The unreliability of Afroasiatic etymologies

The fact that Semitic, Egyptian, Berber, Cushitic, and Chadic all belong to a single family - Afroasiatic - is fairly secure, based on striking correspondences in basic morphology. However, it is often not appreciated just how difficult it is to find reliable lexical comparisons between these families, and just how primitive the current state of AA reconstruction is. The easiest source of AA etymologies online is Militarev's database on Starling, so I'm going to pick on it for this post (Orel & Stolbova and Ehret reveal similar issues, but the latter doesn't even include Berber, and I'm focusing mainly on Berber entries here for convenience.)

Suspiciously many entries are listed as having a cognate in only one Berber language (eg earth, hide, skin, run away); given the general closeness of different Berber varieties, you would expect valid proto-Berber terms to be reflected in more than one place. However, these could always be right. Other issues are more serious.

In several cases, a single proto-Berber root is split across several AA ones, due to mistaken sound correspondences. For example:

• Proto-Berber *i-qăs "bone, (fruit) pit" is split between PAA *ʔayš/ʔawš- "ripened grain, corn" with Zenaga iʔssi (quoted without the glottal stop) "os; grain, graine, baie; comprimé, pilule, cachet, pastille; perle" (Taine-Cheikh), and *ḳ(ʷ)as "bone", with all other reflexes of *iqăs, even though Berber γ (<*q) commonly corresponds to Zenaga ʔ.
• Proto-Berber *ta-Hăli (> *ti-Həli) "sheep" is split between pAA *ʔayl "ram" and *bawil "ram", although Ghadames-Awjila v corresponds regularly to Tuareg h and other Berber Ø. (A couple of forms, like Figuig tili mistakenly glossed as "ram", have even somehow found their way into a third etymon, "proto-Berber" *laH!) The issue is alluded to in a cryptic comment under the Berber section of PAA *waʔil "wild goat/ram; antelope": "Pr. H No. 220 (and Kössm. 193): Ghdm., Audj. Hgr etc. te-hele < *tiHeli, which, on the contrary, is to be connected with *ʔayl- 'ram' 3061 (together with Brb. forms of the t-ili type), as *ʔ > h in Hgr, while *ʕ > Hgr 0".
• 
  Most reflexes of pan-Berber ikərri / akrar "ram" are assigned to PAA *kar(w)- "ram, goat; lamb; kid". (The Semitic parallels listed for this word are rather interesting.) But Zenaga ǝgrǝrh, pl. gurănh 'bélier' (Nic. 156), on its own, is given a supposed proto-Berber form *gur- "ram", corresponding to an AA form *(ʔa-)gʷar "kind of antelope; ram; goat". In fact, however, there is a common correspondence of Zenaga g followed by a sonorant to proto-Berber k (eg ägärgur "chest" = Siwi ikərkər, əməgyih "dine" = Kabyle iməkli etc), and this word is obviously related to the other Berber forms.

Another case is listed as doubtful, eg:

• Most reflexes of Proto-Berber *a-lăqŭm "camel" are under PAA *ʕalVḳ/g- ˜ *lVḳ/gum- ˜ *ḳalVm- "camel"; but the Zenaga one äyiʔm, with regular *l > y (in his source's transcription ǯ) and common *γ > ʔ as seen previously, ends up as PAA *gam-al- (?).
  

Similarly, unrelated forms may be grouped together due to accidental similarity, eg:

• Under PAA *kʷay(-t)- "hen; partridge; dove; chick" is listed a "proto-Berber" form *i-kaHi; but the Ahaggar form listed corresponds regularly to Niger Tuareg tekažit, Mali Tuareg tekazzit, Awjila təkažit "hen" (see Kossmann 2005:60), and as such is unrelated to the Ayr and Tawllemmet forms takəyya quoted.

Another problem is undetected loans; this applies especially in sub-Saharan Africa, where little work has been done on their impact. PAA *ʔa/iw / *waʔ "bull, cow" is supported by Tawellemmet hawu "cow", isolated in Berber and obviously borrowed from Songhay, cp. Zarma haw, Tadaksahak hawú; removing this from the etymology leaves only pan-Tuareg iwan "cows", with no evidence for the desired *H. PAA *bar "cereal, corn" is supported by Zenaga būru "bread"; but this word is isolated in Berber and widespread in West Africa (eg Wolof mbuuru, Soninke buuru, Bambara nbuuru, Peul mbuuru, Zarma buuru), and is more likely a loan from Wolof or Pulaar.

