Syntax borrowing

[Frislander]

Eh, I'm skeptical about the "Ethiopian Sprachbund" in its current formulation. As far as I can tell it seems to be posited as an explanation for why Ethiopian Semitic languages are more like Cushitic syntactically that Asian Semitic which is presumed to represent the "Semitic Prototype", but otherwise has little to define it as a sprachbund. In particular I would like to point out the general lack of phonological parallelisms to go with the apparent morphological ones, and the ones that do exist are mainly in the "wrong" direction, Cushitic languages becoming more Ethio-Semitic like rather than vice-versa, the exact opposite of what we'd expect from a supposed sprachbund where the main direction is Cushitic-to-Ethiosemitic.

[Vijay]

All I know about Afroasiatic is that none of the Semiticists I personally know seem to buy it.

[Whimemsz]

Eh, I'm skeptical about the "Ethiopian Sprachbund" in its current formulation. As far as I can tell it seems to be posited as an explanation for why Ethiopian Semitic languages are more like Cushitic syntactically that Asian Semitic which is presumed to represent the "Semitic Prototype", but otherwise has little to define it as a sprachbund. In particular I would like to point out the general lack of phonological parallelisms to go with the apparent morphological ones, and the ones that do exist are mainly in the "wrong" direction, Cushitic languages becoming more Ethio-Semitic like rather than vice-versa, the exact opposite of what we'd expect from a supposed sprachbund where the main direction is Cushitic-to-Ethiosemitic.

Why would we a priori expect the features in a Sprachbund to all converge on the features of one of the language groups, as opposed to the different groups borrowing various things from each other? I mean what you basically just said is: ES languages are more like their neighboring Cushitic languages syntactically than other Semitic languages are, and Cushitic languages that neighbor ES languages are more like ES languages phonologically than other Cushitic languages are. That sounds like a decent case for some sort of Sprachbund to me!

(But let me admit here that I don't know much about this topic specifically, so by all means correct me if I'm off-base...)

[mèþru]

There is a big bias among Semiticists for data from Central and East Semitic, so the idea that maybe South Semitic better reflects the original grammar in at least some regards is underappreciated.

[Frislander]

Eh, I'm skeptical about the "Ethiopian Sprachbund" in its current formulation. As far as I can tell it seems to be posited as an explanation for why Ethiopian Semitic languages are more like Cushitic syntactically that Asian Semitic which is presumed to represent the "Semitic Prototype", but otherwise has little to define it as a sprachbund. In particular I would like to point out the general lack of phonological parallelisms to go with the apparent morphological ones, and the ones that do exist are mainly in the "wrong" direction, Cushitic languages becoming more Ethio-Semitic like rather than vice-versa, the exact opposite of what we'd expect from a supposed sprachbund where the main direction is Cushitic-to-Ethiosemitic.

Why would we a priori expect the features in a Sprachbund to all converge on the features of one of the language groups, as opposed to the different groups borrowing various things from each other? I mean what you basically just said is: ES languages are more like their neighboring Cushitic languages syntactically than other Semitic languages are, and Cushitic languages that neighbor ES languages are more like ES languages phonologically than other Cushitic languages are. That sounds like a decent case for some sort of Sprachbund to me!

(But let me admit here that I don't know much about this topic specifically, so by all means correct me if I'm off-base...)

The phonological influence is pretty minimal, and is restricted to particular Cushitic languages which have been under intense contact with Ethiopian Semitic - the major Ethio-Semitic and Cushitic languages of the region don't appear to have really influenced each other much at all in terms of phonology, if anything the Cushitic languages in the area are more dissimilar to Ethio-Semitic than their cousins outside of it (e.g. Iraqw). Compare Amharic and Tigrinya with Somali and Oromo and you'll see what I mean.

[Tropylium]

Ethiosemitic is a much deeper and more diverse branch of the family than many people realise, to the extent that it might actually be better to posit the PS Urheimat in the horn of Africa

No, not really. Ethiosemitic probably does have more diversity than any other clear group within of Semitic, but it's also all very recent. A language family can develop an appreciable amount of diversity in just one millennium (e.g. Arabic), quite a bit of diversity in two (e.g. Romance), plenty more yet in three (e.g. Indic) and an absolute shit-ton in four millennia (e.g. Oceanic). On the other hand, already the attested dialects of East Semitic are as old as Proto-Oceanic. I.e. if comparing like time-depth with like, East Semitic alone is "more diverse than Oceanic".

I would have a hard time believing Ethiosemitic is any younger than 3k years. But let's be unnecessarily generous and assume it's 4k years old… well, we still end up with the situation that at that time that there is but one single Semitic language in Africa, versus easily five if not more in the Levant (Akkadian, Eblaite, Amorite, by implication also some other Canaanite dialects and pre-Aramaic) plus likely least one in the Arabian peninsula too (more if you don't buy grouping Arabic, MSA and OSA into a single group). Still no grounds here to privilege the first-mentioned over the half a dozen others. We'd need to assume tons of extinct para-EthS varieties to get anywhere, and where's the evidence for any of that?

Eh, I'm skeptical about the "Ethiopian Sprachbund" in its current formulation. (…)
The phonological influence is pretty minimal, and is restricted to particular Cushitic languages which have been under intense contact with Ethiopian Semitic - the major Ethio-Semitic and Cushitic languages of the region don't appear to have really influenced each other much at all in terms of phonology

A 2000 review article by Tosco cites an apparently pretty good number of cases of Cushitic influence in Ethiosemitic:
– gradual loss of pharyngeals and glottals (strongest in the south)
– introduction of palatals and labiovelars
– presence of ejectives
– reduplicated plurals
– double-marked causatives
– general head-final syntax
– words for native African flora and fauna
– words for higher decads
– words for woodworking tools
(As well as various much less clear cases, such as possessive suffixes, or /f/ but no /p/; both of these found also in Arabic, some of Nilo-Saharan, etc.)

I'd think at least the ejectives may have gone the other way instead though: generally Agaw only has ejectives in EthS loans, Beja and Afar not at all. In HEC they're probably an older Afrasian archaism. Ditto also, as you point out, Southern Cushitic.

Tosco ends up with the following "areality scores" for some languages of Ethiopia and Eritrea:
– Tigrinya, Amharic, Tigre [all EthS]: 23–21 points
– Chaha [EthS], Oromo [LEC], Sidamo [HEC], Afar [LEC], Wolaytta [Omotic], Dhaasanac [LEC], Somali [LEC], Hadiyya [HEC]: 19–15 points
– Kefa [Omotic], Ge'ez [EthS], Beja, Arabic, Xamtanga [Agaw], Awngi [Agaw], Yemsa [Omotic]: 14–11 points
– Kunama [isolate]: 10 points
– Anuak [Nilotic]: 3 points
…which he thinks shows that even if Ethiosemitic has some kind of Cushitic substrate to it, its characteristic features are mostly rather native development.

[mèþru]

Ethiosemitic could easily have had extinct sister languages, and there's also the other South Semitic languages in southern Arabia.
Not saying that Frislander is right, but I think that an Urheimat near Yemen (including on the African side of the strait) deserves equal attention as a possible Urheimat (personally I think it was somewhere around Israel or the Nile)