[b-hebrew] Proto-Semitic, was WAYYIQTOL
bsr15 at cantsl.canterbury.ac.nz
Mon Mar 15 16:07:00 EST 2004
>To distill his three page definition to a paragraph, science can deal
>only with phenomena which can be observed, observed repeatedly. Any
>phenomena which cannot be observed, or observed only once but the
>observation cannot be repeated, cannot be studied as a scientific study.
>In a series of observations, scientists look for patterns, which are
>stated as hypothesis and after more testing (by observations) can become
>theories. In researching the library in physics, chemistry and other
>biology texts, I found none that contradicted this definition.
I suggest you have a look at Baysian statistics.
I've followed the arguments over the WAYYIQTOL and Baysian analysis
might help the statistical evaluation of the linguistic data. Essentially
you summarize your current knowledge or beliefs in the prior and then
use the new observations from your data to evaluate and/or modify the
prior. It might be something Rolf and Vincent might find useful.
Although Baysian statistics has been around for a long time
it has not been very useful because it is computationally expensive.
But with computing power being cheap today Baysian analysis has become
a very worthwhile companion to classical frequentist statistics.
Bill Rea, Information Technology Dept., Canterbury University \_
E-Mail bill.rea at canterbury.ac.nz </ New
Phone 64-3-364-2331, Fax 64-3-364-2332 /) Zealand
Unix Systems Administrator (/'
More information about the b-hebrew