PhDniX's blog

New Article: Noun Prefixes in Eastern Berber

After a fairly long wait, my article on the prefixes in Eastern Berber has finally appeared in Rivista degli Studi Orientali. I am somewhat happy that it has come out on this series, as it was one of the journals that Paradisi first published his work on Awjila Berber. As I owe my title of PhD in part to his work, I am happy to be able to write something about the Libyan Berber languages in this journal.

So what is the article about? Well the title says it all really.

In Eastern Berber (which I take to mean all Berber languages from Libya except for Zwara, and now, presumably Yefreni) we find that the common Berber prefixes sg. m. a- /f.  ta- and pl. m. i- / f. ti- are very often shortened or lost completely, and often in environments where we might not expect this in other Berber languages.

Siwi seems to follow the lines of Zenatic languages (mostly). It loses the prefix vowel of the singular in front of CV, seemingly randomly. In the plural system the productive rule iCC > əCC has shortened the plural prefixes i-/ti- in predictable environments.

Foqahi, which is linguistically close to Siwi (and more distantly to the other Zenatic languages), however, has a surprisingly un-Zenatic look to its prefix system. It loses the singular prefix a- in almost every lexical item, with the exception of nouns with the shape CVC, which is exactly the environment where Siwi and other Zenatic languages would normally lose it. The i- plural prefix occasionally drops out in certain environments. The feminine plural prefix is always t(ə)-, never ti-.

Sokni, another close relative of Siwi, has, is quite similar to Siwi, losing the prefix vowel in front of CV. The plural prefix is t(ə)- and never ti-.

Nefusi, which I think is most closely related to Ghadamsi (discussed below), loses the prefix vowel in front of CV, but also in front of Cə unless the stem is CəCCVC. This distribution puzzles me, but works perfectly if you look at the data. Once again, the feminine plural prefix is always t(ə)-, never ti-.

Ghadamsi's system is more difficult to describe. In general the prefixes a-/ta- are retained (although the are shortened in some environments, it seems). The plural prefix i- is always lost, presumably due to a sound law that dropped it, as suggested by Kossmann (1999b). In the vast majority of cases the feminine plural prefix is tə-, ti- seems to be limited to certain types of plural formations.

Despite Awjili being my main language of expertise in the east, I have to admit I was unable to find any sensible conditioning to the prefix vowels in Awjili. The singular either has a vowle a or a shortened vowel. Awjili never has a masculine plural prefix, and the feminine plural prefix is always t(ə)-.

To sum up, many of the prefix shortenings found in Eastern Berber can be understood as regular phonetic conditioning. The conditioning, however, is quite unusual, and not similar to what you find in the west. I now wonder if they bear any resemblance to the Tuareg shortening of the prefix, which is also quite different from the western dialects.

I conclude the article discussing the issue of the complete absence of the feminine plural prefix ti- in all but one of the dialects discussed. I argue that you can easily show that this does not constitute a sound law *ti- > tə-, as there are examples of the sequence ti- (for example in i-vowel initial stems like Nefusi tinzərt 'nose'). If it cannot be explained in such a way, one must find another solution. I argue that it is unlikely that it would only be the feminine plural prefix that generalized the Etat d'Annexion (none of the Eastern Berber languages discussed have retained the Etat d'Annexion, a type of 'case' marking in Berber, involving a shortening of the prefix vowel).

I finally argue that, if it cannot be a generalization of the Etat d'Annexion, and it cannot be a regular sound change, that it indeed might be ancient, and that all dialects that have the plural prefix ti- simply homogenized the system: *a-, *ta- pl. *i-, *tə >> *a-, *ta- pl. *i-, *ti-. I was quite careful about that conclusion at the time, and I still am, although I think the reasoning is sound. But I might be missing some detail that a researcher one day would clear up. It would certainly make the system 'neater'. Researchers have generally tried to derive these prefixes from pronominal elements (suggesting that they were original definite (and/or indefinite) articles of sorts). There is no obvious candidate for a pronominal element tə, while there are good possibilities for ti.

The point remains, if we wish to explain the feminine plural prefix as coming from an ancient deictic element *ti, we need to account for the Eastern Berber data in some way.

This morning I received an e-mail from Adam Strich, who pointed out to me that the Songhay language Tadaksahak, which is a strongly mixed language with Tuareg in fact has nouns of Berber origin where the masculine plurals have a prefix i- and the feminine plurals have a prefix tə-. This is actually highly unexpected, because the feminine plural prefix of Tuareg is never tə- but always ti-. This situation might suggest that, indeed as a I suggested, the situation with feminine plural prefix tə- is old. I do not, however, pretend to know enough about Tadaksahak to really be able to evaluate whether the tə- could not somehow be a reflex of *ti-.

Can you see the verb 'to see'?

As we have seen in an earlier blogpost, final weak verbs written with a <y> were probably pronounced as ē in the language of the QCT, e.g. ramē <rmy> < Proto-Arabic *ramaya. This is different from the Classical Arabic language which reads this words as ramā. There is however two verbs in the Qurʔān that don't play nice, they are the reflexes of *raʔaya 'to see' and *naʔaya 'to distance oneself'. In Classical Arabic orthography these are spelled راى <rʔy> raʔā and ناى <nʔy> naʔā respectively, but in the QCT they are written without the final <y>: را <rʔ> and نا <nʔ>.

One's first impression is that such spellings may suggest a pronunciation **raʔā, **naʔā, rather than the predicted **raʔē, **naʔē. As all other final weak verbs get their etymologically correct ending <ʔ> for final-w and <y> for final-y verbs, this is not a very satisfactory explanation.

So how do we account for this discrepancy? I suggest that we assume a fairly early loss of the glottal stop, with subsequent lengthening of the vowel in the language of the QCT.

Proto-Arabic *raʔaya > Loss of ʔ *rāya.

