pimi- Third person agreement differs from that of first and second persons in two ways: (i) there are no overt nominative or accusative agreement markers 5 and (ii) there is no singular/plural distinction in dative and benefactive.
The agreement markers have been referred to by various names in the literature. The nominative, accusative, and dative markers have been called, respectively, "subject-personal", "objective-personal", and "objective-possessive inseparable pronouns" (Byington 1870); "active", "passive", and "bound dative personal affixes" (Nicklas 1974); "agentive", "patientive", and "dative pronominal affixes" (Heath 1977); "actor", "patient", and "bound dative echoes" (Jacob, Nicklas, and Spencer 1977; Nicklas 1979); and "I", "II", and "III" affixes" (Munro and Gordon 1982, Munro 1984).6 As will become evident in the course of this study, the terminology adopted here is designed to facilitate cross-linguistic comparison. The agreement markers occur as prefixes in the verbal complex of the clause, with the exception of the first person singular nominative, -Ii, which occurs as a suffix. Benefactive agreement markers are not bound morphologically to. the following morpheme, although they are bound to the preceding nominative agreement marker if there is one (ct. (4a) below).
With many verbs agreement is straightforward: nominative agreement marks subjects, accusative agreement marks direct objects, dative agreement marks indirect objects, and benefactive agreement marks beneficiaries. This is illustrated in ( 2)-( 4).
(2) a. Chi-bashli -Ii -tok. 2ACC cut lNOM PST 1cut you. b. Ano is-sa-hottopali -tok. 1 2NOM lACC hurt PST You hurt me. b. Oka sa-banna -h. water lACC want PRED I want water. The object of the verb alikchi 'doctor' can trigger accusative (8a) or dative (8b) agreement. CHAPTER 1 b. Hattak -at holisso -t im-ihaksi -tok. man NOM book NOM 3DAT forget PST
The man forgot the book.
(13) a. Alia -t chim-iskali 1-kania -tok. child NOM 2POSS money 3DAT lose PST
The child lost your money.
-yat
The child lost your money.
In Chapter 5 these double nominative case marking facts are shown to follow from the Inversion structure, a structure in which each nominal bears the subject relation at a different level of structure.
The data in ( 12) and ( 13) point to an important aspect of Choctaw morphosyntax: the agreement system and the case marking system function independently. A nominal which takes nominative case marking need not determine nominative agreement. In (12), hattak 'man' is marked for nominative case yet determines a dative agreement marker'i. The same is true of alia 'child' in (13). As illustrated in (5b), repeated below, and as is shown at some length in Chapter 2, certain nominals that must be marked for nominative case determine accusative agreement markers.
(5) b. Anakosh sa-yimmi -h. l=FO=NOM lACC believe PRED I am the one who believes (it).
The fact that agreement and case pattern independently is illustrated at various points in the study and is discussed in detail in Chapter 8.
## Switch-reference Marking
Choctaw has a system of morphological marking which indicates whether or not pairs of clauses (linked by conjunction or embedding) have coreferent subjects. Following Jacobsen (1967) and later discussions of similar phenomena, we refer to such morphological marking as switch-reference marking.!! Same-subject (SS) markers or different-subject (DS) markers are suffixed to the predicate of the I For discussion of the classification of Choctaw in the Muskogean language family, see Haas (1941Haas ( , 1946Haas ( , 1973)).
2 Throughout this study I use the terms subject, direct object, and indirect object as pretheoretical labels to identify the arguments of a clause. The theoretical terminology to be used is introduced in Chapter 2.
3 Ulrich (1984) identifies four additional types of verbal agreement morphology marking oblique relations. These markers are not considered in the present study. 4 The allomorphs in the paradigms in (1) are phonologically conditioned. Cf. Nicklas (1974, 1975) for details. 5 Although it is customary in many discussions of Amerindian languages to denote absence of overt agreement morphology by 0, in the transcription of the Choctaw data I follow the tradition of the Choctaw literature and give no overt marking of third person nominative and accusative agreement. Predicates exhibiting no overt agreement morphology should be interpreted as having third person arguments unless explicitly noted otherwise. 6 Many of the cited studies make no mention of the series of benefactive agreement markers. Nicklas (1974, 1979) and Jacob, Nicklas, and Spencer (1977) refer to the benefactive series as the unbound dative. Ulrich (1984) uses the term benefactive but contends that they are morphologically bound to following material in the verbal complex. 7 The morpheme 0 in (3b) is glossed as CONTR, which stands for 'contrastive'. This construction, in which some nominal takes the contrastive morpheme, is referred to as the 'emphatic' construction by Nicklas (1974) and others because it places emphasis on the nominal. I mark this emphasis in the translation, by placing the emphatic nominal in italics. 8 Although there are no third person pronouns such as ano '1', pishno 'we',
Same-subject Marking as a Test for lHood
更多信息……