following beliefs under certain familiar circumstances:

-I see a tree. (upon having a perceptual experience as of a tree)
-I had breakfast this morning. (upon trying to recall what one had for breakfast)
-That person is angry. (upon looking at a particular person's face)

Similarly, one might naturally and spontaneously form the following theistic beliefs under certain familiar circumstances (at least they're familiar to certain sorts of theists):

-God is speaking to me. (upon reading the Bible)
-God has created all this. (upon looking at the starry heavens)
-God disapproves of what I have done. (upon doing something cheap, or wrong, or wicked)
-God forgives me. (upon asking God for forgiveness for the cheap, wrong, or wicked thing)
-God is to be thanked and praised. (when life is sweet and satisfying)

Those who deny that theistic beliefs can be properly basic are adherents of Clifford-style evidentialism, which in turn is grounded in either classical or modern foundationalism. But while such versions of foundationalism entail that belief in God cannot be properly basic, they are self-referentially incoherent: the classical and modern foundationalist theses themselves fail their own criteria of proper basicality; nor can they be properly deductively or inductively inferred from the beliefs they countenance as properly basic. Therefore, Clifford-style evidentialism is unmotivated; as such, it can't serve as a principled basis for ruling out theistic beliefs as being properly basic.

Quinn[2]: Even if belief in God is properly basic for some people (provincial aunt Mabel and pre-critical little brother Timmy), it's not properly basic for contemporary, intellectually sophisticated adult theists. For any basic justification or warrant such beliefs may have enjoyed has been defeated by their knowledge of contemporary objections to/defeaters for theism (the problem of evil, naturalistic accounts of theistic belief, etc.). Thus, to retain justified or warranted belief in God, sophisticated contemporary adult theists require good replies to such criticisms -- defeater-defeaters for the original defeaters for theistic belief. But then their theistic belief is at least partly based on propositional evidence, in which case it is no longer properly basic.

Plantinga[3]: Not necessarily. For this assumes that all defeaters are extrinsic defeaters -- other arguments or evidence. But in addition to extrinsic defeaters, there are also intrinsic defeaters. "When a basic belief P has more by way of warrant than a potential defeater q of p, then p is an intrinsic defeater of q -- an intrinsic defeater-defeater, we might say."[4]

Two examples:
(i) The purloined letter case: Suppose I have means, motive, and opportunity to steal an embarrassing letter that was in fact stolen from the office of my department chair. There is also very strong evidence against me (e.g., I've been known to steal in the past; a trustworthy colleague says he saw someone who looks like me enter the Dean's Office on the day of the incident, etc.). However, I have a clear and vivid memory of being alone in the woods all day on the day of the incident. In this case, I'm rational to retain my belief that I didn't steal the letter because of my memory, even without propositional evidence and argument that could defeat the reasons brought against me; My belief that I was alone in the woods all day on the day of the incident is thus an intrinsic defeater-defeater of the claim that I stole the letter.

(ii) The Moses and the burning bush case: Suppose Moses has an overwhelming and vivid perceptual, auditory, and (no doubt) mystical experience as of Yahweh speaking to him from a bush that appears to burn, and yet is not consumed. Now suppose Freud strolled by and explained to Moses how his belief about Yahweh is the result of neurotic wish fulfillment. In this case, Moses is rational in retaining his belief about Yahweh, even if he were at a loss as to how to show Freud that his experience was veridical and not illusory. Moses' basic belief about Yahweh is thus an intrinsic defeater-defeater of Freud's criticism.

Thus, properly basic beliefs enjoy a degree of warrant that's independent of propositional evidence and argument. And if the degree of warrant of a properly basic belief is greater than that of one of its potential defeaters, then it is an intrinsic defeater of that defeater: it remains properly basic and warranted even in the absence of a propositional argument against that defeater.

Similarly, if belief in God is properly basic for a given theistic believer, and the degree of basic warrant their belief enjoys is greater than that of a given objection to their belief, then their theistic belief is an intrinsic defeater of that defeater; it retains its properly basic status even apart from an argument offered in reply to the criticism to their belief. And if this applies to the contemporary intellectually sophisticated adult theist as well, then they, too, can have properly basic theistic belief.

Quinn[5]: First: while it's certainly true that at least some types of properly basic beliefs have the property of being intrinsic defeater-defeaters for certain sorts of defeaters, it doesn't follow that such types of properly basic beliefs have this property essentially. Rather, properly basic beliefs must meet certain conditions if they are to be intrinsic defeater-defeaters. For example:

(a) Such beliefs must be sufficiently clear (e.g., fuzzy memories, perceptions in dim lighting, etc., have a lower degree of warrant, and thus can be defeated by good evidence).

