Introduction

Focal points for many of the case studies in this book are modality and deixis. We start in this chapter with selected examples of the development of epistemic from deontic meaning within the domain of modality relatively narrowly construed. Themes are developed that will be shown in later chapters to be relevant to modality more broadly construed.

There is little agreement on exactly how to define modality, beyond the observation quoted in 1.7: “The essence of ‘modality’ consists in the relativization of the validity of sentence meanings to a set of possible worlds” (Kiefer 1994: 2515). In work on logic, the fundamental types of modality are necessity and possibility in so far as they relate to the truth or falsity of the proposition. In work on linguistics, two related types, deontic (obligation) and epistemic (conclusion), have received most attention, but at least one other is also recognized: ability/capacity (see e.g. Leech 1971, Lyons 1977, Palmer 1990 [1979], 1986, Coates 1983, Sweetser 1990, Bybee, Perkins, and Pagliuca 1994, Kiefer 1994, 1997, Van der Auwera and Plungian 1998). Some brief working characterizations of the three types of modality follow; more detailed discussion of deontic and epistemic modality is provided in 3.2:

(i) Deontic modality (from the Gk. deon “what is binding”); also widely known as “root” modality (e.g. Coates 1983, Sweetser 1990). This centrally involves obligation or compulsion. In his ground-breaking study of modality in language Lyons identified several characteristics of deontic modality.

Aims of this book

In this book we show that there are predictable paths for semantic change across different conceptual structures and domains of language function. Most especially we will show that, despite century-old taxonomies that suggest that meaning changes are bidirectional, e.g. generalization and narrowing, metaphor and metonymy, when we trace the histories of lexemes cross-linguistically we in fact repeatedly find evidence for unidirectional changes. These changes are of a different sort from those cited in the taxonomies. The taxonomies focus on mechanisms, the kinds of cognitive and communicative processes speakers and hearers bring to the task of learning and using a language. The regularities are, however, shifts from one linguistically coded meaning to another, for example, from obligation to do something to conclusion that something is the case. Such regularities are prototypical types of changes that are replicated across times and languages. They are possible, indeed probable, tendencies, not changes that are replicated across every possible meaningful item at a specific point in time in a specific language, such as the Neogrammarians postulated for sound change. That they recur so often and across totally unrelated languages is, we argue, intrinsically bound up with the cognitive and communicative processes by which pragmatic meanings come to be conventionalized and reanalyzed as semantic polysemies. In particular, they are bound up with the mechanisms that we call “invited inferencing” and “subjectification.” This book, therefore, is a contribution to historical pragmatics as well as semantics.

Introduction

In this chapter we move to the development of pragmatic meanings of adverbials (ADVs). The main function of the subclass of ADVs in question, e.g. in fact, after all, so, then, well, Jp. sate “well,” sunawati “namely, in other words,” in some of their uses, is to signal an aspect of the speaker's rhetorical stance toward what he or she is saying, or toward the addressee's role in the discourse situation. They have little conceptual semantics, and do not contribute significantly to the truth-conditional meaning of propositions, hence some researchers have considered them a subset of “pragmatic markers” or “metatextual particles.” However, they mark the speaker's view of the sequential relationship between units of discourse, that is, they serve as connectives between utterances. They are widely known as “discourse markers” (Fraser 1988) or “discourse connectives” (Blakemore 1987). The ones of primary interest to us here have modal properties, in that they reflect aspects of the speaker's epistemic (information and belief-state) stance toward the sequencing of the discourse; they also have deictic properties in that they index the speaker's viewpoint on the proximity or distance within the world created in the text, and do both anaphoric and cataphoric work (Schiffrin 1990b).

There are several other types of pragmatic markers, including those that signal narrative structure.

