ANTIPASSIVE, ANTICAUSATIVE and REFLEXIVE

NELS 35

University of Connecticut

October 22, 2004

ANTIPASSIVE, ANTICAUSATIVE and REFLEXIVE:

PHASES AND REFLEXIVE MORPHOLOGY

David Basilico

Department of English

University of Alabama at Birmingham

OBSERVATION

MANY LANGUAGES USE IDENTICAL MORPHOLOGY ON THE VERB IN THE ANTIPASSIVE, ANTICAUSATIVE AND REFLEXIVE CONSTRUCTIONS

DATA

(A) Warrungu (Tsunoda, 1988)

ANTIPASSIVE

a. kaya+0 kipa+kali+0

father+abs shave+refl+p/p

Father is shaving someone.

REFLEXIVE

b. kaya+0 kipa+kali+0

father+abs s have+refl+p/p

Father is shaving himself.

ANTICAUSATIVE

c. yuri+0 watyu+kali+n

kangaroo+abs cook+refl+p/p

The kangaroo is cooked.

DATA

(B) Lituanian (Geniušienė, 1987)

ANTIPASSIVE

a. Šu-o kandžio-j-o-si

dog-nom bite-past.3.P.-rm

The dog bit (was in the habit of biting).

REFLEXIVE

b. Jon-as ap-si-rengė

Jonas-nom pref-refl-dressed

Jonas dressed himself.

c. Žied-ai skleidžia-si

flower-nom.pl blossom-out-rm

The flowers were blossoming out.

DATA

ANTIPASSIVE

a. Juan se confiesa

John refl confesses

John confesses.

REFLEXIVE

b. Juan se ama

John refl love

John loves himself.

ANTICAUSATIVE

c. El vaso se rompio

the vase refl broke

The vase broke.

PROPOSAL

A Phase-Based Analysis

What unifies all these constructions is that the argument position of the VP remains unsaturated at the point in the derivation when the strong phase (Chomsky 2001) induced by the light verb v (Chomsky 1995, Kratzer 1996) sends off the VP for LF interpretation. Without RM, the VP contains an unsaturated argument position and the derivation runs afoul of the following condition.

(1) All individual argument positions must be saturated at LF.

Reflexive morphology is present because it existentially closes (EC) the argument position of the verb, saturating the argument position of the verb within the VP and allowing the structure to comply with (1).

ANALYSIS

ANTIPASSIVE

Without special morphology

NP v’

g 3

Juan v VP è sent off for LF interpretation: lxle[confess(x)(e)]

confiesa

This runs afoul of condition (1), since the individual argument position of the verb is not saturated.

ANTIPASSIVE

With Special Morphology

NP v’

g 3

Juan v ClP è sent off for LF interpretation: le$x[confess(x)(e)]

Cl VP

g g

se V

confiesa

Because the individual argument position associated with the verb is saturated by existential closure, this structure is in compliance with (1).

REFLEXIVE

Like the antipassive, there is no NP within VP; the NP that comes to saturate the individual argument position of the verb is located outside of VP, the external argument. Consequently, if the derivation adds the light verb v, the VP will be sent off for LF interpretation. Without the special morphology, there will be an unsaturated argument position. The clitic steps in to existentially close this argument position.

NP v’

g 3

Juan v ClP è sent off for LF interpretation: le$x[loves(x)(e)]

Cl VP

g g

se V

ama

PROBLEM: How can the external argument come to saturate the individual argument position introduced by the verb if it has already been saturated by existential closure?

SOLUTION: Two processes at work (1) existential disclosure and (2) argument identification. These processes are associated with the external argument introducing v head.

Existential Disclosure (DIS) (Dekker 1993, Rivero and Sheppard 2003): This removes the existential quantifier.

Argument Identification (AI): A generalization of Kratzer’s (1996) Event Identification.

In Kratzer’s approach, the light verb v is associated with a theta role predicate that introduces the external argument and an event argument. With Event Identification, the event argument introduced by the verb is identified as the same event argument introduced by the theta role predicate of vP. With Argument Identification, not only are the event arguments identified as the same, but also the individual arguments. Once Argument Identification occurs, the external argument in the specifier of vP can saturate the argument position of both the theta role predicate and the verb by functional application (FA).

