Теорграмматика / Блох М.Я. Семенова Т.Н. Тимофеева С.В. - Theoretical English Grammar. Seminars. Практикум по теоретической грамматике английского языка - 2010

no literary examples at all in this clinical and compendious work. The examples are like lifeless membranes in a laboratory, lacking even the flexibility and unpredictability of living speech. Moreover, people suffering from the "split infinitive" syndrome, those concerned with the dramatic problems of taste, choice, and acceptability described by H, W. Fowler - battered ornaments, pairs and snares, sturdy indefensibles, and all the rest (to use his terminology) - are given little or no help. Such problems, it would appear, do not exist. The choice lies between the older grammars which cite evidence from Swift, Tennyson, and Conrad as if they were contemporaries writing in the same medium, and the quasi-scientific grammars of Randolph Quirk and his colleagues and adherents who seldom get beyond the factual-ity of utterances like "Their safe arrival in Cairo" and "Lobster New-burg is difficult to prepare".

Syntactic Change

In this book I have been much concerned with showing that linguistic rules and attitudes change as the centuries pass. It is self-evi- dent that the same principle applies to syntax. In Old English, an inflected language, customary but not obligatory rules affected the normal subject-verb-object rule: seo cwen beswdc pone cyning "the queen betrayed the king" could be changed to "pone cyning beswac seo cwen " without change of meaning. The endings unmistakably revealed the subject and object. In post-Conquest English the ordering of words can and normally does reverse the meaning. In Old English two negatives strengthen the negativity of a sentence (nads me nsefre gewunelic "it was never customary for me"). In post-Renaissance English one negative normally cancels the negativity of a second one. In Old English the title of a monarch or other person of rank normally followed the name (Alfred cyning), whereas of course the order is now reversed Queen Elizabeth. Old English had no distinctive future tense: the present tense was used to express future time: "gageon mmne winjeard, andic selle eowpset riht bip" "go into my vineyard, and I will give you what is right". The future tense came into being as the verbs sculan and willan lost their ancient power as finite verbs and turned into future auxiliaries. Old English had a present participial form but it ended in -ende or

(in some regions) -Me or -ande. The -ing form emerged after the Conquest from an array of disintegrating and jostling forms, with the process still not fully understood by scholars. Visser in the Second Half of Volume III of his Historical Syntax of the English Language (1973) devotes nearly 200 pages to the development of the second verb as a form in -ing, as in "I've so much enjoyed talking to you", "Have you tried shopping in the Berwick Market?", "He wouldn't have risked killing me", "What are you getting at?", "You are being silly", and so on. His examples are drawn from medieval chronicles and poems and stretch out in great historical swathes down to the works of Aldous Huxley and Kingsley Amis. No construction is everlastingly stable, no cherished rule remains unbroken. At any given time it is safe to assume that permissible patterns of syntax are ascertainable if one has the means of identifying and classifying them. Go back a century or so and the rules are radically different even if on the surface they appear to be the same. Go back two centuries and more and one must call for help from scholars with a particular knowledge of the rules and constraints of the time. It is risky without such help to read the works of any writer whose writings were published more than two centuries ago. And it is unhelpful when scholars yoke constructions together without regard to chronology, geography, type of writing, and social class. We still lack an authoritative grammar based on spoken and written British English of the period since 1945, let alone one that looks further afield. Also lacking is a systematic synchronic treatment of the syntax of (for example) Chaucer, Shakespeare, and Milton. It seems wrong that so much scholarly endeavour is devoted to the algebra and tree-diagrams of impenetrably complex modern syntactic problems when the language of some of our greatest writers remains inadequately analysed.

(pp. 149-158)

Questions:

1.What is the traditional approach to grammatical phenomena?

2.What new methods of grammatical analysis were introduced by N. Chomsky?

3.What proofs of constant syntactic change in English does R.B. Burchfield comment on?

References

Blokh M. Y. A Course in Theoretical English Grammar. - M., 2000. - P. 222229.

. . - M., 2000.

. : . - ., 1969.

. . -

. - ., 1971.

. .

. - ., 1984. .

. - .:

, 2001.- . 69-73.

. . - ., 1976.

., ., . -

. - ., 1981. - . 100-163.

. . . - , 1976.

. . - ., 1957.

. . -

. - ., 1981.

