Pragmatics and Natural Language Understanding

Georgia M. Green

• Publisher: Routledge
• Year: 2012
• ISBN:
  9781136492822
• Edition: 2
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

This book differs from other introductions to pragmatics in approaching the problems of interpreting language use in terms of interpersonal modelling of beliefs and intentions. It is intended to make issues involved in language understanding, such as speech, text, and discourse, accessible to the widest group possible -- not just specialists in linguistics or communication theorists -- but all scholars and researchers whose enterprises depend on having a useful model of how communicative agents understand utterances and expect their own utterances to be understood. Based on feedback from readers over the past seven years, explanations in every chapter have been improved and updated in this thoroughly revised version of the original text published in 1989. The most extensive revisions concern the relevance of technical notions of mutual and normal belief, and the futility of using the notion 'null context' to describe meaning. In addition, the discussion of implicature now includes an extended explication of "Grice's Cooperative Principle" which attempts to put it in the context of his theory of meaning and rationality, and to preclude misinterpretations which it has suffered over the past 20 years. The revised chapter exploits the notion of normal belief to improve the account of conversational implicature.

1 What Is Pragmatics and Why Do I Need to Know, Anyway?

INTRODUCTION

Writing is an act of faith, E. B. White said.¹ Speaking is no less so. Pragmatics is the study of the mechanisms that support this faith, a faith so strong that many can use the term communicate interchangeably with speak or write, never noticing that the term communication presupposes achievement of the intended effect of verbal action upon the addressee, while speaking and writing do not. Contrary to popular belief, communication is not accomplished by the exchange of symbolic expressions. Communication is, rather, the successful interpretation by an addressee of a speaker’s² intent in performing a linguistic act. It is the purpose of this book to give an overview of those mechanisms which allow more to be communicated than is actually said.

Linguistic pragmatics as defined here is at the intersection of a number of fields within and outside of cognitive science: not only linguistics, cognitive psychology, cultural anthropology, and philosophy (logic, semantics, action theory), but also sociology (interpersonal dynamics and social convention) and rhetoric contribute to its domain. This volume does not pretend to do justice to the depth of scholarship in all of these fields, but hopes instead to sketch the interrelationships of the phenomena they variously study.

1. WHAT THIS BOOK IS ABOUT

This chapter discusses the foundations of the analysis of language use. Subsequent chapters treat topics encompassed by successively broader interpretations of the domain of pragmatic theory. chapter 2, therefore, concerns the narrowest interpretation of pragmatics: the interpretation of INDEXICAL expressions (words or phrases like me, here, then whose REFERENCE cannot be determined without taking into account the context of the utterance of a linguistic expression). Minimally the context required for the interpretation of indexical expressions includes the time, place, speaker, and topic of the utterance. chapter 2 also discusses the interpretation of anaphoric expressions like he and next day. chapter 3 concerns problems of reference generally: how a speaker knows what an expression like grass or even Bill refers to. Ultimately this leads us to consider also what the conditions for the use of the definite article the are, and how one understands the use of common nouns and verbs like newspaper and baste or drink. chapter 4 addresses two aspects of meaning that interact with the propositional content of utterances: illocutionary force (what a speaker is doing in making an utterance) and presupposition.

Chapters 5 and 6 focus on means which speakers have available to manage how what they say will be interpreted, and strategies for the exploitation of logical relations in arranging the content of one’s speech. chapter 5 begins with a discussion of principles speakers regularly use to convey more than they actually say. The rest of the chapter illustrates how the exploitation of these principles can be invoked in accounts of discourse coherence, illocutionary force, presupposition, and the relation of literal meaning to metaphor. chapter 6 deals with how a speaker’s discourse goals and attitudes may be reflected in the syntax of the utterances that constitute a discourse.

Finally, in chapter 7, we look at conversational interactions, and some of the devices speakers use to regulate control of conversation and maintain or alter social relations among participants via patterns of discourse. The discussion treats topic manipulation, politeness strategies, uses of questions, and other exploitations of linguistic form to affect the course of the interaction.

