On 1/14/07, K Randolph wrote:
> It was when I digitized my notes from dictionary margins that I was
> struck by how seldom the sin and shin appeared in homonyms, as seldom
> as if they originally were one phoneme, not two as I had been taught. [...]
> This study backed up that observation.
How does the study back up your observation?
I point out, again, that I do not accept that phonemes must occur at a
similar percentage relative to other phonemes. That is, that in order to
differentiate bat and hat as different words, /h/ and /b/ must be found in
the language at the same percentage as any other two phonems such
as /k/ and /p/. However, I am still curious as to how Peter's "study"
backs up your observation.
Yitzhak Sapir
Archive powered by MHonArc 2.6.24.