Introduction

Previous chapters have alluded to the indexical properties of epistemic modals, discourse markers, and performative verbs. In these domains the indexicality is the encoded link between the world of the conceptualized described event (CDE) and the world of the conceptualized speech event (CSE). We now turn to another class of linguistic items that provide such a link: social deictics (SDs). We define SDs as directly encoding within their semantic structures the conceptualized relative social standing (superiority/inferiority, (non)intimacy, in-group versus outgroup status, etc.) of a participant either in the CDE or in the CSE by “pointing” to that social standing from the deictic ground (perspective) of SP/W relative to AD/R and other elements of the CSE.

SDs include contrasting second person singular tu/vous(T-V) pronouns in European languages like French and German; parentheticals such as I pray (you), sentence adverbials such as please; and, in a few languages such as Japanese and Korean, large systems of lexical items, derivational formulae, and affixes that are often termed “referent” and “addressee” honorifics. Referent SDs index the social status of one or more participants in the CDE (here termed “referents”) relative to a deictic ground in the CSE. Addressee SDs, by contrast, index the relative social status of the speech event participants conceptualized independently of their possible roles in the CDE. Although most SDs are politeness markers, SDs also include a much smaller subset of linguistic items that directly encode an attitude of denigration.

Introduction

In this conclusion we briefly summarize the major hypotheses that we have presented, and in doing so essentially reprise the main points of chapter 1. We end with a few suggestions for future research.

Summary of major findings

Our primary goal has been to show that there are general tendencies in semantic change that are widely attested across a range of semantic fields and languages, and that they result from the interaction of language use with linguistic structure. We have hypothesized that semantic change starts with SP/Ws instantiating a code that they have acquired. Drawing on and exploiting, sometimes consciously, sometimes unconsciously, pragmatic meanings, most especially those kinds of implicatures that we have called “invited inferences,” SP/Ws may innovate new uses of extant lexemes. If these new uses spread to AD/Rs and are replicated by them in their role as SP/Ws, then semanticization will take place. We have argued that the main mechanism of semantic change is subjectification (including intersubjectification). This follows from the hypothesis that the seeds of semantic change are to be found in SP/Ws, drawing on and exploiting pragmatic meanings that arise in negotiated interaction. SP/Ws do so for their own ends – to inform, to express beliefs and emotions, and to make explicit what they are (or claim to be) doing with language. Some aspects of what SP/Ws do, most especially projecting innovative metaphorical relations, are primarily self-directed, manifestations of analogical conceptualizing, solving problems of expression, and language play.p

The focus of this work is recent developments in cross-linguistic research on historical semantics and pragmatics, with special reference to the histories of English and Japanese. The framework can be characterized as “integrative functionalist” (Croft 1995) in that we consider linguistic phenomena to be systematic and partly arbitrary, but so closely tied to cognitive and social factors as not to be self-contained; they are therefore in part nonarbitrary. One of the linguist's tasks is to determine what is arbitrary, what is not, and how to account for the differences.

We see semantic change (change in code) as arising out of the pragmatic uses to which speakers or writers and addressees or readers put language, and most especially out of the preferred strategies that speakers/writers use in communicating with addressees. The changes discussed in this book are tendencies that are remarkably widely attested, but that can be violated under particular, often social, circumstances ranging from shifts in ideological values to the development of various technologies. “Regularity” is to be understood as typical change, or frequent replication across time and across languages, not as analogous to the Neogrammarian idea of unexceptionless change in phonology.

Richard Dasher takes prime responsibility for the Japanese data, Elizabeth Traugott for the remainder, but both have discussed all the material presented here in countless meetings over nearly fifteen years. The ideas presented here have been explored in several venues.

Introduction

In this chapter we outline some of the work and issues to date on semantic change, with particular attention to research focused on regularity and on mechanisms of semantic change that will be developed further in later chapters. No attempt is made to cover research into the history of research into semantic change comprehensively. For some overviews discussing different approaches, see Kronasser (1952: chapter 1), Ullmann (1957: chapter 4), Warren (1992: chapter 1), Blank (1997: chapter 1, 1999), Geeraerts (1997: chapter 3), Fritz (1998: chapter 4).

