I'm a big admirer of this young scholar Ahmad al-Jallad, who I think is brilliant (the Ph.D. with distinction from Harvard is probably a good tip-off on that point).
http://www.hum.leiden.edu/lias/organisation/arabic/aljalladam.htmlHe has recently started a new series of articles on an incredibly difficult and interesting subject: What, exactly, is Arabic from a linguistic perspective? How did it develop? And critically, what exactly was pre-Islamic Arabic in the Levant and Northern Arabia? Here is the first article in that series, which just came out:
http://www.academia.edu/7583140/Graeco-Arabica_I_the_southern_LevantHis methodology is ingenious -- by focusing on pre-Islamic *Greek transcriptions* of texts within the Nabatean area, he is able to perform a sophisticated analysis of the Arabic substratum of the region, including its various features. As it turns out, Greek script is terrible for representing Arabic consonants, but quite good for representing complex vowelization.
Numerous fascinating points come out of this analysis, but here are some of Jallad's big arguments (some made in his previous papers). First, there is no clear split between Arabic and "Ancient North Arabian" languages, but rather there was a continuum of Old Arabic languages across the entire region PRIOR to Islam. Second, the modern Levantine dialects were already in place prior to Islam -- these languages did not migrate to the area with Arabs from the peninsula, they were already in the area. Modern Arabic, according to Jallad, represents a 'polygenesis' from several different pre-Islamic Arabic groups, which then mixed to some degree as a result of Islam. Last, it is well known that the Arabic script is derived from the Nabatean (aka Southern Levant) script (which was used initially for Nabatean Aramaic, but also Nabatean Arabic). According to Jallad, however, the base Qur'anic rasm not only reflects the *script* of this region, it also represents the *speech* of this region.
This is important because it relates to a long ongoing debate about the base Qur'anic script, the rasm: Why do so many features of that base script seem so remarkably different than Classical Arabic? The Qur'an appears to have been originally written in a different orthography than it was later read in by Muslim tradition. Certain features -- the lack of medial hamz, the lack of case endings, the peculiar long vowel markings -- seem completely different from Classical Arabic. Yet those same features also seem consistent with Nabatean Arabic. Is this just a case of the Qur'an being written in the 'wrong script', but then read in the 'right language'? In other words, why are words like "salwt" written in an orthography different than Classical Arabic, which reads the 'w' as a long "a"? Why is there such variation in the vowelization of the rasm and the earliest Qur'anic manuscripts? Is this just an alien scribal convention that was 'corrected' even at the time of composition by recitation similar to the later Classical Arabic tradition?
If you look at pages 37 and 38 of his new article, Jallad weighs in with some interesting thoughts. Basically, he argues that the base orthography of the Qur'an agrees with Levantine Arabic as actually spoken at that time, and that Classical Arabic's attempt to read the "y" as mater lectionis for long "a" is erroneous; the Qur'an's vowelization in the base rasm reflects a "non-a" pronunciation. Unfortunately this website doesn't allow me to cut and paste the scripts correctly, but you can see how remarkably different the later "Classical Arabic" reading of this vowelization is relative to the base rasm letters -- Classical Arabic recitation levels out the vowelization incorrectly, reflecting secondary developments. This is not just a matter of the Classical Arabic recitation being written in a 'different dialect' either, because the Qur'an's rhyme scheme is not consistent with that argument -- the rhyme is written to fit the base orthography, not the Classical Arabic recitation.
Apparently Jallad has several further articles that will be coming out which address these issues. Can't wait!
I thought people might be interested in this line of research, and have thoughts about it, because as far as I'm concerned it's some of the most fascinating and important research out there. We still know incredibly little about pre-Islamic Arabic, and about early Arabic orthography -- and as a result of that uncertainty, we still don't know much about the language reflected in the base Qur'anic script. I'm hopeful that guys like Jallad will eventually create a much improved linguistic basis for approaching these issues, and as a result a much better understanding of what the Qur'an is, how it was compiled, where it was compiled, and what its language/script actually means.
Topic: Pre-Islamic Levantine Arabic and the Qur'an