Interestingly, most of the problem cases I've noticed in this quick skim are related to agricultural terminology. I wonder if that has anything to do with the particular interest of such terms for archeologists motivating a more intense search for cognates.

Why they thought the Berbers came from Yemen

A long-standing tradition in North Africa, convincingly rejected by Ibn Khaldūn but perpetuated by poets and curricula alike, claims that some major Berber tribes descend from Yemeni Arabs through semi-mythical pre-Islamic kings and their wholly mythical vast conquests. This idea has little to support it, and probably became popular because it allowed these tribes to claim prestigious connections in the context of a high culture dominated by Arab ideas; but why should the connection be specifically Yemeni, rather than, say, North Arabian or perhaps Persian? Linguistics suggests a possible answer.

In southern Arabia live several groups, most famously the Mehri tribe, whose languages, though Semitic, are only distantly related to Arabic, and quite incomprehensible to other Arabs. (You can hear recordings of it at SemArch.) Recently I borrowed a copy of the recently published Mehri Language of Oman, by Aaron Rubin; looking through it, I could see several points where Mehri resembles Berber but not Arabic that a traveller might seize on, notably:

• -s ـس "her", -sən ـسن "their (f.)"; compare Siwi -nn-əs ـنّس "his/her", -n-sən ـنسن "their (m/f)". A 3rd person in -s was found in proto-Semitic, as shown by Akkadian, but was replaced in Arabic.
  
• əl ال "not" (preverbal first element of negative); compare Tumzabt ul أُل. Again, this is found in Akkadian and hence must be proto-Semitic.
  
• -ət ـت feminine singular; compare Siwi -ət ـت (feminine singular in Arabic borrowings.) Again, the connection is real, but dates back to proto-Semitic rather than indicating any special relationship between the two.
  
• -tən ـتن feminine plural; compare Berber -tən ـتن (plural of some masculine nouns)
  
• a- أَ used as a definite article for some nouns; compare Berber a- أَ(masculine singular noun prefix). A striking case is Mehri a-məsge:d أَمسجيد vs. Siwi a-məzdəg أمزدج "the mosque". However, in Mehri this indicates definiteness, and does not depend on gender; this is probably a coincidence.
  
• tə-...-əm تـ...ـم second person plural imperfective, eg təkə́tbəm تكتبم "you (pl.) write"; compare Berber t-...-m تـ...ـم. The t- is cognate; not sure about the history of the -m offhand.
  
• 'ār آر "except, but"; compare Tuareg ar.
  
• ā آ "oh" (vocative); compare pan-Berber a أ. (This is actually found in Classical Arabic as well, أ, but is not widely used.)

None of these similarities in fact imply any close relationship between Berber and Mehri, of course; some are coincidental, while others can be traced back to proto-Semitic, and hence constitute evidence connecting Berber with Semitic, not specifically with Mehri. However, a medieval traveller between Yemen and North Africa would not have known that, and could easily have observed similarities like these and leapt to the seemingly plausible conclusion that Berber was connected to the language of these Yemeni tribes, who, like many Berbers, seemed to live just like Arabs yet speak totally differently.

The Berber language of Sokna (Libya)

Thank you SOAS library - I finally got a copy of Il dialetto berbero di Sokna! Sokna (they even have a Facebook group) is a small oasis south of Sirt in Libya, whose dialect of Berber, along with that of nearby El-Fogaha, is Siwi's closest relative. There were several surprises inside, including unusual vocabulary like amerru "mountain" or imeγri "Dhuhr (the midday prayer)", and some striking features shared with Siwi; one of the main ones is an unexpected bit of allomorphy. Across Berber, the second person plural ("you guys") is expressed on the verb with t-...-m, except in the imperative; Sokna does the same, so for example "you have" is t-la-m. In the imperative, you have a suffix -t; Sokna again does the same, eg sag-it-ten iyi-leḥbes "(you guys,) take them to prison!" But if you add an indirect object pronoun ("to him" etc.) to the imperative, you replace this t with an m, like the m in the second half of the non-imperative forms: eḍbeḥ-im-as a-na-dd y-used "(you guys) tell him to come to us!" The same thing happens in Siwi, except that in Siwi the prefixed t- of the non-imperative forms has disappeared. I'm doing a paper on the development of indirect object agreement in Siwi for the Berberologie conference in July, and this is a useful pointer to its history. Amazigh readers - have you come across anything like this?

