Diachronic form-to-function mapping

This approach traces individual linguistic items in their historical development. It considers (the emergence of) the pragmatic meaning of these elements. The tertium comparationis is here the linguistic form. It does not have to be constant but the more recent realizations must be seen as direct developments of earlier realizations.

The first four papers classified as diachronic form-to-function mapping in this volume (i.e. de Lima, Bernardez and Tejada, Schwenter and Traugott, and Allen) adopt a perspective that has been described as pragma historical linguistic, rather than historical pragmatic, that is to say they are primarily concerned with pragmatic explanations of language change. Within this perspective, however, they analyze individual linguistic items and describe how these have changed in the course of the development of English or other languages. De Lima uses several examples of lexical change, mainly in German, in order to ascertain to what extent pragmatic maxims can be used to explain language change but he finds that language change induced by pragmatic principles is just a special case of change induced by norm-following, as he calls it; and norm-following is a special case of language change by rational behavior. Bernardez and Tejada (this volume) show how an economy principle and a communicative principle can be used to explain word-order changes from Old to Middle English, and thus to explain these changes from a pragmatic rather than a syntactic perspective.

Traugott has done a lot of pioneering work on the regularities of semantic change (e.g. Traugott 1989 on how again changed its meaning). She describes subjectification as a pragmatic-semantic process whereby "meanings become increasingly based in the speaker's subjective belief state/attitude toward the proposition" (Traugott 1989: 35). More precisely, Traugott relates subjectification to the process of grammaticalisation. Subjectification in grammaticalisation is a gradient phenomenon, whereby forms and constructions that at first express primarily concrete, lexical, and objective meanings come through repeated use in local syntactic contexts to serve increasingly abstract, pragmatic, interpersonal and speaker-based functions (Traugott in press prefinal draft: 2).

Grammaticalisation is defined as the process whereby lexical items or phrases come through frequent use in certain highly constrained local contexts to be reanalyzed as having syntactic or morphological functions, and once grammaticalized continue to develop new grammatical functions (Traugott in press prefinal draft: 2).

Thus, for example, the imperative Let us go developed into the hortative Let’ s go, or the temporal Mary read while Bill sang evolved into the concessive Mary liked oysters while Bill hated them (cf. Traugott in press prefinal draft: 1). Furthermore, in her analysis of the rise of epistemic meaning Traugott (1989) emphasizes the need for a theory of pragmatic inference controlled by principles of strengthening of in formativeness and relevance to account for the semantic changes observed, i.e. the development of both lexical and grammatical items into expressions of epistemicity.

Schwenter and Traugott (this volume) work within grammaticalisation theory, which has its origins in work in historical syntax and morphology within a functionalist framework in which language change is seen as a process and not as a product. This line of research examines historical data in their discourse context and thus differs somewhat from the socio-pragmatic approaches adopted by many of the other contributors. In their contribution Schwenter and Traugott trace the development of three complex prepositional phrases in English expressing substitution, instead of, in place of, and in lieu of.

Allen (this volume) traces the history of the verb please in expressions such as you can do as you please, where the experiencer is the only expressed argument in the construction. She argues that the pragmatic function and text-typological evidence of the construction must be taken into account. Thus, one of the alternative constructions with the experiencer in object position, which remained common in if-clauses until the 18th century, is no longer used in everyday speech except for very specific speech acts in very specific settings such as May it please the court, ... in a courtroom setting.

Wales (this volume) analyses the generalizing-possessive or generic-deictic pronoun your, as in the newspaper headline Just your average French movie star. She traces it from its first appearance in written English in the late 16th century to Modern English and analyses its function, its colligations and collocations, and its frequency of occurrence in a broad range of contexts.

Linguistic items which can be analyzed within a framework of diachronic form-to-function mapping include deictic elements. Fries (1993) proposes a program for the historical study of discourse deixis and offers a preliminary analysis of the Old English discourse deictic elements hēr and nū ('here' and 'now'). By means of discourse deictic elements the speaker or writer expresses in various combinations where he or she is at the moment of speaking or writing.

Kryk-Kastovsky (this volume) investigates the English demonstratives this and that in Early Modern English and compares these findings to the situation in Modern English. It turns out that most Modern usages were already present in Early Modern English, but with some notable exceptions as for instance in the dating of Early Modern English letters (Alicant, this 31st of July, 1706).

Warvik (this volume) traces the development of the conjunctions/ adverbials þa and þonne from Old English to their Modern English counterparts then and when concentrating on the changes that these elements underwent in the Middle English period. In Old English these elements were ambiguous and could be used either as a conjunction or as an adverb. This is again a clear case of form-to-function mapping because the starting point is the linguistic item. The item itself changes too in the course of time; þa and þonne even conflate in Late Middle English to then. But Wårvik investigates the different discourse functions that these elements fulfil.

Discourse markers attracted a lot of interest in this field (cf. Brinton 1990; Stein 1985b; Taavitsainen 1994b; Burger 1980; Henne 1980; Fludernik this volume; and Onodera this volume). One of the major debates focuses on their historical status. In particular, it is not clear whether discourse markers can be characterized as ephemeral due to the fact that they are typical of linguistically transitional periods (cf. Stein 1985b: 300). Stein (1985b) investigates the choice between the two endings s and th of the third person singular present indicative and the use of do or the finite form in declarative sentences. In Stein's (1985b: 300) view the "discourse meanings of s/th and do have to be considered as instances of [...] ephemeral and transitional functionalisations of syntactic and morphological contrasts". He suggests a discourse structure explanation according to which in the s-passages "the circumstances, the agents and the places are described in more detail" (Stein 1985b: 284), "the use of direct speech makes for a more involved and vivid impression of the scene", and "much more descriptive, circumstantial, detailed individuating verbal effort" can be observed. The th-parts, on the contrary, are "distinctly more technical" (Stein 1985b: 288). Furthermore, the use of do seems to be "related to what is the high point of the content, the essential parts" (Stein 1985: 295).