At this point, we still are not at the form that we expect because *rāya would be spelled just like *raʔē as <rʔy>. However, there is a subsequent sound law *āy/wv > āʔv well-attested in cases like the *samāyun 'sky' > samāʔun (as in Classical Arabic), *qāwimun 'standing' > qāʔimun, and finally *ʔaḍāyatun 'watering hole' > *ʔaḍāʔatun.[1]

*rāya > *rāʔa, and this leaves us at the form that we find in the Qurʔān, which may either have been pronounced *rāʔ(a), or simply *rā (with a second loss of ʔ).[2]

A prediction this hypothesis makes, is that the Yemeni dialect of the Minabbih area, which did not undergo the *āy/wv > āʔv shift would have the reflex rāy in the 3sg.m. form, and it does (Behnstedt 1987: 202)! Below is a diachronic Paradigm of the verb 'to see' in the language of the QCT:

	Proto-Arabic	Loss of ʔ	āyv > āʔv	QCT
3sg.m.	*raʔaya	*rāya	rāʔ(a)	<rʔ>
3sg.f.	*raʔat	*rāt	rāt	<rʔt>
2sg.m./2sg.f./1sg.	*raʔayta/i/u	*rāyta/iu	rāyt(a/iu)	<rʔyt>
3pl.m.	*raʔaw	*rāw	rāw	<rʔwʔ>
3pl.f.	*raʔayna	*rāyna	rāyna	Unattested
2pl.m.	*raʔaytum	*rāytum	rāytum	<rʔytm(wh)>
2pl.f.	*raʔaytunna	*rāytunna	rāytunn(a)	Unattested
1pl.	*raʔaynā	*rāynā	rāynā	Unattested

As far as I'm concerned, this conclusively shows that in the language of the QCT *ʔ was lost completely (pace Myself)

The raʔē form remembered?

In Surat an-Naǧm we actually find the Classical spelling راى <rʔy> for this verb twice (and those are the only attestations in this Surah).  This Surah has a clear rhyming scheme, rhyming with an ē <y> from Ayah 1 to 56 (of the 62 Ayah's). Q54:11 uses <rʔy> in the rhyme. [3] This might suggest that the composer of this Ayah invoked a dialectal form, to suit the rhyme. This is, of course, not unlikely, but it is surprising that the second attestation of this verb is not invoked to suit the rhyme, as it appears in the middle of an Ayah (Q54:18). Is this Surah composed in a different dialect altogether? Or is this simply an attempt by scribes to unify the spelling within a single Surah?

Classical Arabic getting it wrong

Another unusual feature of the verb raʔā in the imperfective is that it's the only verb that applies the *Cʔ > C sound law that I have posited even in the reading tradition and Classical Arabic. As such the reflex of *yarʔayu is yarā (or yarē) <yry> in the reading tradition, and not the predicted **yarʔā, **yarʔē. This goes to show just how artificial the reading tradition is, in that it is unable to reconstruct the correct pronunciation in an ambiguous case like this verb. There is one other verb which also underwent this sound law and that is the imperative of saʔala 'to ask', which is read as sal rather than the predicted **isʔal.

[Edit] Lameen Souag pointed out in the comments that this last paragraph was a little confusing. What I meant to say is: In Classical Arabic (and in the Qurʔānic reading tradition) the imperfective of the verb 'to see' undergoes the *Cʔ > C. Within the linguistic system of Classical Arabic, this is irregular, as all other cases of *Cʔ (besides sal) are simply retained. This seems to suggest that somehow the colloquial usage (and/or dialectal usage) of this verb ended up in the Classical Language, whereas this has not happened with less high-frequency words.

---

[1] Interestingly, this sound law appears to not always operate word-internally, while it did operate word-finally, e.g. Moroccan Arabic ḍaya 'lake, pond' < *ʔaḍāyah but sma 'sky' < *samāʔ

[2] This two-step loss of first etymological *ʔ and secondary ʔ < y/w may explain the asymmetry between *aʔu and *āʔu where one yielded <wʔ> spelling from the earliest documents while *āʔu did not. If we assume that the word-final short vowels were around with the first loss, but not with the second loss, we would get the exact forms we find:

*malaʔu > malawu > malawu > malaw > mala
*ǧazāyu > ǧazāyu > ǧazāʔu > ǧazāʔ > ǧazā

[3] Note that the rhyme is broken in Q53:28 by šayʔan. Obviously, in its caseless pronunciation, this word would fit the rhyme. Perhaps it was originally composed without the case ending, and therefore with šay. In the regular reading tradition where Pausal -an is pronounced ā this rhyme works out, but the Warsh reading actually does not break the verse at this point (and therefore the numbering is different), because the rhyme does not work. Still considering the length and syntactic unit, it seems reasonable to suppose the Ayah probably ended here originally.

 

What is the deal with word-final ʔ nouns?

One of the great questions about the language of the QCT, is whether it was originally pronounced with case endings. I believe it's not a point of contention whether the QCT had case, it certainly did for example the sound plural ending -ūn/-īn is clearly distinguished for case and so are words like abū/abī/abā 'father'. It is less clear, however, whether the case system was similar to what we find in Classical Arabic.

Let us look at the paradigm of the regular masculine noun:

	Definite	Construct	Indefinite
Nom.	al-baytu البيت <ʔlbyt>	baytu بيت <byt>	baytu-n بيت <byt>
Gen.	al-bayti البيت <ʔlbyt>	bayti بيت <byt>	bayti-n بيت <byt>
Acc.	al-bayta البيت <ʔlbyt>	bayta بيت <byt>	bayta-n بيتا <bytʔ>

 As already discussed in the my blogpost on nunation, I think it's unlikely that the indefinite forms had final n, first of all, because this n is never written and because there is good internal evidence that word-final -un was lost regularly.

The question now is, where the word-final short vowels there? They are never written, but this is hardly surprising in a language where short vowels are never written.

However, there is some evidence that word-final short vowels were written out in Arabic names in the Nabataeo-Aramaic orthography.  As Arabic orthography ultimately evolved from Nabataeo-Aramaic, we may expect the language to take over this practice too.

The evidence for it is found in transcriptions of the name ʕabdu llāhi 'servant of god' in Nabataeo-Aramaic, where it is generally written <ʕbdlhy>, using final <y> for short final -i. The fact thet the QCT orthography does not transcribe this, might suggests that for the language that lies at the origin of the QCT orthography, that vowel simply was not there. This is hardly complete evidence, and attestations of the name as <ʕbdlh>, which exist too, make this evidence a lot less clear. We must therefore conclude that the final short vowels might simply be invisible in the orthography of such nouns, but were, nevertheless around.

