Friday, February 18, 2011

Kant on what logic is possible, in distinction to Hume

I'm going through the Critique of Pure Reason, and here are some thoughts from my reading today on the section about non-transcendental analytic and dialectical logic.

Kant, not directly by name but indirectly, criticizes Hume's idea that a person can't make conclusions based on empirical evidence, except in a very tentative way based on previous generalizations, by implying that logic itself would have to be invalid for that to be the case. What we're talking about when we talk about empirical conclusions, generalizations, are logical statements about the relationship of one empirical fact to another. Kant's argument suggests that while one can't absolutely say that seeing one billiard ball hit another means that the first one caused the second one to go, neither should we dismiss the possibility that that was so. Instead, we should recognize a diversity of possible logical relationships between two facts, that are then demonstrated by experiment. And going beyond that, saying that it's difficult to generalize between two facts is not the same as saying that there isn't any potential logical connection between them, even if the actual connection is obscure and not obvious.

The difference, though, is that positive logic, that is logic that says that something is true for some reason as opposed to saying that something is not true, is always speculative and needs verification, yet that doesn't mean that we can't work with logic based on examining previous observations in order to come up with possible new connections that may work in upcoming observations.

If many logical connections between two objects are possible, then it should also be possible to combine generalizations from previous experience with possible logical directions gotten from analysis, and then test to see whether these are true or false, in order to produce new knowledge.

All of this is based on the idea that anything that is can be expressed in the form of some logical relationship to another thing, even if that relationship is very, very, complex and almost infinite.

No comments: