In medias res

What do we understand by functional sentence perspective (FSP)? An analysis of the following four versions – French, English, German and Czech – of the first paragraph of the closing chapter of Victor Hugo's Les Misérables will help to answer this question. The concepts used by the analysis as well as others linked with them, and together with them forming a theory of FSP, will be the object of the present study.

(1) II y a, au cimetière du Père-Lachaise, aux environs de la fosse commune, loin du quartier élégant de cette ville des sépulchres, loin de tous ces tombeaux de fantaisie qui étalent en présence de l'éternité les hideuses modes de la mort, dans un angle désert, le long d'un vieux mur, sous un grand if auquel grimpent, parmi les chiendents et les mousses, les liserons, une pierre. (2) Cette pierre n'est plus exempte que les autres des lèpres du temps, de la moisissure, du lichen, et des fientes d'oiseaux. (3) L'eau la verdit, (4) l'air la noircit. (5) Elle n'est voisine d'aucun sentier, (6) et l'on n'aime pas aller de ce côté-là, parce que l'herbe est haute et qu'on a toute de suite les pieds mouillés. (7) Quand il y a un peu de soleil, les lézards y viennent. (8) II y a, tout autour, un frémissement de folles avoines. (9) Au printemps, les fauvettes chantent dans l'arbre (Victor Hugo, Les Misérables, Paris: Nelson, 1935).

Degrees of prosodic prominence

How does the spoken language react to the distribution of the degrees of CD as determined by the interplay of the non-prosodic factors of FSP? Is there a relationship between this distribution and the distribution of degrees of prosodic prominence? But can we speak of degrees of PP? For instance, O'Connor and Arnold's treatment of the English system of intonation and their system of tonetic notation (O'Connor and Arnold 1973) permit the conclusion that the configuration of prosodic features within what is termed by them ‘tune’ and here ‘tone unit’ (‘a stretch of speech containing one intonation nucleus’ CGEL 1356) displays a hierarchy of PP.

Within the tone unit, the section constituted by the head and the nucleus shows greater PP than the sections serving as pre-head and tail. The most prominent feature of the entire tone unit is the nucleus. On the other hand, the lightest feature, occurring either outside or inside the head, is absence of stress. As the head and the nucleus exceed in PP the pre-head and the tail, O'Connor and Arnold regard the stresses inside the head as well as the nucleus as accented, and the stresses occurring in the pre-head and tail as unaccented (1973: 31–6). All this suggests at least four degrees of PP: (i) absence of stress (occurring inside or outside the head); (ii) stress not combined with accent (occurring on stressed syllables in the pre-head or the tail); (iii) stress combined with accent (occurring on stressed syllables inside the head); and (iv) nuclear stress, or for short, nucleus (see also Gimson's (1970: 267) four degrees of accentuation).

The introductory chapter has given a rough sketch of the interplay of the three factors that determine the distribution of degrees of CD over a written sentence and in this way its FSP (see p. n). The present chapter will concentrate on the operation of what for convenience is termed the ‘contextual factor’; what is actually in play is the retrievability/irretrievability from the immediately relevant context (see ibid.). Since context is a complex phenomenon, the chapter will have to deal with a number of questions (see Daneš 1974b: 109–10). As Daneš has pointed out (ibid.), one of them is the relative and very broad (if not vague) character of the notions of given (known, old) and new (unknown) information. This question will be taken up first.

Known information and the structure of context

An examination of the function of the articles in FSP (Firbas 1957, 1966) shows that information marked by the definite article as known cannot always be regarded as such from the point of view of FSP. The following sentence will illustrate.

(For abbreviations following example sentences or passages and indicating the sources, see the list of abbreviations, pp. xiii–xv. The number after the abbreviation indicates the page of the source.)

In view of the development of the narration, the elements the window, the veranda, the grass and the gate convey new information, although they have already occurred in the story and in this sense convey information that is known.

The foregoing discussions enable me to formulate my view of the contextual applicability of the spoken sentence structure, answer the question whether CD can be equated with PP, summarize the role of intonation in FSP, subscribe to Vachek's assessment of the relationship between the written language and the spoken language, account for the concept of potentiality and assess the role of the contextual factor in FSP.

Functional sentence perspective and the contextual applicability of the spoken sentence structure

Discussing FSP in written communication, I posed the question of the contextual applicability in FSP of the written semantic and grammatical sentence structure. Let me now add a note on FSP contextual applicability in the spoken language. An enquirer into contextual applicability is free to choose the structure, the applicability of which he wishes to assess; but in doing so, he has to determine which features are to be regarded as constitutive, i.e. as identifying the structure.

