An Eleventh-Century Karaite Hebrew Grammar

The study of eleventh-century Karaite Hebrew grammar has been enhanced during the last two decades, especially since the Firkowitch manuscript collections in the Russian National Library in St. Petersburg are again accessible to the public. One of the immediate consequences is the publication of the critical edition of Abū l-Faraj's Al-Kitāb al-Kāfi fī al-Luġa l-'Ibrāniyya (hereafter: Al-kāfi), by Goeffrey Khan et al. ¹ Prior to this publication, only small portions of Abū l-Faraj's works and theory have been studied and published. ² For the first time, the text of any of Abū l-Faraj works is presented comprehensively.

Khan's introduction to Al-kāfi is historical and general. He touches little upon the contents of Al-kāfi, yet the text in its comprehensiveness speaks for itself. We are lucky that Al-kāfi survived not only in its entirety but also in several copies, one of which is assumed to be an autograph. Khan's translation into English is brilliant-it makes the text lucid and constitutes in itself a tremendous contribution to the field. Khan and his collaborators have accomplished the enormous task of collating the text of Al-kāfi and preparing the critical apparatus, in addition to the task of translating the original Arabic into English.

Al-Kāfi's Structure

Al-kāfi was not Abū l-Faraj's first grammatical treatise. Initially he composed Al-Kitāb al-Mushtamil 'ala l-'Uṣū l wal-Fuṣūl fī l-Lugha l-'Ibrāniyya (The comprehensive book on the roots and branches of Hebrew language), which is an encyclopedic biblical Hebrew grammar; afterward he presented an abridged form in Al-kāfi; ³ then he went on to make a third and fourth shorter versions in the form of Al-Mukhtaṣar and Kitāb al-'Uqūd (pp. xi-xii). Unlike Al-kāfi, Mushtamil did not survive in its entirety, and its size contributed to its comparably poorer circulation and preservation. It was thus a wise decision on the part of the editors to prepare Al-kāfi first for publication.

This decision is further justified by the fact that Abū l-Faraj's linguistic theory is the same in both Mushtamil and Al-kāfi. Indeed, Abū l-Faraj did not write Al-kāfi from scratch. He rather used the model and style he developed in Mushtamil. The material is presented according to the taqsīm (=classification) method (e.g. I.14.12, I.15.2), so much so that though Al-kāfi is an abridgement it is not as concise and economical as it could have been. For instance, the content of Mushtamil part III-where the grammar is presented in a mnemotechnic way following the Hebrew alphabet ⁴ -is well presented in the same structure in Al-kāfi, though not as an independent part but rather embedded in part I, chapters 24-28 (pp. 228-523). Though shortened to half its original size, this section in Al-kāfi is still extremely large, consisting of over 27,000 words and taking up about a quarter of the book. Many grammatical rules dealing with morphemes consisting of more than one servile letter, for example, are repeated in different sections for the sake of formality, because they are discussed in each of their composing letters. ⁵ Had Al-kāfi been redacted differently or had the rhetoric and dialectical style at least been avoided (as, e.g., in I.27.51 ⁶ ), it would have been much smaller in scope.

Hence Al-kāfi's basic organizational structure is not different from that of Mushtamil. The abridgement was achieved either by omitting...