Lexical Priming in Spoken English Usage

Michael Pace-Sigge

• Publisher: Palgrave Macmillan
• Year: 2013
• Print ISBN: 9781137331892
• eBook ISBN: 9781137331908
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

This book shows that over forty years of psychological laboratory-based research support the claims of the Lexical Priming Theory.It examineshow Lexical Priming applies to the use of spoken English as the book provides evidence that Lexical Priming is found in everyday spoken conversations.

1

Introduction

The argument I would like to put forward is that linguistics has been formed and shaped on inadequate evidence ( ... ). There has been a distinct shortage of information and evidence available to linguists, and this gives rise to a particular balance between speculation and fact in the way in which we talk about our subject. (Sinclair 1990 [2004]: 9)

The evidence that Sinclair considers to be inadequate is naturally occurring language. Since 1990, the amount of, and accessibility to, computing power has certainly made it easier to process huge amounts of text. The one area, however, where only a comparatively little amount of text evidence is available is spoken language. Amongst spoken language, there is little research evidence of how words are used within a particular group of speakers and in how far it does or does not differ from the usage amongst a more ‘general’ speaker. While dialectologists have focussed on differences in sound production and the use of characteristic words or grammatical constructions, in this book I will focus on the differences in the use of highly frequent words and clusters of words.

1.1 Where lexical priming came from

Some ideas need incubation time, new people, new techniques, and new technologies to finally make an impact. It is a bit like the invention of the early aircraft. As long as man remained on the ground, only geographical fixed elevated points (like trees for short distances and mountains for longer distances) could give an impression of what things looked like from above. Today, an outline is available to anyone. The same experience is true when millions, or indeed, billions of words from different sources can be collated and used for concordances which allow for a much finer grained vision of language.

As early as in the 1920s, Harold Palmer developed what would become a cornerstone of British Applied Linguistics. He devised lists of the most frequently used words and phrases, and constructed what he later termed Pattern Grammar. It seemed then that traditional grammar was to take a tumble:

The traditional categories of grammatical description are survivals of a medieval scholastic instrument. They have been used to deal both with the forms and meanings of linguistic constituents in the vaguest of socio-philosophical terms, and judged by modern standards they have been found wanting in both enterprises. ... Is there any more reason to perpetuate them than medieval alchemy? (Firth 1957: 154)

J. R. Firth was a believer of English as both a world language and ‘the greatest social force in the world’ (Firth 1957: 156). Firth described language as a unit that comes out of usage rather than a mere collection of words. This view, like much else Firth said so many years ago, seems to affirm work that would be done much later; Firth appears to have sown a seed for John Sinclair’s corpus work as well as for a lot of the empirical research into language use that is based on the corpora of naturally occurring language.

It is from the analysis of concordance lines that further areas of research stems. Nelson (2000: 122) discusses the tripartite backbone of concordance work and shows how important it is in seeing language as the means of communicating, pointing out that collocation, semantic prosody and colligation cannot be considered to be separate, independent entities or concepts; rather, they are ‘interdependent and together create a network of meaning’.

All three elements indicate that certain words or word forms either attract or avoid each other. While we can find beer + glass as well as beer + mug, a quick Internet search shows that there is clearly greater attraction of beer to glass than beer to mug.¹ Research by Katherine Hirsh and Jeremy Tree (2001) shows this as well. Similar experiments can be undertaken to observe the occurrence of not just word collocations but also their colligations and semantic associations.

Thus far, corpus linguistics has been able to show us these relations between words, yet it provides only a representation of what can be observed as occurring in the use of language. What corpus linguists have not done, however, is to explain why it occurs in the first place. There must be a reason (or reasons) why words (or clusters of words) collocate, group into certain constructions but avoid others (colligate), and have certain semantic associations.

At this point, lexical priming gives us an idea of why it is likely (or unlikely) that certain words (or clusters of words) stand in such relations to each other. Hoey (2005: 7) indicates that collocation is a concept that is pervasive; still it is more than that: it is also subversive.