(b) One must not have decent evidence that the source of such a belief is unreliable (e.g., you have good evidence from your doctor that you have a certain sort of memory disorder that affects the reliability of the sort of belief at issue).

Second: from the fact that

1. some basic beliefs are intrinsic defeater-defeaters.

it doesn't follow that

2. Some theistic beliefs are intrinsic defeater-defeaters.

or even

3. Possibly, some theistic beliefs are intrinsic defeater-defeaters.

(3) may well be acceptable, though, on the grounds that the scenario seems at least possible and, prima facie, it seems as though it would be an intrinsic defeater-defeater for Freud's objection if it were to obtain. Unfortunately, justifying (3) is insufficient in this context. For this only provides reason to think that there is at least one possible world at which belief in God is properly basic for an intellectually sophisticated theist. But what is metaphysically possible doesn't even entail what is epistemically possible (i.e., what could be true of the actual world, given what we already know of it), let alone what is actual.

What about (2)? It'll prove helpful to distinguish several interpretations of (2) before evaluating it (ordered from strongest to weakest):

(2a) For most theists, some thestic beliefs are intrinsic defeater-defeaters.

(2b) For some theists, some theistic beliefs are intrinsic defeater-defeaters.

(2c) It's epistemically possible (i.e., it could turn out, for all we know, that the actual world is such that) for most theists, some thestic beliefs are intrinsic defeater-defeaters.

(2d) It's epistemically possible (i.e., it could turn out, for all we know, that the actual world is such that) for some theists, some thestic beliefs are intrinsic defeater-defeaters.

Which construal of (2) is Plantinga arguing for in his exchange with Quinn? Of course, the one that's of most interest to the typical Christian theist is (2a). For they care (or at any rate, they ought to care) about whether their faith is at least minimally rational. So if they lack both arguments for their faith and adequate responses to criticisms against their faith (or even knowledge that someone, somewhere, has adequate arguments for their faith and adequate responses to such criticisms), Plantinga's arguments will be cold comfort for the vast majority of them if the only conclusion warranted by his arguments is something weaker than (2a). In fact, in the last paragraph of his reply to Quinn, he offers a qualified endorsement of (2a): "I am therefore inclined to believe that belief in God is properly basic for most theists - even intellectually sophisticated adult theists." (The Foundations of Theism, p. 312. Italics mine.)

However, as Quinn argues in his rejoinder, Plantinga offers nothing in their exchange that could adequately support (2b), let alone (2a). To see this, recall Plantinga's two paradigm cases of intrinsic defeaters:

(i) The purloined letter case
(ii) The Moses and the burning bush case

In case (i), we have a clear, fresh memory that serves as an intrinsic defeater-defeater of evidence that one stole a letter out of the department chair’s office. Such a belief has a lot by way of psychological force, vivacity, and warrant. In case (ii) Moses has an overwhelming experience as of God speaking to him from a bush that appears to burn, and is yet not consumed by the flame. It, likewise, has a lot by way of psychological force, vivacity, and warrant. Thus, if you were to be the subject in either scenario, and you met conditions (a) and (b) mentioned earlier (clear, fresh, vivid experiences in normal conditions, and you had no decent evidence that your cognitive faculties are unreliable with respect to forming the relevant sort of belief), then such beliefs have such a high degree of force, vivacity, and warrant for you that they will overwhelm the force of virtually any argument one could offer.

The same is true of the uncontroversially properly basic beliefs Plantinga mentioned in the earlier paper, viz.:

-I see a tree. (upon having a perceptual experience as of a tree)
-I had breakfast this morning. (upon trying to recall what one had for breakfast)
-That person is angry. (upon looking at a particular person's face)

For these are clearly analogous to cases (i) and (ii) in terms of warrant, force, and vivacity. In effect, Plantinga has given a list of examples of (what epistemologists call) Moorean facts. As such, their force, vivacity, and warrant will trump just about any philosophical argument against them, even if one is unable to refute them.

However, the crucial issue is whether Plantinga's examples of the humdrum variety of properly basic theistic beliefs -- the sorts of cases that are applicable to the typical Christian believer (and not just Moses and, perhaps, a handful of other fortunate souls) -- are comparable to cases (i) and (ii) in terms of degree of warrant, force, and vivacity.

