| old_uid | 11935 |
|---|
| title | How to cause a passive state: The role of by-phrases in adjectival passives |
|---|
| start_date | 2012/12/11 |
|---|
| schedule | 14h30-17h |
|---|
| online | no |
|---|
| location_info | salle 108 |
|---|
| details | Projet « La causalité dans le langage et la cognition » de la Fédération « Typologie et Universaux Linguistiques » |
|---|
| summary | The received view for languages like English and German is that in adjectival passives the external argument of the underlying verb is not syntactically active, based on its lack of control into purpose clauses and the absence of the disjoint reference effect. In both respects, verbal passives behave differently, leading to the conclusion that the external argument is present in the syntax, even in the absence of by-phrases (cf. Baker, Johnson, and Roberts 1989; Kratzer 1994, among others). This is illustrated by German, in which a morphological difference is made between adjectival passives, which combine a past participle with sein ‘be’ ((1-a), (2-a)), and verbal passives, which appear with werden ‘become’ ((1-b), (2-b)).
(1) a. *Der Reifen war aufgepumpt, um die Fahrt fortzusetzen. the tire was inflated in order the journey to continue intended:
‘The tire was inflated in order to continue the journey.’
b. Der Reifen wurde aufgepumpt, um die Fahrt fortzusetzen. the tire became inflated in order the journey to continue
‘The tire was (being) inflated in order to continue the journey.’
(2) a. Das Kind war schlampig geka?mmt. the child was slopp(il)y combed ‘The child was combed in a sloppy manner.’
(i) Disjoint reference: Someone (else) (has) combed the child.
(ii) Reflexive: The child (has) combed him/herself.
b. DasKindwurde schlampiggeka?mmt. the child became slopp(il)y combed ‘The child has been combed in a sloppy manner.
Only disjoint reference: Someone (else) (has) combed the child.
These observations, combined with the common assumption that by-phrases syntactically and seman- tically express external arguments, lead to the prediction that by-phrases in English and German should only be possible with verbal passives. The puzzling fact, then, is that the literature reports many in- stances of German adjectival passives combining with phrases headed by von ‘by’ (e.g. (3)) (Rapp 1996; Kratzer 2000; Maienborn 2007; Gehrke 2012) (see also McIntyre to appear; Bruening to appear, for data from English).
(3) a. Die Zeichnung ist von einem Kind angefertigt. the drawing is by a child produced
‘The drawing is produced by a child.’
b. Das Haus ist von Studenten bewohnt. the house is by students inhabited ‘The house is inhabited by students.’
The topic of this paper is how to deal with by-phrases with German adjectival passives. I will discard two possible analyses, according to which such phrases modify the adjectival phrase (AP) i.e. the state underlying adjectival passives (as proposed by Meltzer-Asscher 2011, for Hebrew), or adjectival passives do not differ from verbal passives, in that both contain a Voice layer which licenses external arguments (as proposed by Bruening to appear, for English). Instead I will maintain that these con- structions are adjectival passives, which have implicit external arguments but whose by-phrases are different from those with verbal passives. In particular, I will show that by-phrases with adjectival passives are different from by-phrases with verbal passives. E.g., unlike in verbal passives (with wer- den ‘become’), the complements of by-phrases in adjectival passives (with sein ‘be’) do not introduce discourse referents, cannot control into purpose clauses and cannot be modified (4).
(4) a. Das Bild {wurde / #ist/#war} von einem Kind angefertigt. Es hatte rote Haare.
the picture became is/was by a child produced it had red hairs
b. Das Bild {wurde / *ist/*war} von einem Kind angefertigt, um seine Eltern zu
the picture became is/was by a child produced in order its parents to erfreuen.
make-glad
c. Das Bild {wurde / *ist/*war} von einem blonden Kind angefertigt.
the picture became is/was by a blond child produced
Based on the general restrictions on event-related modification with adjectival passives, I propose that German adjectival passives refer to the instantiation of a consequent state kind of an event kind, and that, as a consequence, event participants of such event kinds do not get instantiated but remain in the kind domain as well. I will argue that by-phrases with adjectival passives are acceptable only if they modify either the event kind (naming an established subkind) or the (consequent) state token (which is only possible with stative verbs).
The overall topic of this paper touches on several interface issues. First, there are well-known issues concerning the lexicon-syntax interface. It is generally assumed that the participles in adjectival pas- sives are ultimately adjectives, but there is disagreement as to whether such adjectives are formed in the lexicon (as implicitly assumed for German by Maienborn 2007, and subsequent work) or in the syntax (e.g. Kratzer 1994). In this paper, I will take a syntactic position and treat adjectival passives as true passives, which contain verbal structure associated with eventivity and which undergo adjec- tivization at some point of the derivation. Thus, even though Bruening’s (to appear) account will be argued to be incomplete, his overall conclusion that word formation in the case of adjectival passives is syntactic in nature also underlies the current proposal.
The main contribution of this paper, however, lies at the syntax-semantics interface. In particular, I will be concerned with the question what it means to have verbal structure within adjectives and what the consequences are for the nature of the underlying event. My answer to this question will be that, as a consequence of the adjectival categorization of verbal material the event associated with the verbal predicate does not get instantiated but instead remains in the kind domain. I will propose that it follows from this that event participants also remain in the kind domain and do not get instantiated. |
|---|
| responsibles | <not specified> |
|---|
| |