Burchfield R. The English Language. - London, 1985. - P. 149-158. Ilyish B.A. The Structure of Modern English. - L., 1977. - P. 171-182. Palmer F. R. Semantics. A New Outline. - Cambridge: CUP, 1977. - P. 94-100.

Seminar 10

ACTUAL DIVISION OF THE SENTENCE.

COMMUNICATIVE TYPES

OF SENTENCES

1.The basic principles of sentence division. Actual division of the sentence. The correlation of the "1" syntactic ("nominative") division and actual division of the sentence. The notion of theme and rheme. The notion of transition. The notions of topic and comment. Topicalization. The notion of presupposition.

2.Language means of expressing the theme.

3.Language means of expressing the rheme.

4.Actual division of sentences with non-finite forms of the verb. Construc tions with the double/triple rheme. Double theme-rheme construction.

5.Classification of sentences according to the purpose of communication: traditional classification, Ch. Fries' classification. Modern classification of communicative sentence types. The problem of exclamatory sentences. Actual division and communicative sentence types.

6.Constructions with mixed communicative features.

7.Classifications of speech acts (J. Austin, J.R. Searle). The basic notions of pragmatics. Context of situation.

1. The Main Principles of Actual Division of the Sentence

From the point of view of the actual division the sentence can be divided into two sections: thematic (theme) and rhematic (rheme). The theme expresses the starting point of communication; it means that it denotes an object or a phenomenon about which something is reported. The rheme expresses the basic informative part of the communication, emphasizing its contextually relevant centre. Between the theme and the rheme intermediary, transitional parts of the actual division can be placed, also known under the term "transition". Transitional parts of the sentence are characterized by different degrees of their informative value.

2. Language Means of Expressing the Theme and the Rheme

Language has special means to express the theme. They are the following: the definite article and definite pronominal determiners, a loose parenthesis introduced by the phrases "as to", "as for", and the direct word-order pattern.

In comparison with the language means used to express the theme, language has a richer arsenal of means to express the rheme because the rheme marks the informative focus of the sentence. To identify the rhematic elements in the utterance one can use a particular wordorder pattern together with a specific intonation contour, an emphatic construction with the pronoun "it", a contrastive complex, intensifying particles, the so-called "there-pattern", the indefinite article and indefinite pronominal determiners, ellipsis, and also special graphical means.

3. Actual Division and Communicative Sentence Types

The theory of actual division has proved fruitful in the study of the communicative properties of sentences. In particular, it has been demonstrated that each communicative type is distinguished by features which are revealed first and foremost in the nature of the rheme.

As a declarative sentence immediately expresses a proposition, its actual division pattern has a complete form, its rheme making up the centre of some statement.

As an imperative sentence does not directly express a proposition, its rheme represents the informative nucleus not of an explicit proposition, but of an inducement in which the thematic subject is

usually zeroed. If the inducement is emphatically addressed to the listener, or to the speaker himself, or to the third person, thematic subjects have an explicit form.

The differential feature of the actual division pattern of an interrogative sentence is determined by the fact that its rheme is informationally open because this type of sentence expresses an inquiry about information which the speaker does not possess. The function of the rheme in an interrogative sentence consists in marking the rhematic position in a response sentence, thus programming its content. Different types of questions are characterized by different types of rhemes.

The analysis of the actual division of communicative sentence types gives an additional proof of the "non-communicative" nature of the so-called purely exclamatory sentences (e.g. "Oh, I say!"): it shows that interjectional utterances of the type don't make up grammatically predicated sentences with their own informative perspective; in other words, they remain mere signals of emotions.

The actual division theory combined with the general theory of paradigmatic oppositions can reveal the true nature of intermediary predicative constructions distinguished by mixed communicative features. In particular, this kind of analysis helps identify a set of intermediary communicative sentence types, namely, the sentences which occupy an intermediary position between cardinal communicative sentence types.

Questions:

1.What are the main principles of the actual division of the sentence?

2.What sentence elements can be called "thematic"?

3.What language means mark the theme of the sentence?

4.What is understood by the rheme of the sentence?

5.What language means are used to express the rheme of the sentence?

6.In what do you see the connection of the actual division and the communi cative sentence types?

7.What actual division pattern is typical of the declarative sentence?