The opposite view postulates that discourse markers "are characteristic of every period, no matter how stable" (Brinton 1990: 62). Accordingly, Brinton (1990: 63) concludes from his analysis of Old English hwcet and Modern English you know that the history of particular discourse markers is not as transitory as Stein suggests in his analysis of discourse markers in Early Modern English. For example, old English hwcet is similar to you know in Modern English, since it indicates knowledge shared by speaker and addressee or common to all; it presents new information as if it were old, it establishes either intimacy or distance between speaker and addressee; it solicits a favorable reception for the following information; it is an attention-getting device; it provides evaluation of the narrative point; and it makes explanatory material salient (Brinton: 1990: 56).

Hwœt disappeared after the Middle English period, but its function has been assumed by a Modern English equivalent. Brinton (1990: 62) interprets the development of hwcet "from interrogative in direct questions to complementizer in indirect questions to discourse marker of shared information" in Traugott's terms. Thus the development of the discourse meaning of shared knowledge from the simple interrogative meaning can be seen as the result of a pragmatic inference. Furthermore, it represents an increased emphasis on the speaker's attitude, from a questioning of what the hearer knows to an expression of the speaker's belief in, or confirmation of, what the hearer knows to an expression of what is known by both speaker and hearer (Brinton 1990: 62).

Therefore, discourse markers cannot simply be attested a temporary nature. They continue to exist beyond transitional periods through shifts in form or function.

Fludernik's paper (this volume) is devoted to the development of narrative discourse markers (in particular po) in Middle English. The old po-pattern, in which þo operates as a discourse marker, undergoes radical changes in Middle English. It becomes functionally ambiguous and has to compete with other discourse markers until it disappears in the late 14th and 15th century, when the when-then construction wins out over þo and other discourse markers. Her analysis traces both the shifts in form (from þo/þa to then, than, thenne, thanne and eventually to then) and the functional changes (from discourse marker to adverbial and finally to conjunction). Her starting point in this very complex series of functional and formal changes is the linguistic item þo, and in this sense her analysis is an example of a diachronic form-to-function mapping.

Onodera (this volume) works within the same general framework as Schwenter and Traugott. She starts out with the question of whether the discourse functions that have been postulated for various linguistic items are a recent phenomenon in the evolution of language - and in particular in the evolution of Japanese - or whether their earlier existence just has not been discovered yet. She analyses two Japanese discourse markers, demo and dakedo. The former was first used as a marker in the 16th century and the latter in the early 20th century.

Burger's (1980) article on German interjections (e.g. ach, ey, ha) in texts from the 17th century outlines a programme for tracing the frequency and functional diversity of interjections. In this context, he suggests to investigate the decrease in interjections and the increase in discourse particles (Abtönungs-partikel), and to ascertain whether this process can be regarded as a shift in written style or language change in both written and spoken communication.

From a more general perspective discourse markers, such as apropos or kurz und gut, can also be studied as linguistic items establishing the broader contours of conversational structure, e.g. as portrayed in Henne's (1980) study of dramas of the Sturm und Drang period.

Exclamations are another type of discourse marker which have already been under investigation from a historical perspective. The common assumption seems to be that exclamations were used much more widely and for a broader variety of functions in Late Middle English than in present-day English. Taavitsainen (1994b) outlines the use of exclamations in Late Middle English, e.g. alas, ey, ah, or harrow, and discusses their function with special reference to text types. Her analysis shows that exclamations act as markers of the fictional rather than the non-fictional mode. Their functions include personal affect, foregrounding, and the marking of turn-taking. In this volume, she continues her work with an analysis of interjections in Early Modern English.

She provides statistical evidence of the frequency of interjections in different Early Modern English text genres, and she analyses the functions that these interjections serve. In traditional accounts based on Present Day English, interjections have usually been relegated to the purely emotive level of language, as the speaker's reaction to the current situation. In her data, however, Taavitsainen found many interjections that were used to appeal to addressees, most commonly as vocatives.

Topic changers such as well, now, by the way or anyhow can also be studied from the perspective of diachronic form-to-function mapping. Their development from the mid-seventeenth century until the early 20th century suggests that they "are more common in informal discourse than in formal discourse" (Finell 1992: 731). In particular, topic changers express a communicator's attitude, his or her wish for interaction, and politeness. In her analysis of topic changers in informal letters, Finell (1992: 732) concludes that it seems that "in present day English there is, proportionally, a higher number of discourse markers that function as explicit topic changers than there was in Early Modern English". This may be taken to express a more urgent need for interaction and to be part of the process of a move towards the subjectification of language. In Modern English the functional specialization and diversification of topic changers led to their conventionalization. Topic changers thus underwent a shift from semantic explicitness to pragmatic explicitness. These results "attest Traugott's hypotheses concerning the direction of functional-semantic shifts (from propositional via textual to expressive); and the tendency towards the subjectification of language" (Finell 1992: 721).