Corpus, Discourse and Society: A Diachronic Approach to Meaning
Lead Research Organisation:
University of Birmingham
Department Name: Department of English Literature
Abstract
With the advent of computers, linguists learned to look at language from new perspectives. Formerly, linguistics had focused on the language system and tended to neglect the empirical evidence we find in the textual contributions people make to the discourse. Corpus linguistics is, however, primarily concerned with real language data and the ways we can describe and explain them. It has taught us that even though the language system, grammar and the lexicon, allow us to say new things all the time, people hardly ever say something new. To a much larger extent they re-use what has been said before: a relatively small set of formulas, phrases and idioms make up the largest part of any contribution to a discourse. For almost any phrase in any text, we find a multitude of duplicates by a simple search in Google. It seems we can normally deal with something new only if it is embedded in a context that is familiar.
Over the past forty years, corpus linguistics has developed a methodology to analyse large corpora, principled collections of texts in electronic form. The combination of query tools and statistical know-how has made it possible to discover forms of lexical patterning which previously were unobservable. It is now generally recognised that single words should not be seen as the predominant units of meaning. This had been anticipated by J.R. Firth more than fifty years ago. He pointed out that we only know what single words mean when they are embedded in a context and surrounded by collocates. John Sinclair, Mike Stubbs, Michael Hoey and many other corpus linguists have investigated how statistically significant co occurrences of words can be translated into lexical patterning, and how these patterns can be interpreted as units of meaning. They show what is common to all their occurrences, regardless of when exactly they have occurred. This synchronic perspective implies looking at the discourse as if all utterances have occurred at the same time.
However, there is more to meaning. Whenever we make a contribution to the discourse, we react to something said before. We may endorse it, we may add to it, we may explain it in slightly different ways, or we may reject it. The discourse has, of necessity, a diachronic dimension. We know what the units of meaning, the objects of the discourse, mean because their meaning has been explained to us at some prior point. Whenever we are aiming for a fuller understanding of a discourse object, we therefore have to go back to the past, to look at those instances when it was discussed before. This means we have to pay attention to two linguistic features which so far have been largely overlooked. One feature is paraphrase, i.e. how the units of meaning are constantly being reformulated, explained, discussed, and evaluated. The other feature is intertextuality, i.e. the intertextual links between texts, in their diachronic succession. Intertextuality allows and relates one text segment to segments in other texts to which it is a reaction, and to yet other texts in which it leaves theses.
The meaning of words is not their reference to a discourse-external reality. It is our negotiations within the discourse which construct and shape the reality of which we are conscious. A look at the world outside does not tell us the meaning of globalisation. It has to be explained to us before we can grasp it. My goal is to investigate the features of paraphrase and intertextuality, by borrowing from two related disciplines, social constructionism and hermeneutics. Particularly on the Continent, these have a long tradition. In four case studies, dealing with the concepts of globalisation, of the sublime, of work and property, and of guilt feelings respectively, I want to demonstrate that corpus linguistics, once it takes account of the diachronic dimension of the discourse, can contribute considerably to our understanding of the world in which we
Live.
Over the past forty years, corpus linguistics has developed a methodology to analyse large corpora, principled collections of texts in electronic form. The combination of query tools and statistical know-how has made it possible to discover forms of lexical patterning which previously were unobservable. It is now generally recognised that single words should not be seen as the predominant units of meaning. This had been anticipated by J.R. Firth more than fifty years ago. He pointed out that we only know what single words mean when they are embedded in a context and surrounded by collocates. John Sinclair, Mike Stubbs, Michael Hoey and many other corpus linguists have investigated how statistically significant co occurrences of words can be translated into lexical patterning, and how these patterns can be interpreted as units of meaning. They show what is common to all their occurrences, regardless of when exactly they have occurred. This synchronic perspective implies looking at the discourse as if all utterances have occurred at the same time.
However, there is more to meaning. Whenever we make a contribution to the discourse, we react to something said before. We may endorse it, we may add to it, we may explain it in slightly different ways, or we may reject it. The discourse has, of necessity, a diachronic dimension. We know what the units of meaning, the objects of the discourse, mean because their meaning has been explained to us at some prior point. Whenever we are aiming for a fuller understanding of a discourse object, we therefore have to go back to the past, to look at those instances when it was discussed before. This means we have to pay attention to two linguistic features which so far have been largely overlooked. One feature is paraphrase, i.e. how the units of meaning are constantly being reformulated, explained, discussed, and evaluated. The other feature is intertextuality, i.e. the intertextual links between texts, in their diachronic succession. Intertextuality allows and relates one text segment to segments in other texts to which it is a reaction, and to yet other texts in which it leaves theses.
The meaning of words is not their reference to a discourse-external reality. It is our negotiations within the discourse which construct and shape the reality of which we are conscious. A look at the world outside does not tell us the meaning of globalisation. It has to be explained to us before we can grasp it. My goal is to investigate the features of paraphrase and intertextuality, by borrowing from two related disciplines, social constructionism and hermeneutics. Particularly on the Continent, these have a long tradition. In four case studies, dealing with the concepts of globalisation, of the sublime, of work and property, and of guilt feelings respectively, I want to demonstrate that corpus linguistics, once it takes account of the diachronic dimension of the discourse, can contribute considerably to our understanding of the world in which we
Live.
Organisations
People |
ORCID iD |
| Wolfgang Teubert (Principal Investigator) |