It should be clear by now that all the components of a text — from morphology to syntax, from semantics to pragmatics, from prosody in the spoken language to punctuation in a written text — interact with all the extra-textual elements (the situational context and discourse strategies) in order to perform their communicative function.

This paragraph analyses both the formal and the semantic-pragmatic aspects of a syntactic device that Otto Jespersen, a Danish philologist, called the cleft sentence. In order to give salience to the information focus, a sentence like

can be transformed into

to highlight the fact that "the wife (not anyone else) is the deciding person." This is parallel to Italian Io prediligo il barolo that can be changed into E' il barolo che prediligo (not another wine).

The sentence is "cleft" (i.e. broken or cut) into two parts: the main clause begins with It + be and includes the focal element; the rest is embedded in a relative clause that, like all defining relative clauses, normally begins with that, never preceded by a comma; when the relative pronoun is not the subject of the relative clause, it can be omitted and a contact clause results:

instead of

Here, again, we have a different focus than in the simple, non-cleft sentence

and the meaning of the cleft sentence is that

We find a that-clause in example 1 and a contact-clause in example 2. In all cases, the simple sentence with one verb becomes a complex sentence with two verbs: to be in the main clause, preceded by anticipatory it, and the original verb in the relative clause. Anticipatory it is obligatory in English, where a subject is always required before the verb, and it is responsible for the 3^rd person singular form of to be. This is the syntactic function of it. Its semantic function is to point at the part of the sentence to which the reader's or listener's attention is drawn.

Jespersen remarks that cleft sentences are less frequent in those languages that are not so rigid as English in keeping to the SVO order; he adds that the disadvantage of a rather rigid word-order is counterbalanced by the use of such syntactic variations as the cleft sentence.

A pseudo-cleft sentence is obtained, instead, when a noun phrase is fronted (that is, shifted towards the beginning of a sentence) and is given the form of a relative clause (or wh- clause), followed by a main clause with to be. In Italian, starting from the sentence A Giovanni piace il barolo, the use of a pseudo-cleft construction may give either Ciò che a Giovanni piace, è il barolo or Colui a cui piace il barolo, è Giovanni.

This type of transformation is called "pseudo-cleft" because it produces an improper form of subordination, a false separation between two clauses that really expand one simple sentence.

It should be emphasised that the origin and the justification for the development of cleft and pseudo-cleft sentences are linked to the speakers' communicative needs: it is sometimes important to select one of the components of the sentence and foreground it, as the following examples also show:

Syntax provides a variety of solutions, each of them being a marked variant of a simple sentence, so that emphasis can be worked into the written language. In spoken English, it is prosody (in particular, stress and intonation) that mostly conveys emphasis.

Another variant, fronting, that was introduced in paragraph 1.7, is a marked construction whose chief purpose is to give salience to one element in the sentence, an element that is placed before the subject — in spite of the fact that the subject is the primary constituent of an utterance.

Example 4a is characterised by the object in the front position. This shows that the grammatical subject and the informational topic in a sentence do not always coincide. However, this device is rhetorically much less flexible than a cleft sentence.

Starting from the same example, let us shift the rheme, the new information in sentence 4, to the left, but let us begin the sentence with it is — the fronting has been turned into a cleft:

And here is what we get if we adopt the pseudo-cleft construction in order to satisfy our communication needs:

Cleft and pseudo-cleft sentences have been called it-clefts and wh-clefts, respectively, by other linguists.(14) These constructions give salience to the structure of information and the sequence is not so "heavy" as it is with fronting. In a sequence like 4b, it is the given, the information that is known or that can be inferred by the hearer, that becomes the new information at the discourse level. (15) The pseudo-cleft 4c is marked by the relative clause filling the position of the subject, followed by the informative climax in the subject complement.

The flexibility of cleft sentences in English is shown by the way its several components can easily be given salience:

Salience on the subject:

Salience on the object:

Salience on the adverbial of time:

Salience on the adverbial of place:

In spoken English, all these examples are pronounced with emphasis on the head of the noun phrase in italics. This means that the tone rises on John, suit, night or dance and then keeps falling as far as the end of the sentence. It is another instance of the general rule stating that the rheme is always marked by intonation.

The dynamics of "thematisation" — which have just been analysed at the sentence level — are mirrored by the dynamics of information processing at the macro-text level. Introductory sentences often inform about the topic/theme of the whole passage and are followed by rhematic sentences; however, the reverse order can also be found. At all levels of text production — from sentences to paragraphs, chapters and whole books —speakers or writers convey their communicative aim by means of devices and techniques that are adequate in terms of the language used and of the context. In particular, the presence of cleft or pseudo-cleft sentences in discourse often performs a thematising role. (16)

Above (paragraph 1.6, footnote 8) a brief mention was made of the terms that are used to refer to how information is organised in utterances. It is not just a matter of terminology — given/topic/theme on one hand and new/comment/rheme on the other — but an important issue because these concepts are essential in order to understand the dynamics of discourse. The analysis of how language is organised as a tool for communication has introduced the concepts of theme and rheme. Mathesius, a linguist of the Prague School, used the terms base and nucleus; when Firbas translated Mathesius and used theme and rheme that lend themselves to being turned into the adjectives thematic and rhematic. Other Anglo-American linguists use different terms: topic/comment; Halliday, an Englishman, adopts the terms given/new.

When the information is conveyed by simple unmarked structures, the sequence theme + rheme corresponds to the syntactic sequence subject + predicate: the subject/theme is the shared information, i.e. the information that is not new for the reader or listener because it can be retrieved either anaphorically in the co-text or pragmatically from the situation, whereas the predicate/rheme is the new information that introduces the notion of change that is inherent to textuality. (17)

The organisation of information reveals how thought proceeds and develops in a text, modifying the relationship between the people involved and the world referred to in the text. When we detect the linguistic-textual correlates that bring about all the changes in a text we learn how to trace the boundaries of the smaller meaningful units. Such boundaries are difficult to define because the deep sequence of a text is not always parallel to the surface sequence. The processes of emphasis and salience vary depending on the communicative aims of the speaker/writer and the consequent ways in which the language is manipulated.

Whatever the syntactic variant may be, in English the element placed at the beginning of the sentence shows how the utterance is to be interpreted — it places the communicative aim in its context (18) and posits it as the theme. If we focus our attention on the thematic part of the information, we still do not know what the speaker/writer wants to say about the theme, the subject he or she is going to deal with. In a text, discourse or message, there is a constant retrieval of known information and an anticipation of new information — the sequences carrying all this information are linked to the co-text and context to make up one connected whole. In order to break a text into meaningful units, a good starting point would be the detection of the rheme — that is what is said or written about the topic, what is new, what motivates the text.

Below are three more examples of the cleft sentences dealt with in this paragraph: cleft sentences proper introduced by it + be followed by what the speaker/writer wants to emphasise, and pseudo-cleft sentences, where the information focus is placed in the second part of the sentence:

The syntactic organisation of the text and the subdivision into meaningful units are made coherent by how the information is organised. For a better understanding of how information is organised in cleft and pseudo cleft sentences, here follow two examples of implicit relative clauses in Italian:

Depending of how the speaker/writer organises the message, the clause with the verb essere may either precede (as in 9a) or follow (as in 9b) the implicit relative clause.

Cleft and pseudo-cleft sentences became fairly common in the 18^th century, under the influence of French, and now they are very frequent in spoken and written English at all levels.