Consonants

The sounds in the chart above are based on the constructed phonology of Proto-Semitic and the phonology of Modern Hejazi Arabic.

Notes:

1. The consonants ⟨ض⟩ and ⟨ظ⟩ were voiced, in contrast with Northern Old Arabic, where they may have been voiceless^[2]
2. The glottal stop /ʔ/ was lost in Old Hejazi, except after word-final [aː].^[3] It is still retained in Modern Hejazi in few positions.
3. Historically, it is not well known in which stage of Arabic the shift from the Old Hejazi phonemes /p/, /g/, /q/ and /ɮˤ/ to Modern Hejazi /f/ ⟨ف⟩, /d͡ʒ/ ⟨ج⟩, /g/ ⟨ق⟩ and /dˤ/ ⟨ض⟩ occurred. However, the change in /g/ and /kʼ ~ q/ has been attested as early as the eighth century CE, and it can be explained by a chain shift / kʼ ~ q / → /g/ → /d͡ʒ/.^[4] (See Hejazi Arabic)

Vowels

In contrast to Classical Arabic, Old Hejazi had the phonemes [eː] and [oː], which arose from the contraction of Old Arabic [aja] and [awa], respectively. It also may have had short [e] from the reduction of [eː] in closed syllables:^[5]

The QCT attests a phenomenon of pausal final long -ī dropping, which was virtually obligatory.^[6]

Example

Here is an example of reconstructed Old Hejazi side-by-side with its classicized form, with remarks on phonology:

Notes:

• Basmala: final short vowels are lost in context, the /l/ is not assimilated in the definite article
• Line 2: the glottal stop is lost in /qurʾān/ (> /qurān), proto-Arabic */tišqaya/ collapses to /tašqē/
• Line 3: /taḏkirah/ < */taḏkirat/ < */taḏkirata/. The feminine ending was probably diptotic in Old Hejazi, and without nunation^[7]
• Line 4: /tanzīlā/ from loss of nunation and subsequent lengthening. Loss of glottal stop in /alarḍ/ has evidence in early scribal traditions^[8] and is supported by Warsh
• Line 5: Elision of the definite article's vowel in /lʿarš/ is supported by similar contextual elision in the Damascus psalm fragment. /astawē/ with fixed prothetic /a-/ is considered a hallmark of Old Hejazi, and numerous examples are found in the Damascus psalm fragment and support for it is found as well in Judeo-Christian Arabic texts. The word /ʿalay/ contains an uncollapsed final diphthong.
• Line 8: Old Hejazi may have had /huww/ < */huwwa/ < */hūwa/ < */hūʾa/ with an originally long vowel instead of /huwa/ < */huʾa/ as in Classical Arabic. This is supported by its spelling هو which indicates a consonantal /w/ rather than هوا had the word ended in a /ū/.
• Line 10: The orthography indicates /rāʾ/ , from */rāʾa/ < */rāya/ < */raʾaya/^[9]

Proto-Arabic

Proto-Arabic nouns could take one of the five above declensions in their basic, unbound form.

Old Hejazi (Quranic Consonantal Text)

The Qur'anic Consonantal Text presents a slightly different paradigm to the Safaitic, in which there is no case distinction with determined triptotes, but the indefinite accusative is marked with a final /ʾ/.

Dadanitic

A single text, JSLih 384, composed in the Dadanitic script, from northwest Arabia, provides the only non-Nabataean example of Old Arabic from the Ḥijāz.

Transitional Nabataeo-Arabic

A growing corpus of texts carved in a script in between Classical Nabataean Aramaic and what is now called the Arabic script from Northwest Arabia provides further lexical and some morphological material for the later stages of Old Arabic in this region. The texts provide important insights as to the development of the Arabic script from its Nabataean forebear and are an important glimpse of the Old Ḥejāzī dialects.

Arabic (Quranic Consonantal Text and 1st c. Papyri)

The QCT represents an archaic form of Old Hejazi.

Greek (Damascus Psalm Fragment)

The Damascus Psalm Fragment in Greek script represents a later form of prestige spoken dialect in the Umayyad Empire that may have roots in Old Hejazi. It shares features with the QCT such as the non-assimilating /ʾal-/ article and the pronominal form /ḏālika/. However, it shows a phonological merger between [eː] and [aː] and the development of a new front allophone of [a(ː)] in non-emphatic contexts, perhaps realized [e(ː)]. (citation?)