From Classical to Modern Arabic.

Kees Versteegh literally wrote the book on Arabic, and he has helpfully uploaded the pdf of his “From Classical Arabic to the Modern Arabic Vernaculars” to; whoever I got the link from (Lameen?) recommended it as a good summary, so I feel confident in posting it for those who might be interested.


  1. ‘. . . pre-Islamic Classical Arabic was . . . based on the Bedouin dialects of the central and eastern Arab Peninsula’

    Given that the Bedouin and the Prophet lived in the Hejaz should that not read ‘central and western parts’?

    And given that there were Hadhramauti (eastern Yemeni) immigrants to al-Andalus, didn’t they have some influence on the Moroccan dialect of Maghrebi when their descendants became refugees there after the ‘Reconquest’ of Granada?

  2. Well, I saw a discussion recently in which someone posited that the speech of Najd at that time formed a “poetic koine” for the Arabs, while the more innovative speech of Hijaz formed most of the basis for the modern dialects. They suggested that the Koran was written in the latter variety but came to be recited in the former, explaining why CA features like nunation aren’t shown in the standard orthography.

  3. One of the nice synchronicities here was that skimming a bit of it and running into the term ‘lugat’ (tribal dialects), I finally realized what Lameen’s site name, Jabal al-Lughat, meant. Climbing the mountain of languages indeed.

  4. Wasn’t me, but it looks good!

    “Central and eastern parts” is correct; judging by the early grammarians’ scattered dialectological comments, or even by the present-day dialects, mainstream Classical Arabic is clearly closer to the dialects of the Najd than to those of the Hijaz. Qur’anic Arabic is of course automatically accepted as Classical by virtue of its status, but it has several features (presumably Hijazi) which are exceptional within Classical materials, such as the replacement of matā “when” by ‘ayyāna < 'ayy- 'ān-a "which time", or the assimilation in forms like tassā'alūna for more usual Classical tatasā'alūna "you wonder".

  5. I always assumed that Jabal al-Lughat was named in playful reference to the mediaeval Arab name for the Caucasus, ‘the mountain of tongues’, but on investigation that turns out to be jabal al-alsun. It’s certainly an apt name: an area the size of Spain (rather larger than California) with fifty-odd languages belonging to six families: Northeast Caucasian, Northwest Caucasian, Kartvelian, Indo-European, Turkic, Semitic, with the first three found nowhere else.

  6. Thanks for setting me straight, Lameen.

  7. By the way, Lameen, what is your opinion of the Jordanian film ‘Theeb’?

  8. John: Your assumption was correct, in fact, but I came across the expression in an article in English, and never managed to track down the Arabic original, so I ended up back-translating…

    iakon: Never watched it. But on the Moroccan-Yemeni connection, there do seem to be a certain number of Yemeni-Maghrebi isoglosses (the one I always remember is juljulān rather than simsim for “sesame”.)

  9. For LH readers who are interested and have registered with (it’s free and easy if you haven’t!), here is a link to a recent treatment of the Yemeni-Maghrebi connection, with relevant bibliography of previous studies of the question:

    Marijn van Putten (forthcoming). The illusory Yemenite connection of Andalusi Arabic

  10. Trond Engen says

    Something tells me that Lameen has seen that paper before*.

    *) I cheat. I read footnotes. At least the very first.

  11. Come to think of it, California two centuries ago had more than 80 languages of at least seven families (or 20 if you don’t accept Hokan and Penutian). It was the Caucasus of North America.

  12. J.W. Brewer says

    California is reasonably close in square miles to Papua New Guinea, which is probably way ahead of the Caucasus in language diversity. (The Caucasus is similar to both in square mileage according to one online source, although obviously the boundaries there are more judgmental/fuzzy.) Hard to compare languages on a per capita basis since estimates of the indigenous population of what’s now California prior to European contact vary wildly and presumably ditto for New Guinea.

  13. David Marjanović says

    From this one onwards, most posts on Phoenix’s blog are about the differences between the spellings and the pronunciations of Classical and Qur’anic Arabic. Note that the list of 5 posts at the bottom hasn’t been exhaustive in a long time.

  14. Of course NG (in linguistic terms the border is irrelevant, and the island is about equally divided by area) is like nowhere else in the world today: ten million people speaking a thousand languages, in round numbers. Even Nigeria, a language diversity hotspot, has only half that many languages and twenty times as many people, spread over ten times the area.


  1. […] Hat links to a paper examining the transition from classical to modern […]

Speak Your Mind