« Prev C. External Rationality and Warrant: Faith is… Next »

C. External Rationality and Warrant: Faith Is Knowledge

The part of Calvin’s definition of faith that is especially striking to contemporary ears is that on his account faith is a really special case of knowledge (“a sure and certain knowledge”; compare also the account of true faith in the Heidelberg Catechism, above, p. 247). Faith is not to be contrasted with knowledge: faith (at least in paradigmatic instances) is knowledge, knowledge of a certain special kind. It is special in at least two ways. First, in its object: what is allegedly known is (if true) of stunning significance, certainly the most important thing a person could possibly know. But it is also unusual in the way in which that content is known; it is known by way of an extraordinary cognitive process or belief-producing mechanism. Christian belief is “revealed to our minds” by way of the Holy Spirit’s inducing, in us, belief in the central message of Scripture. The belief-producing process is dual, involving both the divinely inspired Scripture (perhaps directly, or perhaps at the head of a testimonial chain) and the internal instigation of the Holy Spirit. Both involve the special activity of God.

If faith is such an extraordinary way of holding belief, why call it ‘knowledge’ at all? What about it makes it a case of knowledge? Here we must look a bit more deeply into the model. The believer encounters the great truths of the gospel; by virtue of the activity of the Holy Spirit, she comes to see that these things are indeed true. And the first thing to see is that, on this model, faith is a belief-producing process or activity, like perception or memory. It is a cognitive device, a means by which belief, and belief on a certain specific set of topics, is regularly produced in regular ways.320320   Although this regularity is typical of cognitive processes, it isn’t really necessary; see my reply to Lehrer in Warrant in Contemporary Epistemology, ed. J. Kvanig (New York: Rowman and Littlefield, 1996), pp. 332ff. In this it resembles memory, perception, reason, sympathy, induction, and other more standard belief-producing processes. It differs from them in that it also involves the direct action of the Holy Spirit, so that the immediate cause of belief is not to be found just in her natural epistemic equipment. There is the special and supernatural activity of the Holy Spirit. Nevertheless, faith is a belief-producing process. Now as we saw in chapter 7, what is required for knowledge is that a belief be produced by cognitive faculties or processes that are working properly, in an appropriate epistemic environment (both maxi and mini) according to a design plan that is aimed at truth, and is furthermore successfully aimed at truth. But according to this model, what one believes by faith (the beliefs that constitute faith) meets these four conditions.


First, when these beliefs are accepted by faith and result from the internal instigation of the Holy Spirit, they are produced by cognitive processes working properly;321321   A caveat: as Andrew Dole points out in “Cognitive Processes, Cognitive Faculties, and the Holy Spirit in Plantinga’s Warrant Series” (as yet unpublished), it is not obvious that one can directly transfer necessary and sufficient conditions for warrant from beliefs produced by faculties to beliefs produced by processes. they are not produced by way of some cognitive malfunction. Faith, the whole process that produces them, is specifically designed by God himself to produce this very effect—just as vision, say, is designed by God to produce a certain kind of perceptual beliefs. When it does produce this effect, therefore, it is working properly; thus the beliefs in question satisfy the external rationality condition, which is also the first condition of warrant. Second, according to the model, the maxienvironment in which we find ourselves, including the cognitive contamination produced by sin, is precisely the cognitive environment for which this process is designed. The typical minienvironment is also favorable. Third, the process is designed to produce true beliefs;322322   Though this need not be the only purpose involved. Perhaps the beliefs produced have other virtues in addition to truth: perhaps they enable one to stand in a personal relationship with God, to face life’s vicissitudes with equanimity, to enjoy the comfort that naturally results from the belief that constitutes faith, and so on. and fourth, the beliefs it produces—belief in the great things of the gospel—are in fact true; faith is a reliable belief-producing process, so that the process in question is successfully aimed at the production of true beliefs.

Reliability, of course, demands more than just that these beliefs be true. A thermometer stuck on 72˚F is not reliable even if it is somewhere—San Diego, say—where it is always 72˚F. What it would do if things were relevantly different (what it would do in appropriately nearby possible worlds) is also relevant; a process or instrument is reliable only if it would produce a true output under different conditions. On the current model, this condition is also met. The Holy Spirit doesn’t work just by accident or at random, and there are a thousand ways in which, even if things had been different, the Holy Spirit would have produced the results actually produced. Clearly, any circumstances in which it produces this output are circumstances in which this output is true; hence, under those circumstances, it would have produced a true output. Under what conditions would the Holy Spirit have failed, with respect to a given person, to do this work of enabling one to see the truth of the great things of the gospel? The model need take no stand on this issue, but it is part of much traditional Christian teaching to hold that a necessary condition of my receiving the gift of faith is my acquiescing, being willing to accept the gift, being prepared to receive it. There is a contribution to this process that I myself must make, a contribution that I can withhold.


According to this model, faith as a belief-producing mechanism involves a supernatural element; it involves God’s doing something specially and directly and quite out of the ordinary. Does that compromise the claim that the deliverances of faith constitute knowledge? I can’t see how. There was no suggestion in the original account that cognitive mechanisms must all be natural, whatever precisely that comes to. Must the account be revised because faith doesn’t go just by natural laws or regularities, working instead by way of the free cooperation of a person—God himself—whose speaking in Scripture is, of course, free, as is the action of the Holy Spirit in revealing and sealing the great truths of the gospel? Again, I can’t see why. The same goes for the mechanism Thomas Reid calls ‘testimony’, a mechanism whereby we learn from others; this mechanism too (often) works by way of free human agency. (When you ask me how old I am, I can [freely] tell you, or in a minor fit of pique, freely refuse.)

Why, then, does faith constitute knowledge? Because what one believes by faith satisfies the conditions that are jointly sufficient and severally necessary for warrant. If the degree of warrant (which, given the satisfaction of the above conditions, is determined by the firmness or strength of belief) is high enough, then the beliefs in question will constitute knowledge.323323   On the account of knowledge given in WPF. I leave as homework the problem of showing how to modify the model in such a way as to accommodate the other main accounts of warrant.

« Prev C. External Rationality and Warrant: Faith is… Next »
VIEWNAME is workSection