? Why "Burden of proof" ?

[X99Lynx at aol.com]

In a message dated 8/10/99 10:03:06 PM, ECOLING at aol.com wrote:

<<Rather, the real null hypothesis is something like
"We do not know whether all languages are related
(or whether there was polygenesis)" (etc.)>>

Just a note that the null hypothesis is a technical term.
It is taken as a requirement of scientific testing.  In a basic sense, it
simply means that you cannot prove something true if you cannot also prove it
false.

The classic example is the postulate: everything in the universe
is expanding at a constant rate.  (Not just cosmology, but also lightwaves
and rulers and anything that we would use to measure this expansion with.)

We can't think of a way of disproving this statement, because all
the means of disproving it have been eliminated.  Since it can't be
scientifically disproved, it can't be scientifically proved.

Now the same problem exists with the statement that all languages
are related.  Short of going back in time, how we can we prove the negative?
If we prove that two languages are totally dissimilar, it still does not
prove they are unrelated.  No way to test the null hypothesis.  No way to
prove the hypothesis.

<<"We do not know whether all languages are related
(or whether there was polygenesis)...">>

Short of hard scientific proof, we do have some evidence pointing us in one
direction or the other. First off, we can imagine a world where everyone
speaks the same language, like on StarTrek. Absent other evidence, that would
strongly suggest monogenesis.

If everyone spoke an IE language, you'd have a pretty good case for
monogenesis, even if Albanian isn't that close to English.

But the world we live in is not monolingual or even close. And, historically,
we can't even be sure if humans had language before they dispersed to their
respective corners.

Finally, consider how things would be different than they are now
if 'polygenesis' were the case?  Would we have even more languages?  Would
they be even more different than they are now?  Would we have any stronger
comparative proof of polygenesis than we have now?  What would that proof be
like?

Wouldn't a 'polygenetic' world pretty much look the way things look now?

Given the historical and modern situation, it does feel like polygenesis
might be more likely - though that's not provable either.

Regards,
Steve Long