This is the text for a presentation I gave at the Eastern Division Meetings of the American Philosophical Association in Washington, D.C. on December 28, 1998.
One of the most important aspects of Alvin Plantinga’s paper, and of his religious epistemology generally, is his claim that some Christian beliefs are properly basic.(1) In what follows, I will very briefly sketch, defend, and present for your consideration an alternative picture according to which Christian beliefs are not properly basic.
1. Properly Basic Beliefs
What is it for a belief to be properly basic? Life is short, and my time here today is even much shorter, so let me skip the complicated story I could give about the significantly different things that can be meant here and how these different things compare with various remarks Plantinga makes, and cut straight for an abbreviated account of what I think is the best way, for our present purposes, of construing such a claim. Plantinga often has occasion to speak of “warrant transfer”; for instance, he here speaks of beliefs which “don’t get their warrant by way of warrant transfer from other [beliefs].”(2) Let us distinguish between immediate and transferred warrant, letting immediate warrant be warrant a belief has that it does not have by virtue of a transfer of warrant from another belief. And let a “sufficiently warranted” belief be a belief that has enough warrant to count as a piece of knowledge. A properly basic belief, then, is best construed as a sufficiently warranted belief that is sufficiently warranted by virtue of its immediate warrant, i.e., independent of any transferred warrant it might enjoy. (Thus, a properly basic belief may have some transferred warrant, so long as it doesn’t depend on this transferred warrant in order to be sufficiently warranted.)
2. The Structural Soundness of “Foundherentism”
Now, suppose that, among the beliefs one holds, is a certain system of closely related beliefs, each of which enjoys a significant amount immediate warrant, but not enough immediate warrant to be sufficiently warranted. But suppose further that, in virtue of relations of mutual support that hold among beliefs in the system, these beliefs transfer warrant among themselves, with the result that many of them end up being sufficiently warranted. Note two important things about a belief that is sufficiently warranted in the way described above. First, such a sufficiently warranted belief is not properly basic: Though it enjoys significant immediate warrant, its immediate warrant is not sufficient for knowledge. Second, such a sufficiently warranted belief is not based (directly or indirectly) upon any properly basic beliefs: The beliefs from which its transferred warrant comes are in the same boat that it is in, i.e., they are not properly basic either.
In light of the above two facts, such a belief does not fit well into a foundationalist picture of the structure of knowledge. However, it’s difficult to say how a foundationalist like Plantinga might rationally reject such a possibility.
To show this, consider a very simple system, consisting of just two beliefs, which we’ll call belief A and belief B. Suppose A and B each have a good deal of immediate warrant for the subject in question, but the amount of immediate warrant each enjoys for her falls just short of what’s needed for them to be sufficiently warranted. For reasons I won’t take the time to fully discuss, it’s difficult to see how Plantinga could reject that a situation, as so far described, could arise. Given that he, as a good foundationalist, believes in immediate warrant, and given his views about how immediate warrant is generated, I think he pretty well has to admit that beliefs can enjoy immediate warrant just shy of the amount needed for knowledge.(3) “Admit” might even be the wrong word; I suspect he’d happily accept this much. But things haven’t gotten interesting yet. Suppose further that A and B are mutually supporting beliefs (it would perhaps be more proper here to speak of the propositions that are the objects of these beliefs as being mutually supporting, but again, life is short); that is, A supports B and B supports A. Plantinga accepts this possibility: He sees that “The supports relation is not asymmetrical” (Warrant: The Current Debate, p. 73). But now suppose that our subject considers her two beliefs together, noticing that A supports B and that B supports A. Shouldn’t she then feel more confident about both beliefs, and rationally so? And wouldn’t each then transfer some warrant to the other? After all each enjoys significant warrant, and supports the other, and our subject has noticed this. When you have a (partially or sufficiently) warranted belief, like A, notice that it supports another belief (B, in this case), and, as a result, increase your level of confidence in that other belief, that seems a clear case in which warrant has transferred from A to B.