Examining the contextual applicability in FSP of a written sentence structure on pages 110–14, I regarded the words used, their syntactic implementation, their linear arrangement and – if employed – their typographical intensification (its italicization, for instance) as such constitutive features. A different word and/or another syntactic implementation and/or a change in word order and/or a difference in typographical intensification would have created a different sentence structure.

Examining the contextual applicability in FSP of a spoken sentence structure, I exclude from the features just enumerated the signals of typographical intensifications, but add the PP distribution.

The sentence as a field of relations

Before starting a discussion of the problems of FSP, the question must be answered of how to interpret the relation between the semantic contents and the semantic ties interlinking them within the sentence on the one hand, and the grammatical (syntactic) structure of the sentence on the other; and of how to interpret the relation of this semantic and grammatical complex to FSP. In my approach I have adopted the following assessment of these interrelations.

The communicative needs of the speaker are satisfied through a twofold process: elements of thought – reflecting the extralinguistic reality, concrete or abstract – are denominated by naming units, which are combined to produce sentences. The former process consists in selecting – or if need be, in coining – naming units (and is covered by Mathesius' functional onomatology; see Mathesius 1975); the latter consists in syntactic structuration (and is covered by Mathesius' functional syntax; see ibid.).

Either process involves meaning and form. This may be obvious in regard to the onomatological process, but perhaps less so in regard to the process of syntactic structuration. But the latter does not merely combine forms: as has been emphasized by Reichling (1961: 1) and subscribed to by Dane (1968: 55), syntactic structuration effects a semantic connection, i.e. a connection of meanings. In this way syntactic structure (and grammatical structure in general) is ultimately rooted in meaning (Poldauf 1954). Trost speaks of grammatical structuration effected on a semantic basis (Trost 1962: 268; in Sgall's translation, semantically grounded grammatical articulation; Trost 1987: 147).

In surveying past treatments of apposition, I demonstrate in this chapter that they provide either an inadequate or incomplete definition of apposition, and argue that apposition is best defined as a grammatical relation realized by constructions having specific syntactic, semantic, and pragmatic characteristics (1.1). To study these linguistic characteristics of apposition, I analyzed the appositions in three computer corpora of spoken and written English (1.2) with the aid of a problem-oriented tagging program (1.3).

The inadequacy of past studies of apposition

A survey of the literature on apposition supports Quirk et al.'s (1985:1302) assertion that “Grammarians vary in the freedom with which they apply the term ‘apposition’…”

Some sources take a very conservative approach to defining apposition. Both Fries (1952:187) and Francis (1958:301) restrict the category of apposition to coreferential noun phrases that are juxtaposed:

Others have expanded the category of apposition considerably. Curme (1931) admits as appositions a diverse group of constructions, including predicate appositives (p. 30):

appositive genitives (p. 84):

apposition proper (pp. 88–91), which can be loose (example 4) or close (example 5), categorizations that correspond in this study to, respectively, nonrestrictive and restrictive apposition (see 3.3):

Apposition as a grammatical relation

and appositive adjectives (p. 93):

Jespersen (1961), like Curme, quite liberally defines apposition.

Apposition is a grammatical relation having various syntactic characteristics, characteristics which will serve as the focus of this chapter.

The relation of apposition is realized by a variety of syntactic forms (2.1): noun phrases predominantly (2.1.1) but other syntactic forms as well (2.1.2–2.1.4). Although these forms can have a full range of syntactic functions, they most commonly have two: subject and object (2.2). In addition, the forms making up the two units of an apposition have a linear (2.3) and hierarchical (2.4) structure that differentiates the relation of apposition from other relations, such as modification and complementation. However, like other grammatical relations, apposition is a gradable relation (2.5): some appositions are fully appositional; other appositions behave in a manner that places them on a gradient between apposition and other grammatical relations, such as coordination (2.5.2), peripheral elements (2.5.3), modification (2.5.4), and complementation (2.5.5).

The syntactic form of units in apposition

The three corpora contained a total of 2,841 constructions counted as appositions in this study (Table 2.1). Approximately three-quarters of these appositions were evenly distributed among the written samples of British and American English. A much smaller percentage (27 percent) occurred in the spoken samples. This distribution indicates that overall there is little difference in the occurrence of appositions in written British and American English but that there is considerable variation in their use in spoken and written English. This variation has a pragmatic explanation: appositions are communicatively more important in speech than in writing (4.3).

The previous three chapters described the syntactic, semantic, and pragmatic characteristics that typify units in apposition. In this chapter, apposition is considered within the context of the grammar of English.

