Ayin-less in Gaza

Gaza, Arabic غزّة ghazzah, is another extremely old city of the eastern Mediterranean, having been in existence for at least three millennia. After a period of Egyptian rule, it became a member of the Philistine Pentapolis. Its name has been recorded in several forms over the years, including:

• Hieroglyphic: q3d3ti, g3d3y, g3d3tw (says Wallis Budge);
  
• Akkadian (Tell el-Amarna): Az-za-ti;
  
• Akkadian (Assyrian): Kha-az-zu-tu;
  
• Biblical Hebrew: `azzah;
  
• Greek (Herodotus): Cadytis (probably Gaza, but some dispute)
  
• Greek (Septuagint): Gaza (Γάζα)
  
• Latin (Pliny): Gaza

Some sources derive the town's Hebrew name, `Azzah, from the root `zz "be strong". However, this is a folk etymology. The two proto-Semitic consonants *` (pharyngeal voiced fricative/approximant) and *gh (uvular voiced fricative) merged to ` in Biblical Hebrew as we know it; but `zz "be strong" had ` in proto-Semitic (compare Arabic عزّ `azza), whereas Gaza clearly had gh (note that Akkadian had no gh, so a null/kh alternation in transcribing it is expected.) As a matter of fact, the Septuagint provides evidence that some dialects of Hebrew retained the `/gh distinction well into the classical period; some instances of written `ayin are left untranscribed in the Septuagint's Greek (eg Yehoshua` = Ἰησοῦς; Bet-`Araba = Βαιθαραβα), while others are transcribed as gamma (`Amora = Γομορρα; `Azza = Γάζα), clearly suggesting that the pronunciation was still distinguished.

The interesting thing is that Arabic has preserved the gh in Gaza, which would be impossible if it had taken the word from 7th-century Aramaic, which has no gh either (Hebrew was almost surely extinct as a spoken language by the time Islam arrived.) Could it have been borrowed from Greek? Maybe; but, given that Herodotus notes that "Arabians" dominated the coast between Gaza and al-Arish even in his time, another obvious possibility is simply that the word Gaza entered Arabic from one Canaanite language or another well before the loss of the `/gh distinction, and didn't change.

As an interesting coda, the name Gaza may apparently be the source of the English word gauze.

Sumerian grammatical texts

Sumerian Grammatical Texts available online! The title is a misnomer - most of the texts given are early Sumerian-Akkadian lexica arranged by topic, or just plain Sumerian texts - but there are other interesting things, such as a phonetically organised syllabary (vowel order: u-a-i), and a series called "ana ittišu" (p. 30) with some rather paradigm-looking stuff, such as:

Sumerian	Akkadian, English
ùr	sûnu, lap, bosom
ùr-bi	sûn-šu, his bosom
ùr-bi-šú	ana sûni-šu, upon his bosom
ùr-bi-šú in-gar	ana sûni-šu iškun, he placed upon his bosom

which I guess offer a clue about the teaching methods used. These tablets were used to teach young Akkadian-speaking would-be scribes Sumerian, long after Sumerian itself had become extinct.

Kurdish giving way to Turkish in some areas?

Found a telling first-hand account of language shift. I had no idea the last decade or two had made such a difference.

until the end of 1980s the kurdish language was still preserved, because the kurds were still in their villages [...] most of them would not know one single word turkish and the women, in specific, did not know one single word turkish! [...] but at the beginning of 1990s, and since then going on, we have been losing the kurdish language [...] and it is mainly because around four or some say five thousand kurdish villages were forcibly evacuated, i should use "they were destroyed by the turkish army" instead. and more than three million people(kurds) were displaced! and of course it had its consequences! [...]

all the kurds started to go to school, where they would only speak turkish, and if, in any way someone were to speak kurdish s/he would punished for speaking kurdish and this way it would have a deterring effect on the other children(students) as well! kurdish students were despised and made fun of because of their accent so the families of those kurdish students thought that if they spoke only turkish at home it would help their children and they would be able to speak turkish better, and nobody would be able to fun of them. [...]

they only watched the turkish tv channels! and especially the mothers were very badly affected by this, because they wree the ones who would stay at home and when they did not have anything to do they would watch the tv and improve their turkish, but after a while they started to use turkish words while speaking kurdish, keep in mind that their children were not taught kurdish, so even if some of those children wanted to learn kurdish they would learn it wrong because their parents would not speak appropriate kurdish! i still cant believe that some kurds would say "qapi qepamiş bike" for "close the door" in kurdish: i have a very hard time understanding this, qepi originally is kapı(it is pronounced liek qepi in kurdish) "qepamiş" means nothing, it is supposed to mean "close", they combine turkish root of "to close" and add a kurdish suffix to it and make it kurdish. when i see people using those words, and killing kurdish it really hurts me very badly!

The extreme borrowing is an interesting point - and probably a universal of low-status languages. I can sympathise - excessively Frenchified Arabic really grates on my ears...

Mountains of Lebanon - some etymologies

Lebanon's cities and villages, tragically now in the news, have some interesting etymologies. I always used to wonder about the different names: why is the same city called Tyre in English, but Ṣuur in Arabic? or Byblos in English, and Jubayl in Arabic? The reasons illustrate the sheer length of these towns' history, and the time depth of Greece's contact with them.

Sometime before the characteristic sound shifts of Proto-Canaanite happened - perhaps 1200 BC or so? - Tyre would have been known by a Semitic term meaning something like "peak" or "crag": θ'uur-u. This was borrowed by the early Greeks as tur-os > tyros (when u got fronted to y) > Latin Tyrus > English Tyre. Meanwhile, in Lebanon, that glottalised θ', perhaps unsurprisingly, was among the first sounds to disappear; in Canaanite (that is, Phoenician, Hebrew, and assorted minor languages of the area), it became s' (or ṣ - it's hard to be certain whether the Canaanite emphatics were glottalised or pharyngealised). Case endings also vanished. This gave the Phoenician name: S'uur (or Ṣuur). It was adopted without change into Aramaic, and thence Arabic, as the region's languages shifted over time. The regular cognate of Proto-Semitic θ'uur-u in Arabic would have been ظور đ̣uur; this root is unattested as far as I know. However, in Aramaic *θ' became ṭ, and from this source the root entered pre-Arabic as طور ṭuur "mountain", a rare but well-attested term used in the Quran, notably for Mount Sinai. (In Ugaritic, freakishly enough, *θ' became γ (gh), and γuur- "mountain" is a well-attested Ugaritic word. In Ugaritic, incidentally, Tyre was actually called ṣuur-; so either my etymology here is wrong, which is possible, or Ugaritic borrowed the name after it had already changed *θ' to γ.)

Likewise, Byblos would have started out as gubl-u (attested in Ugaritic and Akkadian), which (judging by possible Arabic cognates) may have meant "mountain" as well. This went into early Greek as gwubl-os > byblos (Mycenaean gw > Greek b, u > y) > English Byblos. In Arabic, the g of course became j; and, for some reason (maybe it was a little town at the time?), it looks like a diminutive got added, turning it from *jubl to jubayl (colloquial žbeyl), which in Arabic just means "little mountain".

Let's hope the day that these towns appear on the news for their history or their beaches, not for the bombs being dropped on them, comes more quickly than looks likely.

Polysemy vs. homonymy: some Algerian Arabic examples

I'm recently back from Algeria (hence the blog gap), so I thought I'd post some more meditations on Algerian Arabic...

Q: Which of the following words from Algerian Arabic are cases of polysemy (different meanings with a shared conceptual core) and which of homonymy (different meanings coincidentally identical in phonetic shape)?

`ṛuṣa عروصة - bride; daughter-in-law
ħjəṛ حجر - stone; lap
bakuṛ باكور - early-ripening figs; young bonito fish

A: `ṛuṣa, from Classical Arabic `aruusah عروسة, is a case of polysemy; a new bride traditionally goes to live in her husband's family house together with her new parents-in-law, so the extension is natural.

ħjəṛ is a case of homonymy: "stone" comes from Classical ħajar حجر, and "lap" from Classical ħijr حجر. Though it would be amusing to try and find a common conceptual core, I can't see any plausible one.

bakuṛ is etymologically a case of polysemy: both derive from Classical baakuur باكور, "coming early, early; premature; precocious" (Wehr). But synchronically, given the two independent restrictions of its meaning - it isn't used to mean first fruits in general, or young fish in general - I can only take it to be a case of homonymy.

Ohlone

I used to live in the Bay Area for a while, so naturally I tried to find out about its pre-colonial language group, Ohlone. This turned out to have been a set of fairly closely related dialects/languages stretching from San Francisco down beyond Monterey, plus the coast of the East Bay. Their only reasonably close relative is Miwok, another small language family spoken to its north and west, although wider relations with languages further north along the Pacific coast are likely. Among the more noteworthy features of Ohlone are regular metathesis processes - for example, the plural suffix can be either -mak or -kma, depending on whether it's preceded by a consonant or a vowel.

Dave Kaufman has just posted some interesting Ruminations on Rumsien, one of the southern dialects; or, if you speak Spanish, you can read a grammar of Mutsun, a southeastern dialect. Wikipedia has a map.

Tunisian Berber

Amazing things turn up at the University of Western Sydney: a complete thesis online offering An outline of the Shilha (Berber) vernacular of Douiret (Southern Tunisia). Check it out; the rather endangered Berber varieties of Tunisia are quite ill-documented.

