Epistemology is the philosophical inquiry into the nature of knowlege. How do we know what we know? How do we know that we know what we think we know? As you can see, it’s complicated just to pose the question, much less answer it.
In Mathematical Epistemology, we are exploring propositions (theorems) deducible from a set of axioms. A specific proposition is true or false according to a particular set of axioms. Propositions are not true or false “on their own merit” so to speak. The proposition that 1+1=2 is true according to the axioms of arithmetic.
An axiom set is either consistent or inconsistent. If the axiom set is consistent, then it is impossible to correctly derive a proposition that is the inverse of another correctly derived proposition; i.e. p and not p are not both true theorems of the axiom set.
Godel proved that it is impossible to use theorems of an axiom set to prove that the selfsame axiom set is consistent. Therefore, to prove an axiom set consistent, we have to create a new axiom set according to which we can prove the first consistent. But we now have an infinite regress; we cannot use the second axiom set to prove itself, and must create a third to prove the second against, ad infinitum.
We also have the “meta-mathematics” of propositional calculus which formalizes the ordinary rules of deduction; for instance “The expression p or q is true if and only if one or both of p and q are true.”
However, being an axiom set, Godel’s theorem applies to propositional calculus as well! One cannot use propositional calculus to prove its own consistency. So not only are arbitary axiom sets not provably consistent, the usual method of deduction is also not provably consistent. Needless to say, these conclusions give mathematitians, who consider themselves seekers after absolute truth, a queasy feeling, and most of them don’t like to discuss the situation.
Scientific Epistemology is somewhat more forgiving. The tests of scientific knowlege depend on predictive power (e.g. physics) or narrative consistency (e.g. cosmology). A scientific theory, an explanation of the nature and behavior of physical phenomena, is considered “strong” if it makes consistently true preditions about directly observable phemonoma. It is considered “robust” if it makes correct predictions under a wide range of circumstances.
The Scientific Method operates much like Sherlock Holmes’ dictum: “When you have eliminated the impossible, whatever remains, however improbable, is the truth.” They seek to create a falsifiable hypothesis to explain a phenomenon, and then try their darndest to prove that hypothesis false.
Scientific Epistemology also has its fundamental difficulties. The most obvious being the “hypothesis creation” problem, that Pirsig eloquently describes in Zen and the Art of Motorcycle Maintenance. There are an infinity of hypotheses that could explain a given phenomenon; how do we choose which to test according to the Scientific method?
And then there’s Quantum Mechanics. QM is most definitely a very strong and very robust theory according the basic definition of Scientific Epistemology. The fact that you can read this message at all on your computer demonstrates that every second, billions of predictions from QM are proving themselves true. The problem is that features of ordinary reality we consider fundamental, e.g. definiteness, locality, causality, do not apply to the quantum microworld. Even time itself can appear as an ordinary spacial dimension in QM. With the Copenhagen Interpretation of QM, scientists finally in effect threw up their hands and said, “we have no idea what is really going on in the microworld; QM is just a mathematical fiction to predict the results of experiments.”
It also turns out, apparently coincidentally, that almost every interesting pure mathematical system (axiom set + theorems) turns out to be a powerful way of generating theories with strong and robust predictive power. It is very unsettling to see this almost perfect correspondence between abstract creations of the human mind, and the complicated behavior of seemingly unrelated physical phenomena.
The human mind has evolved to both seek and create patterns from empirical phenomena. It doesn’t want to stop even when confronted with the limitations of ordinary knowlege. Ultimately the choice between a scientific or religious interpretation of the limits of knowlege is an aethetic choice. The religious interpretation sees the fundamental nature of reality as transcendent and purposeful; the scientific sees it as purposeless and possibly perverse. It seems that not only does God roll dice, but sometimes he rolls them where we can’t see them!