Yes. I like this. And as Rich used the example of astrology, it
directly illustrates the relevance of this to last week's debate.
I think Rich hit the proverbial nail when he says that predictive
validity is the test for a "science". It was Bandura who stated that
the value of a theory is ultimately judged by its usefulness as
evidenced by the power of the methods it yields to effect changes.
Personally, I buy this account.
Take backward causation. A phenomenon which changes the past cannot be
proved nor falsified (changing a time-line means you have followed that
time-line anyway, so who's to say that you have changed... if that
makes sense). This, for me, holds no value. The same goes for Freud's
theory - non-falsifiable. A theory which can be tested and survive -
even successfully predicting an event - now there's something worth
chewing on.
The benefit of hindsight reveals illusory correlations - inappropriate
causal attributions from one event to another. This is what
parapsychology seems to be doing, and not infrequently. Velikovskian
"scientists", for example, are hugely selective in their evidence for
the theory that the Earth has flipped over more than once in the past.
For this reason I will stick to Science with a capital "S". True, these
arguments can apply to standard phenomena (the "file-drawer" effect
occurs in Psychology too), however the maturity of judgement attained
by physics etc. validates these issues.
Having said this, I do not claim that there are many proven truths in
Science. We generally make abductive inferences based on past
observations, and this is not totally reliable (was it Popper who said
a theory can only be disproved?). Sherlock Holmes makes no assumptions
until he has sufficient data, for fear of falling into the trap of
"cognitive lockup". In the real world, the planets do not OBEY Kepler's
laws, however they reliably follow a pattern which we have applied in
order to predict their movements.
So I'm back where I started. Science may or may not find the answers.
In the meantime, I'd rather stick with what is essentially a good
working hypothesis which CAN predict things. Skinner said that science
hasn't changed man, but the chance of doing something about the subject
of analysis has.
This archive was generated by hypermail 2b30 : Tue Feb 13 2001 - 16:23:15 GMT