Linguists, Semitists, historians, and biblical scholars will welcome two new research aids by Charles R. Krahmalkov. A specialist in Northwest Semitic languages and epigraphy, the author has produced brief, accessible, and innovative analyses of the Phoenician language and its vocabulary. Since 1970, Krahmalkov has published fundamental studies elucidating Phoenician-Punic grammar and syntax. Transliterated Punic texts are among the least intelligible in the surviving corpus of Northwest Semitic. Krahmalkov has long pondered the Punic passages of Plautus' Poenulus and the Neo-Punic stelae in Latin letters from Tripolitania, and his achievements have made these vocalized texts fundamental for grammatical description of Phoenician and Punic. (1)
Both books are accessible to specialists and educated non-specialists. All Semitic texts are transliterated, and the writing is clear and precise, avoiding technical jargon and needless formalism. Examples almost invariably include transliteration. English translation, and morphological or phonological analysis. When the author comments on a form or construction, he cites complete phrases or sentences (in transliteration and English translation) in support of the analysis. Italic capitals transliterate Phoenician letters; Latin letters are transliterated with boldface lowercase roman type. (2)
The reader can discern the textual base of both the grammar and the lexicon from the reference lists (PPD 19-21; PPGK xvi-xix). (3) Most published texts are represented, with some omissions. (4) Krahmalkov generally follows established readings, but offers several brilliant new restorations and alternative readings. Specialists may find some readings open to question.
The limited bibliographical citations make these works truly a personal statement. (5) Scholarly consensus is inconsistently represented, compelling readers to weigh the evidence by comparison with other grammars and lexica. I will review the grammar first, then the dictionary, despite the reverse order of publication.
A PHOENICIAN-PUNIC GRAMMAR
THE PHOENICIAN LANGUAGE (1-15). Krahmalkov sets out cultural and geographic terminology first: the indigenous name of Phoenicia was PT /put/; the name of the Phoenicians and their language was /ponnim/ (note the interpretation of Ps. 45:12b-14a). A discussion of Plautus' Poenulus establishes this usage (3-5). Linguistic diversity characterized "Greater Phoenician" (6) as a language in all periods and regions (6). The essay then sketches the southern coastal dialects (7-8), the northern coastal dialects (Arvad, Byblos) (8-9), and western Phoenician (10-15).
ALPHABET, ORTHOGRAPHY, AND PHONOLOGY (16-37). According to the author, the consonantal system for writing Phoenician derives from literary Ugaritic, but Phoenician scribes infrequently employed waw and yod as vowel letters in spelling foreign names and writing certain inflectional morphemes (e.g., pleonastic spelling of the pronoun 'NKY 'aniki "I"; plene spelling of the first-person singular possessive suffix -i "my"). The tendency to employ vowel letters increases with time in Punic and Neo-Punic. In Phoenician, 'alep is infrequently used to represent final vowels in transcribing hypocoristic personal names and foreign personal and geographic names. Krahmalkov delineates orthographic distinctives of Cypriote inscriptions and the increasing divergence of Punic and Neo-Punic spelling practice.
Before surveying phonology, Krahmalkov acknowledges the sporadic and incomplete character of the evidence, cautioning that his own description is "perforce fragmented, incomplete and always problematic" (20). These limitations arise in part because the author makes more extensive use of examples from transliterated Punic and Neo-Punic texts than any previous grammar of the language.
According to Krahmalkov's analysis, the Proto-West-Semitic sibilant series [theta], s, s merged in Phoenician as simple /s/, represented orthographically as S (25-26). Transcriptions such as 'S = /'is/ may confuse readers who fail to apprehend this point. Regarding vowels, note the single example of the so-called "furtive" a-vowel before a laryngeal (32). The discussion of word stress and vowel reduction (33-37) advances beyond previous treatments, bringing system to the apparently disparate spellings of Phoenician words in Greek and Latin letters. Among the truly brilliant analyses are Punic ierasan /ye[r'.sup.a]san/
INDEPENDENT PERSONAL PRONOUNS (38-49). Chapter 3 begins the format that continues through the rest of the book: section A of the chapter or topic presents morphology (followed by comments), and section B takes up syntax and usage (see the rationale, xiv). The vocalization of the first-person singular independent pronoun /'aniki/ is explained from the Latin-letter spellings anec(h) and anic (39-40). (7) The intensive personal pronoun BT or BNT "I myself" is now fully described (47-48). (8) The anaphoric pronoun ("that, the aforementioned") is analyzed with respect to syntactic context (determination of the antecedent noun or of the pronoun, 48), and the emphatic anaphoric pronoun (BT- or BNT-"the/that very, the/that same"; to be distinguished from the intensive personal pronoun is a newly recognized form. (9) Note that the second-person feminine plural independent pronoun remains unattested (39, 41). (10)
SUFFIXAL PRONOUNS (50-74). One of Krahmalkov's lasting contributions to the study of Phoenician and Punic grammar is his description of the morphology and distribution of suffixal pronouns. (11) Krahmalkov worked out the phonological shape of these morphemes in several important studies, updated and supplemented in this grammar. The second-person feminine plural suffixal pronoun remains unattested. Krahmalkov has supplied three attestations of the third-person feminine plural suffixal pronoun (55, 59). The synchronic dimension of this grammar's approach can be seen in its care to document orthography as well as morphology: e.g., QL' and QL' are listed as variants of the third-person masculine singular suffix in Form A (51), although only a spelling variation is involved. The discussion of syntax (72-74) distinguishes five categories of object pronouns (note the interpretation of KAI 89.2).
