CALL and DOT

Two conferences in the last three weeks: my first Colloquium on African Languages and Linguistics (as a speaker), in Leiden as always, and a day and a half of the 35st Deutscher Orientalistentag, in Erlangen.

Both were a lot of fun. I saw many different talks at CALL, too many to summarize, and mostly too off-topic as well. I was there to ask why we think Cushitic forms a single family within Afroasiatic (see also these blog posts). Despite the purposefully provocative title of my talk, I was not assaulted by any angry mobs of Cushiticists.1 The main question seems to be whether we really should disregard the lexicon when looking at subclassification (and then the next question should be whether the lexicon does show that Cushitic is a clade). It was also really cool to see several talks by young researchers whom I taught as first-years and who have now all finished their MAs and partially started PhD projects: shout-outs to Nina van der Vlugt, Melle Groen, and Jeroen van Ravenhorst. Post your slides online, guys!

Kollegienhaus Erlangen.

At the DOT, I co-chaired a panel on Semitic (in practice: mostly Hebrew) reading traditions together with Harald Samuel. While some of our presenters sadly had to cancel, we still had a great line-up, with exciting findings in every talk:

Chanan Ariel (Tel-Aviv University) proposed a highly original new explanation for the Biblical Hebrew phenomenon of dehiq, where consonants following certain unstressed vowels are geminated. According to Ariel, this is an orthoepic feature and applies to vocalic suffixes that alternate with zero, as well as some cases where the geminated consonant had to be kept apart from a following guttural. Works really well IMHO.

Aaron Hornkohl (University of Cambridge) provided a thorough discussion of the ketiv-qere phenomenon, presenting an up-to-date linguistic view of its origins and purpose in hopes of spreading more awareness of this to less linguistically inclined Hebrew Bible scholars. One thing that stood out to me is that words that are present in the consonantal text but left unpronounced in the reading tradition (ketiv wela qere) are sometimes translated in targums and other ancient versions.

Jonathan Howard (Ben-Gurion University of the Negev) presented his ongoing PhD research on the “Palestinian” vocalization system of Hebrew and Aramaic and pointed out that so far, there’s really no good evidence that it’s from Palestine. He’s hoping to find some, but it might be more impactful if he doesn’t.

Johan Lundberg (University of Oxford) walked us through the increasing complexity in Syriac punctuation signs, including the development of something that is roughly equivalent to an exclamation mark! Cool fact: in at least one of the few Syriac manuscripts of the entire Bible, the scribe has simply maintained the punctuation of each source text, resulting in several different systems coexisting in the same final work.

Emmanuel Mastey (Tel-Aviv University) presented a nice statistical inquiry into h-final spellings of 2m.sg. perfect verbs in Biblical Hebrew. Besides the very frequent case of נָתַתָּה ‘you gave’, Mastey finds that this spelling is especially common with verbs that have t as their third radical and, less so, with third-weak verbs. He suggests a phonological explanation for both classes; I wonder whether with the III-t roots, it may rather be motivated by the usefulness of distinguishing e.g. שתה ‘you placed’ from שת ‘he placed’.

Isabella Maurizio (University of Lorraine according to the programme, but I think that may be outdated? Sorbonne soon from what she told me) presented her recently completed research on the Second Column of Origen’s Hexapla, the oldest fully vocalized source (in Greek script!) for Biblical Hebrew. Big shock to me: Maurizio dates the Secunda to the 2nd c. BCE-1st c. CE, not the 3rd c. CE!

Marijn van Putten (Leiden University) appeared virtually to frighten the Hebraists with the tricky history of the Qur’anic reading traditions, with examples like one where a certain reader’s Arabic is notably more archaic than that of his teacher’s teacher. Since we barely know anything about who transmitted the Hebrew reading traditions, how much of this stuff are we missing due to a lack of data?

Harald Samuel (University of Tübingen) continued the sceptical line by noting some features of Tiberian Hebrew that appear to be really late (quoting me[!] from an informal conversation in which I said that a certain change must have taken place “about two hours before Ben-Asher went to work that morning”). How do we reconcile this with the alleged presence of extremely early, First Temple period features in the reading tradition as well?

