UFJF - Universidade Federal de Juiz de Fora

The Saussurean sign revisited: Accounting for form-meaning mismatches in Construction Grammar

You are at: Home > Theme sessions > The Saussurean sign revisited: Accounting for form-meaning mismatches in Construction Grammar


Nikos Koutsoukos | Université Catholique de Louvain

Kristel Van Goethem | F.R.S.-FNRS & Université Catholique de Louvain 

Hendrik De Smet | Katholieke Universiteit Leuven


F. de Saussure (1916/1995) defined the linguistic sign as an association of ‘signifier’ (signifiant, i.e. ‘l’image acoustique’) and ‘signified’ (signifié, i.e. an abstract concept). The language as a whole he envisaged as a system of functional oppositions holding between signs. Since then it has often been tacitly assumed that there exists a symmetry in the relation between the signifier and the signified of any given sign, the so-called one-Form-one-Meaning Principle. However, it should not be taken for granted that all linguistic signs display this symmetric relation. Beard (1998) dubbed this the Correspondence Fallacy. Form-meaning mismatches (or asymmetries) are omnipresent in language (Francis & Michaelis 2002). Three types of form-meaning mismatch can be distinguished: 

Mismatch type 1: Many forms correspond to one function/meaning: Well-described examples of this mismatch usually include synonymous lexemes (such as redouter, craindre, avoir peur ‘to be afraid of’), for which Saussure (1916/1995: 160) already claimed that they “n’ont de valeur propre que par leur opposition”. This mismatch is related to the notion of ‘competition’ between the linguistic signs, in line with the structuralist concept of language as “un système où tout se tient”. A similar type of mismatch can sometimes be observed within a single (morphological or syntactic) construction, e.g. in the form of multiple exponence in inflectional morphology (Harris 2009), affix pleonasm in derivational morphology (see Lehmann 2005) or redundancy in syntax (see for example Croft (2000: 136) on paratactic negation). 

Mismatch type 2: One form corresponds to many functions/meanings: Evident examples of this category are polysemous/polyfunctional affixes, lexemes or syntactic constructions. For example, English to-infinitives have been analyzed as polysemous by Wierzbicka (1988). More recently, Colleman & De Clerck (2011) have analyzed the English ditransitive construction as polysemous. This type of mismatch also applies to conversion, in which one form is used to express different functions (e.g. the bridge/ to bridge) (e.g. Valera 2014). 

Mismatch type 3: The meaning of the syntagm cannot be derived from its signs: In morphology this type of mismatch applies for instance to exocentric compounds: in Italian portalettere ‘lit. carry-letters, postman’ or French tire-bouchon ‘lit. pull-cork, cork screw’, the agentive or instrumental meaning cannot be derived from the meaning of the formal constituents of the compound. In syntax, unexpected functions of verbs or nouns are hard to account for when adopting a word-level analysis, e.g. transitivity of the verb sneeze in She sneezed the napkin off the table (Goldberg 1995) or the mass noun interpretation of cat in There’s cat all over the place (Michaelis 2003). 

Bazell (2004 [1949]) argued that these phenomena have not been sufficiently addressed in the structuralist tradition. One century after the first publication of Saussure’s Cours de linguistique générale (1916), this workshop aims to revisit the nature of the linguistic sign and more specifically form-meaning mismatches from a constructional perspective. In the framework of Construction Grammar, the Saussurean sign as a conventionalized form-meaning mapping has been extended and has become known as a ‘construction’ (among others, Hoffmann & Trousdale 2013). While the Saussurian sign primarily correlates with the word-level (1916/1995: 99), constructions encompass any meaningful regularity in language, so that linguistic structure is composed of “constructions all the way down” (Goldberg 2006: 18). An important argument for a constructional account of form-meaning mismatches is the fact that constructions may have ‘holistic properties’, not derivable from the properties of their constituents and/or structure (cf. Booij 2010). To accommodate mismatches, it may be useful to recognize the complexity of the relations within the constructional network and the possibility of multiple inheritance (De Smet et al. 2013), as well as to accept that rather than one-to-one relationships between form and meaning, many-to-many mappings are the norm in language (Van de Velde 2014). 

Keywords: linguistic sign, form-meaning mismatch, structuralism, Construction Grammar

Accepted Papers:


A quantitative measure of constructional contamination through superficial resemblance

Dirk Pijpops & Freek Van de Velde

Pop goes the weasel: On how full-verb inversion lifts restrictions on the use of the simple present

Astrid De Wit

Disentangling form-meaning mismatches of light verbs: an image-schematic approach

Ferran Suñer Munoz

V1 and V2 Coordination, Multiple Inheritance, and Polysemy

Soteria (Roula) Svorou

Overgeneralization of modal verbs in child language as a result of non-conventional form-meaning mapping

Sara Jonkers

Synchronic mismatches and morphological differentiation: the impact of grammaticalization on form-meaning relations

Livio Gaeta