DEMONSTRATIVE PRONOUNS AND THE DEFINITE ARTICLE (75-92). The grammar recognizes a Z-series (masc. Z, 'Z, Z', esde, esse; fem. Z, 'Z), an S-series (S, si, sy), an S-series with excrescent -t (ST, sith, syth), the neuter hoc (a Latin loan?), and the plural ('L, 'L', ily, illi). The syntactic discussion (77-82) distinguishes pronominal and adjectival uses of the demonstrative, further distinguishing deictic and locative uses. Determination is a free variant, producing the phrases QRT Z, HQRT Z, QRT HZ, and HQRT HZ. Byblian employed two sets of demonstrative pronouns, labeled A and B (82-85). The section on the definite article makes significant steps forward in defining the morphology and phonological realization of determination in Phoenician and Punic.
RELATIVE AND DETERMINATIVE PRONOUNS (93-107). The first part of the chapter (93-103) concerns the relative pronoun. The author mentions but does not describe the old relative pronoun zu, written Z, found in archaic inscriptions from Byblos (KAI 1-7). From the ninth century onward, the relative pronoun 'is replaced the older form in all dialects of Phoenician and Punic (94). (12) The phonetic shape can be determined from "the plene spelling 'YS and Roman and Greek letter spellings es, is, ys, vs." (13) A Hebrew example is cited in Num. 1:4. Krahmalkov stresses that the form s- is a determinative pronoun, not a relative (94). (14) Late in the Neo-Punic phase, a form mu replaced 'is, or combined with it in the relative phrase mu 'is (this form is the key to a stunning interpretation of Poen. 939). A possible example of Phoenician M = mu in a votive inscription on a bronze carinated bowl (15) is problematic, as the author notes (the text as transcribed ignores the numeral 2 in the original).
Discussion of the syntax of the relative pronoun is organized in eleven sections. The first category involves a relative pronoun intro-ducting non-verbal clauses (a) with or (b) without an independent personal pronoun (95-96). In the second category, the relative pronoun introduces verbal relative clauses employing (a) participles or (b) finite verbs (96-97). The third category involves relative clauses with a resumptive pronoun, subdivided into (a) clauses including a pronoun that resumes the indirect object and (b) clauses without such a pronoun. (The last example under category b  appears to be misplaced; see my comments concerning category eight below.) Additional categories involve ellipsis of the antecedent of the relative pronoun, relative pronoun with definite article, locative function of the relative, and relative pronoun as adverbial complement to the jussive/optative (99). This last category is exemplified in a Neo-Punic inscription from Mactar (KAI 147), ingeniously interpreted by Krahmalkov as a report of a prayer service. The damaged condition of the inscription leaves room for uncertainty about its complete restoration, but as a possible example of Punic prayer, the passage has tremendous significance. (16)
Category eight, designated the virtual relative clause, (17) fits the Neo-Punic example from KAI 168 (101), which appeared out of place in category 3b above. The Phoenician example, from the Ur ivory box (KAI 29), involves a restoration. Krahmalkov follows H. L. Ginsberg's apt suggestion that the fourth letter be restored as S rather than Z, an earlier restoration. (18) This new restoration is rejected by some epigraphers in favor of the relative [Z]N. (19) Krahmalkov's syntactic analysis depends on the restoration S, which eliminates the relative pronoun; MGN is the verb (see 167). Firmer examples are needed.
Categories nine and ten examine compound relative phrases: 'S L-...