[Lingtyp] Gumuz languages and Sumerian

[Thomas Goldammer]

Dear all,

I'm currently reading a very interesting chapter by Dimmendahl et al. in
the Cambridge Handbook of African Linguistics (edited by E. Wolff), *Linguistic
features and typologies in languages commonly referred to as ‘Nilo-Saharan’*.
While reading the section about the Koman/Gumuz languages I thought I was
reading a paper about Sumerian typology - especially for the Gumuz
languages. For those not familiar with my favorite subject, Sumerian is a
language isolate spoken until about 1700 BCE in what today is southern
Iraq, written in cuneiform script that was invented for this language. Of
course, it is geographically distant to where the Gumuz languages are
spoken nowadays, but not too far, separated "only" by the Ethiopian
highlands, the Red Sea and the deserts of the Arabic Peninsula. But
nevertheless, the typological similarities are pretty interesting and
definitely cry for further investigation.

I would be extremely grateful for any reading suggestions about any kind of
(typological) comparison of Sumerian and the Gumuz languages, or other
families commonly referred to as part of the "Nilo-Saharan" entity.
Unfortunately, I do not have access to any good library or to papers behind
paywalls, or to institutional journal accesses of any kind, so if you can
spare a PDF file, I'd be even more grateful.

For those who are still with me, here is what I mean in some more detail
(yay, data!)...
Dimmendahl et al. list several typological features for the Gumuz
languages. I'll skip the phonological features, as the Sumerian phonology
is subject to a lot of uncertainty and disagreement among scholars. Some
rarer features include:

1. "Typically, only human and some animate nouns can be marked for plural."
(Dimmendahl et al., p. 12, in section 11.3.3). It is noted in the chapter
that this is an areal feature in the region. The same rule is true for
Sumerian.

2. Gumuz languages have a small set of verbs that inherently mark singular
or plural of participants. Dimmendahl et al. give an example from
Daats’íin: *dugw* ‘run.SG’ vs. *ranɗ* ‘run.PL’. Sumerian has the same for
a couple of verbs, e.g. *gub* ‘stand/put.SG’ vs. *sug* ‘stand/put.PL’. This
plurality marks participant plurality.  Daats’íin (according to Dimmendahl
et al.) also uses reduplicated stems to mark event plurality - so does
Sumerian.

3. Gumuz languages use a deictic directional towards the deictic reference
point (ventive) form in the verb. Sumerian does have such a bound morpheme (
*m(u)-*) as well.

4. Noun incorporation into verb stems, especially of body part terms, but
occasionally also other nouns, to form new lexical meanings is found in
Gomuz languages. Something strikingly similar can be found in Sumerian, but
the noun is not fully incorporated. It is positioned in the S/P position
(immediately preceding the verb) and changes the lexical meaning of the
verb. Some verbs only occur with one of these bound nominals. Examples for
such complex verbs are *ĝeštug2 du3* ‘ear erect’ = ‘listen’, *šu gi4* ‘hand
return’ = ‘repay’, *ki(-ig) aĝ2* ‘xxx measure’ = ‘love’ (the meaning of
*ki-ig* is unknown, it occurs only in this lexeme). These verbs are pretty
abundant in the lexicon of the language. They also exist for making noises,
different types of speaking and similar things: *šud3 ša4* ‘prayer sound’ =
‘pray’.

Numbers in the Sumerian examples do not indicate tones. They are
conventional numberings of lexical entries with the same transcription.

Please note that I see no evidence for cognates between Gumuz languages and
Sumerian, and every such comparison would be pretty weird, anyway, given
the temporal distance between the two data sets. Neither would I dare to
claim any genealogical relationship between the two.

With best regards and greetings from Koblenz,
Thomas Goldammer, PhD.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20201129/70dc0964/attachment.htm>