As Chapters 2 and 3 demonstrated, appositions have a variety of different syntactic and semantic characteristics. However, some of these characteristics are more common than others, making apposition a relation in which certain syntactic and semantic characteristics are dominant (5.1). Because apposition is such a linguistically diverse relation, appositions have, as Chapter 4 illustrated, a variety of different communicative functions, functions which vary mainly by genre rather than dialect (5.2): in the corpora there was little variation in the use of appositions in the samples containing written British and written American English, but considerable variation in the use of appositions in spoken and written English and in different genres of written English. Apposition is also a gradable relation, and if appositions are ranked according to their positions on the syntactic and semantic gradients of apposition, some appositions are centrally appositional, others only peripherally appositional (5.3). Finally, apposition is one of the more important grammatical relations in English, occurring considerably more frequently than most other grammatical relations, such as coordination and complementation (5.4).

The predominance of certain syntactic and semantic characteristics of apposition

Even though appositions have a variety of different syntactic and semantic characteristics, some of these characteristics occurred much more frequently in the corpora than others.

Syntactic characteristics

Syntactically, apposition is most commonly a relation between two juxtaposed noun phrases having a syntactic function (such as direct object) promoting end-weight.

Apposition is a grammatical category discussed in most scholarly grammars, from Jespersen's A modern English grammar on historical principles to Quirk et al.'s A comprehensive grammar of the English language. But despite the fact that apposition has been widely discussed, it remains a category that is poorly understood. An investigation of Jespersen, Quirk et al., or any of the other sources that discuss apposition reveals numerous disagreements about how apposition should be defined and a wide variety of different kinds of constructions that are considered appositions. In this book, I attempt to clarify the confusion surrounding the category of apposition by both defining apposition and detailing its usage in computer corpora of spoken and written British and American English.

In Chapter I, I demonstrate the inadequacies of previous treatments of apposition and argue that apposition is a grammatical relation (like complementation and modification) realized by constructions having particular syntactic, semantic, and pragmatic characteristics. In subsequent chapters, I describe these linguistic characteristics of apposition in detail, using three computer corpora of English as the basis of my study: the Survey of English Usage Corpus of Written British English, the Brown University Standard Corpus of Present-day American English, and the London-Lund Corpus of Spoken British English. In Chapter 2,1 detail the syntactic characteristics of apposition, covering such topics as the various forms and functions that units in apposition have and the relationship between apposition and grammatical relations such as modification and complementation.

The two previous chapters detailed, respectively, the syntactic and semantic characteristics of units in apposition. In this chapter, the emphasis will be the pragmatic characteristics of apposition.

Thematically, apposition is a relation in which the second unit of the apposition either wholly or partially provides new information about the first unit (4.1), new information that can be optionally introduced by a marker of apposition (4.2). Because appositions have this thematic characteristic, they are better suited to some contexts than to others and were therefore distributed differently across the genres of the corpora. In general, appositions occurred most frequently in those genres (such as press reportage) in which there was a communicative need for new information to be provided about the first unit of the apposition (4.3). In addition, certain types of appositions occurred more frequently in some genres than in others (4.4) because certain kinds of appositions (e.g. those within the semantic class of paraphrase) are communicatively more necessary in some genres (e.g. spontaneous conversation) than in others (e.g. fiction).

The thematic characteristics of units in apposition

Semantically, apposition is a relation whose units are related by various kinds of equivalence (3.4): referentially, for instance, the two units of an apposition can exhibit exact equivalence (coreference) or partial equivalence (part/whole relations). Thematically, however, the units of an apposition are characterized by various degrees of non-equivalence: the second unit of an apposition is not equivalent to the first unit but rather either wholly or partially provides new information about the first unit.

In the previous chapter, the relation of apposition was discussed in terms of the syntactic characteristics that typify the units of an apposition. In this chapter, the focus shifts from the syntactic structure of units in apposition to their semantic structure.

The semantic relationship between units in apposition can be described in a variety of ways. The two units of an apposition can be characterized by the semantic relations existing between them, relations that are either referential (3.1.1) or non-referential (3.1.2). In addition, appositions can be classified into various semantic classes, depending upon whether the second unit of the apposition provides information about the first that is more specific (3.2.1), less specific (3.2.2), or equally specific (3.2.3). In appositions containing first units that are noun phrases, the apposition can be categorized according to whether or not the second unit of the apposition restricts the reference of the first unit (3.3). And finally, appositions form semantic gradients (3.4), with some appositions being semantically more appositional than others.

The semantic relations between units in apposition

Traditionally, apposition has been considered a relation consisting of two units that are coreferential (cf., for instance, Fries 1952:187, Hockett 1955:101, Quirk et al. 1985:1301, and Burton-Roberts 1975). However, as is argued in Meyer (1987a), not all constructions considered appositions in the literature consist of units that are coreferential, and to restrict the relation of apposition to only those constructions whose units are coreferential severely limits the number of constructions that can be admitted as appositions.