The subversion can be found in a subliminal, subconscious, and psychological concept: namely that the repeated use of an occurrence pattern reinforces the idea that it is a natural pattern. The ‘repeat occurrence’ primes one’s mind to make automatic connections. Priming can provide for sets of actions, or, in the lexical field, sets of words. For example, a listener hearing the word bread will recognise words like baker, butter, knife far quicker than unrelated words like doctor, mortar, and radiator.

1.2 Lexical priming in spoken use; or, redefining the notion of dialect: the example of Liverpool English

One aim of this book is to test whether the theory of lexical priming – which has so far been investigated only on the basis of written material – is also applicable to spoken language use. Lexical Priming (Hoey 2005) uses data from one written source almost exclusively (ten years of Guardian articles). Spoken language is, however, extremely varied, and there are too few corpora available. Additionally, a spoken language corpus cannot be easily assembled, as transcription is extremely time-consuming. One option would be to look at examples of scripted or semi-scripted spoken language, such as lectures or radio broadcasts. This, however, would only reflect a minor part of the area of spoken language – one which often resembles written text more than naturally occurring language. The course I have taken here is to look at transcripts of casual conversation by one particular speech group – speakers in the City of Liverpool – and then compare this material to a wider and more general corpus based on transcripts gathered throughout the United Kingdom. The aim is to show evidence of lexical priming in spoken language by comparing, and indicating marked divergence between, two sets of informal spoken English.

This book does not deal with discourse analysis as such. However, using a corpus that is both small and thoroughly known by the author, the recorded material shows a number of salient discourse structures (cf. Biber et al. 1998: 106). We also find that spoken language has a far lower density than written language (cf. Stubbs 1996: 72ff.). Wray (2000: 46) and O’Keefe et al. (2007: 159ff.) point out that conversation shows a high degree of recurring formulaic chunks. In the next chapter, I try to show how these can be explained in terms of the lexical priming theory.

One of the issues this book highlights is the extend to which certain discourse structures can be seen as salient for spoken discourse in general. This is shown in investigated areas where there is little difference between all the corpora compared. The focus is, however, on the point(s) of divergence: areas of informal spoken discourse in which the use of the selected speech community is significantly divergent from what would be expected based on the national average. That this divergence can be explained as an expression of lexical priming is the premise of this book.

We will look at the words and sets of words used by speakers in Liverpool to see the degree to which usage is divergent from, or broadly similar to, the occurrence pattern in a general United Kingdom spoken corpus.² Liverpool English is what Kevin Watson (2007), referring to Gerald Knowles (1975), calls a new urban dialect.³ This is a dialect whose characteristics are, to a greater or lesser degree, different from those of the dialect found in the surrounding area. And the difference would only have become apparent over the last 100 to 150 years or so, as certain urban areas have experienced massive population growth and the influx (and influence) of people from further afield. When we look at contemporary Liverpool English, we discover that, apart from the phonetical properties of the words, it has few characteristic features. This means, in theory, that all words and sets of words that diverge from the average should be specific to the speech community of Liverpool speakers.

My investigation follows from the hypothesis that, in casual spoken Liverpool English, it is not just the traditional criteria that identify a particular variety of language as a dialect. I am going to argue that one variety of English may also differ from other recognised varieties of English in respect of systematic variations in the use of collocations, colligations, lexically driven grammatical patterns, and semantic preferences (or semantic associations).⁴ These terms and their respective authors are fully described in Chapter 2, and shall be briefly described here as follows:

Collocation – the company a lexical item keeps. Collocation has been written about and researched by Firth (1957); Halliday (1959); Sinclair (1991); Stubbs (1996); Partington (1998) and Hoey (2003a,b,c, 2005), amongst others.

Semantic preference – the semantic field that a lexical item prefers. The term was coined by John Sinclair (1997). See also Hoey (2005).

