Induction and Inertia

[John Dailey]

Darrell:

~ In your post #67, where you discuss what I now call 'The Swan Problem', you bring up the idea of looking for a 'cause' of swan-color(s) (or 'future' ones) re studying and understanding the swan-genome. Unfortunately, this begs the question, since the very idea of 'cause' is an induction/inference (ie: NOT a sensorial-perception, as are the swans) itself, no? Hate to rain on your arguments, but, these swans are really killer-swans in this subject of induction...since all perspectives to argue against are Humean/Russellian/Popperian. I'm not sure if an 'argument' can really ever suffice...for followers of those guys. After all, as Hume says, there's no tag or label there saying 'cause' on it for them to (literally) see.

LLAP

J:D

PS: A great summary (long, if you read the links therein) of his views, for those not too familiar, can be found here

Edited October 8, 2007 by John Dailey

[John Dailey]

Darrell:

~ In that same post, re your argument about 'arguments' THEMSELVES as being a 'test': fascinating. We're not talking 'emiricalness' anymore, yet, all such (Popperian/whoever) do fall under this idea of arguing against the worth of induction. Well argued!

~ A side thought: has 'falsifiability' been itself tested as being a 'verifiable' criterion for determining the truth/falsity of a conjecture, or, has it merely been conjectured (aka: 'argued'/assumed/asserted) as such? The obviousness of its use seems unarguable, yet, that is merely nothing more than an intuited conjecture, no? I'll have to read more Popper to see how (or if) he rationally justifies his...point (beyond mere Humean/Russellian 'intuition' [aka feeling] that "so far, it's worked that way, so...let's hope it continues to".)

LLAP

J:D

[Mindy Newton]

[quote name='Daniel Barnes' post='37144' date='Oct 3

This is an appealing argument in a sense. If each expert has some probability, say 90%, of finding fault with the theory, then 10 experts would have a probability of only 0.1^10 = 0.0000000001 of not finding fault if one existed.

You can't calculate probability that way unless each expert's "testing" were independent of the others. Experts have mostly overlapping knowledge, so they represent colinear measures.

--Mindy