Christian Stadel (Ben-Gurion University of the Negev) presented on some clearly late and some unquestionably early features of the Samaritan reading tradition and talked about how it relates to the consonantal text of the Samaritan Pentateuch more generally. It reminded me a bit of a presentation I gave on a similar topic several years ago. I only have one semester of Samaritan Hebrew, though—taught by Christian Stadel!—while Stadel is a real expert on the Samaritan languages. So it was reassuring to hear him argue for similar conclusions as well as present a whole lot more interesting data.

Last of all (due to alphabetization, but it worked out alright), I got to present on the project on the construction of the Biblical Aramaic reading tradition that I’ve been doing at Leuven since 2019. I’m not sure the argument I presented is fully sound, so it was great to be able to discuss it with some colleagues afterwards.

The Semitics section continued this morning. In her section keynote, Na’ama Pat-El (University of Texas Austin) presented her SemitiLEX project (recorded talk by another project member, haven’t watched it yet), looking at cognate Semitic lexemes not just in terms of roots, but also looking at morpho-lexical features like gender and pluralization. Unexpected result: building phylogenetic trees based on these data shows Akkadian, Ugaritic, Hebrew, Aramaic, and Arabic clustering as four or five separate branches, instead of Northwest Semitic clustering together and then being closer to Arabic than to Akkadian.

Maria Rauscher (Université Félix Houphouet-Boigny) presented her ongoing work on a dictionary of Arabic verbal nouns, focusing on the difficult case of k-r-h ‘to dislike’. As we had some extra discussion time for both Pat-El’s and Rauscher’s talks, there was time enough for the audience to draw up battle lines and get into the details of linguistic theory (such as: are morphemes even a thing?).

Stefanie Rudolf (Max Planck Institute for the History of Science) presented on two Qur’anic phrases that she suggests are unrecognized borrowings from Ethiosemitic. “The Lord of the East and the West” is attested in an Ethiopian Early Sabaic inscription, while Rudolf proposes the Arabic root f-t-w ‘to judge’ may be borrowed from Ethiosemitic f-t-ḥ. While she acknowledges the phonological difficulty of the last case, maybe we should reckon with the possibility of an unknown (South?) Ethiosemitic language that lost the pharyngeals acting as an intermediary: in the beginning of her talk, she pointed out that early Islamic sources refer to an Abyssinian with a name that is not Ge’ez but pre-Amharic (I think Ababut?), which I found very cool.

Jan Retsö (University of Gothenburg) pulled off the trick of reading out a text with no slides or handout while being perfectly easy to follow and entertaining. After an overview of the scholarship on Semitic–Ancient Egyptian cognates and loanwords, Retsö responded to Alexander Borg’s recent claim that there are lots of specifically Arabic loanwords in Egyptian. Retsö thinks there’s something there but urges for methodological precision.

Mohammad I. Ababneh (University of Halle) presented on some difficulties in Safaitic paleography, including merged letters and ligatures and other weird letter shapes. Nice to see some discussion of former Leiden colleague Chiara Della Puppa’s dissertation!

Finally, Vera Tsukanova (Philipps-Universität Marburg) took a look at the phonological adaptation of Persian loanwords into Arabic from a Semiticist and diachronic perspective. Historical differences in aspiration go a long way in accounting for prima facie unexpected sounds in borrowings.

And now, the conference is kind of on hold for various business meetings, which I took as my cue to leave. In conclusion, I would like to note that I am posting this from a high-speed train, which feels very futuristic. While some discussions in the field stay the same for what seems like forever—Paul Kahle’s lecture at the first DOT in 1922 1921 was referenced multiple times—I take this as a sign that like Deutsche Bahn passengers, no matter the inevitable delays, detours, and frustrations, overall, we are getting somewhere.