2. BELIEF AND INTENTION

The broadest interpretation of pragmatics is that it is the study of understanding intentional human action. Thus it involves the interpretation of acts assumed to be undertaken in order to accomplish some purpose. The central notions in pragmatics must then include belief, intention (or goal), plan, and act. Assuming that the means and/or the ends involve communication, pragmatics still encompasses all sorts of means of communication, including nonconventional, nonverbal, nonsymbolic ones, as for example, when a life guard throws a volleyball in the direction of a swimmer struggling in the ocean.³ The lifeguard believes that the swimmer wants assistance, and that the swimmer will understand that the volleyball thrown in his direction is intended (by the life guard) to be assistance, and that the swimmer will know how to take advantage of the volleyball’s property of being lighter than water. That makes at least three beliefs and one intention on the part of the lifeguard, including two beliefs about the swimmer’s beliefs, and one about the swimmer’s desires.

Reflexive beliefs and intentions (i.e., speaker’s and addressee’s beliefs and intentions about each other’s beliefs and intentions) of this sort are typical of the most ordinary, straightforward uses of language, and such uses cannot be understood without reference to them. For example, belief is what makes the difference (obviously) between a lie and a mistake. When people say something false which they believe to be false, they are lying, but if they say something which is false, but which they happen to believe is true, they are merely mistaken. Belief is also the core of the difference between informing and reminding. If a speaker, S, says to an addressee, A, “The Koreans have a holiday which commemorates the invention of their alphabet,” and believes that A does not already know this, S most likely intends to inform A; if S believes that A does already know it, S most likely intends to remind A. Sometimes speakers are content to be vague about this belief, hoping that the remark will be taken as a reminder if A already knows, as information if A doesn’t. Other times, speakers may want to make their position clear (for instance, to avoid insulting A by implying that A did not know something which A feels everyone should know that A does know). For example, As you know, I remind you, and Of course preface utterances that are intended to be taken as reminders; they indicate the speaker’s belief that the addressee already knows that the proposition expressed by the sentence is true. Similarly, Actually and In fact preface utterances that the speaker believes represent information new to the addressee.

Likewise, intent makes the difference between a lie, and a joke or a figure of speech, between a promise and a prediction. Uttering something false like “She phones her mother everytime you say ‘Boo!’” is a lie if the speaker intends her addressee to believe that she believes it, a figure of speech if she intends him to understand that it is an exaggeration for effect (hyperbole). Uttering “Johnny will be on time tomorrow” constitutes a promise if the speaker intends to be taken as guaranteeing that it will be true, but it constitutes a prediction if it is only intended to represent her current belief. The nature and interrelationship of beliefs and intentions are explored in the study of philosophy of mind (for an introduction, see Bechtel, 1988).

3. PLANS AND ACTS

Beliefs and intentions are not the whole story, of course. The proverb, “The road to hell is paved with good intentions” means that intentions count for naught unless acted upon. A speaker with an intention and a set of beliefs about her audience acts rationally on that intention by forming a plan to effect it that is consistent with those beliefs. Typically such plans are hierarchical, and involve subgoals and mediating intentions as well as an ultimate goal. Plan formation and reconstruction play an important role in artificial intelligence (AI) work on natural language understanding; the papers in Cohen, Morgan, and Pollack (1990) are representative.

For example, I might have the ultimate goal of having you have a low opinion of a certain mutual acquaintance. Since I cannot effect this by just willing it to happen, I must form a plan which will cause you to have this opinion. I could tell or order you to have a low opinion of him, but given my belief that you are rational, and that you believe that I believe this, I know this is unlikely to be effective. I have to choose a plan (from an infinity of possibilities) which will take advantage of your rationality in causing you to conclude that our acquaintance, call him Eks, is a sleaze. For example, I might choose to tell you that he has instituted a policy in the office he manages which I believe you will find offensive, and having decided that, I must choose exactly how to say this. Shall I just blurt out, “He refuses to hire qualified blacks, Jews, and women!” Or shall I tell an anecdote which gives evidence, with real names and quoted dialogue, which I can expect you will take as indicating that Eks is a despicable character? Or maybe just a single sentence with statistics and one or two names? Once this is decided, the construction of each individual sentence requires many similar decisions (cf. Green, 1982a). And all of this usually takes place without the speaker being aware of even thinking about it. In uttering a single sentence, more decisions have to be made than there are words in the sentence (cf. Green, 1982a), so it is obvious that these decisions must be made very quickly, and below the level of consciousness, or the flow of speech would be considerably more intermittent, and exponentially slower than it is.

