Lexical telicity?

Processing evidence for and against verbal telicity

1 2
E. Matthew Husband and Linnaea Stockall
1
Department of Linguistic and Cognitive Science, Brown University
2
School of Cognitive Science, Hampshire College
Introduction & Background

Main Questions Linguistic Theory Lexical Semantics of Verbs

I What role do lexical items (in particular, verbs)
play in composing an event interpretation?
I What are the processing correlates of event
interpretation?
I Telicity is determined by lexical semantic properties of the verb and properties
of its direct object (Dowty 1979, Verkuyl 1972).
I Lexically bounded (achievement) and unbounded (activity) verbs trigger telic
and atelic interpretations respectively (1,2).
I Lexically unspecified (activity/accomplishment) verbs alternate between telic
I Linguistic theory supports the existence of telic verbs (Borer 2005; Mittwoch
1991).
(4) a. Linnaea struck oil on Monday and on Tuesday. (two events only)
b. Matt pumped oil on Monday and on Tuesday. (one event possible)
I Status of atelic verbs is contentious (Borer 2005; Kiparsky 1998; Schein

Current Study
I Can we use the processing correlates of event
interpretation to inform our understanding of
the linguistic primitives used to interpret
and atelic interpretations based on their direct object (3). 2002).
I See Stockall, Husband, & Benatar on processing the mass/count
(5) a. Linnaea pushed a button for 10 minutes. (iterative, telic)
distinction for a discussion of ±bounded nominals (poster #2.35). b. Matt read a word for 10 minutes. (iterative, telic)
I Telic verbs are distinct from unspecified verbs, but atelic verbs are not.

events? Verb Direct Object Event

(1) a. unspecified −bounded atelic inspected gardens # in an hour Psycholinguistic Theory
Background b. unspecified +bounded telic inspected the gardens in an hour I Telic verbs and unspecified verbs with +bounded direct objects incur
I Telicity, a component of event interpretation (2) a. telic −bounded telic completed gardens in an hour processing costs under the scope of a temporal modifier (Piñango et al.
(a.k.a. inner aspect, lexical aspect, situation b. telic +bounded telic completed the gardens in an hour 1999; (Todorova et al. 2000; Brennan & Pylkkänen 2008, in press).
aspect, or aktionsart), has been studied in (3) a. atelic −bounded atelic roamed gardens # in an hour I Recovery from reduced relative clause garden-paths is eased by telic verbs,

both linguistic theory and psycholinguistics. b. atelic +bounded atelic roamed the gardens # in an hour but not by atelic or unspecified verbs (O’Bryan 2003).

Experiment 1: Telic Verbs

Methods
I Verbs which yielded telic events with both bare and definite objects using Dowty’s (1979) tests were classified as telic verbs.
I 60 subjects, 2 x 2 telic vs. unspecified verb by bare vs. definite determiner (48 items, 192 stimuli total, 122 fillers).

I Subjects made an acceptability judgement after completing each sentence.

Verb Type Determiner Type Event Sentence

telic bare telic The expert physicist lost files on the formation of black holes.
telic definite telic The expert physicist lost the files on the formation of black holes.
unspecified bare atelic The expert physicist read files on the formation of black holes.
unspecified definite telic The expert physicist read the files on the formation of black holes.

Results
I Acceptability Rating: Significant main effect of Verb Type (F1(1,59)=20.858, p<.001; F2(1,47)=14.590, p<.001)
I Reading Times
I Significant main effect of Determiner Type on the noun position (F1(1,59)=19.167, p<.001; F2(1,47)=13.051, p=.001) and the

noun+1 position (F1(1,59)=10.262, p=.002; F2(1,47)=16.648, p<.001). Bare nouns > Definite nouns.
I Significant interaction of Verb Type and Determiner Type on the noun+1 position (F1(1,59)=4.579, p=.037; F2(1,47)=3.876,

p=.055). Effect of telicity: Atelic event (Unspecified, Bare) > Telic event (Telic, Bare & other 2 conditions).