8.What actual division pattern characterizes the imperative sentence?

9.What kind of rheme is peculiar to the interrogative sentence?

10.In what way does the actual division help reveal the differential features of intermediary communicative sentence types?

I.Dwell upon the actual division of the sentences and the language means used to mark it.

MODEL: a) The time came for her to dance with Adams.

T2-»R2

This sentence represents a case of double theme-rheme construction:

T2->R2.

b) Asforla_ Falterpna, she had a natural and healthy contempt for the arts.

The antetheme "la Falterona" is introduced with the help of the phrase "as for"; the theme of the sentence is "she", the rheme is "had a natural and healthy contempt for the arts".

a)

1.I must take some definite actions tonight (Doyle).

2.I cannot allow the examination to be held if one of the papers has been tampered with (Doyle).

3.The situation must be faced (Doyle).

4."In that case, your Grace, since you have yourself stated that any unhappiness in your married life was caused by his presence, I would sug gest that you make such amends as you can do to the Duchess, and that you try to resume those relations which have been so unhappily inter rupted." "That also I have arranged." (Doyle)

5.He heard her singing in her snatchy fashion (Lawrence).

6."Teddilinks, light a fire, quick." (Lawrence)

7.Why don't you sew your sleeve up? (Lawrence)

8.With a little flash of triumph, she lifted a pair of pearl ear-rings from the small box (Lawrence).

9.The exterior of the building was a masterpiece of architecture, elegant and graceful (Sheldon).

b)

1.It was Mr. Eccles I particularly wanted to see (Christie).

2.Somebody ought to be getting rich. Somebody ought to be seen to be getting rich (Christie).

3.Baxter Dowes he knew and disliked (Lawrence).

4.For me to get up early was something like a deed.

5.I have never been told to come there to retype the papers.

6."How long have you lived in Hollowquay?" "Barely a month."

7."Well, that's all right. No need to give me a whole account of your literary triumphs in early youth." (Christie)

8.She remained clinging round his neck (Lawrence).

9.Sunday was a holiday for Dad, not for Mum (Leacock).

c)

1.Triumphant, that's what she was at the prospect (Christie).

2.Aunt Ada was silent until Tuppence had gone out of the door with Miss Packard and Tommy followed her. "Come back, you" said Aunt Ada, raising her voice. "I know you perfectly. You're Thomas." (Christie)

3."Red-haired you used to be. Carrots, that's the colour your hair was."

(Christie)

4.Desperately you want something to do to amuse yourself so you try on some public character and see what it feels like when you are it (Christie).

5."You'd be surprised the way she got to know things. Sharp as a needle, she was." (Christie)

6."Miss Fanshawe was never dull. Grand stories she'd tell you of the old days." (Christie)

7.That was when he saw Ginelli wasn't in the car (King).

8.The pie sat on the seat beside him, pulsing, warm (King).

9.It's the people who aren't scared who die young (King).

d)

1.It was then that Constantin Demiris entered Melina Labrou's life (Shel don).

2.Modern hotels and office buildings were everywhere amid the timeless ruins, an exotic mixture of the past and present (Sheldon).

3.In the beginning, she had asked questions (Sheldon).

4.The Blue House was opened to special patrols only (Sheldon).

5.Again he wasn't sure - rather vague, the whole thing (Christie).

6."Isn't it a long time after to be looking for her?" (Christie)

7.Apparently he only heard there was a child quite recently (Christie).

8."She's a striking looking woman, isn't she? Interesting, I always think. Very interesting." (Christie)

9.Who does it actually belong to nowl (Christie)

II.Define the communicative sentence type, dwell on the actual division of the following sentences. Define the speech-act features of these sentences.

MODEL: "What have you got?" "His book. "

___ ^..guoii \jjctmmar

The first sentence is interrogative and its rheme "what have ... got" is infbrmationally open. As it is a special question, the nucleus of inquiry is marked by the interrogative pronoun which is the rhematic peal. The theme of the sentence is "you". The second sentence is elliptical and rhematic. The rhematic peak of the answer ("His book") is the reverse substitute of the interrogative pronoun. As the two sentences make up a thematic unity, the theme in the answer is zeroed.

a)

1."I'd like to know what you think of her. Go and see Dr. Rose first." (Christie)

2.Why not walk down to the village after tea? (Christie)

3."I would strongly advise you, Mr. Worthing, to try and acquire some relations as soon as possible, and to make a definite effort to produce at any rate one parent, of either sex, before season is quite over." (Wilde)

4.Suppose you fetch your bricks and build a nice house, or an engine (Christie).