Colligation – the grammatical company a word keeps or avoids keeping and its preferred positioning and functions. See Firth (1957); Halliday (1959); Sinclair (1991); Hoey (2003a,b,c, 2005).

Lexically driven grammatical patterns – extending the middle ground where grammar and lexis meet as revealed by corpus-driven research. These were first discussed by Palmer and Hornby around 1933, written about by Hornby in 1954, taken up by Halliday and Hassan in 1976, and later still by Hunston and Francis in 2000.

While English accents have been studied for many centuries, the Liverpool English variant – Scouse – has only received attention since the 1970s. Previous surveys of English accents by Stanley Ellis (1974), Wells (1982: 371ff.) and Trudgill (2000: 71) are all agree that Scouse is an accent, based on the fact that it differs mainly from Standard English in its realisation of sounds (particularly vowel sounds and the voiceless plosive consonants). Furthermore, Knowles (1978: 34) points out that Liverpool English is an accent, not a dialect, on the grounds that ‘Liverpool English differs insufficiently in its grammar from Standard English’. Likewise, it contains only a small lexicon of words unique to the area.

A case can be made, however, for taking a different perspective on what counts as dialectal differences in order to explore whether casual spoken Liverpool English can be classified as a dialect. Dialectologists have traditionally concentrated on syntactic and morphological structures to describe a dialect. More recently, however, corpus linguistics has suggested that lexis is a more complex phenomenon than traditional accounts of syntactical and morphological structures allow, and some lexical features that have not previously been studied in a dialectal context may, accordingly, be relevant to a determination of difference.

1.3 Potential value of this work

1.3.1 In respect of dialectology

In my previous research into Scouse lenition (Pace-Sigge 2002) I worked closely with a spoken corpus. I felt then that Liverpool English speakers seemed to use their lexicon in a way that was different both collocationally and colligationally from spoken Standard English. In this book I revisit the question of whether Liverpool English is an accent or a dialect.

I focus on lexical items that have so far not been described in the ways I have mentioned, and whether these have preferences which are distinctively different in Scouse to those in a corpus of spoken Standard English. In this book, I therefore intend to re-visit the question of whether Liverpool English is an accent or a dialect. The focus will be on lexical items that have so far not been described in the ways I have mentioned, and whether these have preferences which are distinctively different in Scouse from those in a corpus of spoken Standard English. The research into Scouse will therefore serve as an example: the method, however, is applicable to all variants of a language when sufficient material is available.

Providing evidence of systematic differences in lexical use might extend the analytical tools of dialectology, in that I will have shown that dialects are distinguished as much by their collocational, colligational and semantic uses as by their grammatical and lexical differences. This demonstrates that it may not only be grammatical or lexical differences that define a set of speakers as a separate speech group, but also the fact that those speakers use the same lexicon in a distinct way. This would show that lexical use, rather than just lexical stock, is a characterising feature of dialects.

1.3.2 In respect of lexical priming in Spoken English

Hoey’s work since 2002 has introduced the concept of lexical priming into the field of language studies. Here, I map out the development of the concept of priming and show how Hoey came to find these principles salient for the use of competent language production. Where Hoey has provided evidence of lexical priming – based on corpora of written texts - this book will shows that lexical priming is a theory that is equally applicable to spoken (English) language. I hypothesise that speakers in a geographically restricted area, through ...

Table of contents

• Cover
• Title
• 1 Introduction
• 2 Lexical Priming: The Theoretical Backbone
• 3 Testing the Theory through Spoken-Corpus Evidence
• 4 Spoken Differs from Written – The Case of YES and YEAH
• 5 Referring to Oneself and Others in SCO and BNC/C
• 6 Use of Intensifiers and Discourse Particles in Casual Speech
• 7 The Uses of JUST and LIKE
• 8 The Most Frequent Clusters Found in Casually Spoken English Corpora
• 9 Conclusions
• Appendices
• Notes
• Bibliography
• Index