This is PhoeniX's Typepad Profile.
Join Typepad and start following PhoeniX's activity
Join Now!
Already a member? Sign In
Interests: linguistics, comparative linguistics, Sino-Tibetan, Indo-European, Tangut, Berber Languages,
Recent Activity
Motivated by the posting of a series of religious early Islamic inscriptions on Twitter written by women, I'd like to discuss once again the spelling of the feminine ending in Arabic. One of the striking things about the Feminine ending, which in Classical Arabic is -at- followed by case endings... Continue reading
Posted 4 days ago at Phoenix's blog
Huh. How did that happen! Thanks for pointing that out. It's a screw up on the side of the journal. Our version on Google Docs simply has "μορς = murṣu", so we wrote it in Unicode and everything. Just checked our proofs, and it wasn't like this in the proofs. So it somehow bizarrely screwed up in between the proof phase and publication, which was about two weeks apart or so... That's annoying. As for the example of French final -e, yes that's somewhat comparable. And actually the somewhat haphazard incorporation of dialectal forms into poetry because it's metrically useful reminds me very much of what people think happens in Epic Greek. The difference both to the French and the Greek parallels is, if the modern transmission of the poetry is to be trusted, the word is always in this dialectal form. Always yarā, never yarʔā. Its occurrences should really be checked in the poetry though, and see if it ever yields metrical irregularities. I've found one Quran document where, surprisingly the word is spelled with a media Alif, suggesting (especially in that document) a post-consonantal glottal stop. A form that never occurs in any of the reading traditions... It's quite puzzling.
A starting assumption present among many Arabists is that the ʕarabiyyah -- or Poetic Koine -- was a form of the Arabic language, quite close to what we now know as Classical Arabic, and that it existed as a kind of inter-tribal form of communication for high poetic culture. The... Continue reading
Posted Oct 21, 2017 at Phoenix's blog
Ack! Seems like you're right. I overlooked the fī "in" in the description of Warš's tradition. So for those who are curious what the text says exactly: Ibn Mujāhid says: "Warš transmits on the authority of Nāfiʕ that he removed the ʔ in the likes of of yuʔminūna and what is similar to that, and likewise for the intervocalic (ʔ), like yuʔaxxira-kum, lā yuʔāxiḏu-kum and yuʔaddi-hī and what is like it." Whereas for Abū ʕamr he says: "He did not hamzate (= apply ʔ) to each pre-consonantal ʔ, like yūminūna, yūminu and yāxiḏūna and whatever is like that." The difference then comes down to the specific mention that each hamza is dropped, and that for Warš it is said "in the likes of" rather than "like". And that is indeed how it is understood in the modern readings, where for Abū ʕamr, e.g. ar-raʔsu is indeed read as ar-rās. The description is technically still somewhat ambiguous, after all what is "in the likes of" that Warš is referring to, but the lack of any mention that Warš did not apply the ʔ to each place where it is expected, makes the interpretation that we find in al-Jazarī probably a likely interpretation.
[NOTE: I'm probably wrong about this whole argument, luckily pointed out by Lameen Souag in the comments, see the comments for the discussion.] For a long time, I've been struck by the irregular nature of the reading traditions of the Quran. The Warš tradition for example, is well-known for its... Continue reading
Posted Jul 24, 2017 at Phoenix's blog
That's an excellent suggestion, I had not really considered that yet! I still think that development is probably not regular in the sense that you can write a sound law to arrive at this development. The fact that you have form like ktibāt-uh (so a different strategy) to solve the same 'issue' would seem to suggest that it's an attempt to retain the integrity of the word-shape. I think this is similar as with *min-uh, which would regularly yield **mn-uh, so to retain the vowel of min > minn-uh. But there's a couple of ways you can go arguing about that.
