Re: Proto-Semitic (was Re: markjjones@HOTMAIL.COM)

On Wed, 9 Mar 2005 03:38:28 +0100, Steven Williams <feurieaux@...>
wrote:

>--- Rob Haden <magwich78@...> wrote:
>> Sorry, what's the difference between laminal and
>> apical?
>
>I believe apicals are pronounced with the tip of the
>tongue, while laminals are pronounced with the broad
>surface of the tongue. English and German both have
>apical [s] and [z]; I believe some dialects of Spanish
>have a laminal [s], which gives the silibants a
>'lispy' feel. Basque, IIRC, tells between both laminal
>and apical [s] as phonemes.
>
>'Retroflex' can be an extreme case of apicality, by
>the way. In one conlang sketch, I had an apical velar
>fricative; a 'retroflex' [x], with the tip of the
>tongue approaching the velum (I have a very flexible
>tongue). Can't think of any natlangs with that
>phoneme.
Thanks for the explanation. :)

>> If there was */ts/, */dz/, and */s/, then we could
>> say that */K/ became /S/ in Arabic (rather than
>> merging with /S/).
>
>For my current conlang, I tried to get away with using
>/shiyn/ for [K], since IIRC, it was possibly
>[K]anyways at one point. Didn't 'look' right, so I
>decided to use /saad/ (using the logic that /saad/ is
>a mere variant of /siyn/, which is an easy and typical
>step to make, since I don't expect the speakers of my
>language to be expert phonologists).
>
>> However, the South Arabian languages have both /K/
>> and /S/.  Perhaps */s/ > /S/, */ts/ > /s/, and
>> */dz/ > /z/ there.  But what would cause /s/ to
>> become /S/?
>
>Palatalization? But in a root language like the
>Semitics, it would mean that [s] and [S] would
>alternate; it would stand to reason that other
>consonants would fall under the influence of
>palatalization as well; i.e., [z] becoming [Z] or [dZ]
>and so on.
Right; and from what I know, we don't see that.

>I feel like I'm beating a dead horse when I say that
>laminal [s] can easily become postalveolar [S]; that's
>what happened in German, at least.
>
>> I don't mean to sound pretentious, but I wonder if
>> the traditional interpretation of written Akkadian
>> is a little incorrect.  Not only does Akkadian show
>> |š| in the S-stems, where Arabic shows ?- and
>> Hebrew h-, but it also has |š| in the personal
>> pronouns: šu: 'he', ši: 'she' (cf. Arabic
>> huwa 'he', hiya 'she').  So the question is, did
>> Akkadian retain an earlier /S/ where Arabic and
>> Hebrew did not?
>
>Good question. I'm going to have to look into that;
>like I don't have enough to think about already.
>Today, at work, I was trying to figure out what an
>object with negative mass would be like. I lead a very
>rich inner life :).
LOL!

>The alternation between [h] in Hebrew and [s/S] in
>Akkadian makes some sense ([s] leniting to [h] is
>pretty common in languages, IIRC), but how does Arabic
>get away with [?]?
>
>Then again, it could be that Arabic dropped [h] under
>certain circumstances (in most dialects, it's a voiced
>glottal approximant, and therefore, very weak indeed)
>and replaced it with [?], since words in Arabic cannot
>begin with an empty onset.
It seems that /h/ was dropped in morphemes.

>I heard it posited that the definite article /al-/ was
>/*hal-/ at some point in both Hebrew and Arabic, and
>the two langauges just dropped a different phomene in
>each of their respective cases. Makes some sense, at least...
Is the definite article reconstructed for Proto-Semitic, or just 'Central
Semitic' (i.e. Canaanite and Arabian)?

- Rob