However, one of the cases is always clearly orthographically distinguished, and that's the indefinite accusative, which, while pronounced as -an is consistently written with a final <ʔ>, suggesting a final -ā pronunciation. In fact, this is exactly the pronunciation that we find in the 'pausal pronunciation' of the indefinite accusative.

I have already discussed some of the problems with the concept of 'pausal pronunciation', which appears to be a category only relevant to the Qurʔānic text, in Pre-Islamic poetry this does not seem to exist (while traditionally they are supposed to be the same language form). To that we may add that, with nunation gone, it is hard to imagine how the indefinite accusative would consistently be written with <ʔ> with it being the same in pronunciation as the construct or the definite form. It seems therefore likely that the language of the QCT pronounce the indefnite accusative as -ā, as found in its pausal pronunciation in Classical Arabic.

Final glottal stop nouns in Classical Orthography [Edit: I was super confused last night, changed this paragraph to actually be right]

So how can we say something more about word-final short vowels? Nouns that end in an etymological glottal stop may provide a clue here. In the Classical Orthography, nouns (often infinitives) with the shape CaCāʔ have a somewhat unusual behaviour.

e.g. the noun ǧazāʔ-u/i/a(n) 'retribution' is always written جزاء <ǧzʔʾ> (I use the sign ʾ here to refer to a 'hamza' sign, added to the QCT later, to denote positions of the original glottal stop, where it was seemingly gone in the language of the QCT). As we have seen earlier sequences like āʔi and āʔu are normally written with a <y> and <w> as 'carriers' of the glottal stop in Classical Orthography when they occur word-internally. I would interpret that, not as a 'carrier' of the glottal stop, but rather the writing of the resulting glide that appears when the intervocalic glottal stop is lost: āʔi > āyi and āʔu > āwu.

The fact that such a word is not written with <y> and <w> suggests that there was no glide, and presumably no pronunciation of the word-final vowels in the language on which the Classical orthography was based. Interestingly, when these nouns are followed by possessive clitics the <y> and <w> reappear. If we follow the idea that the short vowels were lost word-finally, this is of course unsurprising.

Language of the Classical Orthography:

*ǧazāʔun, *al-ǧazāʔu > (al-)ǧazā(ʔ) <ǧzʔ>
*ǧazāʔin, *al-ǧazāʔi > (al)-ǧazā(ʔ) <ǧzʔ>

But:

*ǧazāʔu-hā > ǧazāwu-hā <ǧzʔwhʔ>
*ǧazāʔi-hā > ǧazāyi-hā <ǧzʔyhʔ>

Nouns that end in a aʔ rather than āʔ display the same behaviour in Classical Orthography.

malaʔ- 'chiefs' is written as ملا <mlʔ> when not followed by a clitic, and the <ʔ>  is replaced with <y> when followed by a genitive i and a possessive clitic and with <w> when followed by nominative u and a possessive clitic. So these words too, seem to point to the loss of case vowels in absolute final position, while they are retained when a clitic directly followed it.

*ǧazāʔu > ǧazāʔ > ǧazā
*ǧazāʔu-hā > *ǧazāʔu-hā > ǧazāwuhā

*malaʔu > malaʔ > malā
*malaʔu-hā > malawu-hā >

Final glottal stop nouns in QCT

In the the Qurʔānic text nouns with final āʔ very often, but not consistently write āʔu and āʔi with the glides <w> and <y> that we would expect, not only when followed by a possessive clitic, but also simply when in construct. There are many examples of this, whenever I could, I checked whether older Qurʔān manuscripts have the same spelling.

The sequence āʔu

In the Cairo edition, āʔu is spelled <wʔ> 19 times while not followed by a clitic. This seems to point to an isolated pronounciation /āwu/. If we look at earlier documents however, it seems that the spelling is an innovation. While we do not find any document that has no cases of āʔu spelled as <wʔ>, we find significantly fewer cases, it is normally written with just <ʔ>, as is the case in the Classical Orthography. A few times we find it spelled <ʔw> where the <w> appears to be a later addition.

This seems to be a spelling convention that was in development at the time. For reasons unknown, this spelling convention has not been consistently applied throughout the whole Qurʔānic text, and it has never gained traction in the Classical Orthography, which sticks to the seemingly more archaic form.

Q5:18 ʔabnāʔu (a)llāhi <ʔbnwʔ ʔllh> 'The children of Allah'

    Spelled identically in the Birmingham Fragment (568-645AD)

Q5:29 ǧazāʔu (a)ẓ-ẓālimīna <ǧzwʔ ʔlẓlmyn> 'retribution of the wrong-doers'

    Spelled <ǧzʔw ʔlẓlmyn> in We II 1913, (dated 662-765AD) the <w> perhaps looks like a later addition.

Q5:33 ǧazāʔu (a)llaḏīna yuḥāribūna (a)llāha <ǧzwʔ ʔlḏyn yḥʔribwn ʔllh> 'retribution for those who wage war against Allah'

    Spelled <ǧzʔw ʔlḏyn> in We II 1913, the <w> once again seems to look like a later addition.

Q6:5  ʔanbāʔu mā kānū bihī yastahziʔūna <ʔnbwʔ mʔ kʔnwʔ bh ysthzwn> 'the news of that which they used to ridicule'

    Spelled <ʔnbʔ> in We II 1913.

Q6:94 ʔanna-hum fī-kum šurakāʔu <ʔnhm fykm šrkwʔ> 'that they were among your associates'

    Spelled as such in We II 1913, and Birmingham Manuscript but as <šrkʔ> in the much later Samarkand codex.

Q11:87 mā našāʔu <mʔ nšwʔ> 'what we please'

    Spelled as such in We II 1913.

Q14:21 fa-qāla (a)ḍ-ḍuʕafāʔu <fqʔl ʔl-ḍʕfwʔ> 'And the weak will say''

    Spelled as such already in We II 1913.

Q26:6 ʔanbāʔu mā kānū bihī yastahziʔūna <ʔnbwʔ mʔ kʔnwʔ bh ysthzwn> 'the news of that which they used to ridicule'

    Spelled identically in the Tübingen Manuscript. (dated 649-675)

Q26:197 ʕulamāʔu banī ʔisrāʔīla <ʕlmwʔ bny ʔsryl> 'the scholars of the Children of Israel'

    Spelled identically in the Tübingen Manuscript.

