Reviewed by Elly van Gelderen, Arizona State University

Warihío is a Uto-Aztecan language of the Taracahitic branch of Sonoran. Rolando Félix Armendáriz has done extensive field work on one dialect of Warihío, River Warihío, and the related Yaqui.

The grammar consists of a short sketch of the sound system (5–13), the word classes (14–40), the noun phrase (41–47), relative clauses (48–52), the simple sentence (53–105), negatives (106–11), voice and valency (112–50), complex sentences (151–81), and a typological perspective (182–89). There are uncountable example sentences with excellent glosses but no glossed full text. The book is couched in a (mild) functionalist framework.

Of particular interest is the fact that the language has no case marking on arguments (nouns and pronouns) and has a relatively free word order. Most non-core argument roles are marked by postpositions. In ambiguous sentences with more than one participant, however, the Agent comes first (79), and the subject may be marked on the verb.

The subject is marked as either a free noun or pronoun, as in neé umá-re tapaná ‘I ran yesterday’ (62), or as a clitic on the verb, as in yau-rú=ne tapaná ‘I danced yesterday’ (78). As in other Uto-Aztecan languages, these clitic and pronoun forms seem to be related, with the clitic a more grammaticalized form. As the answer to a question in which the first person would be the new information, the free pronoun is chosen as in the first example so as to come first in general focus, whereas the clitic in the second example would not be licit.

Non-subject pronouns are free, except for one bound first person singular form (29; 76), and typically occur before the verb. There are interesting differences with Yaqui in that the third person object pronouns can be dependent forms (Lilián Guerrero, The structure and function on Yaqui complementation, Munich: LINCOM Europa, 2006) but not first and second person.

Many other suffixes occur, to indicate mood, tense and aspect, valency, and embedding, as well as purposive and desiderative. In the conclusion, there is a short discussion of the relationship between culture and language (188–89) in which the small size of the typical Warihío community is linked to the lack of participant marking. I would have liked more evidence of this since it seems to shape A’s description, e.g. in emphasizing the free word order and lack of argument role marking.