In the scientific method, Occam's razor, or
parsimony, is an
epistemological,
metaphysical, or
heuristic preference, not an irrefutable principle of
logic, and certainly not a scientific result.
[15][16][17][18] As a logical principle, Occam's razor would demand that scientists accept the simplest possible theoretical explanation for existing data. However, science has shown repeatedly that future data often supports more complex theories than existing data.
That's a rather disappointing Wikipedia entry since it misses the point spectacularly. The point is that there are many, many, many
more complex theories than simple ones - exponentially more than there are added degrees of freedom. I am simplifying a lot, but suppose that the odds that the right theory is complex are 80%. What will usually happen here is that there will be
one simple theory, accounting for a 20% probability mass and, say,
eight complex theories, accounting for 80% in
total but 10%
each. If you have a small amount of data, you cannot tell any of these nine theories apart. They all explain the current data fine. Now, if you keep experimenting, you will get new data and that new data will allow you to filter the candidates. You will probably have to reject the simpler theory as well as the great majority of complex theories. Sure, the theory you end up with will probably be complex. But the fact remains that
before seeing that new data you will have to go with the simplest theory because for any set of particular theories it is the simplest that is the most probable. You can't tell in advance how complex a theory will be and you know that any particular simple theory is more probable a priori than any particular complex theory (because for every simple theory, similar complex theories are dime a dozen). That is why
picking anything else than the simplest (working) theory at any given time is lunacy. Picking a theory doesn't mean you are absolutely certain it is the right one. You can always revise it later under the light of shiny new data - when data allows us to discriminate between two similar theories, Occam's razor doesn't even apply anymore!
Occam's razor doesn't say that the correct explanation for a phenomenon is more likely to be simple than complex. It says that if two particular hypotheses X and Y explain a phenomenon equally well, the simplest of the two is inherently more probable. A certain balance in probability is achieved with the very simple fact that complex hypotheses are
many more than simple ones, but that's frankly irrelevant.
"Outlandish supernatural claims"? Sorry but I don't care for such flagrant bias in serious discussion. This is not a debate about vampires and werewolves. There is a massive lack of appropriate technology for discerning any kind of higher being in our universe; we know nothing about our own existence because we simply are not that advanced a race. You cannot use Occam's Razor for the existence of life or the universe because there is no 'simple explanation'. "It just happened" is as complex a theory as that of an omnipotent deity.
Of course there could be a simple explanation. Perhaps a certain kind of cellular automaton initialized on a large matrix of active cells produces something that works exactly like our universe. In any case, the problem is that there are trillions of theories which explain the universe just as well as an omnipotent deity (that is,
not at all). I could make one up on the spot. Or two. When you have trillions of candidates to "explain" something, you sure as fuck should pick the simplest one for reasons I explained above. And since God doesn't explain jack shit about anything in the first place (hint: you can't explain anything without being falsifiable - no wonder why it's part of the definition of a scientific theory), you're better off with a lack of theory.
PS: what does "it just happened" even mean? I'm puzzled.