Sadly, Berber is probably no longer spoken in Sokna. When this article was written in 1911, the shaykh of the oasis reported that only 4 or 5 Isuknan could still speak it, although many more could understand a bit. I don't know whether the people of Sokna today regret the loss of their language or are glad of it - but its disappearance destroys a key not just to Sokna's history but to that of Libya, Egypt, and the whole of North Africa, leaving only this article's fairly short wordlist (and a few even shorter older sources) as evidence for migrations between central Libya and Siwa and early contact with vanished pre-Sulaymi Arabic dialects.

Religious origins of the "Welsh Not"?

A well-known weapon in the arsenal deployed by educational systems the world over against local languages was what in the UK used to be called the Welsh Not - a piece of wood hung around the neck of a student caught speaking their own language, and passed on through the day to anyone that student heard speaking their language, so that whoever was wearing it at the end of the day would be punished. At a talk yesterday I heard that the same idea was implemented in Japan (against Ryukyuan languages) and Sudan (against Nubian.) Coincidentally, I just came across an account that gives interesting insight into the origins of this oppressive practice:

"With a general consent of all our company, it was ordained that there should be a palmer or ferula which should be in the keeping of him who was taken with an oath; and that he who had the palmer should give to every one that he took swearing, a palmada with it and the ferula; and whosoever at the time of evening or morning prayer was found to have the palmer, should have three blows given him by the captain or the master; and that he should still be bound to free himself by taking another, or else to run in danger of continuing the penalty, which, being executed a few days, reformed the vice, so that in three days together was not one oath heard to be sworn."The Observations of Sir Richard Hawkins, Knt in his voyage into the South Sea in the year 1593

Hard to imagine a ship full of sailors submitting to such a practice! But was this the original purpose of the Welsh Not? It would be interesting to find out. If anyone has an older citation to compare, I'd love to see it.

Endangered languages on Aljazeera

Aljazeera English is doing an interesting series on language endangerment and revitalisation:
* Language on the brink, talking with the last speaker of Wichita.
* Saving the language of the Cherokee, in Tahlequah
* French region aims to save language, on Breton
* Turkey's fading linguistic heritage and Saving Turkey's Laz language, on Laz (a close relative of Georgian, not "an ancient tongue that bears no resemblance to any other language in the region".)
* Circassians in bid to save language in Jordan - at a talk this week by Enam al-Wer I heard that, at the start of the twentieth century, the only permanent population in Amman was Circassian.

Manatees and bilingual compounds

In Djenné Chiini, the Western Songhay dialect of Djenné in Mali, the word for "manatee" is ayuumaa. This is clearly a compound of two elements: ayuu, the word for manatee throughout the rest of Songhay (as well as in Hausa), and maa from Bozo máa, which also means "manatee" (Bozo being the original language of the Djenné region.) It's as if the American English word for an elk were "elk-moose". I can't think of any other examples of this kind of half-borrowing, where a native word is "expanded" by adding on its translation into another language; can you?

(Sources: Daget 1953, La langue bozo; Heath 1998, Dictionnaire songhay-anglais-français, tome II: Djenné chiini.)

More on the WOLD Kanuri entry

The World Loanword Database is a great resource, and the Hausa/Kanuri team deserve congratulations for undertaking the Herculean labour of putting together two sets of etymologies. However, there are some issues with the Arabic etymologies in the Kanuri entry. The transcription is inconsistent and sometimes incorrect; more seriously, a few entries give incorrect meanings or impossible etymologies, as in the following cases:

