Quote:
Originally Posted by geonuc
The thread may be 'done', but I'm a bit slow on the uptake. Not that I have anything profound to add to the commentary already presented.
But it seems my original confusion as to the topic might have been justified. Y'all are talking about two things: 1) does a hypothesis/statement/whatever have to falsifiable to be within the realm of science (Popper's assertion) and 2) what is the role and effect of falsification.
As to 1), I think Popper is correct. If a statement is not falsifiable, it's not scientific. That doesn't mean it is useless or beneath discussion, just that maybe it should be handed to the philosophy department.
2): I'll subscribe to Ken G's initial post.
Thanks for your patience. I'll go back to rocks now.  |
There's a very important distinction that has been bubbling along through many posts in this thread, but not once (I feel) brought clearly into the open:
"falsifi
able", as in "{hypothesis Y} is falsifiable" vs "falsifi
ed", as in "{hypothesis Y} has been falsified (by observations {X})".
The last ~30 years of physics history (at least that part of it that has to do with String Theory anyway) notwithstanding, a good test of whether a hypothesis is scientific or not is whether it is falsifi
able, at least in principle; for example, could you do an experiment (in principle) the results from which could rule out the hypothesis? If so, then you're doing science (at least in this regard). Now, historically, whether this rather bland test had to wait for Popper to come along and give it a name, or whether it was an essential part of the working life of just about every scientist anyway (well before Popper was born to boot), may be an interesting question ... but it's not what I myself am particularly interested in.
I started the thread with the intent of exploring the other part - "falsifi
ed". Assuming, for now at least, that "ruled out at the n sigma level" (or similar) is essentially equivalent to "falsified", wrt a hypothesis, the interesting aspects remaining include:
* does the hypothesis so ruled out/falsified simply get dumped into the round file?
* what about "theories" and "laws"; is it meaningful to say one of these has been falsified?
* in this sense, is "falsification" (the historical process whereby a hypothesis/theory/law goes from being {whatever} to "falsified") a one-way street? MUST it be a one-way street?
One purpose of my two neutrino examples (above) is to show that falsification ("the law of conservation of energy has been falsified", for example) is not at all helpful, whether as a guide to the doing of science, or even as a thread in writing a history of science - it's entirely arbitrary, capricious, and (most damning of all) has no explanatory power.