But likewise in reverse — from B to A. And by now the foundationalist’s warning lights have been lit up. Plantinga insists: “But even if [he should have written “even though”] the supports relation is not asymmetrical, the basis relation, in a proper noetic structure, is asymmetrical. If my belief that A is accepted on the evidential basis of my belief that B, then my belief that B must not be based on my belief that A” (Debate, p. 74), and more generally, as we get into systems that include more than just two beliefs, he insists that, “The basis relation, in a proper noetic structure, is noncircular” (Debate, p. 74). But his arguments (see primarily the section entitled “Against Circles,” Debate, pp. 74-77) seem fit to show only the conclusion that “believes solely on the basis of” is asymmetrical (and non-circular) in a proper noetic structure, and have no power against our example, which, though it includes mutual partial basing, violates none of the intuitions Plantinga uses to rule out the propriety of circles. We may suppose that B was already a belief the subject held before it was “brought into contact” with her belief A; the result of that contact was perhaps just an increase in the level of confidence with which B was held. So B is not believed solely on the basis of A. Likewise, A is not believed solely on the basis of B. Rather, each is believed primarily because of whatever it is that generates immediate warrant (and let’s leave it open what that is), and is then bolstered by the support it receives from the other. Plantinga intuits that warrant cannot be initially generated merely by means of warrant transfer. I agree. That doesn’t happen in my example; rather, warrant that is immediate for one belief is then transferred from that belief to the other, and vice versa. Indeed, though B is partially based on A and A is partially based on B, there is no circle of warrant, because no warrant moves in a circle: The warrant A transfers to B was immediate to A and did not come from B, and the warrant B transfers to A was immediate to B and did not come from A. Of course, our subject might start getting over-confident about her beliefs, treating B as if it were supported by a belief that is sufficiently warranted independent of the warrant it received from B, because she loses track of the fact that A (B’s partial basis) is sufficiently warranted only because of the support it received from B. If this happens, our subject’s noetic structure will be defective. But we don’t have to suppose that any such defect occurs, and I’m supposing that it does not.
And, of course, if both beliefs were very close to being sufficiently warranted just in virtue of their immediate warrant, and each transferred enough warrant to the other, then it will happen that each becomes sufficiently warranted in virtue of the partial mutual basing described above — together, of course, with the independent warrant the belief enjoyed since the beginning of our story. So we get the result promised above: sufficiently warranted beliefs that are not properly basic, nor are they based upon properly basic beliefs. Here it’s important not to be misled by the term “warrant transfer,” which to the ears of most of us suggests that the transferrer loses what it transfers to the receiver of the transfer. (If transferers didn’t lose what they transfer, then one suspects that bank transfers would be a whole lot more popular!) That’s not how it works in warrant transfer among beliefs. Take a case of simple, one-way inference (you know, the kind of basing foundationalists like): C is sufficiently warranted, you notice it implies D, and infer D from C. D becomes warranted (perhaps sufficiently so) by virtue of warrant transfer from C, but C’s level of warrant is not thereby reduced (and certainly is not reduced by as much as D’s level of warrant in increased). In our example of partial mutual basing, then, both beliefs will become sufficiently warranted, because they were almost so just in virtue of their immediate warrant, and each received enough transferred warrant from the other to make up the difference without losing the warrant it transferred to the other.
In our two-belief case, I supposed both beliefs were initially (independently of any warrant transfer) just short of being sufficiently warranted. Notice, however, that this is inessential to the case, especially where you get a large system of beliefs, each of which has immediate warrant, and enters into partial basing relations with many of the others (so that each can receive warrant from several other beliefs without any of those others losing the warrant they transfer). Here, the immediate warrant can fall well short of being sufficient, and yet the beliefs comprising the system may end up being sufficiently warranted, because of the warrant they transfer (without losing) to one another.
A structure such as that described above is not foundational — there are sufficiently warranted beliefs, but they’re not properly basic, nor are they based (directly or indirenctly) upon properly basic beliefs — but neither does it suffer from the structural defects that plague coherentism. Susan Haack has given such a structure — alas! — a very unlovely name — “foundherentism”! — but — happily — a very lovely and suggestive metaphor.(4) Every option in the structure of knowledge should have a good metaphor. Foundationalists, of course, love buildings, being apparently especially fond of pyramids. And coherentists have their webs — and rafts. (The web, because webs are “tied down” at many points to the outside, might actually be a better metaphor for foundherentism than it is for coherentism.) Foundherentists should adopt Haack’s metaphor: the cross-word puzzle. Here, the plausibility an answer has in virtue of the clue and number of letters nicely represents immediate warrant, while an answer’s “fitting in” with other answers can nicely represent beliefs’ transferred warrant.
