While translating al-Dānī's taysīr, I ran into a very funny name for Sūrat al-ʾIsrāʾ. While the name it has today in the Cairo Quran is rare in the past, Sūrat banī ʾIsrāʾīl being much more common, the name that Pretzl produced, سجن is one I had never heard of...
So I check Kandil's 2009 article which lists all the different names for the Sūrahs as mentioned in Medieval sources. There was no سجن there, but there was an obvious other candidate! subḥāna.
Hypothesizing that the scribe wrote this Sūrah name defectively سبحن rather than سبحان this can easily be explained, in Naskh script the distinction between the two is rather subtle.
I started looking in some of the manuscripts I have access to to confirm my suspicion.
Indeed, most manuscripts unambiguously have subḥān. And some of the manuscripts even have the hypothesized defective spelling. And funnily enough, the manuscript held in the Paris library (img 4) seems to actually have سجن! (slip of the pen, misreading on the scribe's part?)
Pretzl did not use the Paris Library manuscript, in fact I don't think he used a single one of the manuscripts that I have found digitizations for, but certainly a misreading of the very similar letters or one of these slips of the pen must have led him to this reading.
Months ago, I promised to do a follow-up thread on this series of comparisons between Nabataean Arabic and Old Hijazi. I said I would discuss the so-called Barth-Ginsberg alternation, this concerns the prefix vowel of verbs.
The medieval Arabic Grammarians tell us that the prefix vowel of verbs may be either /i/ or /a/, which is conditioned by the following vowel. If the vowel is /u, i/ the prefix vowel is /a/, and if the vowel is /a/, the prefix vowel is /i/.
- niʿlamu, nistaʿīnu
- naktubu, nafqidu
This alternation affects the prefix 1sg. ʾa/ʾi-, 1pl. na/ni- and the feminine or 2nd person ta/ti-. The masculine prefix ya- is said to be exempt from it (except for some contexts). Thus:
- ʾaktubu, taktubu, naktubu, yaktubu
- ʾiʿlamu, tiʿlamu, niʿlamu, YAʿlamu
The past few days I've been pondering over an interesting terminological conundrum in the use of the term madd 'length'/mamdūd 'lengthened' by al-Dānī (but also ibn Mujāhid), which seem to be mismatched with what he considered to be 'lengthened' in recitation.
So first some basics of Quranic recitation: the long vowels ā, ī and ū (and ē, ǟ and ǖ) are obligatorily made overlong whenever: 1. followed by a hamzah (glottal stop), e.g. السمآء as-samāāʾ "the sky" 2. in a closed syllable, e.g.: دآبّة dāābbah "animal"
This is called madd.
When there is disagreement among readers on such al-Dānī describes the long vowel that precedes the hamzah or consonant as "madd", rather than as ʾalif.
Ḥamzah and al-Kisāʾī: جعله دكا here with madd and hamz without tanwīn (dakkāʾa) and the rest: with tanwīn and no hamz (dakkan)
I'm always conflicted about the question of normalizing spelling in text editions. @bdaiwi_historia is right that this is standard practice for Classical Arabic text editions, but for linguists, this practice erases or distorts the history of a language, including Arabic.
Normalizing of spelling has long been a standard practice in a lot of philological fields, but in Indo-European Linguistics, my original field of study, people have been moving away from it.
This is because essential distinctions between, for example, Old Swedish and Old Icelandic only started becoming salient once text editors stopped normalizing everything towards an ideal "Old Norse" described in the grammars of the early philologists.
ifiɣr pl. ifaɣriwn is one of those words that has unexpected i~a alternation in the stem between the singular and the plural. Compare also igidr pl. igadrn 'eagle', iɣirdm pl. iɣardmiwn. Such alternations also show up in Tuareg tenere pl. tinariwen 'desert'
Kb. izimr pl. izamarən 'lamb'
Tuareg teɣse taɣsiwen 'ewe'
Tuareg eskăr pl. askarăn 'nail'
Whenever such alternations show up, Tuareg consistently gives a reflex with /e/ in the singular and /a/ in the plural. /a/ and /e/ seem to be phonetically conditioned variants elsewhere.
Because there is otherwise no reason to assume historical apophony here, I argue that here too the /e/, that becomes /i/ in most northern varieties must originally come from an *a, that shifts to /e/ in the singular, while this is blocked in the plural.
While there are hundreds of differences between modern print editions of the Quran and ancient manuscripts, this is not the case if you compare ancient manuscripts. They agree with each other in many non-trivial ways. I've done a quick a critical edition of Sūrat al-Raḥmān.
First image is some explanation on my sources and decisions that I have taken. Second image is the critical edition, which required only 12 notes in the critical apparatus. Many these deviations are typical only of later manuscripts. In the early centuries the text is very stable
At some point, some of the rasm is innovated, and in the eastern Islamic world the Uthmanic rasm is dropped altogether in favour of Classical Spelling. But before that the text transmission is remarkably stable.
This was done quickly maybe there's still some mistakes.
In the Quran, God is referred to in three places (Q2:255; Q3:2; Q20:111) as al-ḥayy al-qayyūm "The Living, The All-Sustaining". ʿUmar ibn al-Ḫaṭṭāb , the second Islamic caliph, is attributed as reading al-Ḥayy al-Qayyām, a reading which has an interesting biblical parallel.
This reading shows up in the non-canonical reading collections like Ibn Ḫālawayh, but also in Kufic manuscripts, see previous tweet where a vocalizer has added a yellow ʾalif to indicate al-ḥayy al-qayyām.
The specific al-ḥayy al-qayyām bring to mind a verse from Aramaic part of the Hebrew bible, Daniel 6:27 (thanks for alerting me of it@bnuyaminim!) it reads: הוּא אֱלָהָא חַיָּא וְקַיָּם לְעָלְמִין: hu ʾɛ̆lāhā ḥayyā wqayyām lʿålmīn "he is God, living and steadfast forever".