Originally posted by convect
What do you mean when you say the word, "faith"? What is faith to you, and what role does it play in your life?
It's only fair that I answer first, so here goes: I regard "faith" as "belief in something whether or not I know it to be true." It starts from that nugget, and then the belief becomes encompassing, and I "know" (i.e., I feel like I know, and ...[text shortened]... to convert us, and, please, atheists, let's not try to convert them, either.
Epi and I had a nice discussion about this very topic not long ago. I wrote the following, which perhaps you will find interesting:
It is difficult for me to be clear about all this, primarily because I rarely use the term 'faith' and am typically unsure of the intended sense of the term when I hear it used by others. Further, there is a fine line between excavating and stipulating the sense of terms used colloquially. So, what follows is largely vague and exploratory, so please bear with me, and begins with natural language considerations. Now, many terms that refer to psychological states or properties are not used univocally. Think of all the ways the term 'faith' is commonly deployed in speech. Consider the following handful of examples:
1) S has faith that P.
This reads as though faith is a type of propositional attitude, similar in structure to belief, desire, fear, etc. in that faith is a stance one can take with respect to some proposition. I could believe that it will rain, desire or fear that it will rain, or have faith that it will rain. But the difficulty here is specifying the difference between faith and the other propositional attitudes. It seems clear, to me at least, that the claim 'S has faith that P' analytically entails the claim 'S believes that P'. It would certainly be strange for somebody to claim that he has faith it will rain while sincerely believing that it will not rain. Note that this is one reason it is so hard to understand just what Kierkegaard is getting at when he speaks of the psychology of the Knight of Faith. But it is also strange, though perhaps less so, to construe faith as wholly cognitive, unconnected to conative or affective states. Typically people do not claim to have faith that P when they simply do not care whether it is the case that P. Further, people typically claim to have faith that P only when that take some broadly pro-attitude towards P. It is natural to claim that one has faith that his friends will treat him well, but it sounds strange to claim that one has faith that his enemies will treat him viciously. But, if this is right, then faith is a mongrel state, much like optimism, possessing characteristics of both belief and desire. If faith is construed in this way, then there is no reason to think that faith, by its very nature, is epistemically irresponsible. Whether an instance of faith is epistemically responsible will depend on the evidence upon which it is partly based.
2) For S, P is an article of faith.
Unlike (1), this does not read as though faith is a propositional attitude, but like (1) it entails the claim 'S believes that P'. The term 'faith' here seems to demarcate a class of propositions constitutive of either an epistemic orientation or a normative worldview. If the former, these are the propositions that serve as epistemically basic and are taken to be non-inferentially justified (if one is a foundationalist of either the internalist or externalist variety), or those nearest the center of the web of belief and most insulated against revision urged by the tribunal of experience (if one is a holist or coherentist). If the later, these propositions are those that specify the deepest, most fundamental normative commitments one has; commitments that function as first principles, perhaps tacitly, in practical deliberation. Since, on this reading, 'faith' refers to classes of propositions specified by function within the cognitive and/or evaluative life of agents, rather than to propositions specified by either their content or evidential backing, it will be an open question whether any particular article of faith is epistemically irresponsibly held. Again, it will be the evidence that decides. Of course, settling disputes at this level will often be difficult, and the threat of begging the question in such disputes looms. In many cases the best we can do to settle disputes about the nature of epistemic justification or of the credentials of first-order moral judgments is to construct localized consistency arguments of the form "if you believe that, then you commit yourself to this implausible entailment", where we hope the implausibility of the entailment is evident to our interlocutor.
3) S takes P on faith.
Like (2), this usage of the term 'faith' is closely tied to issues of inference and deliberation, but unlike (2) this claim does not invariably entail the claim 'S believes that P'. I have heard this expression used to indicate that one has taken a proposition as a hypothetical posit; a provisional assumption for the purpose of guiding inquiry. Here, unlike the first two examples, the will is implicated in faith. Nothing in either (1) or (2) entails that S chooses to take any particular stance towards P. But here we have a use of 'faith' that seems to require choice. An agent must choose, perhaps in some limited manner, to deliberate and act as though P was the case. The term "leap of faith" is most clearly an example of this usage, but admonishments to have faith often take this sense as well, at least when they are not mere exhortations in the face of flagging confidence. Now, a case could be made that posits of this sort are epistemically irresponsible, since if one had sufficent evidence for P one would not need to posit P provisionally, and why treat the world as though P were true if there is insufficient evidence that P is true? But there is no guarantee that, for all P, the world is such that the truth or falsity of P is discoverable independently of treating the world as though P were true. Conditional proofs work this way in logic, after all. Further, since provisionally positing P does not commit one to the belief that P, it is always open to one who takes P on faith to cite pragmatic reasons for the positing.
4) S believes P based on faith.
Of course this entails that S believes P, and so it will be an open question whether this belief is epistemically justified. I hear this or similar claims quite often, and admit to being at a loss as to what exactly is meant. Often the claim is simply pejorative; elliptical for "S has no evidence for his belief that P'. If this is how 'faith' is to be construed, then of course it will follow that faith is epistemically irresponsible, since it leads to unjustified believing. But this is not the only way we can construe (4). 'Faith' here could refer to either some other propositional attitude, as in (1) above, from the propositional contents of which the belief that P is inferred, or to some actual proposition taken to be foundational or nomratively central, as in (2) above, from which P is inferred, or it could refer to some psychological capacity or state or trait that functions as a mechanism of belief formation. Although the epistemological details here matter, the upshot of this non-pejorative reading of 'faith', and the basing relation in particular, is that justified faith can itself justify belief (or, if faith is a belief forming psychological mechanism, that it may yield justified beliefs if it reliably produces true ones).