North African language policy

MoorishGirl has an interesting post on an article on a round-table debate on Moroccan Arabic, or Darija, as "a medium of cultural expression". She comments:

I'm fully in favor of using Darija, because of the huge impact it would have on the creation of a reading culture. Imagine: All children's books right now are in Modern Standard Arabic, which is a language no one learns until first grade (i.e. age 6 or 7), by which time reading habits are already in place for many kids.

I think this is a crucial point. Developing a literature of sorts in Darja would allow kids to get into the habit of reading way earlier. A fair number of kids in the West are reading by the age of three; for an Algerian or Moroccan kid to even understand much of the language his/her books are written in at that age would be unheard of. With Darja literature for them to use, they could start reading before they ever started school; it might even lead to them acquiring literary Arabic faster. Moreover, an oral literary tradition already exists, best exemplified by the traditions of melhoun poetry and chaabi lyrics; the language used in these is recognizably a literary register, and all that would be needed would be to write it. My puristic instincts would also rejoice in a move with the potential to stem the tragic loss of inherited vocabulary, and overuse of French, now afflicting Darja. And after all, why should Arabic-speaking kids continue to be deprived of the chance to read in their native language now that Tamazight-speaking ones are finally getting that chance?

However, I would envision Darja as a supplement to literary Arabic, not a replacement. Arabic connects Algeria (and no doubt Morocco), not only to the Arab world but to its own past, not to mention allowing it to engage more fully with its religion. The language in which Amir Abdelkader and Ibn Khaldun wrote - and of which generations were deprived by French rule - should always be a crucial part of an Algerian education. Also - as the ongoing struggle to get adequate higher educational textbooks published even in literary Arabic reminds us - a written Darja would take centuries at least to build up a literature comparable to major languages.

As long as I'm pondering educational policy, what should be done with foreign languages is obvious: end the domination of French. Nothing wrong with French per se, but an all-French policy is a handicap in a global context, isolating Algeria in the ghetto of Francophony at a time when English is a prerequisite to serious scientific work even in Paris, and an embarrassment at home, where it remains a scandal in conservative eyes. From 3rd grade on, have a choice between French and English (and maybe even Spanish) as the second language, and raise a generation of educated North Africans that do not all share a single foreign language; only thus can the domination of French in North Africa, with all its attendant sociological divisions and economic problems, be ended. Of course, in an educational system that has a serious shortage of good teachers as it is, this is a distant dream... but dreaming can be useful.

"-gate" suffix reaches Arabic

Algerian football fans (that is to say, probably most of the population) are up in arms about not being able to watch the World Cup unless they subscribe to ART - a Saudi company which bought up the rights to World Cup footage for the MENA region and is selling it so expensively most terrestrial stations (including Algeria's) can't afford it. I don't particularly care myself, to be honest, but I was impressed to see the following headline in the newspaper Ech Chourouk:

الجزائر على أبواب فضيحة "آرتي-غايت"!

al-Jazaa'ir `alaa 'abwaab faḍiiħat "aartii-gaayt"!
(Algeria is on the verge of an ART-gate scandal!)

The development of "-gate" from a random morpheme at the end of a hotel name into a suffix indicating a political mess (Monicagate, Fostergate, etc.) is remarkable enough; that it should be borrowed into Arabic, even in the weird world of headline idiom, is incredible to me. I guess bound morphemes aren't necessarily as hard to borrow as one might think.

Nandi relatives and Arabic center-embedding

Two random interesting bits thrown up by my current research:

Nandi, a Nilotic language of Kenya with VSO order, would appear to allow you to relativize virtually any constituent of a sentence. I was particularly impressed by examples like:

nikò ce:pyó:sé:t ne â:-nken ci:tà ne kí:-ká:ci kitâ:pú:t
this woman Rel 1s-know person Rel Past-give book
"This is the woman that I know the person that gave [her] a book / that [she] gave a book to."

á-ké:r-é ci:tà ne pè:nt-í: àk la:kwe:-nyi: kâpsá:pit
1sg-see-impfv person Rel 3pl-go and child-his Kapsabet
"I see the person who [he] and his son are going to Kapsabet."

Take that, Subjacency Constraints! (Well, more seriously, I'm guessing ce/ne is probably not a fronted relative pronoun, especially since it agrees in case with the head noun and not with the position of the gap, so maybe no movement is involved - but that just raises other issues, like what does the gap consist of? Surely not pro? And what is ce/ne - a complementizer?)

And, in case you've ever wondered what an Arabic incomprehensibly double center-embedded sentence would look like, here's one:

رأى الولد الذي كتب الرجل الذي عينه الرئيس الرسالة إليه أخاه
ra'aa lwaladu lladhii kataba arrajulu lladhii `ayyanahu rra'iisu rrisaalata 'ilayhi 'axaahu
saw [the-boy [that wrote [the-man [that chose-him the president]] the letter to him]] brother-his.
“The boy the man the president chose wrote to saw his brother.”

Note that Arabic's VSO order renders it less vulnerable to subject- and object-relativization in this regard, but leaves it helpless against relativization of other positions - which is nonetheless permissible.

(Nandi examples from Creider & Creider 1989.)

A little Algerian Arabic folk poetry

I recently came across a nice book (in English for once!) on the Algerian folk poet Muhammad ben Tayeb el-Alili, The Graying of the Raven. It's titled after this stanza, from a poem about a drought:

məššərq ləlməɣrib
fiha lɣ°ṛab yšib
a `aləm əlɣib
wətħənn bəttisir

من الشرق للمغريب
فيها الغراب يشيب
ها عالم الغيب
وتحن بالتيسير

From the east to the west
The raven turns white
O Knower of the Unseen
Grant us respite

(I've substituted my slightly more literal translation.)

His works are not particularly famous, and, while worth a look, are not in the top rank of the genre - but I'll bet they're the only ones available in English. For a perhaps better example, consider Dahmane El Harrachi's famous song - I was going to try and translate the whole thing, but frankly it's not easy, so I'll just give a sampler:

šħal šəft əlbəldan əl`amrin wəlbərr əlxali
šħal ð̣iyyə`t əwqat wəšħal tzid mazal ətxəlli

اشحال شفت البلدان العامرين والبر الخالي
اشحال ضيعت اوقات واشحال تزيد مازال تخلي

How many crowded cities and empty wilds you've seen,
How much time you've wasted - and how much more will you waste?

Incidentally - yes, the pessimism of both examples is characteristic.

Some nice Siouan sites - Omaha, Ioway-Otoe-Missouria, Kansa

A great Omaha language site: Omaha Language Curriculum Development (hat tip to Mark Awakuni-Swetland.) I particularly like the cookbook. Other nice Siouan sites worth looking at are Ioway, Otoe-Missouria Language and WebKanza.

Algerians sure can code-switch

Algerians are rightly renowned for their code-switching wherever they go (or should be). I disapprove of it in general - it often reflects the unjustly low esteem Algerians tend to have for their mother tongue, and encourages the abandonment of less commonly used Algerian Arabic (Darja) terms in favor of unnecessary French loanwords. But you can't help but love an example like this one that I just heard here in London today:

gal-li y-ḥəbb to move
say+PF+3MSg-DAT+1Sg 3sg+IMPF-want "to move"
He told me he wants to move.

What's so weird about that? The thing is, while standard English want requires a non-finite complement, Algerian Arabic ḥəbb "want, like" takes a finite complement. In fact, there are no infinitives in Algerian Arabic - only finite verbs and verbal nouns. So it looks as if the non-finiteness (presumably generated in T) of the complement in the English half is being selected, not by the Arabic verb which precedes it, but by the English translation equivalent of it. I still can't quite believe I heard this sentence.

If you found that fun, you may wish to ruminate over another sentence (Arabic/French switching) from the same conversation:

`ənd-i un problème ta` wəqt
at-me "a problem" of time
"I've got a problem of time."

and, in particular, on what syntactic tree it suggests, and whether this really fits the idea of a DP. Note also that, while Algerian Arabic does have a sort of indefinite article (waḥəd əl-), its distribution is quite different from the French one, and I don't think it would occur in the corresponding code-switching-less sentence.

Center-embedding and Japanese

Lately I've been reading some of John Hawkins' A Performance Theory of Order and Constituency, which puts forwards some very appealing ideas about how to predict the relative frequency of different word orders (both cross-linguistically and within a language) by quantifying how easy they are for humans to parse. (For example, he derives such phenomena as Heavy-NP shift, the relativization hierarchy, and even the relative frequency of the six possible basic word orders SVO/SOV, VSO, etc.) Parsing issues certainly severely affect the grammaticality of sentences, as people who follow titles posts Language Log authors write have know.

I tried out a similar example in Japanese on a friend - going by the grammar books, one would expect "John said Mary thinks Bill came" to be translated as "Jon-wa Merii-ga Biru-ga kita to omou to itta", with three successive subjects followed by three successive objects. She unhesitatingly went for, as I recall, "Biru-ga kita to Merii-ga omou to Jon-ga itta" - moving the subjects to the "wrong" places to make the sentence processable - and said that the three-successive-subject one was "difficult". I can't think of any Arabic parallels offhand - postverbal objects and resumptive pronouns in relative clauses together stop most of the obvious possibilities - and Sylheti turns out to rather cleverly block almost (not quite) all possible ways in which problematic center-embedding might emerge. So my question to you is: in your language, can you think of similar examples of incomprehensible yet nominally grammatical sentences?

National/common/unifying language for the US?

