LINGUIST List 7.1297

Tue Sep 17 1996

Sum: Subjects of resultatives

Editor for this issue: Susan Robinson <robinsonemunix.emich.edu>

Directory

1. Joan Maling, Sum:Subjects of resultatives

It has been repeatedly observed in the literature that a resultative
phrase can only be predicated of an (underlying) object (cf. Levin &
Rappaport-Hovav 1995, Chapter 2). This is clearly true for English. In a
LINGUIST List Posting of 29 July, 1996, I noted that there are languages
like Finnish and Korean, where a resultative can also refer to the subject
of a TRANSITIVE verb assigning a theta role to an overt object. I illustrated
this with the following sentences and asked if anyone knew of other
languages with the same type of resultative construction:

Finnish (M. Vilkuna (p.c.)):

(1) Maria katseli olympialaisia silm"ans"a kipeiksi.
 Maria watched olympic-games-PAR eye-PL:ACC-3Px sore-PL-TRA
 'Maria watched the Olympic Games till her eyes ached.'

Korean (S. Kim (p.c.)):

(2) Moksanim-un chim-i malukey wuli-eykey selkyohasi-ess-ta.
 pastor-TOP tongue-NOM dry we-DAT preach-HON-PAST-DEC
 'The pastor preached [to] us [till his] tongue [became] dry.'

Although my query wasn't worded clearly enough, I was looking for examples
where the 'subject' of the resultative (i.e. the NP the resultative AP is
predicated of) co-occurs with the thematic object(s) of the matrix verb.
The responses are summarized here, grouped by language. Note the
surprising variety in resultative constructions even within Germanic.

 Joan Maling, Brandeis University
 malingvolen.brandeis.edu


KOREAN:
Jae Jung Song (University of Otago, NZ) points out that the element
_malukey_ contains the resultative subordinate marker _key_. Put
differently, the resultative construction, the sequence _chim-i
malukey_ is a subordinate clause. Note that the subject of the
resultative is marked nominative rather than accusative, and is in no
sense a [fake] object of the matrix verb.

Prof. Chungmin Lee (Seoul National University) observes that in Korean
the following ambiguity arises, even if the resultative expression
scrambles to before the object:

(3) Moksa -nim -un wuri -rul son -i pureci-key ttayri -si -ess-ta
 pastor Hon Top we Acc hand Nom break so that beat Hon Pst Dec

 1) The minister beat us till his hands broke.
 2) " " " " " our " "

Alec Marantz (MIT) pointed out that the same ambiguity has been
reported of resultatives in Chinese.

FINNISH:
In Finnish, unlike Korean, the body-part is a "fake object" at least
in the sense that it gets accusative case from the matrix predicate.
The thematic object of the matrix verb gets partitive case rather than
accusative for aspectual reasons: the resultative turns an atelic
predicate into a telic one.

GERMAN: Dr. Wolfgang Koch (University of Lund) pointed to the
existence of the following related construction in German, where an
intransitive verb takes a dative "fake reflexive" object in addition
to an accusative body-part NP:

(4) a. Er redete sich die Zunge lahm 'he talked himself the tongue lame'
 b. Er lief sich die Fuesse wund 'he run himself the feet sore'
 c. Ich starrte mir die Augen blutig 'I stared myself the eyes bloody'

This contrasts with a simple accusative "fake object" resultative:

(5) a. Er ass den Teller leer 'he ate the plate empty'
 b. Er lief das Pflaster duenn 'He ran the pavement thin'

which doesn't license a dative reflexive since "fake object" is not a
part of the subject's body. Interestingly, the body part can be
omitted in certain cases without altering meaning much, in which case
the reflexive pronoun changes from dative to accusative:

(6) a. Ich lief mich wund.
		(it's still the feet that are most likely to be sore)
 b. Er redete sich lahm.
		(implies that body parts other than the tongue got 'lame').


DUTCH:
Jan Odijk (Institute for Perception Research, Eindhoven) points out
the existence of at least one (idiomatic) example of this kind in
Dutch, where the resultative construction (consisting of the
resultative AP and its 'subject') co-occurs with a PP argument of
the matrix verb:

(7) a. Hij beet zijn tanden stuk op dat probleem
 He bit his teeth broken on that problem
 `He tried hard to solve that problem, but failed'

 b. Hij beet zijn tanden stuk op die noot
 He bit his teeth broken on that nut
 `He bit on that nut and by doing so his teeth broke'


NORWEGIAN:
Prof. Helge Lodrup (University of Oslo) reports that the
generalization about the underlying object is problematic also for
unaccusatives. The standard assumption in the literature on English
is that unaccusatives allow subject-oriented resultatives precisely
because their surface subject IS an underlying object. But under the
standard Case-theoretic account, unaccusatives should disallow a
reflexive object or any other kind of "fake" object, because there is
no source of Case. However, just like unergative intransitives,
Norwegian unaccusatives require a reflexive object to take a resultative:

(8) a. Kjoettet har hengt seg moert 'the meat has hung itself tender'
 b. Barna froes seg gule og blaa 'the kids froze themselves yellow and blue'
 c. Doera har staatt seg skjev 'the door has stood itself lopsided'
 d. Hoenene har kokt seg moere 'the hens have boiled themselves tender'

These verbs do not allow a reflexive when they appear without a resultative.
These sentences seem to have the same analysis as a corresponding
sentence with an unergative, like:

(9) Han loep seg svett 'He ran himself sweaty'

Sentences like 'The bottle broke open, The river froze solid, The gate
swung shut' are impossible in Norwegian. The generalization seems to
be that only atelic unaccusatives take adjectival resultatives in
Norwegian. (This means that both with unaccusatives and unergatives
the resultative turns a non-telic predication into a telic one.)


REFERENCES:
Last but not least, a number of respondents provided references:

Kim, Jong-Bok, 1993: Syntax and Semantics of Korean Resultative
	Constructions, in Kuno et al.(eds) Harvard Studies in Korean
	Linguistics V, pp. 471-482 (1993).
Koch, Wolfgang & Inger Rosengren, 1995: Secondary Predications: Their
	Grammatical and Conceptual Structure. Sprache und Pragmatik 35,
	Lund, pp. 1-100
Koch, Wolfgang & Inger Rosengren, 1996: 'Locative Alternations' in
	English and German: Different Lexicalisations of the Same
	Conceptual Structure. Sprache und Pragmatik 43, pp.1-79.
Lodrup, Helge, 1996: Underspecification in Lexical Mapping Theory:
	The Case of Norwegian Existentials and Resultatives, paper to
	be presented at the LFG Workshop in Grenoble, August 27.

************************************************
- 
Joan Maling
Linguistics & Cognitive Science
Volen Center, MS-013				tel: 617-736-3261
Brandeis University				fax: 617-736-2398
Waltham, MA 02254-9110				malingvolen.brandeis.edu