Backgrounds to contemporary work

In Europe and America much of the groundwork of semantic theory was laid by Greek and Roman grammarians, who argued at length about the arbitrariness or naturalness of meaning–form pairs, homonymy and polysemy, and by philosophers and logicians especially from the seventeenth century on, who focused on the nature of reference. Likewise, the nature of the lexicon was discussed by dictionary makers, especially from the eighteenth century on. In the nineteenth century the work on language families that led to comparative Indo-European linguistics focused on sound correspondences and plausible sound change, but crucially required a notion of cognate meaning and of plausible meaning changes as well, and triggered the development of more sophisticated views of semantic change. This is particularly true of the major comparative dictionaries that arose out of this work, much of it in the Neogrammarian tradition, among them Pokorny (1959/69), The Oxford English Dictionary (1989; the first volume appeared in 1884), Wartburg (1928–66).

Speech act and performative verbs

Among the areas of interface between semantics and pragmatics that have attracted particular attention are “speech act verbs” and “performativity.” Nevertheless, there is not yet a great deal of convergence about how to approach them, nor is there extensive understanding of cross-cultural differences with respect to speech acting (but see Blum-Kulka, House, and Kasper 1989). This naturally raises significant methodological questions for historical semantics (Papi 2000). In order to allow for comparison across time and across languages and cultures we have chosen to take a restricted view of the problem, and to focus on the semasiological questions of how certain expressions acquire “speech act” and “performative” meaning over time. Here we outline our assumptions about the domain under investigation.

There is a large number of verbs of speaking in English and Japanese. Some of these refer to ways of speaking, e.g. whine, simper, drawl, Jp. donaru “yell (angrily),” sasayaku “whisper”; these are verbs of “locution.” Some refer to acts of speaking (claim, say, command, threaten, Jp. syutyoo suru “insist,” happyoo suru “announce (in public)”); these are “speech act verbs” (SAVs). A subset of SAVs are “performative” (“illocutionary”) verbs: verbs which, under specific conditions can be used not only to report on sayings but to have the force of a doing (Austin 1962). For example, of the set claim, say, command, threaten, in PDE only claim and command are typically used performatively; in Jp. happyoo suru “announce” is used performatively, but syutyoo suru “insist” is descriptive of a manner of speaking only.

By 1960 structural linguists could look back on thirty or more years of progress. The theory of the phoneme was now widely accepted; and, although there were disagreements in detail between most Americans and most Europeans, few disputed that it was one central unit of language. For many linguists the morpheme or a unit like it was another, and the identification of these units in ‘-eme’, as the elements of articulation on two different levels, was increasingly a basis for the practical description of languages. Such findings had begun to work their way from technical publications into textbooks. I have referred to Hockett's general introduction to linguistics (Hockett, 1958). But it had been preceded by at least one specialist textbook in descriptive linguistics, by H. A. Gleason, whose table of contents gives a clear view of how students, in at least part of the English-speaking world, were taught (Gleason, 1961 [1955]). Martinet's introduction, to which I have also referred, was to have a similarly wide success in continental Europe (Martinet, 1970 [1960]).

These achievements were ostensibly founded in the thought of Saussure, in lectures that had last been given in 1911, and of Bloomfield, formulated in its mature phase in the 1920s. A linguist was therefore describing either an underlying system whose reality was supra-individual (2.1), or a set of potential utterances (2.2). But both accounts had naturally reflected the preoccupations of their own day and, with hindsight, though the structure of a language system had been worked out in some detail, it was time to ask again exactly what reality such systems had.

The last chapter of this book was finished on the day when Bobby Robins, whose A Short History of Linguistics has been admired for more than thirty years, was found dead. It is with sadness and affection that I dedicate it, with its presumptuously similar title, to his memory.

I was initially not at all sure how this history should be written: in particular, how selective and, in consequence, how long it should be. For advice at that stage I am especially grateful to Jeremy Mynott and, in a sense that they will understand, to my fellow editors of the Cambridge Textbooks in Linguistics. Conversations with Kasia Jaszczolt have since helped, at various times, to clarify my thinking.

I am also grateful to Andrew Winnard for waiting patiently for the book to be written.