Language Typology and Syntactic Description, Volume I: Clause Structure, Second edition
Download 1.59 Mb. Pdf ko'rish
|
Lgg Typology, Synt Description v. I - Clause structure
|
4.1.2
Backgrounding passives The function of backgrounding passives is in a sense the core function of passive: to remove the [ +a] argument from prominence in the clause. The degree of downgrading is a parameter that varies 424 William A. Foley quite widely from language to language. We have already seen one language, Marathi, in examples (136–7) in which the [ +a] argument, although realized as a [ +oblique] np, as it must be in a passive construction, still controls a number of syntactic properties due to its being the most prominent argument in the argument structure. Other languages have constructions like the Marathi pas- sive, but in which the now [ +oblique] [+a] argument does cede these proper- ties, and in fact is completely suppressed from the clause; for example Ulcha, a Tungusic language of Siberia: (149) ti d¯use-we h¯on-da ta-wuri dem tiger-acc how-ptcl do-pass ‘What’s to be done about that tiger?’ Nichols (1979) The verb ta- ‘do’ is lexically derived as passive and intransitive via the suffix -wuri. As a passive construction, the [ +a] argument is blocked from linking to [ −oblique]. In fact Ulcha, like many languages, applies a stronger restriction: not only must the [ +a] argument not be [−oblique], it must be removed from the clause entirely, obviously prohibiting any control and binding properties such as are found in Marathi passives. Note, further, that, unlike in foreground- ing passives, the [ −a] remains in accusative case, unable to take on even the nominative case now vacated by the suppressed [ +a] argument. In other languages with backgrounding passives, the [ −a] argument does assume the case marking or other morphological properties of the suppressed or now [ +oblique] [+a] argument. This is because these are tied to the properties of argument structure: the [ −a] argument, now the only [−oblique] argument in the argument structure of the now derived intransitive verb, is treated iden- tically to the sole [ −oblique] argument of an underived intransitive verb and will display whatever morphological properties are proper to that. Chickasaw (150) (Munro and Gordon (1982)) and Fijian (151) illustrate this pattern (in Chickasaw, passive is signalled by an infix -hl-): (150) (a) hattak-at -malili V INTR man-nom 3sg[ +a]-run ‘The man runs’ (b) hattak-at ahi’ --atahli V TRANS man-nom potato 3sg[ +a]-3sg[−a]-prepare ‘The man prepares potatoes’ (c) ahi’-at -a<hl>taha V PASS potato-nom 3sg[ −a]-prepare <pass> ‘The potatoes are prepared’ A typology of information packaging 425 (151) ↓ ↓ (a) e lako na gone V INTR 3sg go art child ‘The child goes’ ↓ ↓ (b) au kau-t-a na gone V TRANS 1sg[ +a] carry-trans-3sg[−a] art child[−a] ‘I carried the child’ ↓ ↓ (c) e kau-ti na gone V PASS 3sg carry-pass art child ‘The child was carried’ Note that, in both Chickasaw and Fijian, the [ −oblique] [−a] argument of a passive construction (the (c) examples) has the same grammatical properties as the sole [ −oblique] argument of an underived intransitive verb (the (a) exam- ples) and is different from when it functions as the [ −a] argument of a transitive verb (the (b) examples). In Chickasaw, the [ −a] argument fails to receive the -(a)t nominative case marker for the most prominent np, if it is an argument of a transitive verb (150b); while as the sole [ −oblique] argument of both the derived intransitive passive verb (150c) and an underived unergative intransitive verb (150a), it does so, for it is now the most prominent argument, there being no other [ −oblique] nps. In Fijian, the [−a] argument of a transitive verb is realized with suffixed pronominals, -a 3sg [ −a] in (151b). However, the sole [ −oblique] argument of underived intransitive verbs co-occurs with proclitic pronominals (151a). Note that the sole [ −oblique] [−a] argument of a derived passive clause in Fijian has the same proclitic pronominals (151c). The evi- dence suggests that languages like Chickasaw and Fijian go much further than Marathi in the backgrounding of the [ +a] argument: its [+oblique] status or complete suppression forces it to cede its prominent position in argument struc- ture to the only [ −oblique] np, the [−a] argument. This is a language-specific side effect of passive in such languages, not a universal constraint, as languages like Marathi demonstrate. These passive constructions in Chickasaw and Fijian are clearly backgrounding passives because both languages are pivotless, and, further, the main function of the construction is to background or suppress the [ +a], not topicalize or foreground the [−a]. Some European languages, notably the asymmetrical nominative–accusative pivot languages of the Germanic family, have an interesting intermediate passive construction reminiscent of Ulcha, in which the [ −a] of the passive construction |