Last night, I discussed in some depth the syllabification of the Bedouin-Type Arabic dialects. I worked mostly from Owens' description, which got me a little caught up in a bind, as I had no way to explain why the feminine ending was behaving like a *-t-, rather than the expected... Continue reading
Posted Jun 16, 2017 at Phoenix's blog
A robust group of dialects that span from the Najd, all the way to the gulf, southern Iraq and Libya are the dialects that may be called the 'Bedouin-Type' (for lack of a better term). These can be recognized most easily by: Proto-Arabic *q > g Loss of the high... Continue reading
Posted Jun 15, 2017 at Phoenix's blog
@Benjamin: I really like this parallel you mentioned about ḥirom/ḥiråm; It does seem very similar to this. The difference being that ḥyrm and ḥyrwm are both actually correct ways of spelling ḥirom. ʔibrahīm simply should not be writable as ʔbrhm. Long ī is always spelled with y. I'm sympathetic to Jeffery's argument. To this we may add ʔismāʕīl as well. So maybe ʔifʕālīl was reinterpreted as a 'vowel pattern for biblical figures'. Still doesn't explain both spellings existing though (Which Jeffery seems to be unaware of? Ignores? SPELLINGS MATTER GUYS!). And not just in isolated Surahs, some Surahs have both and one Aya even has both! Is ابرهم a Nabataeogram? Why this specific name? Why nothing else? Especially since the Nabataeans weren't primarily Jewish, it doesn't make much sense (the name seems to be unattested in Nabataean even). @Lameen: Thanks! Those are very nice parallels. I'll see if there is anything to that. One that comes to mind that could show something is ǧibrīl, but it is read as such by Hišām, and not as ǧibraʔīl that we see in several other traditions. Which honestly makes it look more alien from an Arabic phonotactic perspective. But yeah, the problem remains: why are both spellings present in the first place?
The Islamic name ابرهيم ʔibrāhīm, has always been somewhat puzzling. While this is of course the same name as our Abraham̨ from the Hebrew ʔaḇråhåm, we are left with an unanswered problem in the final syllable, which has randomly shifted the final ā to ī. Many Arabists invoke 'Imālah' to... Continue reading
Posted May 27, 2017 at Phoenix's blog
That's an excellent question. I had not really considered that option. My feeling is that that would not be the most obvious solution, but seems difficult to completely exclude as a possibility. One argument I think one could make is that, in the order that the Surahs are now (which I think is fair to assume is the order in which the Archetypal Quran was written down), you find throughout the text: al-Aykah, Laykah, Laykah, al-Aykah. Assuming the scribe wrote the manuscript in this order, one can imagine them first writing it as al-Aykah, and then making up their mind and switching to Laykah. But then switching back to al-Aykah again for its final attestation would then be a little unusual. Incidentally, I hadn't thought of this analogy yet: This mistake is essentially equivalent to what we find in some German/Dutch/English misparsings of the indefinite article like: Dutch: een adder < *een nadder Belgian Dutch: een nonkel < *een onkel One interpretation takes the definite article as the start of the next word, the other interpretation takes it as the definite article, and assume the word starts with a vowel. It's a little hard to say which of the two was original, although if the intepretation is correct that Laykah comes from Leuke Kome (, than the al-aykah interpretation is the reanalysis.
The Quran speaks in multiple places about a ʔaṣḥābu l-ʔaykah 'Companions of the Wood/Thicket', a people associated with the prophet Šuʕayb, traditionally taken to be the same person as the Biblical Jethro. The name ʔaṣḥābu l-ʔaykah is usually considered to consist of ʔaṣḥāb 'companions', + the definite article l followed... Continue reading
Posted May 24, 2017 at Phoenix's blog
Dear Etienne: For many of these words (e.g. 'bean') it's not so much Proto-Indo-European words that are entering Berber (and Basque and Egyptian etc.), but rather a language of the original agricultural settlers of Europe (and, apparently, North-Africa) that were there already before the Proto-Indo-Europeans arrived. That's certainly the Hypothesis from which Guus Kroonen's database is working. It's elegant because it helps explain why there is such a strong mismatch between what archaeologists generally thing about the settlement of Europe, and what linguists think.
I think you should think in that direction yes. I just remembered another one: *a-agăr 'field', cf. Lat. ager. It might be a loanword from Latin, but it has undergone Mid Vowel Harmony, and therefore shows up as igr in most Berber languages, quite different from the Latin word. DIEAT doesn't mention any cognates outside of europe, but I believe the Sumerian word is quite similar as well...
Yes, the similarity is probably not a coincidence. There are a couple of agricultural and animal husbandry Wanderwörter shared between Berber and Indo-European. I did a little bit of work on Guus Kroonen's wonderful Database of Indo-European Agricultural Terminology. cf. also Proto-Semitic *gady 'id.', which is presumably related. Also in metallurgy you find interesting ones: *a-ẓrəf 'silver' certainly is related to the wanderwort family that 'silver' itself belongs to. ( I should add more cognates to this one) Also: *a-ldun~a-buldun 'lead', uzzal 'iron' (cf. Hebrew barzel) There are probably more, but I haven't found them yet. === *ă is normally thought to be a central low vowel [ɐ], but if I'm correct about mid vowel harmony, it's likely that it was closer to [ə]. It ultimately corresponds to Proto-Semitic short *a (a low front or mid short vowel), so does probably come from something like [ɐ] originally.