Q30:13 min šurakāʔi-him šufaʕāʔu <mn šrkʔyhm šfʕwʔ> 'from among their partners, any intercessors'

    Tübingen manuscript has <mn šrkʔyhm šfʕʔ>

Q35:28 min ʕibādi-hi al-ʕulamāʔu <mn ʕbʔdh ʔlʕlmwʔ> 'from among His servants, who have knowledge.

    Spelled identically in We II 1913.

    Tübingen manuscript has <mn ʕbʔdh ʔlʕlmʔ>

Q37:106 al-balāʔu (a)lmubīnu <ʔlblwʔ ʔlmbyn> 'the clear trial'

    We II 1913 has <ʔlblʔ ʔlmbyn>.

Q40:47 fa-yaqūlu (a)ḍ-ḍuʕafāʔu <fyqwl ʔl-ḍʕfwʔ> 'and the weak will say'

    Spelled as such in We II 1913.

Q40:50 duʕāʔu (a)lkāfirīna <dʕwʔ ʔlkfryn> 'the supplication of the disbelievers'

    Spelled <dʕʔ ʔlkfryn> in We II 1913.

Q42:21 ʔam lahum šurakāʔu šaraʕū ... <ʔm lhm šrkwʔ šrʕwʔ ...> 'Or have partners who have ordained ...'

    Spelled as <šrkʔ> in We II 1913.

Q42:40 wa-ǧazāʔu sayyiʔatin <wǧzwʔ syyh> 'And the retribution for an evil act'

    Spelled <wǧzʔw syyh> in We II 1913, maybe the <w> is a latter addition, not as obvious as with the other cases.

Q44:33 balāʔun mubīnun <blwʔ mbyn> 'a clear trial'

    Ms. Paris BnF Arabe 328 (b) and We II 1913 have <blʔ mbyn>

Q59:17 ǧazāʔu (a)ẓ-ẓālimīna <ǧzwʔ ʔlẓlmyn> 'retribution of the wrong-doers'

    Spelled <ǧzʔ ʔlẓlmyn> in We II 1913.

Q60:4 ʔinnā buraʔāʔu min-kum <ʔnʔ brwʔ mnkm> 'Indeed, we are disassociated from you'

    Spelled as such in We II 1913.

As you can see 8 out of the 19 examples have their unusual spelling in early manuscripts already. The amount of unusual spellings of this kind, thus seem to have been increasing over time. This is a highly unusual. Whenever the Cairo document deviates from older documents, it generally deviates towards a spelling that is closer to Classical Arabic, not further away. Nevertheless, here this seems to be undoubtedly the case.

Besides these, there are countless examples of āʔu that are written as <ʔ> both in the Cairo edition and in old manuscripts. Some examples:

Q17:82 mā huwa šifāʔun wa-raḥmatun li-l-muʔminīna <mʔ hw šfʔ wrḥmḧ llmwmnyn>

    Idem in the Tübingen Manuscript.

Q17:110 fa-la-hu ʔal-ʔasmāʔu alḥusnā <flh ʔlʔsmʔ ʔlḥsny> 'to Him belong the best names'

    Idem in the Tübingen Manuscript (with line break <ʔlʔ//smʔ>).

Q20:8 la-hu ʔal-ʔasmāʔu alḥusnā <lh ʔlʔsmʔ ʔlḥsny> 'to Him belong the best names'

    Idem in the Tübingen Manuscript.

    Idem in the Birmingham manuscript.

Q21:9 wa-man našāʔu <wmn nšʔ> '... and whom we willed'

    Idem in the Tübingen Manuscript.

Q22:5 mā našāʔu <mʔ nšʔ> 'what we willed'

    Idem in the Tübingen Manuscript.

Q22:18 mā yašāʔu <mʔ yšʔ> 'what he willed'

    Idem in the Tübingen Manuscript.

Q24:6 la-hum šuhadāʔu ʔillā ʔanfusu-hum <lhm šhdʔ ʔlʔ ʔnfshm> 'for them (there will be) no witnesses except themselves.

    Idem in the Tübingen Manuscript.

Q24:21 man yašāʔu <mn yšʔ> 'whom He wills'

    Idem in Tübingen Manuscript.

Q24:35 man yašāʔu <mn yšʔ> 'whom He wills'

    Idem in Tübingen Manuscript.

Q24:38 man yašāʔu <mn yšʔ> 'whom He wills'

    Idem in Tübingen Manuscript.

Q24:43 man yašāʔu <mn yšʔ> x2 also Q24:45, and Q24:46 'whom He wills'

    Idem in Tübingen Manuscript.

Q25:25 yašaqqaqu (a)s-samāʔu <yšqq ʔlsmʔ> 'the heaven will split open'

    Idem in Tübingen Manuscript.

Q26:33 hiya bayḍāʔu li-n-nāẓirīna <hy byḍʔ llnẓryn> 'It was white for the observers'

    The edition reads an added <ʔ> <hy byḍʔ ʔ llnẓryn> in the Tübingen Edition. I see a line there but not sure if it is intentional, but it wouldn't make much grammatical sense, to have it there, unless the first <ʔ> is meant to represent the long ā, and the second <ʔ> is meant to represent the ʔ.

Q26:136 sawāʔun ʕalaynā <swʔ ʕlynʔ> 'it is the same to us'

    Idem in Tübingen Manuscript.

Q26:224 wa-(a)š-šuʕarāʔu <wʔlšʕrʔ> 'and the poets'

    Idem in Tübingen Manuscript.

So far I have found no examples of the spelling <wʔ> where the Cairo edition has <ʔ>

Conclusion on āʔu

Although there are quite a few examples of āʔu not followed by a clitic spelled as <wʔ> in the Cairo edition, early manuscripts seem to have way fewer. This is surprising, as one would expect that the non-classical, and more complex spelling, would be the original.