Only by a toddler, possibly for unrelated reasons. ↩︎

#Akkadian #Amharic #Arabic #Aramaic #Beja #Bible #Cushitic #EastCushitic #Egyptian #Hebrew #linguistics #Samaritans #Syriac #Ugaritic

Today I gave my first-ever presentation at the yearly Colloquium on African Languages and #Linguistics. My question to the Africanists, Afroasiaticists, Cushiticists: why do we think #Cushitic is a thing? 1/14

East Cushitic and Omotic passive *-ad’-

I’ve been working on an overview of the morphological arguments for which language families do and do not belong to Afroasiatic. One of the features you find in nearly every branch of Afroasiatic is a system of derivational affixes where *s forms causatives, *m (*n in Semitic and Egyptian) forms some kind of middle, and *t forms reflexives, passives etc. For example, Biblical Hebrew (the *s turns into h or ʔ in most West Semitic languages):

• קָדֵשׁ qāḏēš ‘it is sacred’ (basic verb)
• הִקְדִּישׁ hiqdīš ‘he made sacred’ (causative)
• נִקְדַּשׁ niqdaš ‘he showed himself sacred’ (middle)
• הִתְקַדֵּשׁ hiṯqaddēš ‘he sanctified himself’ (reflexive)

The *m/*n and *t are close in meaning, and sometimes you’ll see *m/*n in one language in a function that *t fulfills in another.

In East Cushitic, the passive suffix *-at- has some allomorphs (variant forms). Hayward (1984, paywall), reconstructs a paradigm where they are distributed according to the person/number/gender of the verb:

1sg.*-ad’-2sg.*-at-3m.sg.*-at-3f.sg.*-at-1pl.*-an-2pl.*-at-3pl.*-at-

Looking at the subject-marking suffixes that would follow this suffix, it becomes clear that the 1pl. form got its *n through assimilation to the following *n. But it isn’t clear where the glottalized (so, ejective or implosive) *d’ in the 1sg. comes from, or why the 1sg. form is different from the 3sg.:1

1sg.*-V2sg.*-tV3m.sg.*-V3f.sg.*-tV1pl.*-nV2pl.*-tVVni3pl.*-VVni

If we look at the matching paradigm in Proto-Agaw (Central Cushitic), or at the historically related prefix conjugation in East Cushitic, Agaw, or Semitic, we see that the 1sg. is marked with ʔ, against y for the 3m.sg. and 3pl. Now the Proto-East-Cushitic passive paradigm makes sense: 1sg. *-at-ʔV assimilated to *-ad’-ʔV, just as 1pl. *-at-nV became *-an-nV. So we’ve got East Cushitic evidence for 1sg. *ʔ as well. That’s cool.

Map of Afroasiatic languages by Wikimedia user Noahedits. Most of the Cushitic south of the Eritrean border is East Cushitic.

Omotic is a group of Ethiopian languages that were briefly considered West Cushitic, but ejected from Cushitic in the 1970s. It’s a diverse group and most scholars think it actually consists of two to four unrelated families. Serious doubts have been raised about whether any of them is even Afroasiatic in the first place.

So far I’ve looked at two of the smaller families, Aari-Banna (= Aroid = South Omotic) and Dizoid (= Majoid = part of North Omotic, allegedly). Morphologically, there’s very little there that looks Afroasiatic. But they have the derivational affixes we’ve been talking about: cf. Hamar (Aari-Banna) causative -(i)s, passive –(a)ɗ– (that’s an implosive), and vestigial –Vm- with a range of mediopassive and imperfective meanings (Petrollino 2016); Sheko (Dizoid) causative -s, passive -t’ (that’s an ejective), middle -n̩ (a syllabic nasal that assimilates to the preceding consonant; Hellenthal 2010). From what I’ve seen, something similar appears to be present in the big Ta-Ne or slightly bigger Narrow Omotic (or North Omotic minus Dizoid) family, again with little else to show for Afroasiatic morphology.

What’s super significant here is that in both (all three? four?) Omotic families, the passive is marked by a glottalized consonant: implosive ɗ in Hamar, ejective t’ in Sheko, something similar in the rest (I think). First of all, this is a typically East Cushitic form of the passive/reflexive affix; I don’t think it occurs elsewhere, you just get reflexes of *t. And second, as we’ve just seen, there’s a beautiful East-Cushitic-internal way to derive the glottalized *d’ from older Afroasiatic *t.

So I’m inclined to see the Omotic derived verb suffixes as borrowings from East Cushitic, languages with which they have a long history of contact and that are spoken right next door. That means that the best morphological argument for counting anything Omotic as Afroasiatic can be attributed to contact.2 For the time being, Aari-Banna, Dizoid, and probably Narrow Omotic as well are ending up on my “Not Afroasiatic” list.

