>
> I don't mean to pick nits with your choice of words, but scientists in my
> experience tend to define science either by method or by its philosophical
> assumptions, if they can be bothered to define it at all. Defining science
> by its product seems circular to me.
>
some thoughts/restatement of what you are probably well aware of:
perhaps the original poster (greg?) is saying things similar to the much advertised difference between the "context of discovery" and the context of justification". given that it is difficult or impossible to demonstrate the existence of a method in scientific research/discovery (as shown by feyerabend, et al), a different way to define science is through the method of justification i.e., through mathematical description etc. so, you can come up with any claim you want (and you could make it up out of thin air or through some colourful philosophy) as long as, say, you can justify it using some accepted notions: mathematical models, experimental verification, conformance to accepted theories in science, double blind tests, whatever.
there is a strong critique of this positions (a critique completely lost of scientists like sokal and science groupies) and as far as i can tell the only definition that seems to hold up is that science is what scientists do.
--ravi