A cleft sentence is a complex sentence (one having a main clause and a dependent clause) that has a meaning that could be expressed by a simple sentence. Clefts typically put a particular constituent into focus. In spoken language, this focusing is often accompanied by a special intonation.
In English, a cleft sentence can be constructed as follows:
where it is a cleft pronoun and X is usually a noun phrase (although it can also be a prepositional phrase, and in some cases an adjectival or adverbial phrase). The focus is on X, or else on the subordinate clause or some element of it. For example:
Furthermore, one might also describe a cleft sentence as inverted. That is to say, it has its dependent clause in front of the main clause. So, rather than: Example:
the cleft would be:
English is very rich in cleft constructions. Below are examples of other types of clefts found in English, though the list is not exhaustive (see Lambrecht 2001 for a comprehensive survey, Collins 1991 for an in-depth analysis of it-clefts and wh-clefts in English, and Calude 2009 for an investigation of clefts in spoken English).
Traditional accounts of cleft structures classify these according to the elements involved following English-centric analyses (such as wh-words, the pronoun it, the quantifier all, and so on). This makes it difficult to conduct cross-linguistic investigations of clefts since these elements do not exist in all other languages, which has led to a proposal for a revision of existing cleft taxonomy (see Calude 2009).
However, not all languages are so rich in cleft types as English, and some employ other means for focusing specific constituents, such as topicalization, word order changes, focusing particles and so on (see Miller 1996). Cleftability in Language (2009) by Cheng Luo presents a cross-linguistic discussion of cleftability.
The role of the cleft pronoun (it in the case of English) is controversial, and some believe it to be referential, while others treat it as a dummy pronoun or empty element. The former analysis has come to be termed the "expletive" view, whereas the latter is referred to as the "extraposition" approach. Hedberg (2002) proposes a hybrid approach, combining ideas from both takes on the status of the cleft pronoun. She shows that it can have a range of scopes (from semantically void to full reference) depending on the context in which it is used.
Similarly controversial is the status of the subordinate clause, often termed the "cleft clause". While most would agree that the cleft clause in wh-clefts can be analysed as some kind of relative clause (free or fused or headless), there is disagreement as to the exact nature of the relative. Traditionally, the wh-word in a cleft such as What you need is a good holiday, pertaining to the relative What you need, is understood to be the first constituent of the relative clause, and to function as its head.
Bresnan and Grimshaw (1978) posit a different analysis. They suggest that the relative clause is headed (rather than headless), with wh-word being located outside the clause proper and functioning as its head. Miller (1996) also endorses this approach, citing cross-linguistic evidence that the wh-word functions as indefinite deictics.
The cleft clause debate gets more complex with it-clefts, where researchers struggle to even agree as to the type of clause that is involved: the traditionalists claim it to be a relative clause (Huddleston and Pullum 2002), while others reject this on the basis of a lack of noun phrase antecedent (Quirk et al. 1985, Sornicola 1988, Miller 1999), as exemplified below:
Finally, the last element of a cleft is the cleft constituent, which typically corresponds to the focus. As mentioned earlier, the focused part of a cleft is typically a noun phrase, but may in fact, turn up to be just about anything:
Clefts have been described as "equative" (Halliday 1976), "stative" (Delin and Oberlander 1995) and as "variable-value pairs", where the cleft constituent gives a variable expressed by the cleft clause (Herriman 2004, Declerck 1994, Halliday 1994). A major area of interest with regard to cleft constructions involves their information structure. The concept of "information structure" relates to the type of information encoded in a particular utterance, that can be one of these three:
The reason why information structure plays such an important role in the area of clefts is largely due to the fact that the organisation of information structure is tightly linked to the clefts' function as focusing tools used by speakers/writers to draw attention to salient parts of their message.
While it may be reasonable to assume that the variable of a cleft (that is, the material encoded by cleft clauses) may be typically GIVEN and its value (expressed by the cleft constituent) is NEW, it is not always so. Sometimes, neither element contains new information, as is in some demonstrative clefts, e.g., That is what I think and sometimes it is the cleft clause that contains the NEW part of the message, as in And that's when I got sick (Calude 2009). Finally, in some constructions, it is the equation between cleft clause and cleft constituent that brings about the newsworthy information, rather than any of the elements of the cleft themselves (Lambrecht 2001).
The shì ... (de) construction in Mandarin is used to produce the equivalents of cleft sentences. Also, certain constructions with relative clauses have been referred to as "pseudo-cleft" constructions. See Chinese grammar -> Cleft sentences for details.
There exist several constructions which play the role of cleft sentences. A very common resource is the adding of "es que" (time-dependent):
Another mechanism is the use of the identificating structure, or relative pronouns, "el/la que", "el/la cual" as well as the neuters: "lo que" and "lo cual".
Furthermore, one can also utilize "cuando" and "donde" when one wants to refer to "that" in a frame of time or place.
The X no wa (ga) construction in Japanese is frequently used to produce the equivalent of cleft sentences.
The construction is frequent in the Goidelic languages (Scottish Gaelic, Irish, and Manx), much more so than in English, and can be used in ways that would be ambiguous or ungrammatical in English: almost any element of a sentence can be clefted. This sometimes carries over into the local varieties of English (Highland English, Lowland Scots, Scottish English, Hiberno-English).
The following examples from Scottish Gaelic are based on the sentence "Chuala Iain an ceòl a-raoir", "Iain heard the music last night":
|"The (one who) bought the house was the woman."|
|"The (one to whom) Pedro had given money was Juan."|
|(or: "The (one who) was given money to by Pedro was Juan.")|
In the examples in (1) and (2), the foci are in bold. The remaining portions of the cleft sentences in (1) and (2) are noun phrases that contain headless relative clauses. (NB: Tagalog does not have an overt copula.)
|"Who bought the house?"|
|(or: "Who was the (one who) bought the house?")|
|"What did Pedro give to Juan?"|
|(or: "What was the (thing that) was given to Juan by Pedro?")|