Well, are they? To help answer this question, let's review Plantinga's examples of what he takes to be the typical, humdrum triggering conditions of properly basic theistic belief for most theists:

-God is speaking to me. (upon reading the Bible)
-God has created all this. (upon looking at the starry heavens)
-God disapproves of what I have done. (upon doing something cheap, or wrong, or wicked)
-God forgives me. (upon asking God for forgiveness for the cheap, wrong, or wicked thing)
-God is to be thanked and praised. (when life is sweet and satisfying)

Reflecting on such cases as these in his own Christian life, Quinn finds that they don't have anywhere near the degree of force, vivacity, and warrant of, for example, that involved in the Moses and the Burning Bush case; i.e., they don't seem to be anything remotely like Moorean facts for him. And of course, very many Christians likewise testify of having, at most, "gentle nudges" with respect to the sorts of triggering conditions of belief Plantinga mentions. But if so, then the relatively weak force, vivacity, and warrant such beliefs enjoy will not be sufficient to function as intrinsic defeaters for the objections to the faith that intellectually sophisticated adult theists are aware of. As such, it's not at all clear that Plantinga has even justified (2c); a fortiori he has not justified (2a). And if that's right, then it looks as though Plantinga has failed to deflect Quinn's charge: even if belief in God could be properly basic for parochial or pre-critical theists, it's not properly basic for intellectually sophisticated contemporary adult theists.

--------------------------------
[1] Plantinga, Alvin. "Is Belief in God Properly Basic", Nous 15:1 (1981), pp. 41-51.

[2] Quinn, Phillip. "In Search of the Foundations of Theism", Faith and Philosophy 2:4 (1985), pp. 469-486.

[3] Plantinga, "The Foundations of Theism: A Reply", Faith and Philosophy 3:3 (1986), pp. 298-313.

[4] Ibid., p. 311.

[5] Quinn, "The Foundations of Theism Again: A Rejoinder to Plantinga", in Zagzebski, Linda, ed. Rational Faith: Catholic Responses to Reformed Epistemology (University of Notre Dame Press, 1993), pp. 14-47.

Saturday, February 13, 2010

Beilby on the Variability-of-Belief Problem for Plantinga's Reformed Epistemology

In his paper, "Plantinga's Model of Warranted Christian Belief"[1}, Christian philosopher James K. Beilby raises an interesting and powerful criticism of Plantinga's latest and most mature version of his Reformed epistemology. First, though, some review and stage-setting:

According to Plantinga's account, a belief must satisfy four conditions if it is to have at least some measure of warrant:
(i) The belief must be produced by properly functioning cognitive faculties.
(ii) The relevant cognitive faculties must be successfully aimed at truth.
(iii) The belief must be produced in an epistemically congenial maxi-environment and mini-environment
(iv) The belief is subject to no undefeated defeaters (i.e., reasons against the belief that have yet to be undercut or rebutted).[2]

So that's what's required for a belief to have any warrant at all. But Plantinga allows that warrant admits of degrees, and he ties the degree of warrant a belief enjoys to the degree of firmness with which it is believed.[3] Thus, for a belief to have a degree of warrant sufficient for knowledge, it must be held with a very high degree of firmness.

Putting it all together, Plantinga's account can be summed up as follows:

I. Conditions of warrant are met + high degree of firmness = high degree of warrant.
II. Conditions of warrant are met + low degree of firmness = low degree of warrant.

So that's Plantinga's account of warranted belief in a nutshell. But how does this account connect to his account of warranted Christian belief in particular?

Very (very!) roughly, on Plantinga’s extended A/C model of warranted Christian belief, the internal instigation of the Holy Spirit acts on the believer by repairing the sensus divinitatis and producing a firm and unwavering faith in the Great Things of the Gospel. And since such belief meets all of the conditions of warrant (viz., (a) the proper function condition, (b) the truth-aimed faculty condition, (c) the epistemically congenial mini- and maxi-environments condition, and (d) the no defeaters condition), the degree of warrant enjoyed by the believer’s belief is sufficient to constitute knowledge. Enough review and setup; on to Beilby's criticism of Plantinga's model of warranted Christian belief.

In the paper mentioned above, Beilby argues that there is a tension between the extended A/C model's depiction of the paradigmatic believer's belief as firm and unwavering, on the one hand, and the actual facts about the typical believer's state of belief, on the other. For contrary to Plantinga's account, the belief of many believers is weak and wavering -- the “I do believe; help thou mine unbelief” sort. This is a potentially fatal problem for Plantinga's latest incarnation of reformed epistemology. For if we can't reconcile the model with the data, then Plantinga's model fails the minimal standards that he himself sets for it, viz., that it be at least epistemically possible (i.e., compatible with what we know or have reason to believe is true about the world.).

