[Corpora-List] Do you think LINGUISTICS is SCIENCE or ARTS?

[John F. Sowa]

The criteria for science include the ability to make predictions
that can be tested against future evidence, to explain the factors
that cause discrepancies between predictions and observations,
and to use those explanations to improve future measurements and
predictions.

YT>> But can linguists reconstruct Latin on the basis of Italian,
 > Spanish, French and other Romance languages? The answer is NO.

MM> I think the answer is partially.

That's a good example because linguistics can also explain why the
reconstruction is partial.  As examples from contemporary languages
show, written forms are more conservative than the spoken forms.

Therefore, the reconstructed language would resemble the spoken
"proto-Romance" more closely than written Latin.  It would have
a high overlap with classical Latin, but it would include words
like 'bellus' and 'caballus' instead of 'pulcher' and 'equus'.

Furthermore, those discrepancies that cannot be explained are
the basis for making new discoveries.  For example, Spanish
'queso', Portuguese 'queijo', English 'cheese', and German 'käse'
could be derived from Latin 'caseus', but French 'fromage' and
Italian "formaggio' are exceptions.  Further search would lead
to an innovation for making "caseus formaticus", which was
shortened to just the adjective in France and Italy.

John Sowa


_______________________________________________
Corpora mailing list
Corpora at uib.no
http://mailman.uib.no/listinfo/corpora