The Arabic Language

lingua franca
of this region, Aramaic, in a variety of the 
Aramaic script. For the history of Arabic they are of less importance, since they do 
not contain many Arabic words, and most of them are proper names. Sometimes 
these are spelled with the same ending 
-w
as in the Nabataean inscriptions.
The testimony of both the inscriptions from Petra and those of Palmyra with 
regard to the history of Arabic is indirect, since in both areas Arabic was the collo
-
quial language, whereas the language of prestige and written communication was 
Aramaic. Consequently, the Arabic elements in the inscriptions remain confined 
to proper names or loanwords, with occasional interference from the colloquial 
language in the written language. The information that we can glean from the 
inscriptions is limited, but we can deduce from them a set of orthographic princi
-
ples that determined the spelling of Arabic names. According to Diem’s (1973a) 
analysis of the material, these principles formed the basis for the orthographic 
conventions of the earliest Arabic script.
The influence of the Nabataean form of the Aramaic script is obvious first of 
all in the arrangement of the Classical Arabic alphabet, in which pairs of letters 
are distinguished by a diacritic dot or dots. This pairing goes back to the writing 
system of the Nabataean/Palmyrene inscriptions. Since the Aramaic script did not 
cover the entire phonemic inventory of Arabic, several letters had to do double 
duty. Thus, for instance, Aramaic 
dālet
transcribed both 
d
and 
ḏ
, 
ṣadē 
transcribed 
both 
ḍ 
and 
ṣ
,
 ʿayin
transcribed both 
ʿ
and 
ġ
, and 
ṭēt
transcribed both 
ḏ̣
and 
ṭ
. This 
principle does not mean that the phonemes in question had merged in the collo
-
quial Arabic of the period, but simply that they were not distinguished in the 
Nabataean script. In the case of the two phonemes 
ḍ
and 
ḏ̣
, which soon after the 
Islamic conquests must have merged in colloquial speech, the inscriptions show 
different reflexes, 
ḏ̣
being transcribed by 
ṭēt
, just like 
ṭ
, and 
ḍ
by 
ṣadē
, just like 
ṣ
. In the writing system of Classical Arabic, the effect of this distribution is still 
visible, since the letters 
ṭāʾ
/
ḏ̣āʾ
and 
ṣād
/
ḍād
form pairs that are distinguished by 
a diacritic dot. Apparently, 
ḏ̣
was perceived as the interdental counterpart of 
ṭ
, 
whereas 
ḍ
represented a different category (cf. above, Chapter 2, p. 24).
The most important convention that was borrowed from the Aramaic spelling 
of Arabic proper names concerns the spelling of the long vowels. Within the word 

---

The Earliest Stages of Arabic 
33
the long 
ā
is spelled defectively and is indistinguishable from a short vowel, while 
at the end of the word it is sometimes written with 
y
and sometimes with 
ʾ
. This 
distinction was probably meant as a device to indicate the morphological struc
-
ture of a word: the preposition 
ʿalā 
‘on’, for instance, is spelled with 
y
, because 
with suffixes it becomes 
ʿalay-ka
. This device was taken over by the Arabic writing 
system, hence the large number of words in which final 
-ā
is spelled with 
yāʾ
. 
The defective spelling of 
ā
within the word is still found in many words in the 
manuscripts of the 
Qurʾān
, for example, 
sulayman
, 
haḏā
, 
allah
; later this defective 
spelling was indicated with the so-called perpendicular 
ʾalif
or dagger 
ʾalif 
above 
the word. In one group of words, 
ā 
within the word is spelled in the Nabataean 
inscriptions with 
w
, for example, the word 
ṣlwh
‘prayer’, probably because in 
Aramaic the long 
ā
in these words had developed to 
ō
(Aramaic 
ṣlōṯā
). This is the 
origin of the Qurʾānic spelling of 
ṣalāh
, 
zakāh
, etc. with 
w
.
We have mentioned above the Nabataean principle of spelling proper names 
with 
-w
or 
-y
at the end. In Classical Arabic, this convention is still found in the 
proper name, 
ʿamr
, usually spelled as 
ʿmrw
. The situation in the Nabataean inscrip
-
tions is as follows (cf. Diem 1981: 336): masculine singular proper names very 
often end in 
-w
, that is, 
-ū
, when they are isolated, for example, 
zydw
(Zayd), 
klbw
(Kalb), 
ʿmrw
(ʿAmr). In compound names, the second member has either 
-y
or 
-w
, 
for example, 
ʿbdmlkw
(ʿAbd Malik or ʿAbd Mālik), 
ʿbdʿmrw
(ʿAbd ʿAmr), but 
ʿbdʾlhy
(ʿAbd Allāh), 
whbʾlhy
(Wahb Allāh). These endings occur independently of the 
syntactic context and are apparently quoted in isolation; this quotation form is 
supposed to be identical with the Arabic pausal form. This is not surprising since 
the Arabic names are intrusive elements in Aramaic, which has no case endings.
The most likely explanation for the compound names ending in 
-w
is that they 
are treated as single units following the same convention as the single names by 
ending in 
-w
. If they are indeed names quoted in their isolated form, this means 
that the endings 
-w
, 
-y
could be regarded as the pausal forms of the names. In 
Classical Arabic, the pausal form of a name such as 
ʿamrun
would be 
ʿamr
, except 
in the accusative singular 
ʿamran
, which has the pausal form 
ʿamrā
. But the 
Nabataean evidence suggests that in this earlier period Arabic had pausal endings 
ʿamrū
, 
ʿamrī
, 
ʿamrā
, of which only the third remained in Classical Arabic. Feminine 
names are usually spelled with the ending 
-t
, sometimes with 
-h
; if this, too, is 
a pausal ending, it could indicate a change in the pausal form of the feminine 
nouns, which in Classical Arabic has become 
-ah
.