Studies in Semitic Grammaticalization.

Kaye, Alan S.

Studies in Semitic Grammaticalization. By AARON D. RUBIN. Harvard Semitic Series, vol. 57. Winona Lake, Ind.: EISENBRAUNS, 2005. Pp. xvii + 177. $32.95.

This book is a most welcome addition to the literature on grammaticalization, which may, following Paul Hopper and Elizabeth Close Traugott, Grammaticalization (Cambridge Univ. Press, 1993), 2, and quoted by the author, be defined as the "process whereby lexical items and constructions come in certain linguistic contexts to serve grammatical functions ..." Based on the author's Harvard doctoral dissertation completed under the supervision of John Huehnergard, it takes up a number of thorny issues within comparative Semitic linguistics.

Chapter one, "Introduction," explains the process of grammaticalization by citing well-known examples, such as French pas, "step," developing into the "negative" via analogy from constructions such as Je ne vais pas, "I don't go a step" (p. 3). From the coining of the term by Antoine Meillet in 1912, one comes to appreciate that Semitists were cognizant of the process without knowing it had a specific label, nor were they aware down through the decades of the twentieth century of the general linguistic literature in this field.

Chapter two, "Classification of Semitic," follows Huehnergard's classification of Afroasiatic (using the older hyphenated spelling "Afro-Asiatic") in Roger Woodard's Cambridge Encyclopedia of the World's Ancient Languages (Cambridge Univ. Press, 2004, 138-59). In his discussion of Arabic dialects, the author is correct in his affirmation that they "are more accurately described as languages, if we use mutual intelligibility as a distinguishing criterion" (p. 14).

Chapter three, "Grammaticalization in Semitic," discusses some examples of grammaticalization in the various Semitic languages. Let us begin with the reflexive pronoun. Many Semitic tongues use the word "soul" or a body part ("bone," "head," etc.) to mark reflexivity: Classical Arabic and Ge'ez nafs-, "soul" = Syriac and Biblical Hebrew naps-, Ge'ez r[TEXT NOT REPRODUCIBLE IN ASCII]'s and Moroccan Arabic ras, "head," and so on. Rubin is right to remark that due to the different lexemes used, "Proto-Semitic had (sic) no (reconstructable) reflexive pronouns" (p. 19).

An interesting development which offers many parallels in the languages of the world is the use of the lexeme "go" to mark the future (cf. English I'm gonna write). Thus the root rwh, "go," in the various Arabic dialects evolves into rah or lah, Egyptian Colloquial Arabic ha- (evolving even further into ha-, which is difficult to explain), and so on. (It is strange that the Classical Arabic dahaba, "to go," does not survive in any Arabic dialect, to my knowledge, as the normal verb for "go.") The root ydw, "to go (away); come to be," serves as the future marker in Moroccan Arabic (some Moroccan dialects use masi, another root for "go"), e.g., in its active participle yadi, shortened to yad- or just ya-(cf. Classical Arabic yadan, "tomorrow"). Demonstrating the mixture of inputting dialects into Maltese, historically an Arabic dialect but synchronically a Semitic language in its own right, the author points to two future prefixes: se ~ sa ~ ser < seyyer, "going," and ha-, cognate with the aforementioned ha- (p. 36). Another common verb developing into a future marker is the verb "want," the imperfect of which is yibya ~ yibyi, contracting to ba- and b(i)- = Yemeni Arabic sa- < 'asti, "I want" < Classical Arabic 'astahi. Still other Yemeni dialects use the particle 'a- or 'ad < 'ad, "still," related to do, "to do again; return," and Hebrew 'od, "still," although Rubin states that the etymology remains unclear (p. 37).

One of the most vexing problems discussed is the development of the genitival exponents in the modern Arabic dialects, paralleling Akkadian sa, Hebrew sel, Ge'ez za-, and Biblical Aramaic di (pp. 51-57). This phenomenon is surely to be related to the general drift from synthetic to analytic structure observable in Arabic dialects as a whole. Iraqi and Kuwaiti Arabic use mal, "property," Yemeni and Saudi Arabian hagg, "property, possession," Tunisian Arabic nta' < Classical Arabic mata', "property" = Egyptian and Sudanese Colloquial Arabic bita' via assimilation (however, note the author's extreme caution below), Chadian and Nigerian Arabic hana < Classical Arabic hana, "thing," Syrian and Lebanese Arabic taba' < Classical Arabic taba', "following" > "belonging." Rubin has this to say about relating Egyptian Colloquial bita' and Syrian-Lebanese taba': "But it seems to me that a metathesis of the taba' root attested in the Levantine dialects to the east is perhaps also a legitimate possibility and cannot be ruled out" (p. 54). I rule out any etymology which connects taba' and bita' because of the long vowel a, which occurs only in mata' and not taba'. I, however, agree with Rubin's affirmation concerning Moroccan and Algerian dyal, "of," often shortened to d-, viz., that they cannot be considered borrowings from a Romance language via Andalusian Arabic (p. 56).

Chapter four, "Definite Articles," musters arguments to support J. Halevy's theory ("L'article hebreu," Revue des Etudes Juives 23 [1891]: 117-21) on the origins of the definite article in Central Semitic from the etyma represented by the Akkadian near and far demonstratives (*hanni- and *?ulli-respectively). The shift of a demonstrative > article is an accepted diachronic tendency (cf. Latin ille, "that" > Spanish el, French le, Italian il, and even postposed Romanian -ul, "the").

Chapter five, "Direct Object Markers," is a detailed discussion of the origin of the various Semitic notae accusativi (Hebrew et; Phoenician 'yt ~ 't; Punic t; Aramaic 'yt, later yat, and l-; Neo-Assyrian ana; Arabic li- ~ la- (Maltese lil-); Ge'ez la-; Tigrinya ne-; Tigre '[TEXT NOT REPRODUCIBLE IN ASCII]l ~ '[TEXT NOT REPRODUCIBLE IN ASCII]g[TEXT NOT REPRODUCIBLE IN ASCII]l). Using parallels from Romance and Hindi, the author subscribes to a dative > accusative tendency. The phonological range of changes exemplified above prompts the following remark by Rubin: "I maintain that there is no need to find regularity in the development of the various notae accusativi in terms of their phonological development. These are clitic words which have been subject to restructuring of usage, and which are likely the product of grammaticalization" (p. 120).

Chapter six, "Present Tense Markers," deals with the origin of various tense markers. It is demonstrated that Iraqi Arabic da-, qa(d) < ga'id, "sitting," and North African ka probably < ka'in, the active participle of kan, "to be." The author is right to hypothesize that the Syro-Palestinian and Egyptian b- marking present is connected with the Yemeni Arabic b- < bayn- < Classical Arabic baynama, "while." The bi- marking future in Kuwaiti Arabic, etc. < yabi < Classical Arabic yabyi, "he wants."

Chapter seven presents a nice summary of the book, while the excellent and thorough bibliography (pp. 155-73) and the useful language index (pp. 175-77) conclude the tome. The publication has been very carefully edited and proofread, and the only error which came to my attention is the misspelling of the name of the famous deceased Arabist from Georgetown University, Richard S. Harrell (p. 18, n. 6).

The author expresses the noble goal of "... help[ing] bridge the gap between linguistics and Comparative Semitics" (p. 1), and we certainly concur that these fields can profit from one another.

ALAN S. KAYE

CALIFORNIA STATE UNIVERSITY, FULLERTON