Sunday, June 13, 2004

Bill O'Reilly and the Analytic-Synthetic Distinction

The other night, on the O’Reilly Factor, Bill O’Reilly remarked, “Reagan did not want to change America, he wanted to improve it.” Call me a pedantic nit-picker, but I can’t let that pass without comment. The reason I can’t let it pass is because it is symptomatic of a trend, the tendency to override the objective connotations of words with one’s own idiosyncratic subjective connotations. Such an override is an offense against intellectual hygiene.

Surely, to want to improve X is to want to change X. Every improvement is a change, though not conversely. ‘Every improvement is a change’ is what Kant and subsequent philosophers call an analytic proposition. Such a proposition is one the truth of which is grounded in the meanings of its constituent terms. By sheer analysis of the concept expressed by the subject-term, ‘improvement,’ one can see that every improvement is a change. Compare ‘Every change is an improvement.’ This is an example of a synthetic proposition. Its truth-value cannot be determined by any amount of analysis of the meanings of its constituent terms. To determine its truth-value, we need to look beyond the proposition. One such ‘beyond’ is the world (or some more restricted universe of discourse). Let the U. of D. be the proof-sheets of an article of mine about to be published. Relative to that U. of D., it is true that every change is an improvement. But relative to the world at large, the proposition in question is obviously false.

To appreciate the difference between analytic and synthetic propositons, observe how they behave under negation. The negation of ‘Every improvement is a change’ is ‘Some improvement is not a change.’ (If you think the negation of the first is ‘No improvement is a change,’ then you need Logic 101.) The latter, though not a formal contradiction, is nonetheless incoherent. The negation of ‘Every change is an improvement,’ however, represent a coherently thinkable possibility, namely, that there be a change that is not an improvement.

Now consider ‘Every effect has a cause.’ This is clearly analytic: an effect is an effect of a cause, by the very meaning of ‘effect.’ There is nothing coherently thinkable corresponding to ‘Some effect has no cause.’ But ‘Every event has a cause’ is synthetic: there is nothing in the concept event that requires that every event have a cause. If we analyze event, we see that events are occurrences, that events are temporally locatable, etc., but not that they have causes.

Given that ‘Every event has a cause’ is true, what could make it true? (If you think that the proposition in question is not true given certain quantum events such as a photon’s undetermined passing through slit A rather than slit B in the Two-Slit experiment, then replace ‘event’ above with ‘macro-event.’) Kant said that experience (Erfahrung, experientia) teaches us what is the case, but not what must be the case. If that is right, then it would seem that the truth-ground of ‘Every event has a cause’ could not be experience. It seems plainly false to say that we know it from experience of contingent matters of fact, by induction from individual cases. We don’t examine events one by one and note that each one so far examined has had causal determinants. Nor would we take anything as refuting the principle in question. Take the explosion of the space shuttle, Challenger. No one would seriously entertain the supposition that that macro-event just happened without cause. If you had a starting problem with your car, and took it to a mechanic, and the mechanic said that your car’s engine represents a counterexample to the causal principle, you would not throw out the principle but find a new mechanic. In Kant’s jargon, the principle in question is synthetic a priori. Its truth-ground does not lie within the proposition, but neither does it lie in experience of contingent matters of fact.

How are synthetic a priori judgments possible? If you understand this question, and what motivates it, then you are in a position to understand the theory Kant develops in the Critique of Pure Reason. But if, like Ayn Rant, you do not understand this question and what motivates it, then you will most likely take Kant’s theory as a bunch of gratuitous assertions designed to “destroy man’s mind” or some such Randian nonsense.

But to return to O’ Reilly and sum up. His remark that “Reagan did not want to change America, he wanted to improve it” is a distortion of the meaning of ‘change,’ one that violates the analyticity of ‘Every improvement is a change.’ He and others do this with other words like ‘philosophy’ and ‘religion.’ But that’s a topic for later.