3.592 àkú parrot: the quoted Arabic form is almost impossible as a Classical Arabic noun (and not in the Lisan al-Arab; the Arabic word is babγā’), and parrots are known in the Arab world only as an exotic import. Assuming the form exists in some Arabic dialect, it must be a loan from a sub-Saharan African language, not vice versa.
9.24 mágàsù scissors: the g and the u both suggest that this word entered directly from (Bedouin) Arabic, not via Hausa.
11.12 hàláltə́ own: if this is correctly transcribed, surely it comes from Arabic ħalāl “licit; one’s lawful property”. Arabic halak means “perish”.
11.79 ríwà dìò to earn: “ribā” means usury, and is strongly condemned in Islam; it is unlikely that this would be adopted as a neutral word “earn”. The more plausible source for both the Kanuri and the Hausa is Arabic ribħ “profit, gain”.
11.78 àlwúsùr wages: Perhaps < Arabic al-`ušr "tithe (< one-tenth)"; surely not from ma`āš.
14.451/6 kàjílí evening: “kajir” is not a possible native Classical Arabic word, and is not attested in Classical Arabic. If it’s in Shuwa, it must come from Kanuri, not vice versa.
16.34 tə́wə́rítə́ regret: Hausa tuubaa does come from Arabic, but clearly from Arabic tūb “repent”; it has nothing to do with Arabic ta’assaf (not *tāssaf) “regret”.
16.69 gàfə̀rtə́ forgive: the connection to Arabic γafar- is obviously correct, but Arabic yaʕfū is equally obviously not relevant; even if ʕ were normally reflected as g in Kanuri, it would leave the r unexplained.
18.33 kàsàttə́/àrdìtə́ admit: the Arabic form “kasat” does not exist. yarḍā means “may He hope/ approve” (as noted), not “admit”, making the connection rather tenuous.
18.45 áwúlò dìò boast: there is no Classical Arabic word “awulo”.
19.47 àmàrtə́ permit: Arabic ʔamar- means “he ordered”, not “permission”.
20.31 súlwé armor: Arabic silāħ means “weapons”, not “armor”.
21.24 àlàptà swear < ħalaf "swear" (not < allāh "the god")
21.37 àzáwù punishment: from Arabic ʕađāb “punishment, torment” rather than jazā’.
21.47 perjury: by what chain of semantic changes could “perjury” derive from “lawful”? And why would l > k?

Probable Arabic loanwords not listed as such include:
11.54 bàyîl stingy: from Arabic baxīl.
4.89 sûm poison: surely from Arabic samm?
4.93 sə̀lé bald: surely from Arabic ‘aṣla`?
5.26 kóló pot: perhaps cp. Arabic qullah (or onomatopeic?)
7.58 kábbì arch: surely from Arabic qubbah?
14.25 bàdìtə́ begin: surely from Arabic bada’?
11.29 lòrùtə́ damage: from Arabic ḍarr (impf. -ḍurr-). Cp. “judge” for ḍ > l.
24.02 wàltà become: perhaps from Maghrebi Arabic wəlli “become, return”.

In some cases, looking more widely allows the etymologies to be improved:
3.11 lə̀mân animal: < al-māl- "livestock, money", rather than al-mann "favor, benefit". For the dissimilation, compare the common Maghrebi Arabic change of n...n to n...l, eg badənjal < bāđinjān, fənjal < finjān.
2.34 lòrúsà wedding: probably from al-`arūs “bride” (Maghrebi Arabic l-aʕṛuṣa), rather than direct from ʕurs. Cp. Siwi aʕṛus “wedding”, with the same semantic shift.

There are also a few cases, many probably originally formatting issues, where the correct form is given in comments, but contradicted elsewhere:

3.25 sheep: the source cited, Kossmann 2005 (67), points out that the form quoted by Skinner, *adaman, is unattested. The correct form, adəmman, is found in Arabic as well as Berber, and refers to a type of sheep said to come from sub-Saharan Africa. Given that it refers to a specifically sub-Saharan sheep breed, 5 would seem a better classification than 4, though 4 is understandable.
3.78 camel: Kossmann 2005, cited, makes it rather clear than an Arabic origin for this word is very improbable. Moreover, there is no such Arabic word as “ləγəmal”; only the form jamal is correct.
4.87 physician: If Shuwa Arabic or some such variety has a term liktaay, there can be little doubt that it is a loan into Shuwa, not from Shuwa. As the comment indicates, this comes from English, not from Arabic.
7.422 blanket: The comments indicate a Berber form abroγ, but the field gives abrok. The Arabic etymology is less implausible than it appears, since the semantic shift to “full body covering” is well-attested, as in English “burka” from the same source.
12.081 above: here it is called areal and probably not Arabic, but under “sky” and “heaven” the same word is listed as “clearly borrowed”. One of these statements must be wrong.
13 zero: the Hausa form is transcribed correctly in comments, but wrongly under “Source words”.
18.51 write: rubuta is Hausa, not Berber, as the sources quoted make clear. The proto-Berber form had no suffix -t (as Kossmann indicates), and neither do any of the equivalent modern Berber verbs.
19.62/20.11 quarrel: If it’s related to “alhilaafu”, the Arabic form is al-xilāf. If it’s related to “judge”, that form is irrelevant. In either case, there is no Arabic word “alwalaʔ” with appropriate meaning.