As you may have heard on Language Log, on May 17th-18th, the US Senate approved not one but two amendments - one Republican, one Democrat - on the status of English. The first amendment, by Sen. Inhofe (R-Oklahoma), amends sections 161-2 of Title 4 of the United States Code to state:

English is the national language of the United States. The Government of the United States shall preserve and enhance the role of English as the national language of the United States of America. Unless specifically stated in applicable law, no person has a right, entitlement, or claim to have the Government of the United States or any of its officials or representatives act, communicate, perform or provide services, or provide materials in any language other than English. If exceptions are made, that does not create a legal entitlement to additional services in that language or any language other than English. If any forms are issued by the Federal Government in a language other than English (or such forms are completed in a language other than English), the English language version of the form is the sole authority for all legal purposes.

The second, by Senator Salazar (R-Colorado), makes the same section rather more reasonably, if vacuously, say:

English is the common and unifying language of the United States that helps provide unity for the people of the United States. The Government of the United States shall preserve and enhance the role of English as the common and unifying language of America. Nothing herein shall diminish or expand any existing rights under the law of the United States relative to services or materials provided by the government of the United States in any language other than English.

The bill is still under debate, so it remains to be seen what, if any, of this will be left - but, after 230 years of doing just fine without one, the USA may or may not soon have a national language. Either way, it's an interesting debate to follow. I remember in San Francisco just about any governmental document seemed to be printed in English, Chinese, and Spanish; that approach - choosing the language according to what people actually spoke on a local level, rather than a national one - strikes me as eminently sensible. What I can't seem to figure out is what the plan is now that both have passed - do they stick both texts in the section, or do they just hash it out later?

Maya and Amnesty International

I went to the AI site just now, without a linguistic thought in my head, and what do I find?

Watemaal: Li risinkileb’ laj nat’ol na’ajej moko a’an ta li xb’ehil re xtuqub’alkil ru li ch’a’ajkilal chi rix li ch’och’

I applaud this, although I should point out that putting an international press release in Mayan (dunno which Mayan language - Chol?) is somewhat self-defeating...

Incidentally, if you haven't already seen it, check out the site Language Hat just found. I particularly liked the San Zi Jing.

Shawi blog

Shawiyya (Chaouia) is a Zenati Berber language of eastern Algeria, spoken inland on the Sahara-facing side of the Atlas Mountains. While spread over a far larger area than Algeria's other main Berber language, Kabyle, it has only about half the population (1.4 million or so). Unlike the Kabyles, the Shawis, as their Arabic name suggests, were traditionally seminomadic (transhumant, to be exact); after independence, many seized the opportunity to settle down in the cities, and, from what I hear, this major change of lifestyle led to widespread language shift to Arabic. Shawiyya, like other Zenati dialects of northern Algeria (Chenoua, Bissa, etc.), but unlike Kabyle or the Berber varieties of the Sahara, has the interesting sound change t > h initially in many contexts. Anyway, I found a Shawi-language-focused blog the other day, to my immense surprise, which I figured was worth linking:

Awal nu Shawi

It seems to mainly post lyrics, sometimes with translations.

A new primate and a nice talk

I went to a nice inaugural talk by Prof. Jaggar here at SOAS yesterday about African linguistic diversity, Afroasiatic, and SOAS linguists' resistance, then ultimate capitulation, to Greenberg's groundbreaking African classification - the talk was rendered especially notable by his getting up along with his choir to sing Nkosi sikelele iAfrika afterwards! However, I haven't really got time to summarize it (for the classification part, you could check out my previous post Beja and beyond), so instead I'll post a link to the discovery of a new species of primate: the kipunji, a close relative of the baboon which lives in trees instead of on plains. No news yet on its communication system :)

Sylheti word order

I've been working on Sylheti - a highly divergent dialect of Bengali / language very closely related to Bengali spoken around Sylhet in northeastern Bangladesh - for my field methods class for a while. The particular point I'm focusing on at the moment is the positioning of complement clauses, which obeys a simple rule: if the complement clause has a separate subject, it follows the verb; otherwise, it precedes the verb. The language is otherwise SOV, I should note, so you get contrasts like:

ami exṭa apol sai.
I an apple want-1.
“I want an apple.”

ami exṭa apol xaitam sai.
I [an apple eat-COND-1] want-1.
“I want to eat an apple.”

ami sai he exṭa apol xaok.
I want-1 [he an apple eat-3-OPT].
“I want him to eat an apple.”

This doesn't fit my Japanese-based expectations of "proper" SOV languages (in Japanese, the subordinate clause would always precede the verb) but it turns out that German has basically the same word order (if you factor out the main-clause V2 order by having an initial complementizer). There are some obvious processing motivations for such an order, but it doesn't really fit the head-position parameter idea so well. I was wondering: has anyone seen similar patterns in other SOV languages?

West African grammars in Arabic script

I want to see this talk by Hiroyuki Eto Nikolai Dobonravine (though I'm not likely to be in Dublin for it):

Arabic and Arabic-script writing tradition in West Africa dates back to the 12th century AD, if not earlier. Local scholars were familiar with the linguistic ideas which formed part of Islamic education. Arabic grammars and dictionaries were popular in the region. The interest in the study of Arabic resulted in the development of local Arabic and bilingual vocabularies, sometimes written in verse, as well as some works on Arabic grammar. A few versified vocabularies and grammars of West African languages were also composed. Almost all of them were written in Arabic and used Arabic linguistic terminology.

In the late 19th and early 20th centuries several works were written in West African languages using Arabic script. One such work, "Littafen nahwowin Hausance" ("The book of Hausa grammar"), is analysed in the paper. The work demonstrates a special approach to the parts of speech in Hausa (the verb deprived of the "person-aspect complex" is seen as a noun, although it may be used independently in the Imperative). This is a larger work of traditional lexicography, with notes on folk etymology, pragmatic rules, grammatical gender and possessive pronouns in Hausa.

The shift from Arabic to Roman script and the decline in the use of Arabic did not lead to the disappearance of the earlier linguistic tradition. New grammatical works and vocabularies in Arabic script (including a Fula-French vocabulary in Arabic script) were published. All these writings have been largely ignored by the linguists working at the universities in West Africa and abroad.

Whorf meets warmongering

Pop Whorfianism (usually in forms that Whorf would have been the first to laugh at) is something I usually associate with a slightly hippy-ish multiculturalism. However, it seems to have a certain appeal to Islamophobes as well.

The thesis they find so appealing is summarized in one James Coffman's question: "Does the Arabic Language Encourage Radical Islam?". Apparently, he did a survey in 1988 in Algiers which confirmed a number of fairly obvious facts - notably, that the younger students that year, who were the first cohort of students whose secondary education had been mainly in Arabic, were more "Islamist" than their predecessors who had gone through a partly or wholly Francophone educational system. From this, he concluded that the Arabic language encouraged "radical Islam" - not, for example, that Arabic-literate students had much easier access to "Islamist" literature (and Islamic literature in general), or that the transition to Arabic had been accompanied by a vast expansion of the school system to cover more conservative rural areas, or that many of the imported Arabic teachers who helped tide Algeria over the transition period were Islamic Brotherhood members fleeing crackdowns in Egypt, or indeed (most importantly) that the collapse of the Algerian economy in the late 1980's was encouraging the growth of anti-government ideologies. It's an old, old saw, but one that apparently still bears repeating: correlation does not equal causation.

Mind you, like most people who cite the Sapir-Whorf hypothesis, he doesn't seem to have a very clear idea of its content. On my reading of Whorf, his core idea is (plausibly enough) that a language might make its speakers more conscious of some grammaticalized categories by forcing its speakers to mark them, or less conscious of them by not providing any simple way to describe them; it would thus render some ideas more intuitive than others. For this sort of deep influence to be plausible, the speaker has to do most of his/her thinking in the language in question. But both classical Arabic and French in Algeria are only ever used by most speakers in writing, or in highly formal contexts - scarcely the sort of situation Whorf had in mind...

(PS: It seems Language Log have also just done another post on "No word for X" fallacies. Another example of ham-handed anti-Arab efforts at Whorfian analysis is alluded to on Linguistic Life.)

How to find linguistic universals

I couldn't resist posting this quote:

[In this book] I examine the general conditions under which verbal complements are licensed, and provide a possible explanation for their limited distribution. The primary reference language is English, though the proposed licensing conditions for verbal complements are assumed to hold universally.

Fortunately, the author adds:

That the main proposals of this study and the analyses do indeed carry over to other languages is shown in Chapter 5, which takes a cross-linguistic perspective.

The title of Chapter 5? "Direct Perception Complements in Other European Languages". The languages considered are German, Dutch, Italian, French, Spanish, and Portuguese, representing a grand total of two neighboring subfamilies of Indo-European.

I don't mean to poke fun at this book specifically - it looks like a very thorough analysis of clausal complements of perception verbs in English - but this so neatly encapsulates what in practice is one of the main problems of the generative program: over-reliance on English in particular and what Sapir used to call "Standard Average European" in general.

Reduplication in Siouan

I've finished, handed in, and now uploaded that essay I was working on, on reduplication in Siouan. The main conclusions were that:

* Proto-Siouan-Catawban (and Proto-Siouan-Yuchi, but not Proto-Macro-Siouan) productively formed pluractionals from verb stems by full stem reduplication. Every branch of the family exhibits reflexes of this process, although these have often been affected by semantic extensions and morphological contractions.
* Stoney "adversative" reduplication is most probably borrowed from a Salish language.

