Smashing new results on aspectual framing
How people talk about car accidents
Teenie Matlock, David Sparks, Justin L. Matthews, Jeremy Hunter and Stephanie Huette
University of California, Merced
How do people describe events they have witnessed? What role does linguistic aspect play in this process? To provide answers to these questions, we conducted an experiment on aspectual framing. In our task, people were asked to view videotaped vehicular accidents and to describe what happened (perfective framing) or what was happening (imperfective framing). Our analyses of speech and gesture in retellings show that the form of aspect used in the question differentially influenced the way people conceptualized and described actions. Questions framed with imperfective aspect resulted in more motion verbs (e.g. driving), more reckless language (e.g. speeding), and more iconic gestures (e.g. path gesture away from the body to show travel direction) than did questions framed with perfective aspect. Our research contributes novel insights on aspect and the construal of events, and on the semantic potency of aspect in leading questions. The findings are consistent with core assumptions in cognitive linguistics, including the proposal that linguistic meaning, including grammatical meaning, is dynamic and grounded in perceptual and cognitive experience.
Imagine that you are in court. A man is being tried for reckless driving, and you saw the accident he was involved in about a month ago. After you are called to the witness stand, the judge asks whether you recognize the defendant, and you respond, "Yes". Next the judge asks whether he was the driver of a red 1970 Pontiac GTO, and you reply, "Yes". She then asks where you were when you witnessed the accident, and you report that you were standing at a bus stop. The judge tells you to do your best to remember what you saw. She asks, "What was happening?" After pausing a moment, you report that you saw the driver race out of the parking lot and into the intersection, where he nearly hit a motorcycle and an SUV. You add that he proceeded to veer off the road and smash into a bus. This description would imply erratic, dangerous driving, and would no doubt have negative consequences for the driver. Would your description of the accident have been any different if the judge had asked you, "What happened?" instead of "What was happening?" The research reported in this article suggests that it very well could have been different.
People spend a lot of time talking about events they have witnessed in the past. In doing so, they integrate lexical items in a particular way to foreground or background temporal information. For instance, in talking about a rainstorm earlier in the day, a person could provide information about the duration of the storm by using language such as, "It rained all morning," or "It rained for a few minutes." The person could specify whether the event was continuous by using language such as, "It rained non-stop," or "It rained off and on." The same individual could also designate whether the action finished prior to the time of speaking by using language such as, "It finished raining," or "It is still raining." In discussing events, people rely on grammatical aspect, verbal markers that work in concert with tense, modality, and other systems, to express information about how events unfold over time (Comrie 1976). For instance, "It was raining," suggests that the rain continued for some time, and implies that it may even continue in the future. "It rained," suggests that the rain ended.
Much is known about how aspect is marked and how it functions as a system within and across languages. However, surprisingly little is known about how aspect influences the understanding of event descriptions in everyday language. The main issue addressed in this chapter is how aspectual framing can bias the way situations are conceptualized and communicated. First, we provide a brief overview of aspect. Second, we discuss a novel experiment that investigated aspectual framing in the context of describing vehicular accidents. Third, we discuss the implications of our results for cognitive linguistic theory and for language in the courtroom.
-  We are grateful to Sarah Anderson, Eve Clark, Herb Clark, Nick Davidenko, Caitlin Fausey, George Lakoff, Michael Spivey, and Bodo Winter, for insightful discussions related to this research, and to the editors of this volume, especially Willem Hollmann. We thank research assistants Ken-Ho Yee, Jasminn Chestnut, and Cayetano Valencia, for help setting up, designing, and running the experiment. The research was partially funded by a National Science Foundation grant, IIS-0915665, Parameterization and Collection of Demonstrative Gestures for Interactive Virtual Humans.