5."The Duke is greatly agitated - and as to me, you have seen yourself the state of nervous prostration to which the suspense and the responsi bility have reduced me." (Doyle)

6."Mr. Holmes, if ever you put forward your full powers, I implore you to do so now." (Doyle)

7."I beg you, Mr. Holmes, to do what you can." (Doyle)

8."You will kindly close the door," said Holmes. "Now, Banister, will you please tell us the truth about yesterday's incident?" (Doyle)

9."Would you please remain in the room? Stand over there near the bed room door. Now, Soames, I am going to ask you to have the great kindness to go up to the room of young Gilchrist, and to ask him to step down into yours." (Doyle)

10 Can the leopard change his spots?

b)

1."I wonder why you never answered her letter." (Maugham)

2.Over the breakfast she grew serious (Lawrence).

3."We can be perfectly frank with each other. We want to know, Mr. Gilchrist, how you, an honorable man, ever came to commit such an action as that of yesterday?" (Doyle)

4."You will show these gentlemen out, Mrs. Hudson, and kindly send the boy with this telegram. He is to pay a five-shilling reply." (Doyle)

5."I wish you, Mr. Holmes, to come to Mackleton with me by the next train." (Doyle)

6."You will kindly sign that paper, Mr. Sandeford, in the presence of

these witnesses." (Doyle)

7."I suppose you haven't such a thing as a carriage in your stables?" (Doyle)

8."Tell us about your last talk with Dr. Wilbour." (Schrieber)

9.Paul felt as if his eyes were coming very wide open. Wasn't he to take Clara's fulminations so seriously, after all? (Lawrence)

10."I hope you won't let him keep the stocking." "You are not going to tell me everything I shall do, and everything I shan't." (Lawrence)

c)

1.Oh, Mr. Holmes, you must save him - you must save him! I tell you

'that you must save him! (Doyle)

2."Mrs. Hudson," I said, going out to her, "I want you to pack my bags, please." (Hardwick)

3.I suppose you were in a convent? (Hemingway)

4."Listen," George said to Nick. "You better go see Ole Anderson." (Hem ingway)

5.Thanks for coming to tell me about it (Hemingway).

6.Don't you want me to go and see the police? (Hemingway)

7."Why don't you try to go to sleep?" (Hemingway)

8."Don't be melodramatic, Harry, please," she said (Hemingway).

9."How do you feel?" she said. "All right." (Hemingway)

10. "Who likes to be abused?" (Sheldon)

d)

1."You don't want to go mixing yourself up in things that are no business of yours -" "There's nothing to be mixed up in according to you," said Tuppence. "So you needn't worry at all." (Christie)

2."And there are people who are terribly unhappy, who can't help being unhappy. But what else is one to do, Tommy?" "What can anyone do except be as careful as possible." (Christie)

3."No, I don't want you to go. After all, the last time, remember how frightfully rude she was to you?" (Christie)

4.Would you like to come up now? (Christie)

5."I'll put them (roses) in a vase for you," said Miss Packard. "You won't do anything of the kind." (Christie)

6."You go away," added Aunt Ada as a kind of postscript, waving her hand towards Tuppence who was hesitating in the doorway (Christie).

7."I hope they brought you some coffee?" (Christie)

8."The old lady I was talking to," said Tuppence. "Mrs. Lancaster, I think she said her name was?" (Christie)

i18 - 3548

9."Can you tell me a little more about her, who her relations were, and how she came to come here?" (Christie)

10. "God help the home of the aged that you go to. You'll be Cleopatra most of the time, I expect." (Christie)

Selected Reader

1.

Dijk T.A. van.

