by Tony McEnery and Andrew Hardie; published by Cambridge University Press, 2012

English Corpus Linguistics at the University of Nottingham

Nottingham is best known for its contribution to the study of spoken English in particular. Nottingham's work explains how the major grammars of ECL came to focus on both spoken and written English, rather than just on writing.

The grammars produced in the 1970s and 80s by UCL all had the written language as their principal focus, in spite of the fact that a great deal of effort was invested in the production of spoken corpus material by the Survey of English Usage. The grammars of this period were very much rooted in the attitude to speech that casts it as a debased form of language, mired in hesitations, slips of the tongue and interruptions.

When, some time later, substantial corpora of spontaneous speech were created, there developed a spectrum of opinion regarding grammar in speech. One end of the spectrum was the view that grammar in speech is present only in some bastardised form (relative to fully-grammatical writing), subject to interference from a host of irrelevant performance features. At the other end of the spectrum was the view that not only is speech grammatical, but it also has a grammar of its own which is quite distinct from writing. The contribution of linguists at Nottingham, notably Ron Carter and Mike McCarthy, was to begin to move the debate to the middle of the spectrum.

Carter and McCarthy (1997) used the five million word CANCODE corpus of spoken English to explore the nature of grammar in speech. Their approach did not call for a distinct grammar of speech. Instead, it focused on those features of speech which appear most at odds with grammars taking the written language as their starting point.

The position developed at Nottingham in turn links clearly to the development of later grammars produced in the UCL tradition: grammars produced since the late 1990s have routinely engaged with speech at a detailed level, and have done so by acknowledging the differences between grammar in speech and grammar in writing – just as McCarthy and Carter argued should be done. The Longman Grammar of Spoken and Written English (LGSWE, Biber et al. 1999) is a case in point. It forms part of the UCL tradition (like the earlier Comprehensive Grammar of the English Language, Quirk et al. 1984) but it engages fully with the differences between speech and writing.

An example of this kind of middle-ground approach to the grammar of speech is the list of features in the table below. These are features which, Leech (1998) argues, are either distinctive of speech, almost exclusively related to speech, or rare in speech and much more common in writing.

Type of feature Feature Example
Distinctive in speech

Front ellipsis deleting subject

Doesn't matter

Near exclusive use in speech

Attention signals

Familiarising vocatives

Omission of auxiliary

Vernacular syntax


honey, mum, guys

what you doing?

My legs was hurting

Rare in speech

Dependent genitive

I met Geoff['s student


This page was last modified on Monday 16 April 2012 at 5:45 pm.

Tony McEnery Andrew Hardie

Department of Linguistics and English Language, Lancaster University, United Kingdom