The task of the discourse-interpreter is (a) to understand what the speaker has said, that is, to construct a mental model of the situation which the speaker is indicating exists, and (b) to evaluate that model and use it to update his own model of the world. The first task is exemplified by the acts involved in understanding what statement, imperative, question, wish, and so on, a speaker has made. The second involves drawing conclusions which add to one’s knowledge of the world and to one’s model of the speaker: for example, inferences about what the speaker knows, what the speaker believes that the addressee knows, what the speaker believes that the addressee believes to be false, how the speaker feels about individuals, situations, and events that have been referred to in the conversation. This is also done largely subconsciously, and ordinary competent speakers are rarely able to reconstruct and articulate the logic involved.

Pragmatics, as described in this section, refers to the study of action deliberately undertaken with the intention of causing the intended interpreter to re-assess his model of how things are, including his system of values and his model of the speaker’s beliefs, attitudes, and intentions. To narrow our study to linguistic pragmatics, or the pragmatics of language use, we need only stipulate that the principles of pragmatics must account systematically for acts involving linguistic expressions.

4. LINGUISTIC EXPRESSIONS AND THE DETERMINATION OF MEANING

A language (natural or artificial) consists of (a) a set of basic (i.e., unanalyzed) expressions, termed a LEXICON; (b) a set of rules for combining these expressions in well-formed constructions, termed a SYNTAX; and, according to the standard view, (c) a set of rules for deriving the meaning of each construction from the meanings of its parts, called a COMPOSITIONAL SEMANTICS (Montague, 1970, 1973; Dowty, Wall, & Peters, 1981). All of these components are conventional, which means that principles of logic and laws of physics do not force the rules to be exactly as they are.⁴ Rather, at least some of the rules are arbitrary to at least some degree, and by custom, speakers act in conformity with them. The issue hardly ever comes to consciousness, but using the language in a way that failed to observe the rules would greatly decrease the likelihood of successful communication. For example, there is no logical reason why English adjectives must go before the nouns they modify (e.g., united states) rather than after, as in French (états unis). Likewise, the semantic rules that guarantee that John is understood as the object of please in John is easy to please (i.e., that make it equivalent to It is easy [for people] to please John), might have been otherwise, allowing John to be understood as the subject (making the sentence equivalent to It is easy for John to please [people]).

In addition to being conventional, the semantics is compositional and basically TRUTH-CONDITIONAL. It is standard to distinguish between what a sentence means and what a speaker intends to convey by the utterance of that sentence (Grice, 1957), and to restrict the role of semantics to explicating the meaning of a sentence in terms of conditions which must be fulfilled for it to be used to truthfully describe a situation. Aspects of the interpretation of utterances which do not involve truth conditions are commonly considered to be outside the domain of semantics, and that is the approach followed here. Thus, whether an utterance is a promise, a prediction, or a question, and how metaphorical utterances like India demanded compensation are understood are considered matters of pragmatics, not semantics (cf. chap. 3-5).

To say that the semantics is COMPOSITIONAL is to say that the meaning of a complex expression will be related in a predictable way to the meanings of the parts from which it is constructed. This is usually expressed by saying that the meaning of the whole is a function of the meanings of the parts, where function has its technical, mathematical sense. Consequently, the parts of every structure defined by a particular syntactic rule must be related in the same way, because they are related by a single function.⁵ This means that, for example, whatever semantic relation a syntactic subject like Sandy bears to its predicate in a sentence like Sandy coo...

Table of contents

• Cover
• Halftitle
• Title
• Copyright
• Dedication
• Contents
• Notational Conventions Used in This Book
• Acknowledgments
• 1. What is Pragmatics, and Why Do I Need To Know, Anyway?
• 2. Indexicals and Anaphora: Contextually Identifiable Indeterminacies oF Reference
• 3. Reference and Indeterminacy of Sense
• 4. Non-Truth-Conditional Meaning: Interpreting The Packaging of Propositional Content
• 5. Implicature
• 6. Pragmatics and Syntax
• 7. Conversational Interaction
• 8. Perspective
• References
• Author Index
• Subject Index