abduction and occam's razor

gosh i don't know if this is gonna sound obvious but i think i finally understand abduction, it's role in science, and why parsimony is so important in scientific hypotheses! okay so very quickly: three methods of logical arguments: deduction, induction, and abduction.

in a deduction, you have an argument like this:

(1) all things that have hearts also have kidneys.
(2) david has a heart
/(3) david has a kidney

in an induction, you have something like this:

(4) all of these jellybeans in my hand are red
(5) i got these jellybeans from that bag
/(6) all the jellybeans in that bag are red.

in an abduction, your argument goes like this:

(7) the glove fits
(8) if he did the crime, then the glove must fit.
/(9) he did the crime.

so in (1-3) we're applying a conditional to get the consequent, in (4-6) we're inferring the conditional from what seems to be an antecedent and a consequent, and in (7-8) we're inferring the antecedent from a conditional and a consequent.

unless you're an extreme skeptic, deduction is valid if the premises and the negation of the conclusion are inconsistent. for induction, lots of people (hume, popper) have shown that it is isn't 100% accurate at all times or try to get rid of it, but it seems to me that we use it so often and it's true most of the time, so i'll just take it for granted that, even though it's not full proof, it's okay to use (a full explanation as to why i think this would take me into phil of language and metaphysics and i just don't wanna go there... maybe i'll do it at some other time).

now the one that seems to preoccupy some people (e.g. bas van fraassen) the most is abduction (or as some people call it nowadays, inference to the best explanation), especially in the context of science.

as i see it, in science there are problems. and in lots of these problems there is the following implied dillema: "X. Why X?" an example given in words: if i throw an apple into the air, it'll come right back down. why is this so?

in short, there is a phenomenon which already has been established to occur in the world and science's job is to figure out what causes it. if you will, there's a B and we're trying to figure out an A such that A -> B. now, to get a little technical, if the B has already been established as occurring in the world, then it has a truth value of 1 (where 1= true, and 0=false). but if B is already 1, then the relation A -> B will be 1 no matter if A is 1 or 0.

here's where the scientist comes in. the scientist proposes a hypothesis, A, to see if B follows, either necessarily or probably. if it does, then the hypothesis becomes a theory supported by observation. but again, the way he does this is by investigating in the real world. if B follows from A, as hypothesized by the scientist, then we have reason to believe A, making the abductive process complete. if not, try again, with a new antecedent (A*).

now comes the parsimony part. take the previous example given by (7-9). if we amend to (9)

(9)' he did the crime AND panda express is awesome

then the argument would still be valid. because if the rule remains that (8) and we have (9)', we could simplify (9)' to just (9) and

(10) panda express is awesome.

so (7) follows. but, as you can easily see, (10) is completely unnecessary and irrelevant (even if it might be true). that's why scientific theories attempt to posit the most barren metaphysics and the least amount of stuff necessary to solve the problem at hand, all the while trying to explain nature in terms of natural kinds and laws.

damn i just decided i'm gonna tag this aesthetics too because this is just beautiful.