Re: The Monovocalic PIE Myth (was Germans have no /w/, ...)

william drewery wrote:
>
> This is perhapsa it off-subject, but it seems fitting.
>   One interest of mine is how quickly human-language
> phonemes began to differentiate. It seems reasonable
> to me that the earliest languages would have had
> sparse inventories due to physiological constraints
> (such as lack of control over breath, incompletely
> developed larynx, etc.) and psycho-neurological
> inability to distinguish similar sounds. But oe would
> think that once our language abilitiesemerged, we
> would have tended toward the route of !Xoo
> (considering how we constantly draw infinitely subtle
> distinctions throughout history), yet languages like
> Rotokas are alive and well today.
Because there's *also* an equally strong tendency to simplify, to merge
sounds, which, one presumes, tends to even out in the long run.  For
example, Japanese is a good example.  In the written history of
Japanese, there've been several losses of phonemic contrasts.  Old
Japanese had a distinction between two kinds of /e/ (or possibly between
/e/ and /je/, it's a controversial issue), likewise two /i/ and two /o/
(which likewise might've been something like /i~ji/ and /o~wo/, the
nature of the contrast is unimportant), those merged in the 9th (?)
century.  Likewise, Japanese has lost the contrast between the syllables
/wi~i/ and /we~e/, between syllabic /m/ and /n/, and merged formerly
distinct /dj/ and /zj/, has merged /ou/, /au/, and /o:/, /iu~ju:/ and
/eu~jo:/, and some dialects have merged /e:~ai~ei~oi/ and /ui~i:/.
Furthermore, in the related Ryukyuan, many dialects have merged /i~e/
and /u~o/.  Some dialects have also neutralized /t~d/ in medial
position, and /i~u/ after alveolar or alveo-palatal fricatives and
affricates.

It's just a matter of whether the tendency to split phonemes or merge
phonemes is stronger at a specific point in a language's history.  I
suspect that, if we could somehow learn everything about every
language's history, we'd find that they all have periods of large
numbers of phonemes, and periods of small numbersof phonemes.

Replies