A comparative linguist of the thirteenth century

I've been reading Empires of the Word recently, a quite enjoyable and informative history of the world's main languages; it skimps on Arabic to an almost absurd extent, but makes up for this by a truly excellent chapter on Sanskrit. Anyway, one surprise it provides is that, in addition to his better known poetic activities, Dante also wrote a treatise about language, De vulgari eloquentia, in which he comments on the nature of language change, specifically attempting to explain how Latin could have gradually changed into the Romance languages, a concept which his audience apparently found hard to accept:

Nor should what we say appear any more strange than to see a young person grown up, whom we do not see grow up; for what moves gradually is not at all to be recognized by us, and the longer something needs for its change to be recognized the more stable we think it is. So we are not surprised if the opinion of men, who are little distant from brutes, is that a given city has existed always with the same language, since the change in language of a city happens gradually only over a very long succession of time, and the life of men is also, by its very nature, very short. Therefore if over one people the language changes, as has been said, successively over time, and can in no way stand still, it is necessary that it should vary in various ways quite separately from what remains constant, just as customs and dress vary in various ways... (p. 321, Empires of the Word; original available elsewhere)

It's easy to forget just how difficult even the basics of historical linguistics must once have seemed, but texts like these help.

Incidentally, I hope to reestablish my website sometime soon - can anyone recommend a good free/very cheap website hosting service other than Geocities?

Free online American Indian linguistics papers

- and if that's not a title to draw the crowds, I don't know what is :) I found this handy Kansas University site while researching my Siouan reduplication essay, which is currently growing at an out of control pace (thanks partly to the help of Siouan List); just a few of the highlights are articles on:

Salinan
Esselen
Inga
Karankawa
Opata, Eudeve
proto-Algonquian
proto-Algic
proto-Algic
proto-Algic
proto-Siouan "one"
proto-Muskogean
Klamath-Sahaptian correspondences

Oh, and while I'm posting links, a couple of non-linguistic links for your perusal - some undeservedly obscure news stories:
Bouteflika back in hospital
Central African Republic
The Israel Lobby

Ibn Hazm on conlanging (and other stuff)

Having established the divine origin of language to his own satisfaction, Ibn Hazm goes on to discuss the vexed question of what language Adam spoke, and concludes - sensibly enough - that there is no way to be certain. However, he figures it must have been "the most complete of all languages, and the most clearly expressive, and the least ambiguous, and the most concise, and the one with the most various names for all various nameable things in the world" since, having been bestowed by God directly, it must naturally have been the most perfect.

He also decides that it almost surely must have been the ancestor of all modern languages, because it was not inconceivable but extremely improbable that anyone would have decided to waste so much time and effort as to "invent a new language, which would be an enormous effort for no reason; such meddling would not be undertaken by any intelligent person... [its inventor would be] a person busying himself with what does not benefit him and neglecting what concerns him", and even if this did happen, it was even less likely that the inventor would be in a position to impose his language on any community. He specifically considers the "Esperanto" case of a multilingual kingdom adopting a common lingua franca, and argues that "it would be easier for him [the king] to make them learn one of those languages that they used to speak, or his own language; this would be easier and more plausible than the invention of a new language afresh."

He concludes by tersely stating that "Some people imagine that their own language is the best of all languages; this is meaningless" and justifying it theologically and logically.

I like this guy. Makes me wonder what other early linguists had to say...

Ibn Hazm on language endangerment and the origin of language

I've been reading more of chapter 4 of the 11th-century work Ihkam Ibn Hazm - "On how languages come into being, whether by (divine) construction or establishment of convention" - and it's great. I found his description of how a language becomes endangered particularly compelling:

So when a community's state is destroyed, and their enemy gains power over them, and they are kept busy with fear and need and ignominy and serving their foes, then the death of their spirits is guaranteed - and that may cause their language to disappear, and their lineages and history to be forgotten, and their sciences to perish. This is both observed in reality and deduced through a priori reasoning. (Arabic begins: وأما من تلفت دولتهم...)‍

The main topic of the chapter is, of course, the origin of language. He argues that language must have been taught to man by God, because he argues that the three other possibilities that he considers - mutual agreement on a convention, instinct, or the influence of geography - are logically impossible. His argument on instinct is the most interesting: if language were an instinct, then we would all speak the same language. Chomsky, of course, inverts this: since language is an instinct, we all do speak the same language (modulo trivial details of vocabulary and parameters.) On mutual agreement, he notes that it is impossible that a languageless community could agree on a language; how would they have explained to each other what each word was supposed to mean? The idea that each place causes its inhabitants to speak a particular language - advanced as an explanation for linguistic diversity - he rejects as absurd, since any one place can, and generally does, have a variety of languages spoken in it.

I plan to describe more of the chapter later - his comments on conlanging are particularly amusing...

A comparative linguist of the 11th century

Ibn Hazm (994-1064) was a polymathic intellectual of Cordoba, equally well-known for his poetry and his religious commentary. Less well-known are his opinions on Semitic linguistics, which turn out to have been rather impressive. In the quote below, he demonstrates a clearer understanding of the process of historical change than Ibn Quraysh, who seems to have seen the mutual similarities as as resulting as much or more from intermixture than from common ancestry, although both ultimately succumb to the temptation of explaining linguistic family trees in terms of religiously given genealogies. As near as I can translate it off the cuff, he said:

...What we have settled on and determined to be certain is that Syriac and Hebrew and Arabic - that is the language of Mudar and Rabia (ie Arabic as we know it), not the language of Himyar (ie Old South Arabian) - are one language that changed with the migrations of its people, so that it was ground up... For, when a town's people live near another people, their language changes in a manner clear to anyone who considers the issue, and we find that the masses have changed the pronunciation of Arabic significantly, to the point that it is so distant from the original as to be like a different language, so we find them saying `iinab for `inab (grape), and 'asTuuT for sawT (whip), and thalathdaa for thalaathatu danaaniir (three dinars), and when a Berber becomes Arabized and wants to say shajarah (tree) he says sajarah, and when a Galician becomes Arabized he replaces `ayn and Haa with haa, so he says muhammad when he means to say muHammad, and such things are frequent. So whoever ponders on Arabic, Hebrew, and Syriac will become certain that their difference is of the type we have described, through changes in people's pronunciation through the passage of time and the difference of countries and the bordering of other nations, and that they are in origin a single language. Having established that, Syriac is the ancestor of both Arabic and Hebrew, and to be more precise, the first to speak this Arabic was Ishmael, upon him be peace, for it is the language of his sons, and Hebrew is the language of Isaac and his sons, and Syriac is without doubt the language of Abraham, blessings and peace be upon him and upon our prophet. (Ihkam Ibn Hazm; see Wikisource for original text, beginning الذي وقفنا عليه وعلمناه يقينا أن السريانية والعبرانية والعربية...)

My attention was originally drawn to this remarkable quote by an article by Ahmad Shahlan, in a rather strange Libyan book fusing pan-Arab nationalism with Semitic philology, at-Tanawwu` wal-Wahdah fi l-lahajaati l-`uruubiyyati l-qadiimati, which probably merits a post in its own right at some point.

More from Qatar

I'm here on holiday in Qatar for a while yet, and it's been great. However, I recently heard (at second-hand) a story I just have to share... Apparently, a teacher came across a kid in her second grade class who somehow hadn't learned to talk. After enquiries, it emerged that the child's parents weren't home much. The father wasn't interested in interacting with babies, and the mother was out working and socializing pretty much all the time. So, of course, the kid was being brought up by the maid... and they had strictly forbidden her from talking to their children, for fear the kids might pick up an uncouth accent or, even worse, a different language!

This may make more sense if you consider the frankly bizarre demographics of this country, one of the world's richest and most multicultural. Of the 576,000 inhabitants over the age of 15, only 110,000 are Qataris (who are well subsidized by the legal requirement of Qatari majority ownership of any businesses formed here, and by the oil money); the rest are expatriates from all over the world (in just these couple of weeks, I've heard or seen Arabic, English, Urdu, Malayalam, Persian, Chinese, Turkish, and Swahili used here). While the Qatari population has a more or less 50:50 sex ratio, the non-Qatari population is 77% male. Among the Qataris, more than twice as many women as men make it through university; male drop-out rates are consistently higher, even in primary school (!) It will be interesting to see how the country copes with this over the coming years.

Naxi in Qatar

Yesterday I watched an excellent Chinese documentary called E-Ya Village at the Al-Jazeera TV Production Festival in Doha. It covered aspects of this isolated Sichuan Naxi mountain village's daily life, but focused mainly on their religion, covering what they did for naming, coming of age, mourning, New Year, various sacrifices...

The film was full of (subtitled) Naxi dialog, but what I found most linguistically interesting was the writing system. As everybody should know :), Naxi has a complex pictographic writing system of some antiquity, called Dongba after the priests of their religion. In the film, no secular books or newspapers featured, and the few signs (at the clinic's entrance) were written in Chinese; but Dongba was used several times, always in a religious context. In particular, its most obvious "practical" use was for prayer flags put up in mourning contexts: whenever these flap in the wind, the wind is said to carry the words written on them, sections of the Naxi holy book, to the realm of the dead. It suggests a functional interpretation of the Dongba writing system as one intended essentially, not for communication with the living, but for communication with the spirit world. This has suggestive if not exact parallels - consider Mandaic's traditional functions, for instance. But obviously one would want to see more than just a film to analyze the issue!

The festival, incidentally, was very international, with numerous Persian, Chinese, Latin American, and French films as well as the Arabic ones. Unfortunately, they were let down by insufficient subtitling: non-Arabic films were subtitled only in English, if at all, while Arabic films were not subtitled, substantially restricting the audience for both. Hopefully next year they'll try to remedy this.