But I have just been defending such a view as a coherent option, and, in accepting it as a coherent possible structure, or even as a structure that’s actually instantiated in important parts of human noetic structures, doesn’t rule out the possibility that there be properly basic beliefs in one portion of a subject’s body of beliefs, while others of her beliefs are warranted in the foundherentist way.
3. Two Accounts of the Warrant of Perceptual Beliefs: Stringent Direct Realism (SDR) and Direct Warrant Realism (DWR)
A Stringent Direct Realist believes that perceptual beliefs are properly basic, and thus don’t have to be properly based on other beliefs — for instance, on beliefs about one’s own sensory experiences or on beliefs about how one is “appeared to” — to be sufficiently warranted. By contrast, a foundherentist about perceptual beliefs — call such a theorist a Direct Warrant Realist — won’t think that perceptual beliefs are properly basic, and thus differs from the Stringent Direct Realist on that issue, but can join the Stringent Direct Realist in rejecting the evidentialist requirement that perceptual beliefs must be properly based on beliefs about one’s own sensory experiences in order to be warranted, and thereby join the direct realist in eluding some of the most powerful skeptical arguments against perceptual beliefs — the evidentialist arguments of Berkeley and Hume, that wield the evidentialist requirement, together with powerful arguments to the effect that the requirement can’t be met, in order to reach their skeptical conclusion. For according to Direct Warrant Realism (DWR), though perceptual beliefs don’t have enough immediate warrant to be sufficiently warranted, they (or at least many of them) become sufficiently warranted in virtue of relations of mutual support that they bear to one another. So, as on SDR, they don’t require support from the outside of the type Berkeley showed they cannot have.
The best reasons I know of for preferring SDR over DWR consist of psychological observations about how we in fact form perceptual beliefs. Most importantly, we seem to form perceptual beliefs in a psychologically immediate way: we don’t consciously infer them from other beliefs and our coming to hold these beliefs doesn’t seem to follow our consciously noticing how well they fit in with other beliefs. (Recall that in our two belief example, I described our subject as noticing that A and B were mutually supporting, and her subsequent rise in confidence was due to this noticing.) Rather, whether it’s due to innate or learned dispositions, we seem set up to just form the relevant beliefs upon the occasions of having appropriate patterns of sensory experiences — without having to entertain any thoughts about the relation of the belief we’re forming to others of our beliefs. To accept DWR is to accept that perceptual beliefs are partially based on one another — they transfer warrant one to another — though they are formed in a way that involves no conscious thought about one another. However, I still find DWR to be a very attractive account of the warrant of perceptual beliefs. This is largely because I think it’s wrong to require that a belief be formed due to a conscious noticing of its relations to other beliefs in order for those other beliefs to transfer warrant to it. Perhaps, though we take no notice of a perceptual belief’s relations of “coherence” (relations of mutual support) to other perceptual beliefs as the belief in question is formed, we are appropriately sensitive to its coherence with these other beliefs, where such sensitivity consists in such facts as that we wouldn’t hold or continue to hold the belief in question, or at least wouldn’t continue to hold it to the degree that we do, were it not for its coherence with our other beliefs. Such a sensitivity could hold even where one gives no conscious thought to the relations in question, and yet it seems to me that warrant might very well be trasnferred among beliefs in virtue of the believer’s sensitivity to the evidential relations that hold among them, even where she gives no conscious thought to those relations. (Foundationalists, too, should make such a move, I think. The general advice I would give is: Don’t require for warrant transfer a conscious noticing on the believer’s part of the relations among the beliefs involved in the transfer. The foundationalist, then, should not require that an inference be consciously performed in order for one belief to transfer warrant to another. It is enough that the believer be sensitive in her holding of the second belief to its evidential relations to the first belief. Following such advice is necessary, I think, to avoid an overly idealistic and overly intellectualized picture of cognitive precessing.)