NP v’

g 3

Juan v ClP

g 3

EXP Cl VP

g g

se V

ama

Derivation:

a. [VP ama ] lxle[love(x)(e)]

b. [ClP se [VP ama ] ] ] ] le$x[love(x)(e)]

c. [vP v [ClP se [VP ama ] ] ] ] i. DIS(le$x[love(x)(e)])à lx le[love(x)(e)]

ii. AI(lyle[experiencer(y)(e)], lxle[love(x)(e)])à

lyle([experiencer(y)(e) & love(y)(e)] )

d. [vP Juan [v [ClP se [VP ama ] ] ] ] FA(lyle([experiencer(y)(e)&love(y)(e)]))(Juan)à

le([experiencer(Juan)(e) & love(Juan)(e)]

ANTICAUSATIVE

Borrowing from Alexiadou and Anagnostopoulou (2004), I consider that the anticausative structure involves a functional projection that combines with and stativizes a vP that carries no agentivity features, generating its target state (Kratzer 2000).

However, I depart from Alexiadou and Anagnostopoulou (2004) in two ways. First, I consider that the NP that saturates the individual argument position of the verb is introduced not within the VP but within the stativizing functional projection. Because there is no NP within VP, special morphology is needed to existentially close the argument position introduced by the verb. Later, the v head can disclose this argument position. Second, this stativizing functional projection combines with another light verb v that introduces an inchoative operator ([fient], following Embick (2004)).

v FP

g 3

fient NP F’

g 3

el vaso F vP

g 3

result v ClP

Cl RootP

g g

se Root

rompio

Derivation

a. [RoootP rompio ] lxls[broken(x)(s)]

b. [ClP se [RootP rompio ] ] ] ] ls$x[broken(x)(s)]

c. [v [ClP se [RootP rompio ] ] ] ] i. DIS(ls$x[broken(x)(e)])à lx le[broken(x)(e)]

ii. lxlsle[cause(s)(e) & broken(x)(s)]

d. [FP result [v [ClP se [VP rompio ] ] ] ] lxls$e[cause(s)(e) & broken(x)(s)]

e. [FP el vaso [F’ result [vP v [ClP se [VP rompio ] ] ] ] ]

FA(lxls$e[cause(s)(e) & broken(x)(s)])(el vaso)à

ls$e[cause(s)(e) & broken(el vaso)(s)]

f. [vP fient [FP el vaso [F’ result [vP v [ClP se [VP rompio ] ] ] ] ] ]

lei (fient(ei), ($s$e[cause(s)(e) & broken(el vaso)(s)])

CONCLUSION

This analysis supports a phase-based view of derivations and the cyclic interpretation for the structures built. If the entire structure were sent for interpretation, there would be no unsaturated argument position and no need for special morphology. This analysis is able to capture the essential insight of Lidz (1996) without the need for a separate level of argument structure.

Bibliography

Alexiadou, Artemis and Elena Anagnostopoulou. 2004. Voice Morphology in the Causative-Inchoative Alternation: Evidence for a Non-Unified Structural Analysis of Unaccusatives. In Artemis Alexiadou, Elena Anangnostopoulou and Martin Everaert, The Unaccusativity Puzzle, 114-136. Oxford, Oxford University Press.

Chomsky, Noam. 1995. The Minimalist Program. MIT Press, Cambridge, MA.

Chomsky, Noam. 2001. Derivation by Phase. Ms. MIT.

Dekker, Paul. 1993. Existential Disclosure. Linguistics and Philosophy 16, 561-588.

Embick, David. 2004. On the structure of resultative participles in English. Linguistic Inquiry 35.3, 355-392.

Geniušienė, Emma. 1987. The Typology of Reflexives. Berlin: Mouton de Gruyter.

Kratzer, Angelika. 1996. Severing the External Argument from its Verb. In Phrase Structure and the Lexicon, eds. J. Rooryk and L. Zaring. Kluwer, Dordrecht, 109-137.

Krazter, Angelika. 2000. Building Statives. Berkeley Linguistic Society 26.

Lidz, Jeffrey. 1996. Dimensions of Reflexivity. Ph.D. Dissertation, University of Delaware.

Mascullo, Pascual Jose. 1992. Antipassive Construction in Spanish. In Paul Hirschbüler and Konrad Loerner Romance, Language and Modern Linguistic Theory: Papers from the 20th Syposium on Romance Languages. Amsterdam: John Benjamins Publishing Company, 175-194.

Rivero, Maria Luisa and Milena Sheppard. 2003. Indefinite Reflexive Clitics in Slavic: Polish and Slovenian. Natural Language and Linguistic Theory 21.1, 89-155.

Tsunoda, Tasaku. 1988. Antipassives in Warrungu and other Australian languages. In Passive and Voice, ed. Masayoshi Shibatani. John Benjamins, Amsterdam. 595-650.