The following is an extract form Hunston (2022, p. 256).
Hunston, S. (2022). Corpora in applied linguistics. Cambridge University Press.
Sinclair made a number of generalisations in the 1980s (Sinclair 1991, 2004; see also Francis 1993; Hoey 2005; Hunston 2002; Stubbs 2001) which might be summarised as follows:
• In describing the meanings of a word, the ‘phrases’ that the word is used in are central to that description (= there is no distinction between form and meaning).
• Those ‘phrases’ are neither fully fixed nor fully open – in fact the distinction between ‘word’ and ‘fixed phrase’ does not hold up; the boundaries of a ‘phrase’ may be indeterminate and the variation resists classification.
• Those ‘phrases’ incorporate associations between individual words that might be discussed under the heading of collocation, but the ‘phrases’ also include aspects of grammar and commonalities of meaning rather than of form (= language is not divided into lexis and grammar).
• Although we commonly think of words as having meaning, and we often talk of a word having several meanings, what actually happens is that a word occurs in several ‘phrases’ and meaning resides in the ‘phrase’ rather than the word (= unit of meaning).
• When we look at text we can observe that a lot of it can be explained as a series of units of meaning and the remainder can be explained in terms of residual grammar (= idiom principle and open-choice principle).
The corpus was created from 301,847 ASR transcripts from 2,572 YouTube channels, corresponding to 154,041 hours of video. The size of the corpus is 1,252,066,371 word tokens.
The channels sampled in the corpus are associated with local government entities such as town, city, or county boards and councils, school or utility districts, regional authorities such as provincial or territorial governments, or other governmental organizations.
The transcripts are primarily of recordings of public meetings, although other genres are also present. Video transcripts have been assigned exact latitude-longitude coordinates using a geocoding script.
This information was distributed through the Corpora-List by Steven Coats, University of Oulu, Finland
To cite the corpus, please use
Coats, Steven. 2021. Corpus of North American Spoken English (CoNASE). http://cc.oulu.fi/~scoats/CoNASE.html.