4. Stringent Direct Christianity, Direct Warrant Christianity, the Evidentialist Objection, and Parity Arguments
Our current concern is not perceptual beliefs, but Christian beliefs. But much of what was said above about the two different accounts of the warrant of perceptual beliefs can be transferred to yield two different accounts of the warrant of Christian beliefs. (And here, many who find SDR more attractive than DWR as an account of perceptual beliefs might prefer the “foundherentist” account of the warrant of Christian beliefs.) Plantinga, since he believes in properly basic Christian beliefs, is a Stringent Direct Christian. But we can now see a foundherentist alternative to Plantinga’s account: Direct Warrant Christianity, according to which, although Christian beliefs are not properly basic, some of them do have significant immediate warrant, and many of them are sufficiently warranted due to relations of mutual support they bear to one another.
One of Plantinga’s main motivations for defending the possibility of properly basic theistic beliefs — and, more relevant to our current concerns, properly basic Christian beliefs — in the paper currently under consideration, but also in his religious epistemology for almost the last twenty years, has been to defend such beliefs against evidentialist objections. But, as was the case with sense perception, the foundherentist option, like the foundationalist, properly basic option, provides an escape from such evidentialist objections. The ascription of immediate warrant to the beliefs in question, even where this immediate warrant does not render the beliefs sufficiently warranted, together with the possibility of mutual partial basing, cuts off the evidentialist objection: The beliefs in question don’t have to be evidentially supported from the outside in the way the evidentialist objector demands.
If Stringent Direct Christianity and Direct Warrant Christianity each provide an escape from evidentialist objections, is there any significance to the differences between them? Yes. Let me mention just this. Parity arguments, of the type William Alston has advanced, in which he argues that “[Christian Practice] has basically the same epistemic status as [Perceptual Practice] and that no one who subscribes to the latter is in any position to cavil at the former,”(5) I think, could not be effectively executed from the “foundherentist” perspective. Very briefly, here’s why. One of the key moves in Alston’s Parity argument is to admit that the perceptual beliefs that result from our engaging in “Perceptual Practice” display stronger relations of positive coherence (“we discover regularities” in the physical world by means of perception to a greater extent than we discover regularities in God’s behavior by means of religious experience), but to argue that this does not ruin the parity argument.(6) Briefly, according to Alston’s argument, this is because, given God’s “wholly other” nature and other facts, we shouldn’t expect to discover regularities in God’s behavior to the extent that we discover regularities in the physical world by means of perception — we shouldn’t expect Christian beliefs to display as much coherence as do perceptual beliefs. And thus, that Christian beliefs don’t cohere to the extent that perceptual beliefs do provides no reason to think “Christian Practice” unreliable. From Alston’s perspective, a lack of positive coherence is relevant to the extent that one would expect positive coherence to be displayed, and thus to the extent that the lack provides reason for thinking the practice in question is not reliable.(7) But from the “foundherentist” perspective, the coherence displayed perceptual beliefs display may be crucial to their having sufficient warrant in the first place. Thus, if Christian beliefs don’t display that same level of coherence, they may well not be warranted to the extent that perceptual beliefs are. This foundherentist perspective yields an attitude different from (and, to my thinking at least, far more plausible than) Alston’s to the result that the entity or entities that an epistemic practice (allegedly) puts one in touch with is such that one wouldn’t expect the beliefs that result to display much in the way of positive coherence. The foundherentist will then suspect that it may be harder for the beliefs resulting from such a practice to be warranted, instead of concluding that positive coherence isn’t important in the case. The prospects for a successful Alston-like parity argument then are quite dim from the perspective of Direct Warrant Christianity. Since, on this model, the mutual coherence of perceptual beliefs enjoy are essential to their being sufficiently warranted in the first place, the admission that Christian beliefs don’t enjoy the same degree of mutual coherence shoots any such parity argument right out of the water. I should hasten to add, however, that a lack of parity with perceptual beliefs in the epistemic status of Christian beliefs does not show that Christian beliefs are unwarranted or even that they’re not sufficiently warranted, especially if one, like me, thinks of perceptual beliefs as being super-warranted — as being warranted to an extent that far exceeds what’s needed for knowledge.(8)