Crow language

I was Googling for an essay I'm planning to write on reduplication in Siouan languages the other day, so I typed in "Crow language". I was somewhat surprised to come across an article on the vocalizations of the American crow as my second hit, so I thought I'd share it. Apparently, 27 different vocalizations have been noted in the scientific literature (with names like "scolding call", "distress call", "courtship vocalizations", and "pre-mortality call"), and many remain undeciphered, so to speak. It would certainly be interesting to get a really good idea of the communication system of an animal as intelligent as the crow; bees are all very well, but perhaps a little too alien to compare sensibly with human language.

I never did find anything very helpful online on the Crow language, but John Boyle's Siouan Languages Bibliography will certainly come in handy, and this sketch of Omaha-Ponca seems good, though of limited use for what I'm researching.

The Quran on linguistic diversity

In these times of widespread language extinction and of "religious" tensions, I thought some readers might be interested to hear what the Quran has to say about linguistic diversity. The most important text is, of course, 30:22:

And one of His [God's] signs is the creation of the heavens and the earth and the diversity of your tongues and colors; most surely there are signs in this for the learned.

As the context makes clear, this is part of a more general Qur'anic pattern in which this universe itself - the normal, everyday events that we look at as just the way things are - is identified as a sign from God; the Creator's nature is reflected in His creation. So the thrust of the verse is that linguistic diversity is a part of nature, and as such a part of God's plan for the world.

Another relevant verse, which, in light of 10:47 ("for every nation there is a messenger"), puts the idea that the Quran being written in Arabic makes Arabic the best of all languages into perspective, is 14:4:

And We [God] did not send any messenger but with the language of his people, so that he might explain to them clearly; then God makes whom He pleases err and He guides whom He pleases and He is the Mighty, the Wise.

Libyan Berber and a remarkable Laghouat archaism

In case you ever wondered where in Libya Berber languages are spoken, check out this map at Tawalt. Note at least two oases that don't get any mention in the Ethnologue - Ubari and al-Fogaha.

I was talking to a guy from Laghouat the other day, and it turns out that in that area people say `ma عما rather than more widespread Maghreb Arabic m`a معا for "with". This is rather interesting, in the light of Aramaic `am and Hebrew `im... wonder whether it's a coincidence, or a Syriac borrowing somehow picked up by an Arab tribe en route to Algeria? Anyone heard of another Arabic dialect where this happens?

Chechnya Day

Apparently, it's Chechnya Day today - commemmorating Stalin's expulsion in 1944 of the entire Chechen people from their homeland to Central Asia (along with a number of other minority groups, such as the Kalmyks). Something like half the population died en route. The survivors were not permitted to return for more than ten years.

So, marking the occasion, here are some Chechen language links:

• Chechen - Johanna Nichols
  
• Scanned Chechen language materials online (possibly illegal, but very helpful)

Like most Caucasian languages, it's quite interesting, with no known relatives outside the Caucasus, a complex gender system, and a remarkably large set of phonemes. Its language family, Northeast Caucasian, may be the closest living relative of two long-extinct languages once written in cuneiform north of the Fertile Crescent, Hurrian and Urartian.

Classical Kanembu

I went to a very interesting seminar on classical Kanembu this week. It's a highly conservative form of Kanembu/Kanuri, written in the Arabic script, used mainly (exclusively?) for commentary on (and translation of) Arabic religious texts. The earliest dated example is a bilingual Quran from 1669, currently being studied here at SOAS. I don't want to comment in too much detail, because I'm not sure how much they've published on it yet, but a couple of things particularly struck me:

• Classical Kanembu is still used and written by Islamic scholars of the area - although, apparently, Western scholars only became aware of this fact quite recently.
  
• It has substantially more cases than modern Kanuri, and possibly an even more complicated verb morphology.
  
• Most strikingly, since vowel length is non-phonemic in Kanuri, it seems to use vowel length to indicate high tone instead; thus, for example Arabic al-'aakhirah "the afterlife" has been borrowed as laxíra, and thus gets spelled as لاخِيرَ. As far as I know, this would make it the only Arabic orthography to mark tone. (Actually, Dmitri Bondarev, who observed this, prefers for the moment the more conservative interpretation that the vowel length commonly corresponds to a modern Kanuri high tone, not ruling out the possibility that such vowels were actually long in the Kanembu of the seventeenth century.)

This already constitutes some of the oldest documentation of any West African language, and quite apart from its implications for the reconstruction of Proto-Saharan, it really makes one wonder what other valuable historical data on other African languages linguists might be missing out on by not studying Arabic/Ajami use. So keep your eyes peeled, and tell me if you spot anything!

Where have you been? - a semantic change in progress?

Just a random observation...

Good sentences: I've been to Finland, I'll have been to Finland five times, he's never been to Finland...

Terrible sentences: *I am to Finland, *I will be to Finland, *I was to Finland, **he never is to Finland...

It seems that "been", in some cases, can act as an alternative past participle for "go", replacing "gone". A linking environment is provided by sentences with a bare locative adverb: I've been there / I am there / I will be there are all fine. Presumably, since "I've been there" normally implies "I've gone there" (unless you've been locked up there since birth or something), the "been" was reinterpreted as an even more irregular past participle of "go".

"I'll be there" equally implies "I'll go there", so it's odd that this wasn't extended similarly to allow "*I'll be to Finland" - or so I thought, before checking Google. Google does reveal a couple of instances: "I'll be to bed in a minute", "I'll be to work way early, and perhaps most strikingly, "I've been to more than half of the counties, and in the next six weeks, I'll be to the other half of the counties". So it seems we have a change in progress. Does this depend on the region? Will it culminate in a complete merger of "go" and "be"? Are there any parallels to this outside English? What do you think?

In Languages We Live

Just watched an interesting new film last night, called In Languages We Live/In Sproget Jeg Er (I don't vouch for the second title's accuracy.) It's a film about linguistic diversity, essentially, with cameos from a number of communities including Mla'bri, Totonaco, and Pitjantjatjara, as well as larger languages, such as a rather fun Arabic Hamlet (with Claudius as an Arab dictator, of course), a newscaster who speaks a dialect of Mandarin (Xiang, I think) natively but wants to bring up her kids only speaking Standard Mandarin, and "sheng", the Swahili-English-other street slang of Nairobi teenagers - not to mention, of course, the English and Danish of the narrators. It also had a brief meeting with the last(?) native speaker of Livonian - who apparently has more people to talk to than you might think, what with the steady stream of Finnicists beating a path to his door! But the most memorable bit was the brief narration, in the original language (Arrernte? I'd have to rewatch the film), of the Australian government's 1950's policy of forcibly separating Aboriginal kids from their parents, as witnessed by one of the parents: parents trying to physically hold on to their kids as the police tore them away; parents running with their kids to hide in local ravines, and being tracked down by the police; children crying as they were driven away... It would be hard to believe that such a policy was being practiced just fifty years ago, if the twentieth century weren't already full of such cases.

Tonkawa

There is little information on Tonkawa, a language once spoken in southern Texas. The earliest data is a word list from 1829, but nothing else was recorded until 1876. The odd thing is that the words recorded from 1829 are either identical to the later ones or, very often, totally different; in the latter case, they are usually transparently derived from different words, usually verbal forms or compounds. No such rapid vocabulary turnover is observed in later work; between 1876 and 1928, nothing much changed. Why would something like this happen?

Apparently, the explanation is remarkably simple, if tragic. The Tonkawa, like many Native American and Australian groups, had a strong taboo against mentioning the name of a dead person, and, when a dead person's name resembled a word in common use, would replace that word. (To reduce this problem, they often gave their children Comanche names.) And, on October 24, 1862, 167 Tonkawas were massacred by an alliance of enemy tribes near Anadarko, Oklahoma. (Given the circumstances of the time, I wouldn't be surprised if this wasn't the only massacre between 1829 and 1876 either.) One side-effect was the sort of rapid vocabulary change that should normally be just about impossible. This is one reason why you don't want to rely on glottochronology too much.

(For more details, see Ives Goddard in ed. Campbell and Mithun, The Languages of Native America, University of Texas: Austin 1979, pp. 358-363.)

Desertification in Algeria

Nothing much to do with linguistics, but... Apparently (the book I was reading cites Dresch 1986), the turning point in the desertification of the High Plateaus, the dry zone forming the northern boundary of the Sahara in Algeria and Tunisia, came with French colons' introduction of European-style farming methods. Traditional cultivation methods there left the land fallow one year out of every two, and used nothing sharper than a hoe, leaving numerous clods in the soil, thus keeping erosion to a minimum. The French methods, using a much deeper digging plough and completely weeding the soil, left the soil finely powdered and ready to disappear Dust Bowl-style in the event of any major windstorms, while their expropriation of the land pushed the Algerian farmers into more and more marginal areas, removing their plant cover and rendering them more vulnerable to erosion in turn. The US' Dust Bowl formed in a rather similar way - "sodbusters" removing the crust and vegetation cover that kept the land from blowing away - if on a far larger scale. The process has continued even since independence, due to the obvious short-term financial incentive to plant more land; the result has been "increasing sensitivity to wind erosion and a general degradation of the steppe ecosystem". Nothing so far about Boumedienne's "green line"; I wonder if it made any difference?

I find this interesting in general, but particularly telling given that I recently found some online would-be expert (I don't think I'll bother to link him) boasting that French contributions to Algeria included the "introduc[tion of] modern agriculture". It's amazing how you can put a good face on something just by picking the right words - and on that note, I think I've finally found a linguistics connection for this post.