This is Appleyard’s (2004, paywall) reconstruction of Proto-Lowland-East-Cushitic. It could be that the Highland East Cushitic evidence changes the reconstruction, I don’t know. ↩︎

Dizoid has some independent personal pronouns that look nice and Afroasiatic, but you don’t get the typically AA paradigms with t‘s and k‘s interchanging in the second person, for instance. So it could just be a chance resemblance. ↩︎

#Afroasiatic #Cushitic #Hebrew #linguistics #Omotic

The Semitic languages show a regular correspondence of p in some languages and f in others. For instance, ‘mouth’ in Akkadian is p-ū; Biblical Hebrew pe; Biblical Aramaic pūm; Ge’ez ʾäf;1 and Classical Arabic fam-. (Modern South Arabian should have an f too, but has replaced this word.) This sound is uncontroversially reconstructed as Proto-Semitic *p, as in *p-ūm ‘mouth’.2 Traditionally, the change of *p to f was taken as a diagnostic feature of the South Semitic languages.

This figure and the next adapted from Huehnergard & Rubin (2011).

[p] to [f], a plosive changing into a fricative, is an example of lenition. Lenition is a common type of sound change, so we tell our students, so it makes sense that *p is the older sound and it changed to f. So far, so good.

While preparing my first couple of classes for Comparative Semitics this year, I suddenly wasn’t so sure about this anymore. Two things bother me:

The examples of p > f I know about are all part of a larger change affecting other plosives too, like Grimm’s Law (Proto-Indo-European *p, *t, *k, *kw > Proto-Germanic *f, *þ, *h, *hw and related changes) or Aramaic and Hebrew BGDKPT-spirantization. Is just p turning to f really so common? How about just f turning into p?

Most scholars don’t accept the family tree above anymore. In the current model, the changes look more like this:

Now we need three or four separate instances of *p > *f—just as I’m starting to doubt how common that change is. Huehnergard & Rubin (2011), who argue for this second family tree, explain this as an areal change that spread through contact. But what kind of a contact scenario should we think of here? Did f spread from Ancient South Arabian (if those languages even had it) to all its neighbours? It’s not like we see enough other shared contact features to confidently posit a South Semitic language area or something.

Looking at Afroasiatic, things don’t get better:

• Berber has f, not p
• Cushitic has f, not p
• Egyptian has p and f, but we don’t know which one corresponds to Semitic *p (if either)
• Chadic: same as Egyptian, to my knowledge
• (I’m not sure Omotic is Afroasiatic, still reading up on this)

So if we posit Proto-Semitic *p, either we need two more independent cases of *p > *f (Berber, Cushitic),3 maybe more (Egyptian? Chadic?), or we reconstruct *f for Proto-Afroasiatic and say Proto-Semitic changed *f to *p. At which point, why not cut out the middleman and keep *f, then change it to *p in East and Northwest Semitic? Just two changes instead of the minimum of six you need otherwise.

So, are there any good arguments to reconstruct Proto-Semitic *p—or should we press *f and leave behind this relic from theories that believed in a South Semitic subgrouping?

Probably influenced by Cushitic, but we can still take it as related to the other Semitic words. ↩︎

In my opinion, the only word known so far with a superheavy syllable, exceptionally permitted because the word is monosyllabic. ↩︎

I’m also really starting to doubt that Cushitic is one family. So maybe make that four (Berber, Beja, Agaw, East/South Cushitic). ↩︎

https://bnuyaminim.wordpress.com/2024/11/07/froto-semitic/

#Afroasiatic #Agaw #Akkadian #Ancie #Arabic #Aramaic #Beja #Berber #Chadic #Cushitic #Egyptian #GeEz #Hebrew #linguistics #ModernSouthAr #Omotic #ProtoSemitic

Trying to learn more about Ethiopia(n Semitic languages), I just finished reading William A. Shack’s The Central Ethiopians. Amhara, Tigriňa and related peoples (1974; London: International African Institute). It’s 50 years old, many of the sources it uses are over 100 years old, and I’m sure it’s full of inaccuracies I didn’t recognize besides the ones I did, but it’s a place to start.