If it had been the original, it would have been fairly strong evidence that after nouns ending in āʔ indeed has a case vowel u following it in the language of the QCT. All spellings that agreed with the Classical Norm would then be later corrections. The situation is, however, flipped. This seems to suggest that the text was being corrected to fascilitate some kind of reading that did have case vowels where the original language of the QCT did not.

 

The sequence āʔi

The sequence āʔi is spelled <ʔy> 6 times in the Cairo edition of the Qurʔān, and every time it is found in construct. Again, in old manuscripts, we often find just <y>.

Q10:15 tilqāʔi nafs-ī <tlqʔy nfsy> 'my own accord'

    <tlqy nfsy> in We II 1913.

    Perhaps this should not be read as a construct, but as a double possessive: tilqāʔ-ī, nafs-ī 'my accord, my own' (thanks to Ahmad Al-Jallad for this suggestion).

Q16:90 wa-ʔītāʔi ḏī (a)lqurbā <wʔytʔy ḏy ʔlqrby> 'and giving to relatives'

    <wʔyty ḏy ʔlqrby> in the Samarkand codex. No early dated documents have this verse.

Q20:130 wa-min ʔānāʔi (a)l-layli <wmn ʔnʔy ʔlyl> 'and during the periods of the night'

    Spelled as <wmn ʔnʔ ʔlyl> in the Tübingen manuscript with a later added <y> squeezed between the two <ʔ>'s.

    Spelled as <wmn ʔnʔy ʔlyl> in We II 1913. May have had an added <y> but definitely not certain.

Q30:8 bi-liqāʔi rabbi-him <blqʔy rbhm> 'in the meeting with their Lord'

    Written <blqʔ rbhm> in the Tübingen manuscript.

Q30:16 bi-ʔāyātnā wa-liqāʔi (a)l-ʔāxirati <bʔytnʔ wlqʔy ʔlʔxrh> 'Our versus and the meeting of the Hereafter'

    Written <bʔyytnʔ wlqʔ ʔlʔxrh> in the Tübingen manuscript.

Q42:51 min warāʔi ḥiǧābin <mn wrʔy ḥǧʔb> 'from behind a veil/partition'

    <mn wrʔ ḥǧʔb> in We II 1913.

Of the 5 that are attested in early documents, 4 of them do not seem to write <ʔy> but simply <ʔ> (with later additions in two cases). Q10:15 can plausibly be interpreted as not representing a genitive with yāʔ, but rather the 1sg. possessive clitic -ī.

Conclusion on āʔi

It's fairly clear that the spelling <ʔy> is innovative, where <ʔ> is original. This again points to ab absence of a case vowel in the language of the QCT orthography, which was corrected to represent the proper pronunciation with a glide /y/, leading to the case vowel.

The sequence aʔu

While the long vowels seem to originally not write the glides (unless followed by a possessive suffix), as in Classical Arabic orthography, the situation of the aʔ + case vowel deviates strongly from the classical orthography. There are 21 cases of aʔu being written as <wʔ>, while there are only 16 cases that follow the classical orthography <ʔ>.

Imperfects

I'll divide the cases into two groups, as it concerns many more verbs than the group above. Imperfects that end in a aʔu in the imperfect, in Classical Arabic are simply written with a final <ʔ> e.g. yabdaʔu <ybdʔ>. In the Cairo edition, the most common spelling is <ybdwʔ>. This seems to point to a prounciation /yabdawu/. The spelling with just a single <ʔ> is only attested once, but its attestation is in the oldest known Qurʔān fragment, which is a bit of a problem. Orthographically, these verbs are indistinguishable from Final weak a/u verbs like nadā <ndʔ> impf. yandū <yndw> 'to call'. However, although these merged orthographically, there's not necessarily an obvious reason why they merged linguistically.

 

Imperfect verbs with this spelling are:

Q10:4 ʔinnahū yabdaʔu (a)lxalqa <ʔnh ybdwʔ ʔlxlq> 'indeed he begins creation'

    Idem in We II 1913.

Q10:34 man yabdaʔu (a)lxalqa .... allāhu yabdaʔu alxalqa <ybdwʔ ʔlxlq .... ʔlh ybdwʔ ʔlxlq> 'who begins creation? ... Allah begins creation'

    Idem in We II 1913.

Q12:85: ta-(a)llāhi taftaʔu taḏkuru yūsufu <tʔllh tftwʔ tḏkr ywsf> 'By allah you will not cease remembering Joseph'

    Idem in We II 1913.

Q16:48 yatafayyaʔu ẓilālu-hū ʕani (a)l-yamīni wa-(a)š-šamāʔili <ytfywʔ ẓllh ʕn ʔlymyn wʔlšmʔyl> 'Their shadows incline to the right and to the left'

    Idem in We II 1913. and Wetzstein II 1919 (Carbon dated 670-769)

Q20:18 ʔatawakkaʔu ʕalay-hā <ʔtwkwʔ ʕlyhʔ> 'I lean upon it'

    Idem in We II 1913 and Tübingen Manuscript.

    Spelled <ʔtwkʔ> in the extremely old Birmingham manuscript.

Q20:119 wa-ʔannaka lā taẓmaʔu fī-hā <wʔnk lʔ tẓmwʔ fyhʔ> 'And indeed, you will not be thirsty therein'

    Idem in We II 1913 and Tübingen Manuscript.

Q24:8 wa-yadraʔu ʕanhā (a)lʕaḏāba <wydrwʔ ʕnhʔ ʔlʕḏʔb> 'But it will prevent punishment from her'

Idem in Tübingen manuscript and We II 1913.

Q25:77 mā yaʕbaʔu bikum rabbī <mʔ yʕbwʔ bkm rby> 'What would my Lord care for you ...?'

    Idem in We II 1913 and Tübingen Manuscript.

Q27:64 ʔamman yabdaʔu (a)l-xalqa <ʔmn ybdwʔ ʔlxlq> 'Is He who begins creation ....'

    Idem in Tübingen Manuscript and We II 1913 

Q30:11 allāhu yabdaʔu (a)lxalqa <ʔllh ybdwʔ ʔlxlq> 'Allah begins creation'

    Idem in Tübingen Manuscript.

Q30:27 wa-huwa (a)llaḏī yabdaʔu (a)lxalqa <whw ʔlḏy ybdwʔ ʔlxlq> 'And it is He who begins creation'

    Idem in Tübingen Manuscript.