El-Fogaha

El-Fogaha (الفقهة, or to its own inhabitants el-Foqhat) is a small oasis in central Libya, southeast of Sokna between Waddan and Zwila, whose few hundred inhabitants speak (spoke?) a Berber language. It is not quite the easternmost outpost of Berber - Awjila and Siwa are both further east - but a case could be made that it is the most obscure one. Its vocabulary is heavily influenced by Arabic, but retains some archaic Berber words lost in most westerly dialects, such as isin "tooth", azal "day". Unlike the Ghadames dialect or Tamasheq, but like most Berber varieties, it has reduced proto-Berber *B (Tuareg h) to semivowels or null: aiyaḍ "night". A couple of the plurals end in -aw, whereas in more westernly varieties they would end in -a:

• tamûrt "land" > tmu:râw (Contrast Kabyle tamurt > timura, Tumzabt tamuṛt > timuṛa; compare Ghadamsi tămmurt > tmuro)
• tanâst "key" > tnisâu (Contrast Tumzabt tnast > tinisa; compare Ghadamsi tonest > təniso)
• talîlt "terrace" > tli:lâu

The dropping of final semivowels in Northern Berber is well-known, but I had been wondering where the -o in the Ghadamsi plurals came from. I guess now I know. What I don't get yet is why this set of plurals ends in -a/aw/o in any case, rather than the regular feminine plural -in.

For more on this language, see Umberto Paradisi, "El-Fogaha, oasi berberofona del Fezzân", Roma:Bardi 1961.

PS: A belated Eid Mubarak to everyone!

Istanbul, bishops, Rohingya, and Tamezret

For this week, I thought I'd share two curiously parallel reanalyses I've come across recently:

• Istanbul, apparently, derives from the Greek phrase eis ten polin, "in the city";
  
• 'usquuf, "bishop" in Arabic, which apparently derives from a Coptic reinterpretation of Greek episkopos "bishop" as e-pi-skopos "to the skopos", due to which skopos was reanalyzed as meaning "bishop".

And a couple of interesting language sites I've come across:

RohingyaLanguage.com is a speaker's effort to promote the Rohingya language. The Rohingya are a Muslim minority group of the western coast of Burma. Like virtually all Burma's inhabitants, they have been seriously mistreated by the government. Apparently, their language is most closely related to (a dialect of?) Chittagongian Bengali. If anyone figures out what the acute accent is meant to indicate, do tell me...

Atmazret.info (in French) is all about the endangered Berber language of Tamezret in southern Tunisia, written by a descendant of speakers. Though he's not a trained linguist, this qualifies as quite an important documentation effort in its own right; as far as I know, the only other thing ever published on the Tamezret dialect was Märchen der Berbern von Tamzratt im Süd-Tünisien in 1900, more than a century ago.

Andamanese Phrasebook

Some time ago, I found a copy of perhaps the only five-language phrasebook for the Andaman Islands. The Andaman Islands are a remote island group south of Burma belonging to India. Up to the 19th century, they were inhabited by a number of tribes with Stone Age technology, no significant contact with the outside world, and languages extremely different from any spoken anywhere else. Some of their descendants remain there today, but are a tiny minority except in a couple of areas in the south, and most of their languages are extinct. This phrasebook was written by one A. J. Portman in 1887, when the islands had been turned into a British penal colony, for the use of government officials. Some of the entries paint an interesting picture of life in the colony; for economy's sake, I have given only the Aka-Bea equivalents.

"That woman is wearing his skull." Kát apáil lá ót chetta ngāūrók-ké.

"Some convicts have escaped, you must search for them." Jó chāōga lá kájré, áb átaká."

"This village is very dirty." Ká báraij lót láda-da.

"You will be bitten by sandflies and mosquitoes." Nyípá, ól bédig téil ngáb chá-pinga.

"Don't sing, or there will be a storm." Ngódá rámitóyo-ngayábada, élér-wulké.

"He is a boy, and may not eat turtle." Kát áká kádekada, óda yádi mék-nga yábada.

"Do you eat grubs alive?" Án wai ngó butu ligátí mék?"

"You must bring them in by force." Ngó ítár pórawa.

And finally:

"Is there anyone here who understands his language?" Tén kárin míjólá áká teggí gádí-áté?"

They just don't make phrasebooks like this any more...

Oldest African dictionaries

Some time ago, I came across a web page characterizing a dictionary of Kenzi Nubian dating from 1635 as "the oldest dictionary of an African language". Much as I appreciate their work in getting this very interesting material online, that claim is out by at least 500 years, if not 1000.

The oldest arguable dictionary of an African language that I am aware of so far is the Greek-Coptic Glossary of Dioscorus of Aphrodito, which apparently* dates back to the 6th century. Ibn al-'Assal's Arabic-Coptic sullam muqaffa, written in the 1200s, can quite unhesitatingly be described as a dictionary; following a then-current Arabic tradition, it was arranged alphabetically from the last letter of the word backwards (so, for instance, "apple" would be close to "people" but far from "apricot".) This arrangement was meant to aid in the composition of rhymed prose and verse. Other examples, many arranged semantically, are given by the Encyclopedia of Islam article Sullam (literally "ladder"). In Ethiopia, traditional Geez-Amharic lexicons are titled Sawasew, or "ladders"; I thus assume they are of Coptic inspiration, though I haven't been able to find any detail about when they started to be written.

After Coptic, the next oldest is an Arabic-Berber lexicon written in 1145, containing some two thousand words. Its writer, Abu Abdallah Muhammad ibn Ja'far al-Qaysi, better known as Ibn Tunart, was born in Qalaat Bani Hammad (modern Algeria) and wrote the work in Fez (Morocco) - a second little-known Algerian medieval linguist to add to my list after Ibn Quraysh! The book contains some paradigms and verbs, but consists principally of a list of Arabic nouns with Berber equivalents or glosses, arranged by semantic field; it inspired several later Moroccan lexica. Nico van den Boogert is working on republishing it.

What other African dictionaries predate Carradori's? I don't know, but I can hazard some guesses - Geez, Swahili, Kanuri, and Nubian itself would certainly be worth checking.

* According to Adel Y. Sidarus, “Coptic Lexicography in the Middle Age, The Coptic Arabic Scalae,” in The Future of Coptic Studies ed. R. McL. Wilson (Leiden: E.J. Brill, 1978), 123

A comparative linguist of the 10th century

Yehudah ibn Quraysh was a rabbi of the late ninth/early tenth century from Tahert (modern Tiaret, in Algeria.) Shocked to hear that the Jews of Fez in Morocco were neglecting the study of the Targum (an Aramaic translation of the Bible), he wrote a letter to them intended to establish that they could not and should not get by on the Hebrew alone - because other languages, especially Aramaic and Arabic, are essential in elucidating the Hebrew. In the process, he casually noted most of the correct sound correspondences between Hebrew and Arabic, and ended up writing what amounts to an extensive comparative dictionary of the three languages, even throwing in 9 Berber comparisons and 5 Latin ones at the end. He definitely hedges his bets on the cause of this obvious similarity between the three languages, but seems to come surprisingly close to the correct explanation - common descent - at times... something to bear in mind next time you read about Sir William Jones having founded comparative linguistics in 1798.

Here is what he had to say about it, as far as I can translate it:

I then resolved to put together this book for people with understanding, so that they should know that Syriac [Aramaic] expressions are scattered throughout the whole of the Holy Tongue in the Bible, and Arabic is mixed with it, and occasionally bits of Ajami [Latin] and Berber - and principally Arabic in particular, for in it we have found many of its strangest expressions to be pure Hebrew, to the point that there is no difference between the Hebrew and the Arabic except the interchange of ṣād and ḍād, and gīmel and jīm, and ṭet and đ̣ā', and `ay(i)n and ghayn, and ḥā' and khā', and zāy and dhāl. The reason for this similarity and the cause of this intermixture was their close neighboring in the land and their genealogical closeness, since Terah the father of Abraham was Syrian, and Laban was Syrian. Ishmael and Kedar were Arabized from the Time of Division, the time of the confounding [of tongues] at Babel, and Abraham and Isaac and Jacob (peace be upon them) retained the Holy Tongue from the original Adam. The language became similar through intermixture*, just as in every land adjoining a land of a different language we see intermixture of certain expressions between them and the spread of language from one to another; and this is the cause of the similarities we have found between Hebrew and Arabic...

The original was written in classical Arabic using the Hebrew script; I retranscribe it into Arabic script here:

فرأيت عند ذلك أن أؤلِّف هذا الكتاب لأهل الفطن وذوي الألباب، فيعلمو أن جميع לשון קדש (لغة القداسة: العبرانية) الحاصل في المقرأ (الكتب المقدسة) قد انتثرت فيه ألفاظ سريانية واختلطت به لغة عربية وتشذذت فيه حروف عجمية وبربرية ولا سيما العربية خاصة فإن فيها كثير من غريب ألفاظها وجدناه عبرانيا محضا، حتى لا يكون بين العبراني والعربي في ذلك من الاختلاف إلا ما بين ابتدال الصاد والضاد، والجيمل (حرف عبراني: ڱ) والجيم، والطِت (حرف عبراني: ط) والظاء، والعين والغين، والحاء والخاء، والزاي والذال. وإنما كانت العلة في هذا التشابه والسبب في هذا الامتزاج قرب المجاورة في البلاد والمقاربة في النسب لأن תֶרח (تِرَحْ) أبو אברהם (ابراهيم) كان سريانيا وלבן (لابان: حمو يعقوب) سريانيا. وكان ישמעאל (اسماعيل) وקדָר (قيدار) مستعرب من דוֹר הפלגה (زمان الاختلاف)، زمان البلبلة في בבל (بابل)، وאברהם (ابراهيم) وיצחק (إسحاق) وיעקב (يعقوب) عليهم السلام متمسكين بـלשון קדש (لغة القداسة: العبرانية) من אדם הראשון (آدم الأول). فتشابهت اللغة من قبل الممازجة، كما نشاهد في كل بلد مجاور لبلد مخالف للغته من امتزاج بعض الألفاظ بينهم واستعارة اللسان بعضهن من بعض، فهذا سبب ما وجدناه من تشابه العبراني بالعربي...

