ON THE WAY TO BOUNDEDNESS: FROM GIVENNESS TO LEXICAL ASPECTUAL CLASSES�

Andreas Schramm, Hamline University, St. Paul�Meghan Salomon, DePaul University, Chicago�Michael C. Mensink, University of Wisconsin at Stout

​

​

Text Group, UMN

March 29, 2024

​

​

Background

• Time in language is expressed through tense and aspect
• Tense expresses the temporal ordering of events or situations relative to the speaker: past, present, future
• Aspect expresses the temporal relationship between situations (or events) and their parts
• Finished: “Ann walked to the park. She sat down.”
• Ongoing: “Ann was walking to the park. She turned a corner.”

​

Background

• Situations themselves (expressed by sentences) can be described temporally in terms of their parts (lexical aspect):
• time periods (phases) and
• their endpoints (bounds)
• “seem”= no endpoint; “win”= clear endpoint
• Situations can be classified into aspectual classes based on endpoints (=degrees of boundedness): states, activities, accomplishments/achievements

​

Background

• States: “The statue seems tall” – no bounds *”for 5 minutes”
• Activities: “The heiress walked in the park.” – 2 bounds “for 2 hours”
• Accomplishments – 4 bounds:
• “The runner was winning the race.” “…and was interrupted while winning” 🡪 race is not won
• “The runner won the race.” “…and was interrupted” �🡪 race is still won

​

Function of Boundedness

• Boundedness signals narrative plot (Hopper, 1982), i.e., change of events
• Boundedness even appears central in theorizing about time in quantum physics (Rovelli, 2018)
• In fact, in physics time has been replaced by change (= boundedness of information)

​

Function of Boundedness

• Functionally, time in language thus can be viewed as providing expressions for change
• Each noun phrase as the argument of a verb comes with its own “time” (phase & bounds; Klein, 2012) and potential for change
• “Shelley passed the pickup” 🡪 accomplishment
• “Shelley passed pickups” 🡪 activity
• “Women passed pickups” 🡪 state

​

Experiment 1�Meghan Salomon-Amend

• Even on basic level of bare verbs, speakers have intuitions about boundedness and its degree
• Quantifies to what extent bare infinitives are perceived as bounded
• Participants & Setting:
• 171 (125 female) native English speakers
• attending university
• ages from 18 to 66 years old (M = 23.25, SD = 9.96)

​

Materials & Procedure (Exp. 1)

• Corpus from list of 316 most common verbs (MacMillan Essential Dictionary) was compiled into survey (Survey Monkey)
• Bare verbs were put into 4 randomized lists for boundedness ratings
• On a 7-point Likert scale, participants evaluated how “finished” situations were (1=unbounded to 7=bounded)

​

​

Results (Exp. 1)

• Participants conceived of verbs as along wide continuum of boundedness:

​

​

• Attention to boundedness indicates central role
• Rankings generally follow lexical-aspect classifications of verbs and situations into:
• states, activities, accomplishments (Comrie, 1976; Dahl, 1985; Smith, 1991)

Givenness Hierarchy

• One factor that may influence the time of a noun phrase is the cognitive status of its referent, that is, how much it is in focus:

“this pickup > that pickup > the pickup > a pickup/pickups”

• The referring expression used with the noun phrase appear to impact its amount of focus
• The cognitive status of the noun phrase, and thus its boundedness, may be influenced by the referring expression used

​

Givenness Hierarchy

• Thus, boundedness in situations may interface with elements proposed in Jeannette Gundel’s Givenness Hierarchy (1993):
• “the form of referring expessions […] depends on the assumed cognitive status of the referent, i.e., on assumptions that a cooperative speaker can reasonably make regarding the addressee’s knowledge and attention state in the particular context in which the expression is used” (p. 275)

​

​

Research Question

• Is there a continuum of boundedness in situations with referential forms from Givenness Hierarchy just like in verbs in Experiment 1? [object replaced first]

“Parts of the apartment needed work.”

