(Re)construing Meaning in NLP

Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 5170–5184 July 5 - 10, 2020. c⃝2020 Association for Computational Linguistics 5170 (Re)construing Meaning in NLP Sean Trott University of California, San Diego sttrott@ucsd.edu Tiago Timponi Torrent Federal University of Juiz de Fora tiago.torrent@ufjf.edu.br Nancy Chang Google ncchang@google.com Nathan Schneider Georgetown University nathan.schneider@georgetown.edu Abstract Human speakers have an extensive toolkit of ways to express themselves. In this paper, we engage with an idea largely absent from dis- cussions of meaning in natural language un- derstanding—namely, that the way something is expressed reflects different ways of concep- tualizing or construing the information being conveyed. We first define this phenomenon more precisely, drawing on considerable prior work in theoretical cognitive semantics and psycholinguistics. We then survey some di- mensions of construed meaning and show how insights from construal could inform theoreti- cal and practical work in NLP. 1 Introduction Natural language is a versatile tool for allowing humans to express all manner of communicative intents, from simple descriptions of the entities and situations in their direct experience to elaborate rhetorical flights of fancy. Many NLP applications, such as information extraction, question answer- ing, summarization, and dialogue systems, have restricted their scope to what one might call objec- tive information content—relatively uncontrover- sial facts that systems can infer from an utterance, store in a database and reason about. While it is tempting to equate such information with the meaning of an utterance, a large body of lit- erature in linguistics and psycholinguistics argues that an utterance conveys much more than a simple set of facts: it carries with it a halo of intimations arising from the speaker’s choices, including con- siderations of perspective, emphasis, and framing. That is, linguistic choices subtly color meaning; far from merely conveying objective facts, they reflect how speakers conceptualize meaning and affect listeners’ interpretations in predictable ways. Take, for example, this metaphor-rich portrayal of a newborn as a tyrant over her parental subjects: (1) Nora’s arrival brought a regime change. Life under her adorable tyranny was filled with squawking, swaddling and ceaseless sleep- input-output cycles. We were relieved when she relaxed her tiny iron grip. This report of new parenthood describes a major life change along with everyday caregiver routines, but its emphasis is on the parents’ experience of being suppressed (under) and controlled (grip) by a creature who is cast, variously, as a tyrant (regime), a bird (squawk), and a relentless machine (sleep- input-output cycles, iron grip)—albeit a (subjec- tively) adorable one. The power of linguistic choices to shape under- standing is also evident in more mundane (and well- studied) examples: (2) a. Chuck bought a car from Jerry. Jerry sold a car to Chuck. Jerry paid Chuck for the car. b. I work at Microsoft. I work for Microsoft. c. The statue stands in the plaza. The statue is standing in the plaza. Each set includes sentences that convey roughly the same facts—i.e. they could describe the same scenario—but nonetheless differ in various …

Content truncated to 500 words