Experiment 2: Atelic Verbs

Methods
I Verbs which yielded atelic events with both bare and definite objects using Dowty’s (1979) tests were classified as atelic verbs.
I 60 subjects, 2 x 2 atelic vs. unspecified verb by bare vs. definite determiner (36 items, 144 stimuli total, 164 fillers).

I Subjects made an acceptability judgement after completing each sentence.

Verb Type Determiner Type Event Sentence

atelic bare atelic The careful engineer displayed equipment for the meeting with his supervisors.
atelic definite atelic The careful engineer displayed the equipment for the meeting with his supervisors.
unspecified bare atelic The careful engineer assembled equipment for the meeting with his supervisors.
unspecified definite telic The careful engineer assembled the equipment for the meeting with his supervisors.

Results
I Acceptability Rating: Trend towards a main effect of Verb Type by subject (F1(1,59)=3.641, p=.061; F2(1,35)=1.899, p=.177)
I Reading Times
I Significant main effect of Determiner Type on the noun position (F1(1,59)=9,194, p=.004; F2(1,35)=16.718, p<.001) and the

noun+1 position (F1(1,59)=4.122, p=.047; F2(1,35)=10.382, p=.003). Bare nouns > Definite nouns.
I No significant interaction of Verb Type and Determiner Type on the noun+1 position (F1(1,59)=.330, p=.568; F2(1,35)=.245,

p=.624). No effect of atelicity: Atelic event (Atelic, Definite) = Telic event (Unspecified, Definite).

Summary Conclusions

Experiment 1: Telic Verbs Psycholinguistics

I Lexically telic verbs are distinguished from unspecified verbs in processing measures. I The parser uses verbal telicity and direct object boundedness to commit to an event interpretation.
I Evidence that the parser uses VP internal factors to license event interpretation online. I Supports the existence of telic verbs and fails to support the existence of atelic verbs.

I Supports current linguistic theory in recognizing the existence of telic verbs (Borer 2005; Mittwoch I The processing profile of telic verbs is distinct from unspecified verbs.

1991). I Atelic verbs have the same processing profile as unspecified verbs.

Experiment 2: Atelic Verbs Linguistic Theory

I Lexically atelic verbs are not distinguished from unspecified verbs in processing measures. I Only two verb classes: Lexical telicity may be represented by a privative (single-valued) feature: [telic].
I Also supports current linguistic theory refuting the existence of atelic verbs (Borer 2005; Kiparsky 1998; Telic verbs: {discover, complete, loose, strike, . . . }
Schein 2002). Unspecified verbs: {assemble, display, inspect, investigate, pump, push, read, roam, . . . }

References

Borer, H. (2005). Normal Course of Events. Oxford University Press. Brennan, J. & Pylkkänen, L. (2008). Processing events: Behavioral and neuromagnetic correlates of aspectual coercion. Brain and Language 106, 132–143. Brennan, J. & Pylkkänen, L. (in press). Processing psych verbs:
Behavioral and MEG measures of two different types of semantic complexity. Language and Cognitive Processes. Dowty, D. (1979). Word Meaning and Montague Grammar: The Semantics of Verbs and Times in Generative Semantics and in Montague’s PTQ. Springer. Kiparsky, P. (1998). Partitive
case and aspect. The Projection of Arguments: Lexical and Compositional Factors, CSLI, Stanford, CA 265–307. Mittwoch, A. (1991). In defense of Vendler’s achievements. Belgian Journal of Linguistics 6, 71–84. Piñango, M., Zurif, E., & Jackendoff, R. (1999). Real-time processing implications of
enriched composition at the syntax–semantics interface. Journal of Psycholinguistic Research 28, 395–414. Schein, B. (2002). Events and the semantic content of thematic relations. Logical Form, Language and Semantic Content, 263–344. Todorova, M., Straub, K., Badecker, W., & Frank, R.
(2000). Aspectual coercion and the online computation of sentential aspect. Proceedings of the 22nd Annual Conference of the Cognitive Science Society, 3–8. Verkuyl, H. (1972). The Compositional Nature of the Aspects. Dordrecht: Reidel.

http://eyelab.msu.edu/people/matt matthew husband@brown.edu lstockall@hampshire.edu