Q43:18 ʔa-wa-man yunaššaʔu fī (a)lḥilyati <ʔwmn ynšwʔ fy ʔlḥlyh> 'So is one brought up in ornaments?'

        Idem in We II 1913.

Q75:13 yunabbaʔu (a)l-ʔinsānu <ynbwʔ ʔlʔnsn> 'man will be informed'

     No attestations in the oldest manuscripts.

The Classical spelling is attested 4 times in the Cairo edition:

Q4:140 yukfaru bi-hā wa-yustahzaʔu <ykfr bhʔ wysthzʔ> 'they are denied and ridiculed'

    Idem in We II 1913 and Birmingham fragment and ms.or.fol. 4313 (Carbondated to 606-652AD).

Q12:56 yatanawwaʔu minhā <ytnwʔ mnhʔ> 'to settle therein'

    Idem in We II 1913.

    The expected spelling would have required the spelling <ytnwwʔ>. repetition of <ww> and yy seems to be avoided usually. This might be the reason for this irregularity.

Q39:74 natabawwaʔu mina (a)lǧinnati <ntbwʔ mn ʔlǧnh> 'We may settle in Paradise'

    Idem in We II 1913.

    Same as above, this might be an avoidance of writing <ww>

Nouns

There are not as man attestations of this in nouns, simply because the stem CaCaʔ appears to be quite rare.

Q14:9 ʔa-lam taʔti-kum nabaʔu ʔlḏyn mn qblkm <ʔlm tʔtkm nbwʔ (a)llaḏīna min qablikum> 'Has there not reached you the news of those before you?'

    Idem in We II 1913 and Wetzstein II 1919.

Q23:24 fa-qāla (a)l-malaʔu <fqʔl ʔlmlwʔ> 'the chiefs said'

    Idem in Tübingen manuscript.

Q27:29 qālat yā-ʔayyuhā (a)l-malaʔu <qʔlt yʔyhʔ ʔlmlwʔ> 'She said 'O chiefs''

    Idem in Tübingen manuscript.

Q27:32 qālat yā-ʔayyuhā (a)l-malaʔu <qʔlt yʔyhʔ ʔlmlwʔ> 'She said 'O chiefs''

    The transcription of the Tübingen manuscript is marked as having corrected this spelling. The hand does look different for the <wʔ> the original sign has been erased. [Edit: Strangely the transcription is no longer displayed...]

    Idem in We II 1913.

Q27:38 qāla yā-ʔayyuhā (a)l-malaʔu <qʔl yʔyhʔ ʔlmlwʔ> 'He said 'O chiefs''

     Idem in Tübingen manuscript and We II 1913.

Q38:21 nabaʔu (a)lxaṣmi <nbwʔ ʔlxṣm> 'the news of the adversaries'

    Idem in We II 1913

  Q38:67 nabaʔun ʕaẓīmun <nbwʔ ʕẓym> 'great news'

     Idem in We II 1913.

Q64:5 nabaʔu (a)llaḏīni kafarū min qablu <nbwʔ ʔlḏyn kfrwʔ mn qbl> 'the news of those who disbelieved before'

    Idem in We II 1913.

There are a however also quite a few cases of nouns of this type that are written in the classical orthography in the Cairo Edition:

Q9:70 ʔa-lam yaʔti-him nabaʔu (a)llaḏīna min qabli-him <ʔlm yʔthm nbʔ ʔlḏyn mn qblhm> 'Has there not reached you the news of those before you?' (cf. Q14:9).

    Idem in We II 1913.

All other attestations are the word al-malaʔu

Q7:60 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:66 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:75 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:88 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:90 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:109 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q7:127 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q11:27 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q12:43 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q23:33 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

Q28:38 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in Tübingen manuscript.

Q38:6 al-malaʔu <ʔlmlʔ> 'the chiefs'

    Idem in We II 1913.

We are left with the highly unusual situation that there are two orthographical practices being used in the Qurʔān. These two orthographical practices have been stable over hundreds of years, as they align perfectly between the older Quran document and the Cairo edition. The two spellings do not overlap. A Surah either has <mlʔ>, <nbʔ> or <mlwʔ>, <nbwʔ>. It is very difficult to decide what exactly this means (save for the fact that the copiests were clearly copying from a written copy of the Qurʔān, and not a transcription of a memorized recited form). The two different spellings occur in almost identical environments, and even sometimes in identical environments, e.g. Q14:9~Q9:70.

Is <mlʔ> an etymological spelling while <mlwʔ> represent a phonetic spelling?

The sequence aʔi

This sequence is written as <ʔy> once. This is a somewhat unusual spelling, and is reminiscent of the <ʔy> spelling sequence in miʔah <mʔyh>. The other two times it occurs, it has the Classical spelling.

 Q6:34 min nabaʔi (a)lmursalīna <mn nbʔy ʔlmrslyn> 'from the news of the messengers'

    Idem in Ms. paris BnF Arabe 328

    Idem in We II 1913

Q28:3 min nabaʔi mūsā <mn nbʔ mwsy> 'from the news of Moses'

    Idem in Tübingen Manuscript.

Q78:2 ʕan (a)n-nabaʔi (a)lʕaẓīmi <ʕn ʔlnbʔ ʔlʕẓym> 'About the great news'

    Idem in Ms. Sanaa 15-27-1 (dated 622-822)

The word malaʔ occurs several times in the genitive followed by a possessive clitcs. In all of these case the word is spelled <mlʔy>, this is different from the Classical Arabic spelling which requires <mly>. This seems to be related to the unusual <ʔy> spelling in words like miʔah, etc.

Q28:32 malaʔihi <mlʔyh>  idem Q23:46 and Q11:97 idem Q10:75 idem Q7:103

    Spelled the same in Tübingen manuscript for 23:46 (also We II 1913) and 28:32

       Q11:97 is spelled as such in Gotthelf-Bersträßer-Archiv: Kairo, Dār al-kutub: qāf 47 (dated 606:652).

    Q7:103 is spelled as such in We II 1913.

Q10:83 malaʔihim <mlʔyhm>

    Idem in We II 1913.