Unfortunately, it's hard to see how the tension between his model and the data can be resolved in principle. So, for example, it’s implausible (and blasphemous) to say that the Holy Spirit fails in his job to produce sufficiently warranted Christian belief in many believers; nor is it plausible to say that the multitude of believers with less-than-fully-firm faith are actually non-believers.[4] What, then, can explain the data of variability in degree of belief among Christians? Beilby points out that Plantinga chalks up the less-than-maximal belief in the typical believer to the noetic effects of sin. Will this reply solve the tension?

As Beilby points out, it will not. For this would mean that a certain portion of the relevant cognitive faculty's mini-environment (their sin-racked body and mind)) isn't epistemically congenial, in which case the belief fails the “congenial epistemic environment” condition of his account of warrant.[5] As such, those with less than maximal faith have a warrant-defeater for their Christian faith.

----
[1] in Peter-Baker, Deane. Alvin Plantinga (Cambridge University Press, 2007), pp. 125-165.

[2] "Put in a nutshell, then, a belief has warrant for a person S only if that belief is produced by cognitive faculties functioning properly (subject to no dysfunction) in a cognitive environment that is appropriate for S's kind of cognitive faculties, according to a design plan that is successfully aimed at truth." (Warranted Christian Belief, p. 156)

[3] "We must add, furthermore, that when a belief meets these conditions and does enjoy warrant, the degree of warrant it enjoys depends on the strength of the belief, the firmness with which S holds it." (Ibid.).

[4]As Beilby notes on a related matter: "While it is undoubtedly easier to describe and defend the warrant of "epistemological saints", because the Extended A/C model describes the ideal, fully formed faith of paradigmatic believers rather than the usual, in-process faith of typical believers, Plantinga's attempt to use the Extended A/C model to provide a good way for Christians (including, I assume, typical Christians) to think about the epistemology of Christian belief is in jeopardy. Since the faith of typical believers looks very different from that described in Plantinga's model, they have a choice between questioning the warrant of their belief about God or rejecting Plantinga's model as a good explanation of the warrant of their religious beliefs. Since Plantinga himself argues that the beliefs of "most Christians" are "both externally rational and warranted", the most reasonable option for the typical Christian is the latter." Ibid., p. 146.

[5] Alternatively, the problem could be chalked up to malfunction in the relevant cognitive faculties, due, again, to the ravishes of sin. But then a different condition of Plantinga's account of warrant isn't met -- viz., the proper function condition -- in which case, again, the belief isn't warranted, according to the conditions of warrant laid down by Plantinga himself.

Wednesday, February 10, 2010

Has Plantinga Shown that Classical Foundationalism is Self-Refuting?

John DePoe critiques Plantinga's arguments on this score in his paper, “In Defense of Classical Foundationalism: A Critical Evaluation of Plantinga’s Argument that Classical Foundationalism is Self Refuting".

A Neglected Version of the Great Pumpkin Objection

Here's an argument I'm toying with. (Very rough and partial draft. Comments welcome!)

There's an interesting version of the Son of Great Pumpkin Objection to Plantinga's reformed epistemology that has been raised in passing by Christians (William Lane Craig) and atheists (Keith Parsons) alike. However, so far as I've been able to tell, no philosopher develops the criticism in detail. Here's a first pass at a more explicit statement of the argument.

Setup: Plantinga follows Roderick Chisholm in using a particularlist, inductive method of generating criteria of proper basicality. As Plantinga puts it:

"We must assemble examples of beliefs and conditions such that the former are obviously properly basic in the latter, and examples of beliefs and conditions such that the former are obviously not properly basic in the latter. We must then frame hypotheses as to the necessary and sufficient conditions of proper basicality and test these hypotheses by reference to those examples." (Plantinga, Alvin. "Reason and Belief in God", in Faith and Rationality: Reason and Belief in God (U of Notre Dame Press, 1983), p. 76.)

Now as is well-known, Plantinga intends his use of "obviously" in the passage above to be relativized to epistemic communities ("obvious to us folks"), so as to allow controversial beliefs that are nonetheless strongly held in a given epistemic community to qualify as "obvious", and thereby to allow for correspondingly relativized, theism-friendly criteria of proper basicality. So understood, call Plantinga's relativized version of Chisholm's particularist, inductive method of generating criteria of proper basicality 'PIM' for short.