(Source: D. Becker, Ha-Risala shel Yehudah ben Quraysh, Tel Aviv University Press, Tel Aviv 1984.)

* I previously mistranslated this, having misread qibal as qabl.

Update: Read more at http://lameen.googlepages.com/ibn-quraysh.html.

Curiosities of Semitic articles

As David Boxenhorn noted in his comment to the previous post, the definite articles of Hebrew and Arabic display two odd-seeming properties:

• agreement: if a noun has a given article, so does any adjective modifying it. Thus "a short boy" in Arabic is walad-u-n qaSiir-u-n (where -n marks indefiniteness, and -u marks the nominative case), whereas "the short boy" is al-walad-u l-qaSiir-u. (The vowel of al- elides when preceded by another vowel.)
  
• In direct compounds of two nouns (possessed-possessor, or more generally modifier-modified), the first noun cannot take any article. Thus in Arabic you can say yad-u l-walad-i "the boy's hand" or yad-u walad-i-n "a boy's hand" (-i marks the genitive case) but not *al-yad-u l-walad (intended to be "the hand of the boy") or *yadun al-walad (intended to be "a hand of the boy").

The second property isn't actually all that "exotic" - English does the same thing! You can say the man's hat, but never *the man's the hat or *the man's a hat; just as in Arabic or Hebrew, to make the full range of possible definiteness distinctions you have to resort to prepositions.

Definiteness agreement between nouns and adjectives is more unusual, but at least one Indo-European language has it: Norwegian. No question of substratum influence there, certainly... Anyone have another example?

However, in determining whether or not the article represents a shared innovation, the question is whether other relatives have it. I recall that Biblical/Imperial Aramaic did (later varieties lost it), but I'm not sure of the detailed behavior of its definite article. The Berber obligatory noun prefixes probably derive from an original article (see my post Beja and Beyond) but, though distinctly similar to the Beja definite article, don't seem directly comparable to the Arabic and Hebrew ones.

Demonstratives in Semitic and beyond

Rishon Rishon just posted a table comparing the words in my previous post to Hebrew. Most of them are correct; mo`ed and g'vul are not cognate, and SaH and loa` I'm not sure about. However, one is particularly interesting: ha- = 'al- "the". You often find this seeming cognate cited in works on Semitic: after all, ha- induces gemination of a subsequent non-guttural consonant - suggesting a lost consonant in the prefix assimilating to the subsequent letter - and Hebrew h- occasionally seems to soften to '- in Arabic (the causative measure hiph`iil corresponds to 'af`ala, for example.) Trouble is, the only letter in Hebrew that regularly assimilates to a following consonant is n (although l does admittedly assimilate in the verb laqaH), and the Safaitic inscriptions seem to reveal an early pre-Islamic northern dialect of Arabic which did have a definite article h- (hn- before gutturals, thus hn'lt for Al-Lat.) So are there any other possibilities?

I think so. ha- corresponds pretty well to Arabic haa "here is", which is also the obligatory prefix to the demonstrative "this" (haadhaa, haadhihi, haa'ula'). Compare also Syriac haanaa "this (m. sg.)" - which appears to reveal an added n which could explain the Hebrew doubling (and the Safaitic form) nicely. Conversely, 'al- corresponds well to Hebrew 'elleh, Arabic haa-'ulaa', Syriac haaleyn "these". The vowel doesn't correspond exactly, but then it doesn't in the certain cognate 'elleh = haa-'ulaa' either. This idea has probably already been put forward (or indeed knocked down) somewhere in the literature, for all I know, but there you go.

In either case, both definite articles would derive originally from unstressed demonstratives - a process so common it's barely worth commenting on. For example, all the Romance languages derive their definite articles from Latin demonstratives - usually illum/illa "that" (before the noun except in Romanian), but istum "that" in Sardinian. Likewise, the Coptic definite article pe- (m.)/te- (f.)/ne- (pl.) derives from the ancient Egyptian demonstrative pn (m.), tn (f.), nn (pl.) "this". Indeed, English "the" derives from the same old English word as "that". Come to think of it, I don't know of any definite articles offhand that don't derive from demonstratives; can you think of any?

Eid Mubarak!

Eid Mubarak عيد مبارك, or, as they say in Algeria, Sahha Eidek صحّا عيدك to everybody!

Today is Eid al-Fitr, the day on which the Ramadan fast ends and the subsequent feasting begins. The Arabic term (`iid al-fiTr عيد الفطر) means "Festival of Fast-breaking"; the original meaning of the root fTr seems to be "cleave, cut open", from which it acquired the senses of "form, make" on the one hand and (in a metaphor somewhat similar to the English one) "break fast" on the other. In North Africa, it is more generally known as El Eid Es Sghir (l`id SSghiR العيد الصغير), "the small festival" (as opposed to Eid al-Adha, the "big festival").

PS: luggi turns out to come from a widely attested Berber word ileggwi, meaning a spiny plant (variously broom or needle-furze) - as Salem Chaker was the first to suggest.

Mpre

A tantalizingly brief note of 1931 in the Gold Coast Review describes an ethnic group called the Mpre, found only in the village of Butie in central Ghana (8° 52' N, 1° 15' W) near the confluence of the White and Black Voltas, apart from a few emigrants in Debre. According to the author's description, the Mpre people, once more widespread, were reduced to a single village in the course of comparatively recent wars with the Asante. Noting that their language was “different to that of the surrounding tribes”, he lists 106 words of Mpre. This short vocabulary appears to be the only existing record of the language, which is believed to be extinct. The gap is all the more unfortunate because Mpre turns out to be of some taxonomic significance. It is not closely related to any of its neighbors, and Heine and Nurse (2000) treat it as unclassified. A friend of mine's paper dealing partly with this will be appearing sometime soonish, but I won't spoil the surprise...

You might think, given all this, that it was impossible to retrieve any information on its grammar. However, you would be wrong! Fellow language geeks may find it an interesting exercise to try their hand at extracting grammar information from the wordlist, which Blench gives a copy of, before reading on...



The wordlist strongly suggests a noun class prefix system still at least partially productive. The highly lopsided initial letter statistics would alone suggest this: 31 entries begin with e-, 21 with a-, and 12 with n-, together accounting for the majority of the wordlist. This speculation is confirmed by distributional analysis for the e- which appears in the numbers 1-5, but disappears in 11-13 and 20-30; it is presumably to be identified with the Ga prefix é- observable in the same numbers. (The change of ekpe “one” to mpe in “11” is noteworthy, if it is not a typo.) Likewise, comparison of kelafa “100” with lefanyo “200” reveals a prefix ke- - with precise analogues in Ch./Kr. kʌ́-, Na. gʌ́-, and Go. ká- in the same numbers. Of the 21 entries with a- (corresponding to 19, or possibly 18, distinct words), five are glossed as plural in English, while another four are glossed as collective nouns; no entries not beginning with a- are glossed as plural. I therefore conclude that a- is a marker of plurality - suggesting that ado (the formative element in “20”, “30”, ...) is the plural of edu “ten”. This jibes nicely with other languages of the area: a plural prefix a- is found in Gonja, Twi, Lejana, Akpafu, and Avatime, for example.

Identifiable compounds include zingilzi-nogha “bush cow” (cf. zingelza “bush”, nogha “cow”), sunko kawuseggi “earth owner or tindana” (cf. sunko “earth”), nkemnzui “son” (cf. nzui “child”), lefanyo “200” (cf. enyo “2”, kelafa “100”), eputo nasi “foot” (cf. eputo “leg”) ; all suggest a word order type Modifier-Modified. “Lion” (jikpajikpakoseggi) must surely be a compound, in which I would identify the final koseggi with kawuseggi “owner (?)” above? Also, ataza “finger” and atazai “toe” are clearly related, but it is unclear whether one is a compound form or whether both are simply different transcriptions of the same word.

One short sentence is given - agbem aba “it rains” (cf. agbem “God”). Assuming that this is of the form SV, this could be taken to suggest verb agreement in gender (or at least number) with the subject; however, this is by no means certain.

SOAS, epiglottal trills, Sergei Starostin

Today I attended my first lecture as an MA student here at SOAS - on phonology. Nothing much to report yet; the highlight had to be our lecturer's demonstration of an epiglottal trill (which, believe it or not, actually occurs in some Caucasian languages; I think she named Aghul.) It's a remarkable sound - impossible to confuse with any sort of pharyngeal.

In other news, I was sorry to hear that Sergei Starostin has died. I met him briefly at the Santa Fe Institute, and am one of many to have benefited from his online comparative databases. He will be missed, particularly at the EHL project.