 Conclusions on the aʔu/aʔi sequence

I don't know what to make of this spelling. The nominative spellings with <wʔ> remind one of the spelling of the indefinite/construct form of ʔimruʔun 'man'. This noun, according to classical grammar, supposedly has an internal vowel that harmonizes with the case vowel:

ʔimruʔu(n) <ʔmrw> (<ʔmrwʔ> in the Qurʔān)
ʔimriʔi(n) <ʔmry>
ʔimraʔa <ʔmrʔ>

A strange part of this noun's behaviour in Classical Arabic (and QCT), is that in the definite ofrm this has a full vowel a between the first and second root consonant: al-marʔu/i/a. This behaviour is also found in the feminine form imraʔatu/i/a(n) but al-marʔatu/i/a. I don't think this alternation is ever attested in the Arabic dialects, and Old Arabic doesn't show evidence of it either. The Namara inscription mentions an imruʔu (a)lqays but spells it <mrʔlqys> mar al-qays.

ʔimraʔu would presumably have the same spelling <ʔmrwʔ>, which makes one wonder whether this noun belongs to the aʔu nouns under discussion here. A point that speaks against this is the spelling of the genitive. Every time that we find the case vowel spelled out in the QCT, it always has <ʔy>, while imriʔin is spelled <ʔmry> (also) not <ʔmrʔy>.

General conclusions

There is very little evidence that there were glides conditioned by the following case vowel after āʔ in word-final position. Presumably the language of the QCT already lost these word-final vowels. There are, even in the earliest documents, already some examples of this spelling, however. When directly followed by a possessive clitics, this case vowel was retained.

After the sequence aʔ the situation is ambiguous. There is ample evidence both in the verbal system and in the noun that aʔu yielded awu and was thus spelled <wʔ>. This means that the final vowel must still have been there when *ʔ was lost. But what do we make of the many spellings with <ʔ>? Are those archaic spellings?

Why don't the spellings overlap in a single Surah? This seems to suggest that different Surah's came from different scribal traditions and were only compiled into a single Qurʔān later? Could those different scribal traditions also have had different recitation traditions? Some with vowels in Pause, while others did not?

Only questions. No answers.

Of kāfir and kafirūn

In an earlier post, I pointed out that the writing of ā word internally in the QCT is very inconsistent, and that sometimes it is written and sometimes it is not. I argued that because words that have internal glottal stop are always written with <ʔ>, unlike many other cases of ā, this might suggest that the internal glottal stop was not lost in these words.

I have been looking into the distribution of written long ā and unwritten long ā in the QCT, and it turns out that there are a lot more patterns than I initially thought, and this puts my conclusions about the internal glottal stop on quite uncertain grounds.

Active participles, which have the shape CāCiC are basically never written with an <ʔ> when they are in the plural with the suffix -ūna/-īna:

kāfir-ū/īna <ʔlkfrw/yn> 'unbelievers', there is a single exception:

It is a little bit messier in the indefinite form, but only for the accusative singular, it is otherwise identical:

kāfir-ū/īna <kfrw/yn>, but kāfir-u/in <kʔfr>: dāḫirīna <dʔḫryn> 'humiliated ones' (Q40:60), it's three other attestations are written <dḫry/wn> as expected.

The distribution for the construct plural suffixed -ū/-ī, For the construct form CāCiC-ū/ī,  Writing with <ʔ> occurs 11 times, while the writing without the <ʔ> occurs 14 times, which is about as close to a 50/50 distribution as you would expect in a corpus.

CāCiC nouns with the feminine plural ending -ātu/i(n) show the same distribution. Words like that occurs 99 time spelled as ṣāliḥātu/i(n) <ṣlḥt> 'good deeds', and once as bāsiqātin <bʔsqt> 'tall'.

For the singular form, which has case ending -a/i/u or with nunation -an/in/un, the picture is similar to that of the construct form (but different to the non-construct plurals), both spellings occur: CāCiC with ʔ occurs 203 times, CāCiC without ʔ occurs 186 times.

Only the indefinite accusative plural seems to have a somewhat significantly skewed distribution, where the spelling without ̣ʔ occurs 59 times, while the spelling with ʔ occurs 33 times. Still some general patterns arise, e.g. wāḥid is never written with <ʔ>, nor is wāsiʕ neither of those two words are actually active participles, but a numeral and an adjective with the same pattern, this might be relevant somehow).

So how do we account for this difference in distribution? Why is 'anything goes' in the singular, while the plural has a very strong distribution? The most obvious solution seems to be taking it as a sign of stress, following the classical Arabic stress rules, the long ā would always be unstressed in the plural (accent the first heavy syllable from the right, excluding the first syllable it encounters, if there are no heavy syllables, accent the leftmost syllable):

kāfirū́na, kāfirī́na

kā́firū, kā́firī

kāfirā́tu/i(n)

kā́fira/i/u(n)

The spelling rule thus seems to be that unaccented ā can never be written with <ʔ>, while for an accented ā, it is essentially optional whether it gets written, although there are especially several high frequency words that never get written with <ʔ> (wāḥid 'one', hādihi 'this (f.), ḏālika 'that', wāsiʕ 'wide').

One cannot help but wonder whether this doesn't represent somekind of sound law, perhaps long ā was shortened when unaccented in the language of the QCT, yielding *kafirū́n(a), *kafirī́n(a) and *kafirā́t(u/i(n)).

A similar distribution is found for nouns with the feminine plural suffix -ātu/i(n), which is almost always written with a simple <t>, and not with <ʔt> as is regular in the Classical Orthography, whenever a long ā precedes it in the syllable directly in front of it. In such cases the long ā is never spelled with <ʔ>.

The most common example of this is the noun samāwā́tu/in 'heavens, sky', which is consistently spelled <smwt> 189 times in the Qurʔān (and once as <smwʔt> with the Classical feminine plural suffix spelling, in Q41:12). The 23 other occurrences of this suffix don't spell the preceding long vowel ā either:

ʔamānā́tu/i(n) 'trusts'
ḫālātu/i(n) <ḫlt> 'maternal aunts'
risālātu/i(n) <rslt> 'letters'
maġārā́tu/i(n) <mġrt> 'caves'
ʕalāmā́tu/i(n) <ʕlmt> 'landmarks'
šahādā́tu/i(n) <šhdt> 'testimonies'
ṣāfā́tu/i(n) <ṣft> 'wings spread'
ṣāffā́tu/i(n) <ṣft> 'those lined up'
an-naffāṯā́ti <ʔlnfṯt> 'the blowers'

This is of course very similar to a feature we discussed earlier of Qurʔānic orthography in older documents where words like ʕaḏāb 'punishment' and šāy 'thing', which in the indefinite accusative were spelled as <ʕbdʔ> and <šyʔ>, presumably *ʕaḏabā́ and *šayā́.

