Clutch

First problem: Like Ockham’s Razor, this is a methodological guideline, and not a conjecture or hypothesis itself. Like “Drive on the right” or “Floss regularly”, it cannot be confirmed or disconfirmed.

Well, suppose you rephrase it as an empirically supportable generalization. Notice that you owe an account of what counts as simplicity, if you claim to adduce empirical support for the principle. And simplicity is, notoriously, not simple. Does simplicity allude to the number of laws, kinds, particulars, instances, events, or steps in argumentation? Does it include consilience? But never mind. Let’s grant that science has followed the principle quoted above, and succeeded as a result, and hence that this success *confirms* something like the claim, “Following the Principle of Parsimony leads to scientific success.”

The question is, does this require, or even suggest, some sort of supernatural explanation of a surprisingly simple universe? And the answer is, No.

We don’t actually use Parsimony (I’ll use this label to mean the methodology) to pare away the infinite class of logically possible explanations for some phenomenon. That’s not how hypothesis/theory formation works; we generally only get round to formulating a few competing theories, and it is between these that Parsimony chooses, and within this typical arrangement that Parsimony leads to success. So it is important to understand just how much is built into the notion of two or three theories counting as mutual *rivals* of the sort that parsimony is useful to differentiate. Coming up with just a few theories (out of all those conceivable), each with a practical shot at becoming our working hypothesis, is a process implicating vast antecendent information. Scientific application of the notion of simplicity includes implicit reference to what we already believe to be warranted; of any two candidate theories, both will at least consistent with a wide class of propositions from that domain of discourse and many others, and will probably entail and be entailed by many such propositions as well.

In short, by the time we get round to weighing two theories by simplicity, we have already brought to bear a great deal of knowledge of how the world works; so the odds of *either* being approximately correct are disproportionately good.

Think again of some empirically plausible rephrasing of Parsimony, say:

(S) “Following the Principle of Parsimony leads to scientific success.”

The general success of science, based upon its allegiance to Parsimony, does support (S). But (S) doesn’t say that the world is surprisingly simple. It says that a decision procedure between competing theories is most successful if it begins with the simplest candidate. And this is hardly mysterious, since in context the simplest theory is, as we’ve seen, not the simplest theory tout court. It is really the simplest theory capable of explaining the phenomenon while taking into account what we generally know about how this case, similar cases, and things in general, work. Since all the candidates will be explanatorily *sufficient*, Parsimony amounts to the stricture not believe more than you have evidence for. (Don’t postulate aliens who visit Earth if you don’t have to, for instance.) Is it surprising that science succeeds by enjoining its practitioners not to believe more than they have evidence for? Absolutely not; and the more of scientific realist one is, the less surprising it is.

GunnerJ

Actually, Tercel, there is nothing at all similar about the way you find no merit in Koy's reply as compared with how he finds no merit in your article; he gives a reasoned demonstration of why your article has no merit, you merely assert your opinion.

Koyaanisqatsi

Don't get all snippy, Terc ol' bean. Nothing of merit to your arguments (as I demonstrated), not nothing of merit to you personally.