Text and Context. Explorations in the Semantics and Pragmatics of Discourse

Topic, Comment, Focus, and the Functions in Discourse

In this and the previous chapters the notions TOPIC OF CONVERSATION and TOPIC OF DISCOURSE have been used in order to define connectedness of sentences and coherence of discourse. It has been assumed that such topics are to be defined in terms of propositions, sets of propositions and/or propositions entailed by such sets. At the level of sentence structure another notion of TOPIC has been used in recent linguistics, often in combination with the notions of COMMENT and FOCUS. In that research a sentence may be assigned, besides its usual syntactic and semantic structures, a binary TOPIC-COMMENT STRUCTURE. The definition of such structures is specified both in semantic and pragmatic terms of information and information distribution in sentences and their canonical or transformed syntactic and morpho-phonological expression. The intuitive idea behind the assignment of such structures in a grammar is that in a sentence we may distinguish between what is being said (asserted, asked, promised...) and what is being said "about" it, a distinction closely parallel to the classical SUBJECT-PREDICATE distinction in philosophy and logic. Thus in a sentence like: [42] John is rich.

the part "John" is topic because it denotes the thing about which something is asserted, whereas "is rich" is the comment or focus of the sentence, denoting the thing (property) said about (predicated of) John. This comment may be much more complex as in sentences like: [43] John inherited a large estate from his old uncle who lived in Australia.

where John could be assigned the topic function and the rest of the sentence would be assigned the comment function.

Now, although our linguistic intuitions about the topic-comment distinction may be correct, the theoretical reconstruction is by no means straightforward. Confusion about the levels of description and about their appropriate definition is widespread in the literature. Some of the questions arising are, for example, the following:

(i) is the topic-comment distinction to be defined in syntactic, semantic or pragmatic terms, i.e. do these terms denote parts or functions of syntactic structures of sentences, of meaning or reference of propositions, or of contextual structures of speech acts, knowledge and information transmission? (ii) do all sentences have such a structure, and by what explicit

rules and procedures can topic and comment be assigned? (iii) do sentences have topic-comment structure independent of text structure and/or of their use in communicative contexts? In other words: can the "same" sentence have different topic-comment structure in different (con-)texts? (iv) what are the relationships to notions such as "subject" (grammatical, logical, psychological) and "predicate", presupposition and assertion, etc.?

(v)which grammatical, in particular morpho-phonological and syntactic, structures are systematically related to the topic and comment functions?

(vi)what are the relationships to notions like topic of a conversation or of a discourse as used semi-technically above? These questions cannot possibly be answered here in a systematic and explicit way. Some of them relate to characteristic properties of sentence structure which are outside the scope of this book. Our attention, therefore, will be focused upon the role of the topic-comment distinction in the account of discourse coherence.

Seminars on Theoretical English Grammar

However, some preliminary remarks about the theoretical status of topic and comment are necessary. From sentences such as [42] and [43] it seems as if the topic of a sentence coincides with, or is expressed by, the subject of the sentence, which in turn is normally associated with the left-most (or first) noun phrase of the sentence, as also in:

[44] The estate John has inherited from his rich uncle is in Australia. where the topic is expressed by the complex noun phrase. The comment, thus, would in that case be related to the predicate, or the predicate phrase, of the sentence. This general, informally formulated, rule holds for what could be called the NORMAL ORDERING of sentences in English, but not for sentences such as: [45] London is a town I like! [46] No, Peter has stolen the book.

where the first noun phrases have particular stress. For such sentences the grammatical subject or the first noun phrase does not carry the topic function: the first sentence is not about London but about towns I like, the second not about Peter but about someone who has stolen a book, intuitively speaking, whereas it is asserted that London and Peter are individuals satisfying the particular property or relation, respectively. That is, comments are normally in second (predicate) position or in positions with particular stress. In the latter case, the cleft sentence construction (it was ... who/which ...) may also be used to make comments out of categories with topic function. By particular stress assignment or cleft sentences, nearly any grammatical category can thus be assigned comment function, the rest of the sentence becoming topic:

[47]a: Harry paid for the book with a ten-dollar bill.

b:Harry paid for the book with a ten-dollar bill.

c:Harry paid for the book with a ten-dollar bill. and so on for the major categories (and in some cases also for prefixes, suffixes, prepositions, articles, demonstratives, etc.).

Without giving a more precise analysis and syntactic description of these examples, it will be assumed that the notions of topic and comment cannot possibly coincide with or be identical to particular syntactic categories, and that they must at least have a semantic status. This semantic status most clearly manifests itself in a further

Seminar 10. Actual Division of the Sentence ...