On the traditional religion of the Western Gurage, Shack writes (p. 113):

Yəgzär is the supreme god of the Gurage, the creator of the world. However, there is no cult addressed to Yəgzär, as there are to lesser deities, the most important of which are the cults of Waq, the male “Sky-god,” of Dämwamwit, the female deity, and Božä, the ‘Thunder-God.” Each clan has its own local Waq; Dämwamwit and Božä are central deities for the säbat bet federation. … In Gurage belief, Yəgzär handed over to Božä the responsibility of regulating the daily conduct of Gurage and affording ritual protection against theft and the destruction of property by arson.

Two things stand out to me here:

The creator god as a “high god” who is not the most commonly worshiped one and has handed over control to another god, specifically the god of thunder. This mirrors the relationship between Ilu and Ba’lu at Ugarit. But also compare Kronos and Zeus in Greek mythology, or maybe Odin and Thor in Germanic religion.

“Each clan has its own local Waq“.1 This sounds very Iron Age West Semitic to me. Think of Israel and Judah worshiping YHWH, the Ammonites worshiping Milkom, the Moabites and Kemosh, the Edomites and Qaws… We also find this in Ancient South Arabia, as I learned from Imar Koutchoukali during the last Leiden Summer School: there, everyone venerated Athtar, but each kingdom again had its own particular tutelary deity, like Almaqah for the Sabaeans and Wadd for the Minaeans. We seem to have an explicit description of this theology in Deut 32:8–9:

When Elyon apportioned the nations,
    when he divided humankind,
he fixed the boundaries of the peoples
    according to the number of the children of God;2
YHWH’s portion was his people,
    Jacob his allotted share.

(adapted from NRSV)

Feature (1) occurs in some shape or another in a lot of religions, especially ones from the Near East, and it may well have spread through contact. The Gurage Zone is far enough away, though, that I wonder whether this points to an inheritance from Proto-West-Semitic times. Feature (2) seems less common to me, although that could just be my ignorance speaking. Also, I’m not really sure how the difference between a thunder god and a sky god works out in practice; maybe I should read the other publications by Shack he refers to in this passage. But for now, creator-god-appoints-thunder-god-as-ruler and each-political-unit-has-its-tutelary-sky-god as reconstructible elements of Proto-Semitic religion makes for an exciting hypothesis.

Traditional Gurage dwellings looking out on the sky and, potentially, a thunder storm. Creator god not pictured.

The name Waq is borrowed from (Lowland?) East Cushitic, but from what I’ve read on Wikipedia he’s more important there and the localized aspect may be missing. ↩︎

MT: “the children of Israel”; commonly reconstructed like this based on LXX “the angels of God” ↩︎

https://bnuyaminim.wordpress.com/2024/10/20/gurage-evidence-for-proto-semitic-religion/

#AncientSouthArabian #Bible #Cushitic #Deuteronomy #Gurage #Hebrew #ProtoSemitic #religion

Earlier this year, I had two fun conversations with the team of the then newly-founded Kedem YouTube channel, which popularizes scholarship on the Ancient Near East and the Hebrew Bible. The first video was published yesterday. We talk about the concept of a language family, what languages constitute the Semitic language family, where Semitic comes from geographically and linguistically, how we can reconstruct earlier ancestors of the attested languages, and a few things this kind of reconstruction tells us about Proto-Semitic.

Stay posted for my second video with this channel, to be released sometime next year, on the different modern and—especially—ancient pronunciations of Biblical Hebrew.

https://bnuyaminim.wordpress.com/2023/12/30/video-intro-to-the-semitic-language-family/

#Afroasiatic #Akkadian #Amharic #AncientSouthArabian #Arabic #Aramaic #Beja #Berber #Chadic #Cushitic #Egyptian #GeEz #Hebrew #linguistics #Moabite #ModernSouthArabian #news #Omotic #Phoenician #ProtoSemitic #Tigrinya #Ugaritic

Prospects in comparative Cushitic

https://bnuyaminim.wordpress.com/2023/10/24/blog-post-prospects-in-comparative-cushitic/

#Afroasiatic #Beja #Cushitic #linguistics