From: gts (gts_2000@yahoo.com)
Date: Fri Apr 18 2003 - 15:40:03 MDT
Harvey Newstrom wrote:
> Great. I must have misunderstood you.
> I was under the mistaken belief that you were disputing nutritional
> science with a diet dictated by paleo history.
I'm glad we cleared that up!
> Yes, but I disagree with this method of debate. I think it
> is invalid to choose a default position which does not have
> to be proven.
But I have offered evidence that supports the paleodiet hypothesis. However,
regardless of whether you consider it proven, this is how science makes
progress. It is impossible to perform a meaningful experiment until one
first has some initial default hypothesis to test against the new competing
hypothesis.
> There is no "default" position or "burden-of-proof" in science.
I disagree completely. There is always a default position and the burden of
proof is always on those who contest it to show with statistical certainty
(typically defined as 95% confidence) that the default hypothesis is
incorrect. For example it was the default at one time to believe that the
earth is flat. One might say this hypothesis was never first proven (though
it the minds of most of the ancients it seemed perfectly obvious). It was
however the default null hypothesis.
I think what you're really arguing here is that the default null hypothesis
should be "Harvey Newstrom's personal diet hypothesis" or perhaps "The USDA
recommended diet hypothesis" or perhaps the "The diet hypothesis as
exhibited by the eating habits of the typical modern person" or perhaps your
default hypothesis is the hypothesis of some nutrition scientist you know
and respect. Certainly a nutrition-minded person like you has *some* general
diet hypothesis in mind, which in your mind must be disproved before you
will abandon it.
We must start from some theoretical foundation, and then build on it by
comparing it to new competing hypotheses. When a competing hypothesis better
explains the data than the default null hypothesis, with 95% confidence, it
wins the prize and becomes the new default working hypothesis against which
new hypotheses must be tested.
-gts
This archive was generated by hypermail 2.1.5 : Fri Apr 18 2003 - 15:47:15 MDT