• “Pat first painted a bedroom (as promised).”
• “Pat first painted this bedroom (as promised).”
• “Pat first painted the bedroom (as promised).”
• “Pat first painted that bedroom (as promised).”
• “Pat first painted THIS bedroom (as promised).”
• “It’s the same bedroom Pat painted before.”

​

​

​

Research Question

• Is there a continuum of boundedness in situations with referential forms from Givenness Hierarchy just like in verbs in Experiment 1? [subject & object replaced]

“Parts of the apartment needed work.”

• “A worker first painted a bedroom (as promised).”
• “This worker first painted this bedroom (as promised).”
• “The worker first painted the bedroom (as promised).”
• “That worker first painted that bedroom (as promised).”
• “THIS worker first painted THIS bedroom (as promised).”
• “It’s the same worker that painted the bedroom before.”

​

Predictions

• We predict a continuum based on givenness
• But aspectual classes also appear to be selected
• How do speakers get from a boundedness continuum to aspectual classes?
• Perhaps there is a pruning mechanism as found elsewhere in emergence from complexity (Morowitz, 2002)
• This pruning mechanism could be based on amount of boundedness information

Predictions

• Why may selection criteria for three classes be based on boundedness information?
• states have no bounds: N = 0 🡪 S = 0
• activities have 2 bounds: N = 2 🡪 S = 1
• accomplishments have 4 bounds: N = 4 🡪 S = 2
• In Information Science, amount of information is expressed by the Shannon number S = log₂ N [logarithm in base 2 of N]
• We suspect that boundedness continuum has three segments corresponding to S = 0, 1, 2

Discussion

• My questions to textgroup:
• Would experimental survey method adapted from Salomon-Amend (manuscript) work to investigate a potential continuum of boundedness in situations? [cf. Materials slide]
• Can this data support our claim of pruning from continuum to three classes?
• How can segments of continuum be identified statistically?
• How does selection of aspectual classes based on information quantity fit with current cognitive theories?

​

​

Materials

• Rate from 1 (incomplete) to 7 (complete) how finished or completed the ‘timeless’ event in the second sentence is:
• Imagine the following: “Parts of the apartment needed work. �- Pat first [paint] a bedroom (as promised).” preferred�- Pat first was painting a bedroom (as promised).”�- Pat first [paint] a bedroom (as promised).”prototypical
• “Pat first [paint] this bedroom (as promised).”
• “Pat first [paint] the bedroom (as promised).”
• “Pat first [paint] that bedroom (as promised).”
• “Pat first [paint] THIS bedroom (as promised).”
• “It’s the same bedroom Pat [paint] before.”

​

​

​

​

Thank you!

​

Suggestions? Questions?

​

I can be contacted at: aschramm@hamline.edu

​

Abstract

On the Way to Boundedness: From Givenness to Lexical Aspectual Classes

(with Mike Mensink)

Time is deeply embedded in language (Bohnemeyer, 2009; Comrie, 1976; Dahl, 1985; Hopper, 1982; Klein, 2018; Smith, 1991), and it has been known since Aristotle that it is expressed in two linguistic components: by the verbal auxiliaries and endings of the predicate (e.g., has been painted) and by the verb and its arguments of the expressed event (e.g., Pat (Arg 1) paint (Verb) the room (Arg 2). Its exact means of expression, however, are still under investigation (Bohnemeyer, 2009; Klein, 2018; Kuteva, et al., 2019). In this presentation we investigate the second component. Specifically, we study the temporal interface between the verb and its arguments. The meaning element under investigation is boundedness. Every event is expressed with a certain amount of boundedness. For example, Pat painted the room seems more bounded (completed) than Pat painted rooms or Workers painted rooms. We intend to look at the interaction between the Givenness of verb arguments (a/the/that/this room; Gundel, 1993) and the event’s amount of boundedness that results as well as at evidence for aspectual classes. Rather than present data in this presentation, we are looking for textgroup input at the current conceptual stage.