Wseeb says: If there were no exceptions, nobody would talk about a rule. You don't say: "the rule is for people not to walk through walls."
Ah, but surely here's another confusion of the two senses of rule. There certainly is a natural rule that people cannot walk through walls. If there were an exception, our rule would be sorely probed indeed.
On the other hand, a prescriptive rule is one regarding moral. or legal. behaviour. Again, in this case, any exceptions do not prove the rule, but violate it (often, in an ideal world, leading to punitive action against the rule-breaker).
There are, of course. more general rules in science and other studies, that attempt to outline 'natural laws' but which may, indeed, have exceptions. The simple 'law of averages' for instance, would lead you to believe that if you toss a coin a million times it will come up heads or tails in approximately equal proportions. But if Rosencrantz and Guildenstern are on their way to being dead, it may come up heads only, for as many as a million times in a row. This exception certainly tests, or probes (I like that etymology - thanks Faldage and NickW) the rule, and allows us to be circumspect about its application.
In no circumstances does an exception actually prove (in the modern use of the term) the validity of a natural law.
As to the word 'prove' meaning test - we still have (albeit relatively archaic) terms using that meaning. Proving grounds for armaments are testing centres. And alcohol levels of 'proof' also relate to the alleged practice of 'proving' that a particular liquor contained the requisite levels of alcohol by seeing if it would burn (which it won't at less than about 40% v/v). This is why 40% v/v is 'one hundred degrees' proof - it simply passes the test of being able to be used as a fuel. From there, of course, you can then downgrade the standard and talk about 50 proof - meaning half the alcohol levels that will enable the old proof to be successful. And so on. Of course, in these standardised days, % by volume is more common, and easier to use. But it does raise the interesting point - 250 proof would, presumably, be pure alcohol.
Can anybody confirm or debunk this story of mine (which I seem to recall reading many cycles of the Sun ago) about how HM Customs 'proved' the quality of brandy the French were sending us?
cheer
the sunshine warrior
