Control (linguistics)
From Wikipedia, the free encyclopedia
| This article is in need of attention from an expert on the subject. Please help recruit one or improve this article yourself. See the talk page for details. Please consider using {{Expert-subject}} to associate this request with a WikiProject. (December 2008) |
| It has been suggested that this article or section be merged with Argument control. (Discuss) |
In linguistics, a control construction (also called equi-NP deletion) is a clause that contains a main clause (or matrix clause), the predicate of which has two complements — an embedded clause (or subordinate clause) complement and a nominal complement that acts as the semantic argument of the main clause and of the embedded clause. This nominal argument is the controller (or antecedent). Verbs that occur in control structures are control verbs (or equi verbs).
For example, in the English sentence
- George tried to escape
the control predicate is tried, a control verb. Its complements are the subject controller George and the embedded nonfinite clause to escape. George is the semantic agent argument of tried. The verb escape also has an agent argument, but in English this argument does not appear in the embedded clause (i.e. it is covert). Since this is a control structure, the covert argument of escape is the same as the overt argument of tried, which, in this sentence, is George.[1] (In other words, the covert argument is a bound variable.) The only semantic interpretation of this sentence is that the person trying and the person escaping are the same — both are George. An interpretation where George is trying and a different person is escaping is impossible.
One way to represent this is to use a bound variable in the embedded clause that is coreferential with the controller. This coreference can be show with co-indexation:
- [ Georgei tried [ Xi to escape EMBEDDED] MATRIX]
Here the bound variable (the covert subject of the embedded) is represented as X. It is also coindexed with the controller George (the subject of the main clause) with a subscript i indicating that they both point to the same referent. In other words, necessarily X = George. Semantically, the meaning of this sentence is George tried (George) to escape. The coindexing also shows that X cannot point to a different referent. It is impossible in this control sentence that X = Baron, for example, which would have the impossible meaning of George tried (Baron) to escape.
In English and many other languages, the embedded clause in a control structure are nonfinite. That is, the embedded clause either has an infinitive verb or a gerund verb. For example:
- Georgina expected [ to win the race ] (infinitive)
- Georgina remembered [ winning the race ] (gerund)
However, other languages have embedded clauses that are finite (i.e. the embedded verb is inflected for tense and agreement).
Contents |
[edit] Description
[edit] Object vs. subject control
In English and many other languages, control verbs may be classified as either subject control, meaning that it is their (syntactic) subject that is semantically shared, or object control, meaning that it is their (syntactic) object that is. For example, try is subject control (as seen above), while convince is object control: in "She convinced him to do that," him is syntactically only the direct object of convinced, but semantically both the patient of convinced and the agent of to do.
Note that a verb may be subject control in one sentence and object control in another. For example, in "She asked to be left alone," asked is subject control, while in "She asked him to leave her alone," asked is object control. Generally, when such a verb takes an object, it is implied that it is acting as an object control verb; for example, speakers of English would likely find a bit odd the sentence "She asked him to be allowed to stay" (with the intended meaning of "She asked him for permission to stay"), initially attempting to interpret it as "She asked of him that he be allowed to stay," but they would likely understand it after a moment.
[edit] Obligatory vs. non-obligatory control
Many linguists also make a distinction between obligatory and non-obligatory control (e.g. Hornstein 1999, Landau 1999). Roughly put, obligatory control (or OC) is distinguished from non-obligatory control (or NOC) on the basis that the implicit subject of the control complement (i.e. the element that is the subject of an infinitival clause like to read the book) obligatorily refers to an argument in the main clause. In non-obligatory control, the implicit subject, or the controllee, is free in its reference (e.g. in To read a book is awesome). Below are some examples of OC (a-c), and of NOC (d-f).
a. I prefer getting a raise.
b. I asked him to dance with penguins.
c. I wondered what to say.
d. Riding squirrels is difficult.
e. He said that visiting polar bears would be productive.
f. To sing off-key is easy.
More precisely, OC and NOC are distinguished on the basis of four criteria (a-d) (Hornstein 1999). These restrictions apply in OC cases, but not in instances of NOC.
a. The controller must c-command the controllee.
b. The controllee must have a local controller.
c. The infinitival clause allows only a sloppy reading and not a strict reading.
d. The infinitival clause allows only a de se reading.