Now for the argument:

1. If PIM is legitimate, then it's possible for an epistemic community EC to blamelessly use PIM to generate criteria that entail that belief in God is not properly basic.
2. If it's possible for EC to blamelessly use PIM to generate criteria that entail that belief in God is not properly basic, then EC has a legitimate excuse for not believing in God, contrary to the teaching of the apostle Paul.
3. If EC has a legitimate excuse for not believing in God, contrary to the teaching of the apostle Paul, then conservative Christian theism is false.
---------
4. Therefore, if PIM is legitimate, then conservative Christian theism is false.

Call this 'The Conservative Christian Son of Great Pumpkin' objection to Plantinga's reformed epistemology. What could be said in reply to it?

One could reject (1) by arguing that every possible blameless epistemic community would in fact generate criteria that entail properly basic theistic belief. But this strikes me as implausible; aren't there actual epistemic communities like this? Now I suppose one could reply that such actual communities are in fact blameworthy in forming criteria that preclude properly basic theistic belief, on the grounds that we all know, deep in our internalistically accessible hearts, that God exists. That strikes me as false; in any case, to advance this sort of reply is at odds with Plantinga's decidedly externalist reformed epistemology.

Alternatively, one could reject (2). There are two ways to go here. First, one could argue that even if an epistemic community could blamelessly generate criteria that preclude properly basic theistic belief, they may yet be blameworthy if there are non-basic evidential grounds for belief in God with respect to any possible set of blamelessly formed criteria of proper basicality. The problem with this response is that while this may or may not be so, it is the response of one who rejects Plantinga's reformed epistemology.

On the other hand, one might criticize (2) by arguing that an epistemic community could be blameless in not believing in God in a way that's not contrary to the teachings of the apostle Paul. For example, such a community might live in a time or place in which no one has heard of God, in which case the conditions required for properly basic or properly based theistic belief do not obtain in such a community. And perhaps Paul's teachings are compatible with saying that such a community has a legitimate excusing condition for not believing in God.

However, it seems to me that such a reply requires rejecting the most natural reading of Paul's words in Romans 1: 18-23:

"The wrath of God is being revealed from heaven against all the godlessness and wickedness of men who suppress the truth by their wickedness, since what may be known about God is plain to them, because God has made it plain to them. For since the creation of the world God's invisible qualities—his eternal power and divine nature—have been clearly seen, being understood from what has been made, so that men are without excuse." Furthermore, it's perhaps worth noting that WIlliam Lane Craig agrees with me about the truth of (3) [n.b. add link!]

Finally, one could reject (3). Thus, one might argue that one could still be a conservative Christian even if one rejects Paul's teaching that no normal and sufficiently mature human being has a legitimate excuse for not believing in God. In order to cut debate short about what constitutes conservative Christianity, I'm happy to use a stipulative partial definition: as I'm using the term, 'conservative christianity' requires, at a minimum, accepting Paul's epistle to the Romans as a part of the NT canon, and accepting the doctrine of biblical inerrancy. If some reject this definition, I'm happy to limit the scope of my target audience to those who accept my partial stipulative definition; that's exactly the audience I'd like to address.

Therefore, if the argument above is on track, then it looks as though Plantinga's reformed epistemology isn't an option for conservative Christians. In closing: Plantinga says that "followers of Bertrand Russell and Madelyn Murray O'Hair may disagree [about the Christian's initial data set of particular cases of "obviously" properly basic beliefs, and the criteria of proper basicality that result from Christians employing Chisholm's particularist inductive method], but how is that relevant?" Reply: because it entails that the Apostle Paul was wrong. And while I'm happy with that, I'm not confident that Plantinga would be.

Sunday, February 07, 2010

Thomas Crisp's Refutation of Contemporary Historical Apologetics

I'd like to call attention to Thomas Crisp's[1] recent paper, “On Believing that the Bible is Divinely Inspired”, in Analytic Theology: New Essays in Theological Method (Oxford University Press, 2009). In the paper, Crisp offers his own account of how such belief can be rational. But what I found most interesting about his paper was the rigorous and apparently fatal critiques of the prominent contemporary models for rational belief in biblical reliability and inspiration: those of Richard Swinburne, Alvin Plantinga, and Timothy McGrew. Required reading for those interested in historical apologetics.

