Atheism has recently reasserted itself into the public consciousness through a bellicose flurry of books - Richard Dawkins' The God Delusion(2006); Daniel Dennett's Breaking the Spell: Religion As A Natural Phenomenon(2006); and Christopher Hitchens' God Is Not Great: How Religion Poisons Everything(2007) are a few examples - with an effect on the religious world not unlike that of Hitler's forces smashing through the Allied lines in the Ardennes during the famous Battle of the Bulge.
In its scientific form this atheism can be described as "metaphysical naturalism" and views science as the only pathway to knowledge. Faith, on the other hand, can only produce belief. Knowledge, being scientifically verifiable produces certainty, while faith/belief, which is by nature unverifiable, is uncertain at best, delusional at worse. While this dichotomy makes for powerful rhetoric, it does not seem to hold up to scrutiny. On the contrary, faith makes up a huge part of that knowledge which science claims to produce.
In the first place, it should be recalled that science is inductive in its approach. That is, it reasons from the particular to the general. A fish removed from the water dies. Another fish removed from the water also dies. In fact, five out of five fish we take from the water die. A general theory can be proposed that all fish removed from the water die. We've moved inductively from the particulars of five individual fish to a general conclusion that all fish removed from the water die.
But we should also realize that inductive reasoning yields only probability rather than certainty. It is likely that all fish removed from the water die, and the probability of that being true increases relative to the size of the experiment. If I left off after the first two fish, the probability that I was right would be very low, relative to the number of fish in the world. But if I was after a high probability of being right I would have to increase the number of fish pulled from the water - again, relative to the number of fish in the world.
Absolute certainty could only be achieved if I pulled every fish in the world out of the water and see if they'd die. Once I had done that I could conclude my observations with certainty; my theory would be established fact. On the other hand, one exception would disprove my theory or at least cause me to revise it. (By the way, I think there is a species of fish that has physical features which enables them to survive out of water).
This kind of investigation is very practical and works (produces useful knowledge) in our everyday world. One can see it would be very impractical if not impossible for science to try and achieve certainty by testing a theory exhaustively. I haven't the time or ability to pull every fish on the planet from the water - not to mention the resulting extinction of fish would render my theory moot.
However, two limitations of empirical investigation (science) should be evident here. First, it is limited to the range of our sense perception. It's one thing to investigate the fish in the very limited sample of my backyard pond, but that is the extent of my knowledge. Second, given how small a sample of the Earth's waters my backyard pond represents, the probability of my conclusions about fish being certain on a global scale (let alone universally) is astronomically small.
Therefore, for scientists to assert the factuality of their theories beyond the laboratory (where they do have direct empirical data) to a global or even universal extent requires them to rest on something other than the empirical method. In this case they must rely on faith - faith that their conclusions will hold true in regions where empirical verification hasn't or can't be done.
For instance, to continue the fish experiment, though fish on Earth may die when removed from water that is no guarantee that they do so on planets across the galaxy. What fish do in other worlds when removed from water is literally beyond the reach of scientific verification and my convictions can only be based on by belief that all things being equal they too would die. But notice the word - "belief". I cannot know it empirically. I have to trust (trust!) that what knowledge I've gained about fish on Earth applies also to fish on planets 15 billion light years away.
Now comparing the amount of knowledge I have empirically about fish on Earth with the amount of knowledge I hold by faith regarding fish throughout the galaxy, my empirical knowledge is dwarfed in size compared to the amount I hold by faith!