5. Faith, Plantinga Faith, and Warrant
Plantinga’s account of faith gives us a good clue I think as to why he would gravitate toward a picture according to which Christian beliefs are properly basic. Following Calvin, Plantinga defines faith (or faith of the type he wishes to discuss) as a “firm and certain knowledge” of “the central teachings of the gospel” (pp. 244-249). This certain knowledge is not arrived at by the workings of any of the faculties we were created with, but is the result of the work of the Holy Spirit (p. 245), who “causes us to believe these great truths of the gospel” (p. 245). Let “Plantinga faith” be faith as Plantinga describes it, leaving out the part about the beliefs in question constituting knowledge. (I want to identify “Plantinga faith” in a non-evaluative way in order to leave as an open question the status of beliefs that are the result of “Plantinga faith.”) Plantinga faith then is a very strongly held belief in the great truths of the gospel, produced in accordance with God’s plan by the Holy Spirit in the way Plantinga describes. How strongly held? Very strongly: It is a “firm and certain” belief (p. 244), one finds the beliefs so caused to be “compelling” (p. 250).(9) Would the fortunate recipient of Plantinga faith have properly basic beliefs? Even after reading Plantinga’s defense, I don’t find this matter nearly as clear as Plantinga seems to find it, but it does seem plausible to me to suppose so.
I, however, have not been blessed with Plantinga faith. I believe that I have been blessed enough to have had experiences that are in some ways like those Plantinga describes, but for me, the most I have received directly from the Holy Spirit have been gentle nudges toward belief, certainly nothing even approaching the firm and certain conviction of which Plantinga speaks. And if the people I’ve talked to are to be believed — and they are — there are many who would be thrilled to receive faith as Plantinga describes it, but who have not, despite Plantinga’s claim that faith — presumably as he defines it, as a firm and certain conviction — “is given to anyone who is willing to accept it” (p. 244). Perhaps all will eventually be blessed with Plantinga faith, but I am here to report from personal experience that not everyone who has been willing to accept that gift, not even everyone who has been willing for many years to receive that gift, has received it. And from encounters I’ve had with others with many others, I think I can safely report to you that some who are willing to accept that gift never receive it in this life. Rather, we have to muddle through with a form of faith that resembles a leap in the dark (see p. 247) a lot more than Plantinga faith seems to. (Resembling more closely a leap in the dark, but still not a leap in the dark. A leap in the twilight, perhaps.)
Such faith can grow, however: The degree of belief can increase. It has in my own case. I think rationally so. But my best hope for such rationality is that I’m being appropriately sensitive to relations of mutual support that these Christian beliefs bear to one another, and to support they receive from how well they fit in with my warranted (sufficiently or partially) beliefs from elsewhere in my belief structure. It does seem to me that those gentle nudges toward belief play a crucial role — both in an account of why I in fact believe what I do, but also in an account of why these beliefs are warranted, to whatever extent they are warranted. But, as you can gather, it’s seeming to me that, as compared with the view that Christian beliefs are properly basic, the “foundherentist”, “Direct Warrant Christianity” model of justification seems much better suited as an account of the justification of the Christian beliefs of one who, like me, God has chosen to experientially leave in the twilight.
Given time constraints, rather than arguing as best I can for the conclusion that the “DWC” model is superior for a wide range of believers — those who have received weaker experiential “input” rather than a full dose of Plantinga faith — let me instead close with two questions for Prof. Plantinga to see what he thinks. First, does he find a “foundherentist” account of the warrant of Christian beliefs of the type I’ve here sketched a (structurally and otherwise) coherent option? And, second, if so, would he find that coherent option to be an attractive one as an account of the warrant of the Christian beliefs of one blessed only with gentle nudges toward belief?