[edit] Partial control
| This section requires expansion. |
[edit] Nonfinite and finite control
| This section requires expansion. |
- Hebrew
- Balkan
[edit] Backward control
| This section requires expansion. |
- Tsez
- Malagasy
[edit] Copy control
| This section requires expansion. |
[edit] Control vs. raising
| This section requires expansion. |
Control and raising constructions appear to be superficially similar in the surface forms of sentences are actually quite different.
As with many technical terms, both control verb and raising verb are used somewhat differently in different papers, partly because different linguistic theories may group verbs in somewhat different hierarchies.
[edit] Theoretical analyses
[edit] Equi-NP deletion
| This section requires expansion. |
[edit] Control Theory in Government and Binding
| This article may be confusing or unclear to readers. Please help clarify the article; suggestions may be found on the talk page. (December 2008) |
In the Government and binding theory framework the above-mentioned sentence is analysed as follows:
- He tried [CP PRO to do that]
The existence of PRO ([Spec, IP] in the embedded clause) is required by the Extended Projection Principle, which says that all clauses must have a subject. The main characteristic feature of PRO is that it must be ungoverned (the PRO Theorem). In terms of features it is [+anaphor, +pronominal] (see Binding (linguistics)). That is why such sentences as *I expect [John to meet PRO] are ungrammatical: PRO would be theta-governed by the verb.
The expletive test (using non-referential there or it) helps to distinguish between object-control and other types of verbs:
- persuade: an object-control verb taking a CP complement
- I persuaded John [CP PRO to read the book]
- *I persuaded [there to be a problem]
- *I persuaded [it to rain]
- expect: an ECM verb
- I expected [IP John to read the book]
- I expected [there to be a problem]
[edit] Control as movement
| This section requires expansion. |
[edit] External links
[edit] Notes
- ^ The obligatory coreference is a typical of control constructions. However, see the distinction between exhaustive obligatory control and non-obligatory control and partial control (sections: obligatory vs. non-obligatory control, partial control).
[edit] Bibliography
- Boeckx, Cedric; & Hornstein, Norbert. (2004). Movement under control. Linguistic Inquiry, 35 (3), 431-452.
- Chomsky, Noam. (1981). Lectures on government and binding. Dordrecht: Foris.
- Chomsky, Noam. (1986). Knowledge of language: Its nature, origin, and use. New York: Praeger.
- Chomsky, Noam; & Lasnik, Howard. (1993). The theory of principles and parameters. In J. Jacobs, A. von Stechow, W. Sternefeld, & T. Vennemann (Eds.), Syntax: An international handbook of contemporary research (pp. 506-569). Berlin: Walter de Gruyter.
- Hornstein, Norbert. (1999). Movement and control. Linguistic Inquiry, 30 (1), 69-96.
- Hornstein, Norbert; & Lightfoot, David. (1987). Predication and PRO. Language, 63, 23-52.
- Huddleston, Rodney; & Pullum, Geoffrey K. (Eds.). (2002). The Cambridge grammar of the English language. Cambridge: Cambridge University Press.
- Landau, Idan. (1999). Elements of control, (Doctoral dissertation, MIT).
- Landau, Idan. (2003). Movement out of control. Linguistic Inquiry, 34 (3), 471-498.
- Landau, Idan. (2004). The scale of finiteness and the calculus of control. Natural Language & Linguistic Theory, 22, 811-877.
- Landau, Idan. (2006). Severing the distribution of PRO from Case. Syntax, 9 (2), 153-170.
- Manzini, Maria Rita. (1983). On control and control theory. Linguistic Inquiry, 14, 421-446.
- Martin, Roger. (1996). A minimalist theory of PRO and control. (Doctoral dissertation, University of Connecticut).
- Petter, Marga. (1998). Getting PRO under control. The Hague: Holland Academic Graphics.
- Polinsky, Maria; & Potsdam, Eric. (2002). Backward control. Linguistic Inquiry, 33, 224-282.
- Rosenbaum, Peter. (1967). The grammar of English predicate complement constructions. Cambridge, MA: MIT Press.
- Rosenbaum, Peter. (1970). A principle governing deletion in English sentential complementation. In R. Jacobs & P. Rosenbaum (Eds.), Readings in English transformational grammar (pp. 20-29). Waltham, MA: Ginn.
- Williams, Edwin. (1980). Predication. Linguistic Inquiry, 11, 203-238.

