Non-Canonical Grammar!?

The papers collected in this special issue originated from a workshop held at the Annual Meeting of German University Teachers of English (Anglistentag) in Hamburg in September 2016. Contributors and participants at the workshop were invited to probe into the usefulness – and the limitations – of the notion non-canonical grammar for their respective fields of interest, and the present volume is a lively testimony to an engaging discussion.

The notion of non-canonical grammar is intimately tied to the notion of information structure or information packaging, that is “aspects of natural language that help speakers to take into consideration the addressee’s current information state, and hence to facilitate the flow of communication” (Krifka and Musan 2012, 1). Starting from a point of (presumed) shared knowledge or common ground, speakers maintain and enhance each other’s information state by establishing what is given in the current discourse, by backgrounding or foregrounding topics, by highlighting new information, or by placing a particular discourse referent in focus. In this sense, non-canonical grammar refers to the manipulation of canonical, i.e. “syntactically more basic or elementary” (Huddleston and Pullum 2002: 36) word order patterns to achieve specific effects in discourse. Most prominent are syntactic patterns that differ, more or less systematically, from “canonical” constructions that apply established constituent inventories and follow conventional constituent order(s), for example frontings, dislocations, cleft- and existential there-constructions as contrasted with more established SVX type clauses. Some examples may serve to illustrate the range of non-canonical constructions:

• Left dislocation: Civilization – I’ll do anything to protect it. (Auntie (Tina Turner) in Mad Max III: Beyond Thunderdome)

• Right dislocation: They fuck you up, your mum and dad. (Philip Larkin, This Be The Verse)

• Inversion: In another moment down went Alice after it, never once considering how in the world she was to get out again. (Lewis Carroll, Alice’s Adventures in Wonderland)

• Pseudo-cleft: All I want you to know is how I feel (Colbie Caillat, What I wanted to say)

• Existential there-construction: There is no alternative to facts (March for Science, 22 April 2017)

Apart from generating a great deal of theoretical discussion which cannot be considered here, such non-canonical constructions are regularly treated in standard reference grammars of English (e.g. Quirk et al. 1985, 1356–1418, Huddleston and Pullum 2002, 1363–1447) as “represent[ing] different ways of saying the same thing. More precisely, they have the same propositional content, but differ in the way the information is presented – or ‘packaged’” (Huddleston and Pullum 2002, 46). However, it has to be kept in mind that “syntax is not the only formal level at which information structure is coded. What syntax does not code, prosody does, and what is not coded by prosody may be expressed by morphology or the lexicon” (Lambrecht 1994, 31). If, for example, we want to foreground the information that Mary passed the test (to the exclusion of other alternatives relevant in the discourse context), we could resort to a prosodic, syntactic, or lexical strategy to explicitly mark the focus of our utterance:

• MARY passed the test.

• It was Mary who passed the test.

• Only Mary passed the test.

Languages in general and both, varieties and registers of English in particular may display different preferences for the formal realization as well as the frequency of individual information-packaging strategies.

Three papers in the present volume deal explicitly with non-canonical grammar in this sense. Teresa Pham’s contribution “‘Hard to beat Dickens’ characters’ – Non-canonical syntax in evaluative texts” explores in how far non-canonical sentence patterns support the expression of writers’ attitude or stance. Sandra Götz and Sven Leuckert both focus on (South) Asian varieties of English, that is, Postcolonial Englishes situated in dynamic contact scenarios. Götz’ paper “Non-canonical syntax in South-Asian varieties of English: a corpus-based pilot study on fronting” compares and contrasts written newspaper language derived from six South Asian countries with patterns and frequencies found in the historical input variety British English. Leuckert’s article “Typological interference in information structure: The case of topicalization in Asia” investigates non-canonical syntax in spoken language, with specific reference to possible substrate influence.

The conventional application of the notions canonical and non-canonical in these papers presupposes that canonical forms are a) more frequent or b) more basic than non-canonical forms. This is an accepted view which we would like to challenge in this volume. Non-canonical syntax and information structure are interrelated notions (cf. Birner and Ward 1998), but they do not exhaust each other: information structure may be expressed by means other than word order patterns, as a consequence of which a non-canonical grammar may extend to other linguistic phenomena and domains. Since non-canonical grammar belongs to the realm of discourse (pragmatics), it is typically (but not exclusively) found in actual communicative interaction. However, while e.g. cleft and existential there-constructions occur just more frequently in spoken rather than written language, some forms such as the ‘hanging topic’-construction (e.g. My work, I’m going crazy) are considered ungrammatical in writing. Form this perspective, frequency is not a very robust criterion but may be overridden by contextual or even societal factors.

Thus, there is a second sense of non-canonical grammar, namely that which encapsulates phenomena which occur in less than conventional shapes, when compared to more “sanctified” alternatives. This perspective is based on the idea that established, i.e. canonical, linguistic phenomena contrast with forms and structures which are not part of an accepted or codified standard and are “non-canonical” for that reason. This is probably the lesser understood perspective, since there are few studies that inquire into the theoretical status of non-canonical forms. Generally, this perspective investigates the ways in which linguistic phenomena acquire sanctification, and traces the origins of (perceived) non-canonical constructions.

In this special issue, three papers explore the explanatory value of the concept canonical/non-canonical as a scalar rather than a binary notion in different domains of language and language use. Depending on context, canonical/non-canonical may be conceptualized as centre – periphery, prototypical – less prototypical, standard – nonstandard, or even idiomatic – unidiomatic. Thomas Kohnen’s paper “Non-canonical speech acts in the history of English” considers the speech acts of boasting and apologizing in Old English and traces differences to contemporary usage to different norms of politeness in Anglo-Saxon England. Markus Freudinger’s “Shoulda, coulda, woulda – non-canonical forms on the move?” provides a profile of the frequencies and distribution of the colloquial contracted forms that are moving away from being unaccepted or nonstandard, but have not reached ‘canonical’ status yet. Finally, Ilka Mindt’s paper “Chosen” is an account of a little studied phenomenon at the interface of grammar and the lexicon; her paper also raises the interesting question of the relationship between frequency of occurrence and the status of an item/construction as canonical or “more basic or elementary” (Huddleston and Pullum 2002, 36).

In contrasting works within the traditional framework of non-canonical syntax with more innovative approaches that extend the notions to other levels of description and new phenomena, we hope to provide a fresh look at the processes of language use, variation and change. After all, what is frequent, canonical and basic at one particluar place and time for one set of speakers need not be the norm to which other speakers adhere in some other place and time. All papers collected in this volume agree on this point and, taken together, make valuable suggestions of how to acknowledge this fact in a theoretical framework which is based on the notion of canonicity in some form or other.