1. See especially pp. 9-10, and thereabouts of Plantinga’s paper, “Warranted Christian Belief: The Aquinas/Calvin model,” which appears to be a chapter, or at least a part of a chapter, of Plantinga’s upcoming book, Warranted Christian Belief, or, now, pp. 177-179 of the book itself. I received Plantinga’s paper as an attachment to an e-mail message, and am not certain that as my computer printed it out the page numbers will match up with others’ copies of the paper, though I believe that they will. In what follows, all page references that don’t specify that they’re to other works will be to this paper, and will be to it as it printed out on my computer.
Update 7/29/2005: Now all the page references to this chapter are Plantinga’s book, Warranted Christian Belief (Oxford UP), which came out in 2000.
2. P. 10. (I haven’t yet located this quotation in the published version of Plantinga’s book.) I’ve substituted “beliefs” here for Plantinga’s “propositions”, because I think that’s what he meant: not transfer from other propositions, but from beliefs in other propositions.
3. Further, I think this is so on all plausible accounts of how immediate warrant is generated, and thus, not just Plantinga, but any good foundationalist will have to admit that a belief might have immediate warrant that falls shy of the amount needed for knowledge.
4. For a brief presentation/defense of Haack’s “foundherentism,” see her “A Foundherentist Theory of Empirical Justification,” in L. Pojman, ed., The Theory of Knowledge, second edition (Wadsworth, 1998): 283-293; also in E. Sosa and J. Kim, ed., Epistemology: An Anthology (Blackwell, 2000): 226-236.
For a more sustained presentation, see her book, Evidence and Inquiry: Towards Reconstruction in Epistemology (Blackwell, 1995), and these elaborations/responses to criticism:
- “Prècis of Evidence and Inquiry” (pp. 7-11) and “Reply to Bonjour” (pp. 25-35), in Synthese 112 (1997)
- “Prècis” (pp. 611-614) and “Reply to Commentators” (pp. 641-656), Symposium on S. Haack, Evidence and Inquiry, in Philosophy and Phenomenological Research 56 (1996).
5. Alston, “Religious Experience and Religious Belief,” Noûs 16 (1982): 3-12, p. 12. Alston actually writes “no one who subscribes to the former is in any position to cavil at the latter,” but this is clearly a mix-up of “former” and “latter”.
6. See “Religious Experience and Religious Belief,” especially pp. 8-12. One of the several respects in which my argument here is overly compressed is this: I should, if I had more time, say something about the nature of “positive coherence” — where this is not just the lack of contradictions or other conflicts among beliefs (it’s not just negative coherence), but some kind of positive mutual support — and how beliefs that are related to each other in the way Alston describes — they allow us to discover regularities in the objects putatively observed — would enjoy more in the way of “positive coherence” than do beliefs which are not so “regular”.
7. In the terminology of “Religious Experience and Religious Belief,” whether a lack of positive coherence has the result that the practice fails to be JNW, where that stands for “justified in the weak, normative sense” — i.e., there’s not adequate reason to think the practice is unreliable. (JNS, justified in the strong, normative sense, means for Alston that there is adequate reason to think the practice in question is reliable.)
8. Here, to avoid issues extraneous to our current concerns, I suppress here my contextualist views about knowledge (or, more precisely, about knowledge attributions). Taking these contextualist views into account, I’d describe super-warranted beliefs as beliefs that are warranted to an extent that far exceeds what’s needed to satisfy the standards for knowledge that are set by most ordinary contexts.
9. I worry that I may be misreading Plantinga here. He does write, “This process can go on in a thousand ways” (p. 251), and approves of Aquinas’s term (in translation, of course) of “invitation” to describe the work of the Holy Spirit. “Invitation” seems more fitting for the gentle nudge toward belief of which I’m speaking, and less fitting for the kind of firm, compelling, and certain belief that comes not through the workings of one’s own faculties but is rather caused directly by the Holy Spirit about which Plantinga often seems to writing. (If some spirit were to so zap me with a firm, compelling, and certain belief that P, I certainly wouldn’t be tempted to describe this as a case of a spirit “inviting” me to believe that P!) He does, however, define the kind of faith he is speaking of as a firm and certain knowledge that is caused by the Holy Spirit, so that’s how I’ll read him.