I recently sat down to read the wonderful article by the Naima Louali & Gérard Philippson (2004) "Berber expansion into and within north-west Africa: a linguistic distribution", Afrika und Übersee, pp. 105-130. A wonderful article that discusses some of the chronological issues of the reconstruction of Proto-Berber. The internal coherence... Continue reading
Posted Feb 9, 2017 at Phoenix's blog
A recent article by Ali ibn Ibrahim Ghabban & Robert Hoyland, discussing the earliest discovered Islamic Inscription, a short footnote (#16) states that the Khashnah inscription should be dated to 52AH and not as 56AH as the edition princeps [1] They do not elaborate on why this reading (citing Fahmi)... Continue reading
Posted Oct 18, 2016 at Phoenix's blog
I don't quite know what it means to "introduce patterns of poetry" into a text. So it's hard to say whether that would correspond to point #2. Which paper did Manfred Kropp write this in? Do you have a reference? I'll have a look. #1 Good question. It's difficult to conceive of evidence for that. Some options: really early evidence of comments on that the Quran should be read as classical Arabic / a highly unusual early Quran manuscript whose transcription requires such a reading would be reasonable proofs to make that case.
It is difficult to say this more clearly. In the language in which the Quran was written down, the word ṯamūd is clearly of a different conjugational class than it is in the reading tradition. Therefore, there is a palpable disconnect between the reading tradition and the Quranic text. There are two possible solutions to this: 1. The recitation traditions represent the original prophetic dialect. The Quran was written down in a different dialect from the prophetic dialect. 2. The Quranic text represents the original prophetic dialect. The reading traditions have become more classicized over time. Considering how strong the focus becomes on Classical Arabic around the start of the 9th century, I find the second option more likely, but a scholar of Islam is sure to be able to develop a more nuanced view on this than me.
This is true of course. The QCT does not particularly give the impression it is trying to approximate any of the traditional readings though (if anything, Warsh seems closest to me). It also does not seem to be a 'compound' of different readings. Most spelling conventions are surprisingly consistent across the whole document. And whenever the spelling convention is inconsistent (as with writing construct feminines with tāʔ) those inconsistencies happen rather randomly even within a single Sura, and sometimes even in aya's basically next to each other. So there a 'multiple source' explanation seems to not work particularly well. That point, of course, speaks quite strongly for the traditional narrative of the Uthmanic Canon (and the Sanaa Palimpsest then probably represent a pre-Uthamnic written tradition?).
In my blogpost Is Thamud a triptote? I suggested that the name of the people ṯamūd was a triptote in the language of the QCT, rather than a diptote as it is in Classical Arabic. A triptotic inflection of this noun would be Nom. ṯamūdun <ṯmwd> Gen. ṯamūdin <ṯmwd> and... Continue reading
Posted Aug 27, 2016 at Phoenix's blog
Dear Al-Jallad, Your criticisms are of course valid, and I would have to do special pleading for stuff like ʕmrw /ʕamrū/ (a spelling pronunciation??; Tihama-like dialect influence??) The reason why I liked Jadhima's solution is that it can derived the QCT case system (and a quirk of its orthography) from the Nabatean case system. If we take the En Avdat Arabic def. nom -w as a "real" -u, that must mean En Avdat Arabic had one of the following two case systems: 1. Tihama Yemeni-like with loss of nunation and subsequent lengthening: Definite: *-a, *-i, *-u Indefinite: *-ā, *-ī, *-ū. 2. Loss of nunation without lengthening, definite/indefinite become identical: Def/Indef *-a, *-i, *-u A final possibility is situation 1. which afterwards loses final short vowels, and spreads the indefinite system to the now unmarked definite nouns. This would be indistinguishable from system 2. In the case of system 1. We would expect an orthographical representation: Definite/indefinite -ʔ, -y, -w (which can plausibly be deduced from the En Avdat inscription). If then, short vowels are lost, we would end up with a system: Definite -Ø; Indefinite: -ʔ, -y, -w. It is difficult to argue that final -ū, and -ī were lost, as this would obviously yield eronneous predictions about, e.g. katabū as that would yield **katab for the 3pl.m. But we would have to argue that these vowels were somehow lost, to explain why the QCT orthography only writes the indefintie accusative -ʔ, but not the indefinite nominative and genitive. There is room for special pleading, e.g. The indefinite case vowels were nasalized, and *-ã, *-ĩ, *ũ; Then, first word-final non-nasalized vowels were lost; and subsequently the word-final nasalized high vowels were lost. System 2 would fail to explain why the definite accusative isn't also marked with *-ʔ, moreover it would fail to explain why Diptotes do not have wawation. Jadhima's approach starts with the Classical Arabic system: *-a, *-i *-u, *-an, *-in, *un; 1. Gets rid of final short vowels Definite/diptotes -Ø; indefinite *-an, *-in, *-un; 2. Gets rid of nunation Definite/diptotes -Ø; indefinite *-a, *-i, *-u Orthographically: -ʔ, *-y, -w. 3. *-i and *-u perhaps merged to /ə/ yielding JSNab 17-like Arabic: Definite/diptotes -Ø; indefinite *-a, *-ə Orthographically: -ʔ, -w; 4. Word-final *-ə is lost: The now purely orthographic -w is spread to every non-construct noun, but not to indefintie accusative since it is properly pronounced with -a, and written with -ʔ. Alif al-Wiqāyah gets introduced to distinguish genuine word-final /aw, ū/, from silent /w/. 5. wawation is removed in the orthography, yielding the QCT system. Forms like ʕbdʔlhy (and ʕbdʔlhyw) would then be names from a different dialect that has not yet lost final short-vowels. This is perhaps confirmed by the fact that ʕbdʔlh also simply exists.
