Reason as a Way of Knowing

Reason is the way in which we console ourselves that the universe is congruent with our brain's model of it. It seems to work. But it is not so much a way of knowing as a way with which we define a certain kind of indirect knowing.

Reason as most people have it is apparently the linking of facts and ideas so that they appear to meet certain tests. Knowledge based on reason will generally meet the tests of validity and reliability, transferability and generalisability, and perhaps the tighter tests of coherence and congruence demanded by pure logic.

Two main processes operate in what we call reason: induction, which is based on experience and empirical knowledge; and deduction, which is based on rules and axioms. The former assumes that if we have observed prior cases, we can use those cases to formulate rules and generate norms; the latter assumes that if we have rules, we can use those to classify and determine (or delineate) the cases we are presented with.

The provisional conclusions we must draw are called inferences; the possible conclusions we may draw are called implications. This roughly outlines the concept of reason, and also fixes the boundaries of the kind of knowledge it can provide.

Reason suffers from two problems: simply put, deduction requires rules and axioms, and we assume these are right before we apply them; induction requires a critical number of cases, but this is not always known, and we do not always know what cases are relevant. In theory, both should cancel each other's defects out. In practice, this is not always so.

Nevertheless, apart from even more abstruse considerations concerning the nature of reality, reason (by its own self-defined coherence and consistency) is the best bet we have of making useful sense of the world — which doesn't say much, but says enough.