Welcome back to "Doggerel," where I ramble on about words and phrases that are misused, abused, or just plain meaningless. This version updated, following Dave's largely civil commentary.
Many woos of all stripes talk about having "faith" in their preconceived conclusions. The most vehement promoters of this doggerel are, of course, the fundies. How they use the term is seldom consistent.
First, many refer to their preconceived conclusions as their "faith." The problem with this is that a conclusion must have a line of evidence that leads to it. Calling an unsupported conclusion "faith" is essentially an admission of jumping to said conclusion. For some reason, however, they expect us to nod and smile politely, rather than question. That's some double standard. Welcome to Moonside!
Second, they sometimes refer to "faith" as a means to determine the truth. If that's the case, they should explain why it leads to so many contradictory results. Science usually doesn't have that problem. When experimental results do conflict, the experiment is refined and repeated until an explanation can be reached for the results. If science is wrong about something, we can tell: All scientific theories are falsifiable. "Faith," as far as I've been able to tell, doesn't have that feature. If you've got the wrong one, there'd be no way to tell.
Third, they sometimes talk about "faith" in a way more consistent with the word, "hope," as in, "I hope that X is true." In this form, they're dealing with wishful thinking. A little wishful thinking isn't bad. When you base your life on wishful thinking, it's a very bad thing. I'd also rather struggle for confidence or certainty than live in hopeful ignorance. If there's something about reality I don't like, I'd rather learn about it: Identifying a problem is the first step in solving it.
So, let's review: Woos, and especially fundies use faith to arrive at their faith and have faith that they aren't wrong about their faith. Got it? No? Well, you will. Have faith.
In many cases, some people use "faith" in a manner more consistent with "confidence" and "trust." A scientist can claim that he has "faith" in the Big Bang, but he's more likely guilty of poor word choice than he is of religiousity. A scientist confident in a theory or a hypothesis has evidence to support his conclusions. Faith is uninvolved, or at least unnecessary. For it to be a matter of faith, there would need to be either an absence of evidence, or evidence that contradicts the hypothesis.
I don't have faith that the sun will rise tomorrow: I have confidence. All the experimental data we have supports the heliocentric model of our solar system thus far, along with the sheer repetition. I don't have absolute certainty, but 99.99999999999%+ confidence will do.
Some woos and fundies, however, like to deliberately confuse the matter.
See also: Skeptico's take on this equivocation of the term.
Ver. 16.1, Updated 7/11/2006