“| 737 has the sense of Arab. haja, ‘to articulate,’ in several passages : Pr 87, Is 598. 13,
** It seems difficult to separate the word yw used with 83} Hos 97, 833nD Jer 9938 (cf. 2 K 911), from the Arab. sij, ‘rhymed prose,’ the traditional style of the Kahins. The Heb, 127 was compared by Meier with the Arab. hija. Syn and mathal appear to be also independent.
tt Some examples are given by W. Wright in his Arabic Grammar (2nd ed.) and his Comparative Grammar. The form wpbya (Nu 205) is vulgar (kataltiina for kataltumiina). The uses of ]) can be illustrated by those of ya'niin languages that borrow from Arabic. The use of 12's as a final and explanatory
particle would seem to be a vulgarism. col is so used in
some Arab. dialects, and likewise in modern Armen. the relative wor has taken the place of yethé ‘that.’ Perhaps the Heb. yy ‘to do,’ is the Arab. ghashiya vulgarly used; cf. Lisan al-arab, xix. 363. 5
LANGUAGE OF THE OLD TEST.
LANGUAGE OF THE OLD TEST. 27
ancient language, they contain many relics of the classical rules. The classical language from which both are derived must therefore have flourished long before the 15th cent. B.c., for which time the existence of the later language is certified. The elaborate syntax and accidence which the early poetry of the Arabs exhibits would seem to have been codified more than two thousand years before that poetry was composed. It is in favour of this result that the Arabs have no accurate noticn of the commencement of their literature, or of the time when any of their classical metres was invented. Yet those metres imply the whole of the grammatical system, which can only have been the product of organized study. That all trace of the schools and colleges of early Arabia should have perished is noteworthy, but scarcely extraordinary, if we consider what such isolated monuments as the Mesha stele or the Iguvine tablets imply as to the extent of literatures that have wholly perished.
The evidence for the priority of Arabic grammar to the de- velopment of the Canaanitish language is to be found partly in what may be termed the residues which Canaanitish exhibits. Of these, examples are to be found (1) in the spelling, (2) in the grammatical forms, (3) in the syntax.
(1) As examples of orthographic residues, we may notice (a) the employment of δὲ to represent the sign of prolongation of the vowel o in a number of words in which the Arabic has the consonantal & preceded by a short ὦ (6.5. WN, INS, ON’; see Bottcher, Lehrbuch, i. p. 245). In some other words the letter & is still written without affecting the pronunciation. It would seem clear that the tribes who migrated from Arabia to Canaan had already found diffi- culty in pronouncing the consonantal Aleph, which indeed many still regard as the hardest of the Arabic consonants. They pronounced @ for a’, a pronunciation which indeed the Arabic grammarians tolerate in poetry. But while this ἃ in Arabic was either retained or reduced in the direction of @, the immigrants pronounced it as well as other Arabic @’s (with rare exceptions) as 6. The writing jN¥ for zon therefore is a case in which an old spelling is retained after it has become doubly unsuitable to represent the correct pronunciation ; and in all cases where this letter represents anything but the soft breath- ing, it must be regarded as a remnant from an earlier language, or due to false analogy. The perpetual interchange which we notice in the OT between roots δ Ὁ and roots 77’) shows that the consonantal & could no longer be pronounced at the end of a word. But from etymological orthography of this sort we can infer with certainty the existence of a literature in which the orthography agreed not only with etymology, but with the actual pronunciation ; in other words, the existence of written documents in Arabic earlier than the Canaanitish migration.
(b) Of no less interest as an etymological remnant is the em- ployment of the letter 4 at the end of words to represent the lengthening of a vowel, a peculiarity which the Phonician dialects apparently do not share witb the Hebrew and Moabitic. This mode of writing has two obvious sources. In Arabic the pausal form of nouns ending in atwn is ah, and in this form the h is pronounced as a consonant (Heb. 7), as we learn from its treatment in verse: thus martabah is made to rhyme with intabah, in which the h is radical (Hariri, ed. 1, p. 64), ete. This pausal form has in Hebrew ousted the other, That it is every- where pronounced ἃ for ai is a phenomenon to be easily illus- trated from Hebrew itself (in which the ah of the feminine suffix has a tendency to sink into @), and from many other languages. But the Phenicians did not adopt this pausal form, retaining the ¢ in the absolute as well as in the construct state. Hence one of the sources of this employment of the letter A was wanting in their language.
The second source of this phenomenon is to be found in the masculine suffix of the third person. Relics of the Arabic hu are not infrequent, but ordinarily (as in modern Arabic locally) that suffix is reduced to 6. When modern Arabic is written, the ἃ is retained (see e.g. Kata’if al-lataif, Cairo, 1894, p. 51, etc.), and the same is the case frequently in Hebrew and in Moabitic. In all these cases, however, it is an etymological remnant.