[2] The term was first used in the semantic literature in 1988 by Marc Moens and Mark Steedman, who adopted it due to its "loose analogy with type-coercion in programming languages.”[3] In his written framework of the generative lexicon (a formal compositional approach to lexical semantics), Pustejovsky (1995:111) defines coercion as "a semantic operation that converts an argument to the type which is expected by a function, where it would otherwise result in a type error."
[4] A commonly used example of complement coercion is the sentence "I began the book.” The phrase "I began" is assumed to be a selector which requires the following complement to denote an event, but "the book" denotes a noun phrase, not an event.
"[4] An example of aspectual coercion involving temporal connectives is "Let's leave after dessert" (Pustejovsky 1995:230).
[5] Coercion is a well-discussed topic in the field of linguistics, especially in semantics and construction grammar.
An example is Yao-Ying Lai’s 2017 study on the effects of coercion on mental processing; results showed that phrases involving aspectual words (such as “start”) required longer reading times to understand than did phrases with psychological words (such as “enjoy” and “love”).