The expression of possession (review)

New investigations continue to provide insights into the linguistic realization of possession. Stassen (2009) is one of the new typologically motivated studies of predicative possession. The current book is yet another which offers a variety of analyses to contribute to our knowledge of possession from typologically diverse language families. This book consists of an introduction, nine additional articles, a list of contributors, and indices of subjects, languages, and persons.

In the introductory chapter (pp. 1-12), William B. McGregor overviews the concept of possessives and defines possessives as conceptual relations between entities, between people and their body parts and products, between people and their kin, between people and their representations (such as photographs), between people and their belongings, between people and things that they have control over (such as usership rights), between persons and cultural/intellectual products (such as copyrights), and so on. McGregor then summarizes the articles in the volume.

In "English possessives as reference-point constructions and their function in the discourse" (pp. 13-50), Peter Willemse, Kristin Davidse, and Liesbert Heyvaert focus on possessive noun phrases such as his car and argue against the claims that they are definite NPs and/or new information in discourse. Their data come from the COBUILD Bank of English and the Lancaster-Oslo-Bergen corpus and consist of 400 instances of possessive noun phrases. The analysis shows that, in addition to a subsequent mention (coreferential, 11.86% of the data) and the first mention (new, 28.03% of the data), they function in three more ways: text reference (a subsequent mention of a text referent, 8.62% of the data), inferable (inferable from an associated referent or scenario in the preceding context, 26.95%of the data), and anchored (anchored to an element in the preceding context, 24.52% of the data). Thus, they propose that these five categories form a continuum from discourse-given (coreferential) to discourse-new (new).

In "On the co-variation between form and function of adnominal possessive modifiers in Dutch and English" (pp. 51-106), Jan Rijkhoff investigates possessive modification with of in English and its Dutch counterpart with van which can modify a common noun such as car, dog, or tree. The data are mostly constructed, but in some cases the author consults search engines on the internet and examples from other languages. Rijkhoff argues that, with respect to the functions modification, predication, and reference, three out of the four possible types of adnominal possessives occur in English and Dutch, i.e., classifying, qualifying, and localizing, but not quantifying. The author pays considerable attention to the functions of the adnominal possessives with of/van which are functions of natural classes such as adjective, verb, and noun.

In "Is possession mere location? Contrary evidence from Maa" (pp. 107-142), Doris Payne reports on a study of how Maa, an Eastern Nilotic, Nilo-Saharan language spoken in Kenya and Tanzania, encodes location, possession, and existence. The author focuses on two lexemes: tii and ata. Tii occurs in both transitive and intransitive constructions with locative and existential meanings, whereas ata occurs in both transitive and intransitive constructions with possessive and existential meanings. The author reports an analysis of the first 130 tokens of tii and ata that are found in a corpus consisting of 20,000 clauses (mostly spoken and some written materials). The data show that tii and ata in Maa are in complementary distribution suggesting that possessive ata may not be derived from locative tii, contrary to what one might expect from a localist perspective.

In "Learning to encode possession" (pp. 143-212), Sonja Eisenbeiss, Ayumi Matsudo, and Ingrid Sonnenstuhl explore how possessive relationships (adnominal as in my chickens, predicative as in I have chickens, and external possessive constructions as in I tapped him on the shoulder) are encoded in German child language. The authors also provide some child language data available from other languages such as English, Japanese, Greek, and Hebrew, as well as adult German. Their German data come from 64 recordings...