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
|
4.2.2
Backgrounding antipassives Parallel to backgrounding passives, backgrounding antipassives represent the core function of antipassives; block- ing the linking of the [ −a] argument to [−oblique] status, without the necessary side effect of shifting pivot properties to the [ +a] argument diagnostic of fore- grounding antipassives. Consequently, backgrounding antipassives are much more common than foregrounding ones, being found in languages of more diverse typologies than foregrounding antipassives; in particular, they are quite wide-spread in pivotless languages. Because the [ −a] argument in a back- grounding antipassive construction, if present at all, is necessarily [ +oblique], the clause is formally intransitive. If the language distinguishes morphologi- cally transitive from intransitive verbs, the verb of the clause will be intransitive, and, further, the sole [ −oblique] argument of the derived antipassivized verb, the [ +a] np, will exhibit whatever grammatical properties are proper to the sole [ −oblique] argument of intransitive verbs. Unlike backgrounding passives in languages like Marathi, Ulcha, or Dutch, which exhibit transitive proper- ties, e.g. the [ −a] argument remains in accusative case, this is not possible for backgrounding antipassives – they are necessarily formally intransitive: the ergatively case-marked [ +a] np of a transitive verb must, in the correspond- ing antipassive, assume the case proper to the sole [ −oblique] argument of an intransitive verb. It would seem that the notion of transitivity is defined cross- linguistically in terms of the presence of the [ −oblique] [−a] argument. Because backgrounding passives block the linking of the [ +a] argument to [−oblique], they can leave the [ −a] argument unaffected, so the derived passive clause can remain formally transitive. This is not possible for antipassives: blocking the 434 William A. Foley linking of the [ −a] argument to [−oblique] status by definition makes the clause intransitive. The most prototypical and wide-spread use of the backgrounding antipassive is for the complete suppression of the [ −a] argument. Not only must it not be [ −oblique], it cannot be overtly mentioned at all: (166) Bandjalang of Australia (a) mala-yu d a·dam-bu mala bulan d a-ila dem -erg child-erg dem .abs meat.abs eat-pres [ +a] [ −a] ‘That child is eating meat’ (b) mala d adam d a-le-ila dem .abs child.abs eat-antipass-pres [ +a] ‘That child is eating’ Crowley (1978) The (a) example is fully transitive with two [ −oblique] nps, the [+a] and [−a] arguments. The (b) example is the corresponding backgrounding antipassive. The clause is formally intransitive, and the [ −a] argument is fully suppressed. In other cases, the now [ +oblique] [−a] argument of an antipassive con- struction need not be fully suppressed. Its interpretation is, however, affected by its change in status and there are at least two ways in which this manifests itself. The notion of being totally affected by an action or undergoing a change in state is prototypically associated with [ −a] arguments. In many languages, the effect of backgrounding antipassives is to strip this meaning from the [ −a] argument; when [ +oblique] in an antipassive construction, the [−a] argument is interpreted as partially affected in contrast to its normal interpretation in a full transitive clause: (167) Kabardian (a) e-m q’ w pˆs re-r jedzaq’e dog-erg bone-abs bite | | [ +a] [ −a] ‘The dog bites the bone [through to the marrow]’ (b) e-r q’ w pˆs re-m je-w-dsaq’e dog-abs bone-instr [antipass]-bite | | [ +a] [ −a] | | [ −oblique] [+oblique] ‘The dog gnaws at the bone’ Catford (1976) A typology of information packaging 435 (168) Chamorro (a) un-patek i ga’lagu 2sg.erg-kick det dog.abs | | [ +a] [ −a] ‘You kicked the dog’ (b) mam-(p)atek hao gi ga’lagu antipass -kick 2sg.abs [ +oblique] dog | | [ +a] [ −a] | | [ −oblique] [ +oblique] ‘You kicked at the dog’ Cooreman (1988) In each language, the [ −oblique] [−a] np in the transitive (a) constructions is fully affected (the bone is bitten through and the dog is actually kicked), but in the antipassive (b) examples this does not hold (the bone is only being gnawed at the surface and the dog may have escaped the kick). Note that though the English translations show the same type of syntactic alternations and corresponding meaning shifts, they do not illustrate true antipassives, as there is no requisite derivational morphology for antipassivization in the verb or verbal complex. In other cases, the effects of the backgrounding antipassive may be more pragmatic than semantic: most commonly, the now [ +oblique] [−a] argument loses its ability to be referential or definite: (169) Chukchee of Siberia (a) ʔ aaˇcek-a kimit ʔ - ə n ne-nl ʔ etet- ə n youth-erg load-abs 3pl.erg-carry-3sg.abs.aor ‘The young men carried away the load’ (b) ʔ aaˇcek- ə t ine-nl ʔ etet-g ʔ et kimit ʔ -e youth-abs.pl. antipass-carry-3pl.abs.aor load-instr ‘The youths carried away a load’ Kozinsky, Nedjalkov, and Polinskaja (1988) Some languages, notably Mayan languages like Mam (England 1983b), allow nonreferential [ −a] arguments in backgrounding antipassive constructions to be incorporated into the verbal complex: |
