Agreeing and disagreeing with construction coercion

【Abstract】Coercion theory illustrates the conflict between constructional meaning and lexical meaning both semantically and grammatically. Are there still other functions of coercion besides semantic and pragmatic functions？ Do they function simultaneously or respectively？ This paper， by sorting out the features of coercion， points out that both a lexical item and a construction can be coerced to change their properties to fuse with each other. Meanwhile， coercion relations between constructions and lexical items and coercion relations between constructions are proposed.

【Key words】coercion； constructional meaning； interaction

【作者簡介】沈艷萍（1980- ），女，漢族，貴州麻江人，碩士研究生，講師，云南師范大學外國語學院，研究方向為：認知語言學及語言哲學。

1. Introduction

“Coercion” is a concept originating from computational linguistics， which is used to describe the conflict between constructional meaning and lexical meaning. Since Construction Grammar appeared， “coercion” has turned into an important term in construction grammar， and coercion principle has become one of the major principles used to discuss semantic relations.

2. Studies of construction coercion

Linguistic Gestalts， written by Lakoff （1977）， is the first work about construction grammar， in which he argues that meaning of a grammatical whole is not combined from its component parts， but that construction itself counts. This statement has implied the function of “coercion”. Coercion， as a concept， derives from the interactive relationship of lexicon meaning and construction meaning. Many scholars， such as Sag & Pollard （1991）， Swart （1998）， Moens & Steedman （1998）， Michaelis （1996，2005）， Goldberg （1995， 2006）， Croft （2009）， and Taylor （2003）， have stated the function of coercion principle， but they analyzed it from different angles. Sag & Pollard （1991） put forward “coercion principle” mainly in terms of lexicon coordination， without mentioning the issue of construction. Michaelis also defined “coercion principle” （also called it ‘Override Principle） as the following： If a lexical item is semantically incompatible with its syntactic context， the meaning of the lexical item confirms to the meaning of the structure in which it is embedded （Michaelis， 2004， p. 25）. Goldberg （1995） argued that constructions coerced lexical words （mainly verbs） semantically and pragmatically from the point of view that constructions could change the argument structures of verbs， and further concluded that “constructions coerce lexical items into having systematically related meanings” （Goldberg， 1995， p. 238）. WangYin （2011） gives a complete definition to “coercion”： when a word superimposes or is imbedded into a construction and is incompatible or in conflict with the construction semantically and pragmatically， the construction is always in control of the whole situation and imposes its meaning and usage on the word for accommodation. This is called coercion （Translated by the author）. Judging from statements about coercion， though varying each other to certain content， we can proclaim that they consider coercion as the process of constructions control of words as a whole both semantically and pragmatically.

Goldbergs researches of construction grammar are mainly conducted in two monographs： Constructions： A Construction Grammar approach to Argument Structure in 1995 and Constructions at Work： the Nature of Generalization in Language in 2006， and other related articles. However， most of the studies about coercion are in the first one in 1995， in which she elaborates this principle from the angle of interaction between construction and words. From her point of view， construction can coerce the main verb to change its argument structure and meaning. Construction has its meaning independent of verbs， and she illustrates the interaction relationship between construction and verbs.

3. Discussion

According to the previous review of coercion theory， “coercion” is a concept deriving from the interactive relationship of lexicon meaning and construction meaning. Although many scholars， such as Sag & Pollard （1991）， Swart （1998）， Moens & Steedman （1998）， Michaelis （1996，2005）， Goldberg （1995， 2006）， Croft （2009）， and Taylor （2003）， have stated the function of coercion principle from different angles， they mainly discussed it from perspectives of semantics and pragmatics. Furthermore， they treat coercion and inheritance theories independently as if these two theories have no relationships with each other. In fact， they are interrelated with each other. When coercion relation is discussed between two constructions， inheritance relation between them should also be considered. A lexical item can be seen as a member of a construction category. The fact that there exists conflict between a lexical item and construction with different information and features answers the question why a lexical item or a construction is coerced to change its properties to fuse with a construction. Coercion and inheritance interact with each other. Besides， the functions of coercion are not investigated adequately， with the following questions still left to be answered： Are there still other functions of coercion besides semantic and pragmatic functions？ Do they function simultaneously or respectively？ Meanwhile， coercion relations between constructions and lexical items have not been explored adequately， with the fact that coercion relation between constructions and lexical items has been favored by most scholars， with coercion relations between constructions being ignored. We propose the coercion relation between constructions， i.e. construction coercion on construction， and further propose multiple coercion-inheritance principle， which includes multiple coercion， multiple inheritance， and coercion-inheritance interaction.

In addition， Michaelis discusses construction coercion mainly from perspective of semantics， which can be judged from her definition of coercion： If a lexical item is semantically incompatible with its syntactic context， the meaning of the lexical item confirms to the meaning of the structure in which it is embedded. （Michaelis， 2005， p. 51）

4. Conclusion

To conclude， coercion and inheritance are interrelated with each other. When coercion relation is discussed between two constructions， inheritance relation between them should also be considered. Both a lexical item and a construction can be coerced to change their properties to fuse with each other. Besides， the functions of coercion are not investigated adequately， with the following questions still left to be answered： Are there still other functions of coercion besides semantic and pragmatic functions？ Do they function simultaneously or respectively？ Meanwhile， coercion relations between constructions and lexical items and coercion relations between constructions are proposed.

References：

[1]Croft，W.（2009）.Constructions and generalizations.Cognitive Linguistics，20（1），157-165.

[2]Goldberg，A.E.（1995）.Construction：A construction grammar approach to argument structure.Chicago&London;：The University of Chicago Press.

[3]Goldberg，A.E.（2006）.Constructions at work：The nature of generalization in language.Oxford：Oxford University Press.

[4]Lakoff，G.（1977）.Linguistic gestalts.In proceedings of the Thirteenth Regional Meeting of the Chicago Linguistic Society.

[5]Michaelis，L.A.&Lambrecht;，K.（1996）.Toward a construction-based theory of language function：The case of nominal extraposition.Language，72（2），215-247.

[6]Michaelis，L.A.（2004）.Type shifting in construction grammar：An integrated approach to aspectual coercion.Cognitive Linguistics，15（1），1-67.

[7]Michaelis，L.A.（2005）.Entity and event coercion in a symbolic theory of syntax.In J.?stman&M.Fried;（Eds.）.Construction grammars：Cognitive grounding and theoretical extension（pp.45-87）.Amsterdam&Philadelphia;：John Benjamins Publishing Company.

[8]Moens，M.&Steedman;，M.（1998）.Temporal ontology and temporal reference.Computational Linguistics，14（2），15-28.

[9]Sag，I.A.&Pollard;，C.（1991）.An integrated theory of complement control.Language，67（1），63-113.

[10]Swart，H.D.（1998）.Aspect and coercion.Natural Language&Linguistic; Theory，16（2），347-385.

[11]Tayor，J.R.（2003）.Linguistic categorization：Prototypes in linguistic theory.Beijing：Foreign Language Teaching and Research Press.

[12]王寅.構式語法研究（上卷）理論思索[M].上海：上海外語教育出版社，2011.

[13]王寅.構式語法研究（下卷）分析應用[M].上海：上海外語教育出版社，2011.