Athabaskan language studies: Essays in honor of Robert W. Young Ed. by Eloise Jelinek, Sally Midgette, Keren Rice, and Leslie Saxon (review)

BOOK NOTICES 441 encies' under study—pauses, repetitions, code switching and the like—are objectively correlated with loss of specific L1 grammatical competence. With the exception of word reordering (for which no examples are given), the author deliberately eschews 'textualist' analysis, i.e. loss of particular grammatical or phonological structures, on the grounds that these cannot be appropriately quantified. As a result, there is no demonstration that speakers who e.g. pause frequently, code switch, repeat words, speak more slowly, etc. have actually lost any grammatical competence at all. 'Fluency' as used in this book is in reality a measure of rhetorical abilities such as speaking steadily without pauses or false starts rather than a gauge of actual language loss: ' "L1 competence ". . . [is] aggregate verbal planning, processing and recall ability of a speaker' (6). 'Macro-fluency' is then "the degree to which a speaker is able to carry-out [sic] these basic linguistic processes as expressed by the relative rate of hesitation and speech error frequency' (6). Given this self-fulfilling definition , the notion that L1 competence has deteriorated in the test group is a natural conclusion; most who have studied language attrition would not fully accept this definition of linguistic competence, to the exclusion of fluent use of particular linguistic structures. Equally problematic is the claim that code switching represents language erosion, particularly since the switches observed in the author's corpus were almost entirely from English (presumably the weaker language ) to Arabic (the native language); this type of lopsided code-switching more properly suggests shifting to the stronger language (in this case, L1 ) in moments of linguistic stress, which contradicts the claim that increased code switching represents decreased L1 fluency. From a stylistic viewpoint, the text is marred by an overwhelming use ofhedge words, such as 'would seem', 'perhaps', 'could be', 'intuitively', which qualify all attempts to interpret the data in terms of the speakers' life histories and cognitive strategies. At no point is a unified account of the phenomena put forward; rather, the accumulation ofhedges yields an ever-changing set of speculations. Finally, the quantitative analysis (which is overburdened with an enormous number of graphs and charts) is not as robust as the author claims, since the .05 significance level used as a threshold is higher than most psychometric researchers would find comfortable; many ofthe agegraded charts are less than compelling. Given the methodological shortcomings, the lack of demonstration that hesitation phenomena represent language attrition, and the contradictory results of L1 fluency depending on which variable is chosen, this book falls short of a 'picture of the temporal dynamics of language in speech communities' (231). It is, however, highly suggestive of future study and should not be dismissed by researchers in bilingualism and language loss. [John M. Lipski, University of New Mexico.] Athabaskan language studies: Essays in honor ofRobert W. Young. Ed. by Eloise Jelinek, Sally Midgette, Keren Rice, and Leslie Saxon. Albuquerque: University of New Mexico Press, 1996. Pp. xv, 490. This volume contains sixteen essays dedicated to one of the foremost scholars of the Navajo language. Working in conjunction with William Morgan, Young has produced definitive grammars and lexicons of Navajo as well as contributing to language standardization and literacy programs, and the articles in this festschrift attest to the range of scholars who have benefitted from Young' s unflagging devotion to the Navajo language. Although Young' s own work did not often carry him beyond Navajo, his writings have important implications for comparative Athabaskan linguistics, as reflected in many of the contributions. The first seven articles deal with syntax and morphology , arguably the richest and most complex component of the Athabaskan languages. Kenneth Hale and Paul Platero offer 'Navajo reflections of a general theory of lexical argument structure' (1-13), based on overtly marked transitivity alternations in verbs which have both transitive and intransitive variants. The presence of a subject internal to the projection of V is tentatively implicated in the alternations. Eloise Jelinek and Maryann Willie discuss ' "Psych" verbs in Navajo' (15-34). Navajo has a few psych verbs with Experiencer subjects and a large set of psych verbs with Oblique Experiences. The particular Navajo argument structures are related to the claimed status...