Turning owners into actors: Possessive morphology as subject-indexing in languages of the Bougainville region

Project Overview


Turning owners into actors: Possessive morphology as subject-indexing in languages of the Bougainville region

Project members:

Prof Greville G. Corbett
Dr Dunstan Brown
Dr Bill Palmer

Period of award

February 2005 - February 2008


Arts and Humanities Research Council (AHRC)

A fundamental communicative task for all languages is to show which participant in a sentence is the subject. Languages have various ways of identifying the subject, including word-order, agreement, and case-marking. However, there is another unique and strange method, almost entirely unknown until now, found only in Northwest-Solomonic (NWS), a group of Oceanic languages of the Solomon Islands and Bougainville. In some constructions, these languages indicate subject using word-forms normally indicating possessors of nouns. This use of possessive morphology to mark subjects is theoretically highly significant. To define language fully we must understand the limits on subject-marking. This almost unresearched phenomenon is crucial to our understanding of the fundamental issue of how subjects can be marked.

This project investigated this almost unresearched phenomenon: how it works, how it varies, what it does, and where it comes from. Since the key languages investigated are highly endangered, the project was timely, and as a by-product, resulted in some partial primary documentation work. In most Oceanic languages of the North West Solomonic subgroup (spoken in Bougainville and the western Solomon Islands), some use is made of apparent possessive morphology to index subject and encode aspect on verbs.

Most languages of the North-West-Solomonic (NWS) branch of Oceanic make some use of possessive morphology to mark verbs and index subject. In Mono, for example, a distinct construction employs two separate auxiliaries carrying subject-agreement suffixes:

(1) a. eli sa-ria                  b. soipa ma-mate e-na  
    dig AUX-3PL.SUBJ                Soipa REDUP-die AUX-3SG.SUBJ  
    'They went on digging.' (Wheeler 1926)   'Soipa is dying.' (Ross 1988:250)

The highlighted forms in (1) do not reflect Proto-Oceanic verbal morphology. Instead, they reflect nominal possessive forms, the auxiliaries reflecting classifiers distinguishing general possession (GP) and consumable possession (CP). This nominal morphology also occurs synchronically in Mono:

(2) a. soipa sa-na auau                 b. soipa e-na
    Soipa GP-3SG.POSS dog                 Soipa CP-3SG.POSS banana        
    'Soipa 's dog' (modified from Boch)   'Soipa 's banana'  (Ross 1988:250)

This paper surveys this phenomenon, then considers how a functional shift from nominal-marking to verbal-marking morphology could have occurred. It argues that the construction arose through the reverbalisation of former nominalised clauses in which an event was possessed by one of its participants.

In this hypothesis, the use of general and consumable possessive classifiers reflects a distinction in the originating nominalisations between dominant possession (possession of states or events over which the possessor has control) and subordinate possession (possession of states or events over which the possessor has no control). This distinction occurs synchronically in some Oceanic languages. In such languages, possession of nominalisations expressing controlled states and events uses general possessive morphology, while possession of uncontrolled states and events uses consumable possessive morphology, as in Fijian (Geraghty 1983):

(3) a. no-mu i-vacu   b. ke-mu  i-vacu
    GP-2SG.POSS DEVERBAL-punch           CP-2SG.POSS DEVERBAL-punch   
    ''your punch'  (which you threw)       'your punch'  (which you received)

Evidence suggests that such a distinction existed in Proto-NWS, and that the innovated construction arose when possessed nominalised adverbial clauses were reanalysed as conjoined main clauses in progressive aspect, dragging the possessive morphology with them into the verbal construction as subject-indexing. This casts light both on typological possibilities for subject agreement, and on one possible cause of diachronic functional shift in morphology.