Words have meaning, and if we are going to communicate with each other we need to make sure to use words as carefully as we can. Otherwise, misunderstandings abound. It seems very common that a word like "faith" is used by different people for different ends, and the definition shifts even within an argument. Take for example, this video:
Here there is a distinction drawn between "faith" and "belief", using an analogy of a roller-coaster - belief in the ride being safe vs trusting it being safe enough to ride on. As with many things, I find it far more useful to describe terms in probabilistic vocabulary. Let's start with belief.
We say we believe a proposition A when P(A)>0.5. We say we believe strongly in a proposition A when P(A)>0.95 or some other, somewhat arbitrary, high number. The strength of a belief is a scale.
What is knowledge? I've heard philosophers give the following definition of knowledge:
Knowledge is justified true belief. [note: bad definition, IMO]
I am not satisfied with this definition because, it seems to me that, in order to justifiably label anything as knowledge with this definition we'd need to be able to independently determine that the proposition is true. This presupposes that there is some "outside" knowledge, which I feel comes too close to assuming a religious justification. I do believe that there is probably a truth to be known, but that we can never truly know what it is for certain - but this is not a problem. It is a red herring to bring up 100% certainty for knowledge, because it is never achievable, and isn't what we practically call knowledge. I prefer a definition inspired by Stephen J Gould:
In science, 'fact' can only mean 'confirmed to such a degree that it would be perverse to withhold provisional assent.' I suppose that apples might start to rise tomorrow, but the possibility does not merit equal time in physics classrooms.
Where it says "fact", read "knowledge". Where it says "science" read "life". We label things as knowledge in our lives when they are "confirmed to such a degree that it would be perverse to withhold provisional assent." Thus, P(A)>0.9999 \~ 1. Notice that we don't need 100% certainty to claim knowledge, and that it is possible for the "knowledge" to be wrong.
What about faith? Is faith "belief without evidence", as atheists like to suggest? Is it "trust", as in the video above? We could start with the Biblical reference,
Hebrews 11:1, “Now faith is confidence in what we hope for and assurance about what we do not see.” [NIV]
which I read as a synonym for "wishful thinking". Others may read it as "trust". Still others as "belief without evidence", and perhaps there are more. The problem here is that it doesn't have a clear definition, so it fails in its usefulness. What are people really claiming here?
Let's go back to the roller-coaster example. I would submit that, in this case, strong belief must be a prerequisite to "faith" in the sense he's using it. If you don't believe the roller-coaster is safe, or you don't believe the roller-coaster exists, then you cannot have trust it in. Once you believe it is safe, do you trust it to ride? This brings in decision theory, where we mix probabilities with utility measures. You could believe it to be safe at the P(safe)=0.9 level, but still not trust it "with your life" because of the cost associated with being wrong. So trust requires both belief and a sufficiently positive net utility. Placed in these terms it is much more clear how the argument is set up.
- when the religious say that "faith" is like "trust", they are already approaching the problem with strong belief, and are assessing utility - and they rightly claim that belief is not enough.
- when the atheists say that "faith" is "belief without evidence", they are addressing the strength of the evidence to obtain strong belief - and saying that it is not sufficient.
- when the religious say that "faith is rational" they are either talking about utility, not belief, or they are claiming that the evidence is in fact good enough for strong belief, and then consequently high utility.
In all cases, it seems as if for the religious, utility and belief are muddled when using the word "faith". For the atheist, "faith" is always about belief. Perhaps if we ditch the term "faith" altogether and focus on what we're actually claiming it would make communication a lot easier!