Poetic grammars

Grammars come in many flavors nowadays - Chomskyan, functionalist, structuralist... However, grammars in verse are something you don't see too often nowadays, so I was recently pleased to come across the ''Alfiyyat Ibn Mâlik'', a 1002-line poem describing Arabic grammar; as it says in line 3:

وأستعين الله في ألفية * مقاصد النحو بها محوية

Wa-'asta`înu llâha fî 'alfiyyah * maqâsidu nnahwi bihâ mahwiyyah

And I seek God's help in a thousand-line
Poem in which grammar's basics are outlined

It was written in the 13th century by one Muhammad Ibn Mâlik, a native of Jaen in Spain who emigrated to Syria. The poem was memorized in order to aid the student in recalling the more obscure details of Arabic grammar (strictly prescriptive, of course...) Unfortunately, the poem proved somewhat obscure to prospective students, prompting the writing of commentaries on it, such as Sharh Ibn `Aqîl, in which each verse or group of verses was explained in greater detail. As a sample of the style, I present verse 229:

ويرفع الفاعلَ فعلٌ أُضمرا * كمثل "زيدٌ" في جواب "من قرا"؟

Wa-yarfa`u lfâ`ila fa`lun 'udmirâ * kamithli "zaydun" fî jawâbi "man qarâ?"

And an implicit verb makes its subject nominative
Like "Zayd-NOM" in answer to "Who read?"

(Ie, the subject of a verb implied by context but not actually present in the sentence at hand takes the nominative.) I wonder what parallels exist in other grammatical traditions.

Incidentally, I'm back from Algeria now, and plan to report on more linguistic tidbits - as well as more luggi, on- or off-topic - shortly; I'm also starting at SOAS soon.

Hi from Algeria

In case I have any regular readers, I thought I should explain that I'm currently enjoying a holiday in Algeria and nearly incommunicado as far as the Internet is concerned. The local dialectological situation - a conservative qaf-dialect of Arabic in Dellys itself, a "Bedouin-type" gaf-dialect in the villages immediately around it, and Kabyle beyond it to the east and south - is quite historically suggestive, and I'm still looking into its origins; however, to be honest, I'm spending rather more time on the beach, which brings me to a question: does anyone have any idea what the etymology of "luggi", the local Arabic word for sea urchin, might be? Or even know of another area where this term is used? My guess would be Berber, but I haven't found a convincing answer yet. Many local seafood names seem to be found in Corriente's Dictionary of Andalusi Arabic (often with Vulgar Latin etymologies), but not this one.

Shakespeare was a hobbit...

or, anyway, that was my reaction to the reconstructed pronunciation of a Shakespearean accent provided by the BBC. Apparently, the Globe is planning to stage Troilus and Cressida in its original pronunciation soon - thus bringing a little life back into Shakespeare's dreadful puns! I suggest that for their next challenge they should try reconstructing Fluellen's pronunciation - Elizabethan English with a thick Elizabethan Welsh accent, presumably.

The American Language

I've been reading Mencken's The American Language (Supplement I, 1945), and find it tremendously entertaining in small doses:

Since the earliest days the two Houses have devoted immense amounts of time and wind to pursuing such wicked men and things as Bourbons, slavocrats, embargoroons, gold-bugs, plutocrats, nullifiers, war-hawks, embalmed beef, ..., economic royalists, princes of pelf, land-grabbers, land-sharks, mossbacks, the open shop, the closed shop, and labor and other racketeers. Even Washington made a contribution to the menagerie with his foreign entanglements; as for Jefferson, he produced two of the best bugaboos of all time in his war-hawks and monocrats. From 1875 onward until the late 80s waving the bloody shirt was the chief industry of Republican congressmen, and from the early 90s onward the crime of '73 engaged the Democrats.

It's also genuinely informative at times, providing, for instance, an extensive list of words of Algonquian origins, and revealing that the term African-American (whose modern popularity, of course, came long after the book was written) is not a pure neologism, but has roots in a term that was popular around 1835, Africo-American, and one from 1880, Afro-American. (In a footnote to that section, he quotes a Liberian diplomat as noting that "Liberians consider the term Americo-Liberian opprobious as reflecting upon their [ancestors'] condition of servitude in the United States. Hence they prefer to be called civilized or Monrovian Liberians to distinguish them from the natives of the hinterland..." Diplomatic speech does change!)

In his discussion of social attitudes towards the emerging American dialect, he gives an 1820 quote from a British reviewer, Sydney Smith, that, apparently, "rankled in American bosoms for many years":

In the four quarters of the globe, who reads an American book? or goes to an American play? or looks at an American picture or statue? What does the world yet owe to American physicians or surgeons? What new substances have their chemists discovered? or what old ones have they advanced? What new constellations have been discovered by the telescopes of Americans? Who drinks out of American glasses? or eats from American plates? or wears American coats or gowns? or sleeps in American blankets? Finally, under which of the old tyrannical governments of Europe is every sixth man a slave, whom his fellow-creatures may buy and sell and torture?"

What an embarrassment for the poor fellow - to have been significant enough to give such offence, yet to be remembered two hundred years later principally for the shortsighted arrogance of his sneering asides! Even his last sentence, a justified blow at the time, would soon be made obsolete by a titanic effort. Let the Kilroy-Silks of our own day take note.

Claim of responsibility for the London murders

Language Log has recently posted twice on the bizarre name of the organization claiming to have carried out the attack. An apparently accurate screenshot of the claim can be found on Wikipedia.

The first interesting thing about this statement is the bizarre phrasing of its opening: والصلاة والسلام على الضحوك القتال سيدنا محمد صلى الله عليه وسلم. The Guardian renders this as "may peace be upon the cheerful one and undaunted fighter, Prophet Muhammad, God's peace be upon him." The doubling of "peace be upon him" (a formula added to the prophet's name as a matter of course) is unusual [because of its redundancy] and stylistically flawed, suggesting an imperfect command of Arabic literary style. The phrase الضحوك القتال (ad-Ḍaḥûk al-Qattâl), rendered by the Guardian as "the cheerful one and undaunted fighter", is composed of two words in apposition which Hans Wehr's dictionary renders as "frequently, or constantly, laughing; laugher" and "murderous, deadly, lethal". This extremely unusual epithet is so weird that at first sight I assumed it must be some kind of prank; it may potentially provide some clues to the identity of the killers.

Such an opening has been used at least once before in Europe: the assassin of Theo van Gogh left a note on the body opening after the standard invocation of God's name, with Vrede en zegeningen op de Emir van de Mujahideen, de lachende doder Mohammed Rasoeloe Allah (Sala Allaho alaihie wa Sallam), ie "Peace and Blessings from on the Amir of the Mujahidin, the laughing killer Mohammed the Prophet of God (God's peace be upon him)", which is almost identical, right down to the doubled "peace be upon him". A similar but less repetitive formula was used by Zarqawi in a purported claim of responsibility for the killing of the governor of Nineveh last year on CNN, and a Google search suggests that (again without the repetition) it occurs in other Iraqi insurgent notices. The term itself is probably copied from the 14th-century Hanbali writer Ibn Taymiyya's as-Siyasa ash-Shar'iyya, whose author, living at the height of the Mongol threat, spent much of his time urging people to fight; it does not seem to occur in any of the accepted hadith books.

The third really weird thing about the message is the phrase ابشرى با أمة الاسلام ابشرى يا امة العروبة : "Rejoice oh community of Islam, rejoice oh community of Arabdom". This collocation itself appears to be well-established, if rare - the phrase "community of Arabdom" (ummat al-`Urûbah) gets only 37 google hits, but many are collocations of one sort or another with "community of Islam", and come from speeches or interviews by well-known politicians. However, it does not seem to form any part of the standard rhetoric of so-called "jihadists".

Finally, it's worth noting that the Qur'anic quote at the end (47:7) contains a typo, if an easy one to make: it has لله lillâh "to God" for الله Allâh "God", omitting an alif. (I looked again, and the alif is there; it's just thinner than the adjacent letters, so my eye processed it as part of the subsequent lam. Oops!)

PS: Juan Cole explores, among other things, the implications of the "Arabdom" phrase.

PPS: Shibli Zaman also examines the linguistics of the issue; his summary of the "urubah" issue is more detailed than mine.

Negative convergence

Among the many shared characteristics that make the Maghreb proper (Algeria, Tunisia, and Morocco north of the Atlas Mountains) a linguistic area, in my view (albeit a somewhat trivial one, given that only two fully separate languages are involved) is that of double negation: like French, most languages of the area have a negative particle both before and after the verb.

In Algerian Arabic, this is ma ... sh(i), which derives transparently from Arabic ma:, "not (past)" and shay', "thing" (as in constructions like ma: ra'aytu shay'an, "I didn't see a thing".) In Kabyle, the corresponding construction is ur ... ara, which is purely Berber but exactly parallel; ur or ul meaning "not" is found throughout the family, and ara comes from a root meaning "thing" or, as in Tuareg, "child". By contrast, Tuareg and Tachelhit, south of the Maghreb proper, both use negations based on cognates of ul alone, without any postverbal element. So when I came across the Chenoua negative - u ... sh - I naturally assumed this must be a rather interesting Arabic-Berber hybrid, with the Berber preposed negative and the Arabic postposed one. The Tamezret negative is similar, ul ... sh, and seems to fit the idea nicely.

However, it turns out that, despite appearances, this may not be the best explanation. Tarifit uses war ... sha, and Middle Atlas Tamazight optionally uses ur ... (sha). At first sight these seem to work, but the vowel seems odd if they derive from Maghreb Arabic shi. However, kra happens to be a well-attested Berber word meaning "thing", found also in Kabyle, and its expected form in Zenati dialects like Tarifit, Tamezret, and Chenoua would be *shra (this may be a real form, though I haven't come across it.) And what more natural environment to simplify a consonant cluster than in an unstressed grammaticalized particle?

The issue is examined from a rather different, syntactic, perspective in a paper by Ouali online, which ironically reveals an alternative construction in Tarifit which does seem to be half-borrowed from Arabic: ur ... shi. However, its data seem somewhat at odds with those I've found in other sources; there is substantial dialectal diversity within Tarifit, which may explain this.