277

analysis of the "intuitions" referred to above: a topic is some function determining about which item something is being said. Similarly, a topic is often associated with what is "already known" (to the hearer) in some context of conversation, or what is "presupposed" (to be identified) by some sentence. The comment, then, associates with what is "unknown" (to the hearer) and asserted. An explication of these terms would have to be framed in a referential semantics and a pragmatic component.

The link between topic and presupposition in the given examples shows in the fact that, for instance, [47]a presupposes the proposition "Someone paid for the book with a ten-dollar bill", and [47]c presupposes "Harry paid for something with a ten-dollar bill", where it is asserted that the variables "someone" and "something" are identical with "Harry" and "the book", respectively. Note also that comments do not simply denote "unknown" individuals (objects, properties, relations or facts): both Harry and the book are "known" in the given examples: they are identified by the hearer (the speaker uses, characteristically, the definite article in the phrase the book). It is only unknown that Harry and the book have the specific (complex) property referred to.

By examining the semantic functions of normal sentence orderings or of stress distribution, we may often decide which sentence part expresses the topic and which part expresses the comment. This is less easy in the normal form of [47]a-c:

[47] Harry paid for the book with a ten-dollar bill.

It is not at all obvious whether this sentence is about Harry, about the book, or even about both, especially since both referents are "known". Could a sentence have two topics or should we perhaps speak of one compound topic, e.g. the ordered pair ("Harry", "the book") of which it is asserted that the first bought the second with a ten-dollar bill?

A typical test for establishing the topic-comment structure of sentences is to use preceding questions. If [47] is used as an answer to the question

[48] What did Harry do?

we may conclude that "Harry" or "Harry did something" is the topic of [47]. If the question were:

Seminars on Theoretical English Grammar

[49] What happened to the book?

it would be "the book" which would be the topic. Similarly, after a question like:

[50] What did Harry do with the book?

the ordered pair ("Harry", "the book") would be the topic. What is being established by questions can be established by PRECEDING DISCOURSE in general:

[51] At last Harry found the book he wanted to give Laura as a present. He paid for it with a ten-dollar bill.

Characteristically, noun phrases with topic function may then, or must be, pronominalized. Thus topic can be associated with the logical category of BOUND VARIABLES, ranging over both individuals and properties or relations. Less strictly speaking, it may be said that topics are those elements of a sentence which are BOUND by previous text or context. We should therefore investigate how topic-comment structure is to be determined relative to (con-)textual structure. 6.3

In order to understand the topic-comment articulation of sentences and their (con-)textual dependence, some remarks are necessary about the COGNITIVE BASIS OF INFORMATION PROCESSING in communicative contexts.

As will be shown in detail in the next part of this book, sentences (discourses) are uttered within the framework of specific speech acts and speech interaction. Thus, one of the purposes of the act of asserting a proposition is that the hearer be informed about a certain matter. This information increase is an enlargement or other change in his set of knowledge and beliefs, brought about by understanding of the meaning of the perceived utterance. The details of the actions involved here are less important for the moment. The point is that all "new information" is usually integrated into information already known. Thus, when I say that Peter is ill, it is assumed that my speech participant already "knows" Peter, i.e. knows that Peter exists, and knows his main properties. In this case, general or specific knowledge about Peter is "enriched" with the proposition "that he is ill (now)", to be attached to the complex "Peter" concept already present in the hearer's knowledge.

Now, the topic of a sentence has the particular cognitive function of selecting a unit of information or concept from knowledge. This may be a more general concept (like love or renting a flat) or an individual concept (Peter, this particular book, etc.). In the latter case, the individual referred to may already have been "introduced" into the context of communication either by direct action or perception of certain objects (That chair must be painted red) or by previous sentences of the discourse. In such a way many objects may be introduced into the context, and for each sentence it must be established which of these objects is (again) "picked up", i.e. referred to, in order to make a statement about it.

Cognitively, this "topicalization" of certain phrases is probably a process whereby knowledge of certain individuals is "foregrounded", i.e. taken from long-term memory stores to some working memory, in which the established information may be combined with the incoming new information.

