Friday, May 07, 2010
Hume observes that our inductive practices are founded on the relation of cause and effect, but when he analyses this relation he finds that all that it is, from an empiricist point of view, is the constant conjunction of events, in other words, the objective content of a posited causal relation is always merely that some regularity or pattern in the behaviour of things holds. Since the original problem is that of justifying the extrapolation from some past regularity to the future behaviour of things appealing to the relation of cause and effect is to no avail. Since it is logically possible that any regularity will fail to hold in the future, the only basis we have for inductive inference is the belief that the future will resemble the past. But that the future will resemble the past is something that is only justified by past experience, which is to say, by induction, and the justification of induction is precisely what is in question. Hence, we have no justification for our inductive practices and they are the product of animal instinct and habit rather than reason. If Hume is right, then it seems all our supposed scientific knowledge is entirely without a rational foundation.
- James Ladyman, Understanding Philosophy of Science
"The philosopher Nicholas Maxwell...insist[s]that there are even stronger grounds for justifying modern scientific practice. He suggests that a fundamental assumption underlying science is that the world is comprehensible that it has a certain internal consistency. In other words, the phenomenal world we experience is not wholly chaotic and lacking in pattern, we can find out about it, we can understand how it works. Given this is a central assumption for doing any kind of science at all, he suggests that the most rational way to find out whether or not it is true (and hence whether or not we can really do science)is to assume it to be true and see how far we go. If we find that we do obtain sensible mutually consistent theories that work, then we can be fairly sure that the assumption was indeed true and science is possible; if we fail to obtain consistent results then we will know the assumption is false (and hence the whole enterprise of science is pointless) He argues that we have more chance of finding out whether or not we are right about science by proceeding in this way than by agonising in our armchairs over logical possibilities the way so many intellectuals have tried to do."
-Robin Dunbar, The Trouble with Science
[W]hen we ordinarily use a term like rational we are taking it to have some normative (or prescriptive) as well as descriptive content. In other words, we suppose that reasoning is rational because it conforms to some sort of standard and that it is the sort of reasoning that will tend to lead us to truth and away from falsity.
Merely being called 'rational' is not enough to make a mode of reasoning justified, for it does not establish that the reasoning in question has the other properties that we take rational reasoning to have.
In short, Maxwell's line of argument "reduces" science to a kind of "rough-and-ready" investigation of the world with not much more to recommend it than what chiropractors and homeopaths appeal to.
And how do we judge what "consistent results" are? For over 200 years the epitome of science was Newton's mechanics but they were, in significant part, wrong. Indeed, if you look at how much science has changed over the last 100 years (no matter how hard you try to sweep it under the rug as "refinements" of prior "knowledge"), its hard to see how we can justify any special status for science.
My "solution" is to say that, based on my impression of how the world works, science is (by far) the best sort of knowledge that we have. You may get different mileage but, if so, leave science to go its own way and let others fairly form their own impressions by the outcome.
I don't use the scientific method because of high philosophical ideas. I use it because it has been shown time and again that it works with a relatively high degree of certainty.