Interdisziplinäres Zentrum für Kognitive Sprachforschung

Links und Funktionen



Abstract Wang

What Distinguish Synonymous Constructions? A Corpus-based Study of Chinese lian…dou and lian…ye Constructions

While most versions of construction grammar (CxG) embrace the assumption that the semantic component subsumes the semantic and pragmatic properties of a construction (Croft and Cruse 2004: 258), it may not be clear what actually distinguishes synonymous constructions as separate constructions. Past research in CxG (e.g. Goldberg 1995) has shown that they may differ in their productivity and semantic constraints. Therefore, questions must be asked as to whether there are other semantic-pragmatic properties that may differentiate synonymous constructions. In this case, Chinese focalizaton constructions, lian…dou and lian…ye ‘even’, stand as an appropriate target of investigation, since they have been considered synonymous and virtually undifferentiated in discourse functions (Paris 1998). An example from Paris (1998: 142) is cited below in (1) (my boldface):

(1) lian Zhangsan ye/dou yao lai ma?

 even Zhangsan also/all want come F.P.

 ‘Does even Zhangsan want to come?’

Although dou, which denotes totality, and ye, which denotes association or resemblance, diverge in their meanings, their functions appear to merge when collocating with lian, as evidenced by the fact that most previous studies do not discriminate between lian…dou and lian…ye constructions. Our corpus analysis, however, shows that their primary distinction resides in the argument roles they tend to focalize (Du Bois 2003). A manual statistical analysis of the Academia Sinica Corpus of Modern Chinese reveals that among all the argument roles occurring as the focalized element which include S, A, O, V, and T(opic), 43.64% of lian…dou’s 676 tokens and 45.75% of lian…ye’s 400 tokens go to O and S respectively (see Table 1 below). Thus, the two synonymous constructions, although equivalent in meaning and discourse functions, display a crucial split preference in their argument structures. If any construction with unique discourse properties must be represented as an independent node in the constructional network to capture a speaker’s knowledge of their language (Croft and Cruse 2004: 263), we shall argue that CxG must recognize and account for distinction between constructions drawn with regard to preferred argument roles (Du Bois 2003). Furthermore, while this kind of subtle disparity in actual usage has not yet received much attention in the current CxG paradigm, we suggest that the usage-based model (Barlow and Kemmer 2000) may hold the key to an adequate explanation.


  S A O V T Total
Lian…dou 146/21.6% 143/21.15% 295/43.64% 17/2.51% 75/11.09% 676
Lian…ye 183/45.75% 83/20.75% 95/23.75% 2/0.5% 37/9.25% 400

Table 1. Argument roles occurring as focalized elements in lian ...dou and



Barlow, Michael, and Suzanne Kemmer (eds). 2000. Usage-based Models of Language. Stanford: Center for the Study of Language and Information.

Croft, William, and D. Alan Cruse. 2004. Cognitive Linguistics. Cambridge: Cambridge University Press.

Du Bois, John W. 2003. Argument structure: Grammar in use. Preferred Argument Structure: Grammar as Architecture for Function, ed. by John W. Du Bois, Lorraine E. Kumpf, and William J. Ashby, 11-60. Amsterdam: John Benjamins.

Goldberg, Adele E. 1995. Constructions: A Construction Grammar Approach to Argument Structure. Chicago: University of Chicago Press.

Paris, Marie-Claude. 1998. Focus operators and types of predication in Mandarin. Cahiers de Linguistique—Asie Orientale 27.2: 139-159.