For 'he', I'm sure there are spellings with the final aleph, look at JSNab 18 in the same article: w dkyr ʿdmn hwʾ ktb ktbʾ dʾ bṭb w šlm And may ʕdmn who wrote this text be remembered for good and may he be secure. Cantineau also lists the form in his grammar besides hw, while he only lists the feminine form hy.
In my recent blogpost on the alif al-wiqāyah I argued that <wʔ> denotes final vowels /ū/ and /ō/ (or /aw/) in the orthography of the QCT. While final <w> denoted actual consonantal /w/. While it is fairly clear that this is the rule that is operating in the QCT orthography,... Continue reading
Posted Aug 21, 2016 at Phoenix's blog
Hey there, thanks for your interesting comments! The solution that you suggest has the enormously useful value that it explains why the وا was introduced. It's just such a shame that we don't have any evidence for such a stage (but that might just be an accident of history). I also think the use of silent و was preceded by a phase in the spoken language that must have had long final vowels as case endings at least in pause, and these were represented in writing such as in theophoric names like عبدلهى etc. When final /u/ and /i/ were deleted that phonological distinction was obliterated so و was generalized to the genitive. I don't think it's necessary to assume that the final vowels were long (in pause). Aramaic did not have final short vowels at all, and when the Aramaic spelling conventions were adapted to write Arabic, Arabic final -i may have just been felt to be closer to Aramaic -ī than writing nothing at all. This also helps explain why final -u and -i were eventually lost on the nouns, but not on, say the verb. If those endings were really pronounced -ū and -ī we would expect them to have been retained like other cases. So perhaps, the original وا vs. و distinction was to mark the difference between long ū and short u, a distinction Nabatean Aramaic never had any use for. So then we are left with four phases: 1. Classical Arabic-like: Triptotes: def. -u, -i, -a; indef. -un, -in, -an Ditptotes: def. -u, -i, -a; indef. -u, -a 2. Loss of final short vowels: Triptotes: def. no ending; indef. -un, -in, -an Diptotes: no endings. 3. Loss of nunation: Triptotes: def. no ending; indef. -u, -i, -a Diptotes: no endings. 4. Loss of final -u/-i: Triptotes: def. no ending; def. -0, -0, -a When the Nabatean script was adapted for Arabic writing, the main dialects must have been in stage 3, and therefore they chose to write the final indefinite case vowels as ـو، ـي، ـا. When they transitioned into stage 4, the vast amount of words that were written with final ـو now had a "silent waw", and to distinguish that from a real waw they introduced وا. Spellings like تيمالهي would then reflect either an archaic pronunciation of a name from a time that Arabic was still in phase 2, or alternatively, a dialects that was still in phase 2. (And spellings like عبداله which are also found in Nabatean would then be pronunciations of phase 3). Phase 2 dialects must have certainly existed for some time, as the En Avdat inscription is certainly a reflection of such a dialect. The En Avdat dialect presumably had: def. -u, -i, -a; indef. -ū, -ī, -ā (loss of nunation with lengthening?). Very good insights! Will you publish on this/have you published on this? If not: What name should I put in the acknowledgements when I undoubtedly write a paper on this at some point?
Toggle Commented Aug 17, 2016 on The Alif al-Wiqāyah in the QCT at Phoenix's blog