------------------
[1] Thomas Crisp is one of the brightest young stars in the constellation of contemporary Christian philosophers, and is very highly regarded in the broader philosophical community. He did his M.A. in Philosophy of Religion and Ethics at Biola University, studying under J.P. Moreland. While there, he was generally regarded as the best Philosophy student they ever had. After graduating from the program, he went on to Notre Dame for his doctoral studies, where he quickly impressed Alvin Plantinga and the other faculty and graduate students. He completed his dissertation (which was a defense of presentism) in 2002; Plantinga was his dissertation advisor.

Three years into a tenure-track position at Florida State University, Crisp returned to Biola, where he is currently an Associate Professor of Philosophy. His past and current work focuses on issues in metaphysics, epistemology, and philosophy of religion.

Friday, February 05, 2010

Alvin Plantinga is Retiring

Alvin Plantinga (Notre Dame), the greatest living Christian philosopher of religion, is retiring. Despite my disagreements with his views, he is one of my favorite philosophers of religion. His work is of the highest caliber, and his style is a model of clarity and rigor. I have learned more from his work than I can say.

On a happier note, a conference is scheduled in celebration of his retirement. Details about the conference can be found here.

P.S., Plantinga's latest reply to criticisms of his evolutionary argument against naturalism ("Content and Natural Selection", Philosophy and Phenomenological Research (forthcoming)) is now available at his department website. Here is the link.

Porcupinefish send out mixed messages

Porcupinefish send out mixed messages

Posted using ShareThis

Tuesday, February 02, 2010

plantinga's proper functionalism

In a previous post, I emphasized the fact that on Plantinga's account of warranted belief, the degree of warrant a given belief enjoys depends on the degree of firmness with which one holds it. Thus, Plantinga's view entails the following two (roughly stated) theses:

I. Conditions of warrant are met + high degree of firmness = high degree of warrant.
II. Conditions of warrant are met + low degree of firmness = low degree of warrant.

Now Plantinga characterizes his account of warrant as externalist. But while thesis (I) is in accordance with his externalism (e.g., one can be warranted in firmly believing that Christianity is true if it meets all of Plantinga's proper functionalist conditions of warrant, even if one cannot know that (say) the faculty responsible for that belief is successfully aimed at truth), it appears that thesis (II) is not. For one can, at least sometimes, be aware by armchair reflection alone that one does not have a sufficiently firm belief that Christianity is true, and thereby infer that one's belief has little by of warrant. And if that's right, then Plantinga's theory of warrant isn't purely externalist.[1]

One implication of this is that if Plantinga's theory of warrant is correct, then one can tell, at least under certain conditions, whether one's Christian faith is warranted. So, for example, if one is aware that one's belief in Christianity is of the fairly common weak and waivering sort ("I do believe; help thou mine unbelief"), then one can rule out that one's belief has enough warrant sufficient for knowledge.[2]


(UPDATE: I have since read an old paper by Linda Zagzebski, in which she makes virtually the same point about asymmetry I bring up here. See her "Religious Knowledge and the Virtues of the Mind", in Zagzebski (ed.) Rational Faith: Catholic Responses to Reformed Epistemology (University of Notre Dame Press, 1993), pp. 199-225, esp. pp. 201-202.)
-----------
[1] There are of course very many characterizations of internalism and externalism in the literature, but discussion of such would distract from the point I'm making. It's sufficient for my purposes to speak in these rough and intuitive terms.
[2] Michael Sudduth has made a somewhat related point re: the partially internalist character of Plantinga's account in (e.g.) "The Internalist Character and Evidentialist Implications of Plantingian Defeaters", International Journal for Philosophy of Religion 45 (1999), pp. 167-187.
[3] I'm often astonished, though, at how frequently Christians underestimate how low their degree of belief in the truth of Christianity often really is. My guess is that they often think of belief in terms of an implausible mental-assent account (what Christian philosopher Dallas Willard memorably calls "barcode faith": the believer mentally assents to a set of propositions about their faith, and then God "scans" their minds and grants them salvation), when belief (especially the degreed notion of belief Plantinga has in mind in his account of warrant) plausibly has satisfaction-conditions that are much more stringent than that. The point I'm making in this post can thus be seen more clearly against the background of some of the most important recent literature on the nature of belief. I therefore refer the reader to these papers on belief by philosopher Eric Schwitzgebel (UC Riverside). If Schwitzgebel is anywhere near the truth about the nature of belief, then, again, it's implausible to think that many Christians believe with sufficient firmness to have warranted Christian belief, even if Plantinga's account of warrant is correct, and even if they meet all the other conditions of his account. The faith of a mustard seed may or may not be sufficient for salvation, but it's nowhere near sufficient for warrant.
Site Meter