The feminine plural suffix spelled simply as <t> (793x) vastly outnumbers the spelling as <ʔt> (a couple of dozen times) in the Qurʔān, and this of course raises the question whether this is not the result of a reduction somehow. The word ḫālāt <xlt> 'maternal aunts' at least precludes the possibility that it was unstressed, as there simply is no syllable left to stress if the ā in the first syllable was also unstressed (unless we place the stress on the case vowel, which would be highly unexpected, considering its ubiquitous loss in all Arabic dialects).

Nouns that write -āt as <ʔt> are a limited set of nouns:

banāt 'daughters'.

One may consider that within Arabic this word was not considered to have a plural suffix -āt, as it is the plural of bint, with an Old Semitic feminine suffix -t no longer productive in Arabic. Therefore this may have been considered a type of broken plural instead, and therefore avoided the orthographic standard of writing the plural suffix as <t>, alternatively the fact that banāt and bint would become homographs <bnt> sticking to that spelling may have been the motivation.

as-sayyiʔāt  <ʔlsyʔt> 'the evil deeds', the ʔ in this word is probably /ʔ/ rather than the long ā.

al-munšaʔātu <ʔlmnšʔt>  'elevated (of sails)', the ʔ in this word is probably /ʔ/ rather than the long ā.

marḍāti (a)llāhi 'approval of Allah', clearly part of religious vocabulary, may have an Archaic spelling.

    Already spelled as such in Wetzstein II 1913, carbondated to 662-765 AD.

fī rawḍāti (a)lǧannāti <rwḍʔt ʔlǧnʔt> 'in the lush meadows of the heavens', part of religious vocabulary, may have an archaic spelling. (Q42:22)

    Already spelled as such in Wetzstein II 1913.

ṯubātin 'groups', plural of ṯubah unattested in the Qurʔān. No obvious explanation presents itself which this would be spelled with <ʔ> (Q4:71).

    Already spelled as such in Wetzstein II 1913.

samāwātin <smwʔt> 'skies', as mentions shows up once like this and as <smwt> 189 other times. (Q41:12).

    Spelled as the expected <smwt> in Wetzstein II 1913.

naḥisātin <nḥsʔt> 'misfortune', occurs as such once, no obvious explanation (Q41:16).

    Spelled as the expected <smwt> in Wetzstein II 1913.

So of the few words that appear to be spelled with <ʔt> for -āt, two don't show up in the earliest documents, and 3 forms show up early. Two of which clearly are part of religious vocabulary, and it should be noted that outside this one attestation of ǧannāt is always just spelled <ǧnt>.

We can conclude that the spelling <-t> for -āt seems to be a well-established spelling convention in the QCT. It's difficult to decide whether this represents a phonetic reality of the Qurʔānic language, or is simply a spelling convention. The fact that other final -āt sequences are spelled with <ʔt>, at least occasionally, suggests that there was no specific sound law that caused word-final -āt to shift to -at.

Another insight I had on this long ā spelling is that  all nouns and verbs with the shape CāC(a) are written with an internal ʔ, regardless of whether they had an etymological glottal stop. The exceptions to this are staggeringly small:

There are 6 examples in total with an unwritten internal ʔ for a CāC stem, consisting of only two lexical items: qāla 'he said', ḥāša, and al-Lāt. qāla 'he said' is actually found much more often in this spelling in older Qurʔān documents (it's basically regular), and this seems to suggest that it was proclitic and unstressed somehow. This is not unusual, quotative verbs "and he said: "x"" are often proclitic (and thus accentless), like the classical Greek φημι (Benjamin Suchard p.c.). That this word was originally proclitic and 'special' in some way in the language of the Qurʔān may be inferred from the fact that some reading traditions require you to not pronounce the final a in qāla, and assimilate the l to following resonants (Ahmad Al-Jallad p.c.).

The other word ḥāša occurs twice in the Qurʔān, in the fixed phrase ḥāša li-llāhi <ḥš llh>, which is supposedly from ḥāšā li-llāhi from the stem III verb of the root ḥšw/y which has the meaning 'to except/exclude' (from s.o.)', Lane explains this phrase as  'I ascribe unto God remoteness from every imperfection or the like, or freedom thereform; generally implying wonder or admiration, so that it may be rendered how far, or how free, s God from every Imperfection!;  or this phrase means I seek protection by God; or, as often used by later writers, and in the present day, God forbid!'. Surprisingly, both long ā vowels are not written. I think that this fixed phrase may have simply shortened the vowels in colloquial usage, which spread to the orthography.

al-lāt, one of the Pre-Islamic Arabian goddesses, is spelled as <ʔllt> once in Q53:19, presumably an Archaic spelling.

To conclude: While accented ā seems to have been optionally written with <ʔ> or without, the unaccented etymological ā is never written with <ʔ>. This suggests that it was shortened.  It seems doubtful that this is part of the Orthography of the QCT, as such a specific phonetic rule seems difficult to consistently get right if it is not part of the language of the Qurʔān.

A final thought to ruminate over: The 1pl perfective suffix -nā <nʔ> is always written as <n> when it is followed by an object clitic. This suggests that it was unaccented in, e.g. ǧaʕalnā-hu <ǧʕlnh>. This is a little bit surprising if we follow the rules of stress assignment, as this is exactly the environment in which, e.g. kāfirūna has an accented long vowel (vvCv#). It is however also the exact same environment in which the feminine plural suffix is written as <t>. So perhaps there is a sound-law -ā́C# > -áC# at work anyway (assuming word-final vowels were gone).

The other option is that the object clitics did not partake in the stress assignment, in which case the stress would happily fall on the vowel preceding nā.