This new information, usually called the "COMMENT" or also the "FOCUS" of the sentence, may be in various forms: it may assign a general or particular property to a known and identified individual, or a relation between individuals of which one or more are known (Peter met a girl. He kissed her), or the instantiation by one or more individuals of a known property or relation (Peter hasn 't committed the murder), or the assignment of various higher level properties or operators to events or propositions (The robbery had been planned cleverly. Your appearance was really unexpected, you know). From these assumptions it follows that in principle any phrase of a sentence may express topic function, or even several, discontinuous phrases like subject noun phrase and (in-)direct object noun phrase.

6.4

This is roughly the general theoretical basis for the topic-com- ment articulation in natural language: it is mainly a result of the constraints of effective information processing. Now, what are the implications for the structure and interpretation of discourse?

The first point to be made here is that, according to the principles adopted, certain sentences beginning a discourse or a section of discourse (e.g. a paragraph) may not always have a topic, viz in those

cases where no individual object or property known to the hearer is selected for "comment", as in:

[52] A man was walking slowly along a beach. Here, individuals (person, place) and a relation are introduced at the same time. Although, intuitively, we might say that this sentence is "about" a man, according to the canonical topic-comment mapping onto the subject-predicate structure of the sentence, there is, formally speaking, no topic in [52] but topic introduction. In cognitive terms: the hearer's knowledge "slate" is still clean with respect to a topic of conversation. Note, however, that sentences like [52] are rather unusual, and occur more in literary narrative than in everyday, natural narratives, where we would have something like: [53]a: This afternoon a strange man came to my office (...) Again, we could speak of topic introduction, but there is already established knowledge (time: a specific afternoon, determined by time of context of communication, and place: a particular, known, office), which is formally the topic of [53]a. In other words, [53]a is not primarily about a strange guy, but rather about what happened this afternoon, to me, in my office. We see that the notion of ABOUT-NESS is not very precise, and, at least for sentences, not always de-cidable. A sentence like [52] may be about a man, his walk or about a beach, or about all of them. More in general, aboutness should be established in (con-)textual terms, perhaps in such a way that a discourse or a passage of the discourse is about something if this "something" is referred to by most phrases with topic function. In this case, however, we no longer deal with the topic of a sentence but with a TOPIC OF DISCOURSE or a TOPIC OF CONVERSATION. We here find ourselves at a more global level of discourse description, to be discussed in the next chapter. Such a topic may be "a strange man" even if in the individual sentences the topics may be "his cigarette", "his trousers", "I", etc., i.e. those referring phrases of which the referents are associated with the strange man. It will appear, however, that aboutness at this more global level is again ambiguous: a story may be about Romeo, about Juliet, about both, about a specific (forbidden or impossible) love or about certain political structures in the middle ages. Often, however, the "aboutness" pertains to a given individual object or person, if most properties and relations are as-

I signed to one permanent referent or to those objects/persons introduced in relation to it.

Topics are established not only with respect to explicit previous information but also with respect to implicit information as defined above. If we continued [53]a with a sentence like

[53]b: His nose was nearly purple (...)

the phrase his nose would be assigned topic function even if its referent has not been explicitly referred to before. However, the concept "man" entails the meaning postulates of being a human adult male and of having a nose. The proposition "a has a nose" is therefore implied by [53]a, referred to definitely (by possessive pronoun) in [53]b, and therefore implicit. In cognitive terms: the hearer already knows that if there is a man he also has a nose. Topics, thus, may be expressed by any phrase referring to an individual (con-)textually identified by the hearer, but also by all other expressions for individuals or properties belonging to what may be called the EPISTEMIC RANGE of that object.

In this semi-formal framework, topic function may be related to any object of previous models, also to facts or possible worlds. This would explain the notorious difficulty of assigning topic-comment structure to such sentences

[54]// is hot.

[55]It was raining.

It would express a topic by referring to some particular time-place or world. Similarly, in sentences like [52] which have no apparent topic part, but in which some particular real, fictitious, or narrated world is taken as the (implicit) topic. In fact, the sentence specifies a number of properties of such a world (that there is a man, that the man is walking, that the man/his walking is slow, and that the walking takes place along a beach, in the past).

Note that this textual approach to the problem of sentential topics does not always guarantee that the subject of a sentence is automatically the topic of that sentence, even in normal ordering. After the question "What happened to the jewels?", we may have

[56] They were stolen by a customer.

where the topic function is indeed assigned to the first noun phrase (subject), but we may also have a sentence like