Thread by @PhDniX

As promised, here's a follow up to my series of ongoing comparisons between Nabataean Arabic and Old Hijazi. This time we will look at the Deictic system within the Arabic of the Nabataeans and the grammarians.

https://twitter.com/PhDniX/status/1358376447035998210

The medieval Arabic grammarians report an astounding amount of variation in the deictic system. And a good amount of this is dialectal variation.

The base deictics are:
m.sg. ḏā
f.sg. tī, ḏī, ḏih, ḏihī
pl. ʾulāʾ, ʾulā (or ʾulē)
loc. hunā

The grammarians tell us that the Hijaz is that the near deixis 'this, these, here' is *always* combined with a presentative hā-. Thus hā-ḏā "this (masc.),", hā-ḏihī "this (fem.)", hā-ʾulāʾi "these" and hāhunā "here".

Farrāʾ only reports the difference explicitly for the plural

In Classical Arabic the forms with hā- have become practically standard. You do not find ḏā, ʾulāʾi or ḏihī. However the unprefixed form hunā 'here' is quite frequent besides hāhunā. The Quran clusters completely with Hijazi here, and only has hāhunā (and hāḏā etc.).

Quranic Arabic distinguishes itself in this regard from the pre-Islamic Nabataean inscriptions. For example the Harran inscription which dates 568 CE has ذا المرطول ḏā l-marṭūl "this Martyrion", not هذا المرطول as we would expect in Quranic/Hijazi Arabic.

The feminine form hā-ḏihī is also considered a special Hijazi isogloss by Sībawayh. The banū tamīm are said to have hāḏī instead (and hāḏih in pause). In the Quran we only find the Hijazi form. In Classical Arabic prose hāḏihī is also standard, but poetry frequently has hāḏī

The feminine deictic of Nabataean Arabic seems to use neither ḏī or ḏihī as its base, but the third possibility of the feminine: tī. For example in the Namarah inscription (328 ce) we find: تي نفس مرالقيس where Hijazi would have هذه نفس امري القيس.

The plural deictic is yet another place where a difference is reported. Although admittedly more explicitly for the distal form, al-Farrāʾ tells us that the Hijazis say ʾulāʾika "those" (and hā-ʾulāʾi for "these") while other tribes say ʾulāka (or ʾulēka?) and (hā-)ʾulā/(hā-)ʾulē

It is indeed the Hijazi forms that we find in the Quran: هولا, اوليك not هولى, اولاك.
In Classical Arabic prose, the Hijazi long ʾulāʾi stem has won out, and even in poetry I don't think the short form occurs often (though I'm sure it does at times).

For Nabataean Arabic, we simply do not know what the plural deictic looked like. @phillipwstokes recently published an article that shows evidence for the plural base in Safaitic, but that script does not allow us to distinguish between ʾulāyi and ʾulayi.
academia.edu/44769456/2020_…

Now turning to the far deixis. The grammarians tell us that there are two strategies for the singular. The people of the Hijaz add a -l(i)- to the far deixis: ḏālika 'that (masc.)', tilka 'that (fem.)', whereas other tribes have ḏāka, tīka.

As al-Farrāʾ points out explicitly: the Quran exclusively has the Hijazi form. In Classical Arabic prose ḏālika is certainly dominant, but ḏāka occurs with some frequency I think it would be quite interesting to study if a functional difference can be discerned.

In Classical Arabic poetry ḏāka frequently occurs besides ḏālika. Interestingly, tīka does not seem to occur besides tilka. This is probably because ḏāka and ḏālika are metrically difference whereas tīka and tilka are metrically identical, so serve no practical purpose.

The li- expansion even applies to the locative deictic in the Quran: hunālika "there", a feature likewise identified as Hijazi, whereas the Tamīm had hunāka. In Classical Arabic, hunāka seems to be used far more frequently than hunālika. This brings up an interesting issue:

While most of the deictic system in Classical Arbaic follows the Hijazi pattern (hāḏā, hāḏihī, hāʾulāʾi, ḏālika, tilka, ʾulāʾika) for some reason the Najdi forms for the locative deict, i.e. hunā, hunāka won out over hāhuna and hunālika.

As of yet, no Nabataean Arabic inscriptions have been found with the distal deictics. Although in Safaitic they do show up, indeed lacking the characteristic Hijazi -l(i)- expansion. It is for this reason that @Safaitic argued this to be one of the typical Hijazi innovations.

So as should be clear from this thread, the deictic system brings several isoglosses that distinguish Old Hijazi from other regions (including Nabataean Arabic). The Quranic system is perfectly Old Hijazi; In Classical Arabic, however, we see non-Hijazi forms with some frequency.

That's the thread for now. Next thread will look at another clear innovation of Old Hijazi, the lack of Barth-Ginsberg alternation.

If you enjoyed this thread, please consider buying me a coffee: ko-fi.com/phdnix

Or become a patron on patreon: patreon.com/PhDniX