Historical Spoken Language Research

Corpus Perspectives

Ivor Timmis

• Publisher: Routledge
• Year: 2017
• ISBN:
  9781315390024
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

Historical Research on Spoken Language: Corpus Perspectives uses historical sources to discuss continuity and change in spoken language. Based on two corpora compiled using data from sociological and anthropological studies of Victorian London and 1930s Bolton, the author shows how historical spoken corpora can illuminate the nature of spoken language as well as the attitudes, values and behaviour of the specific community represented in a corpus. This book:

• 
  
  

• demonstrates how spoken language can be examined using material collected before the advent of sophisticated recording equipment and large-scale computerised corpora;
• shows how other written sources such as diaries, letters and existing historical corpora can be used to analyse informal language use as far back as the fifteenth century;
• provides insight into the longevity and resilience of many spoken language features which are often regarded as vernacular or non-standard;
• comes with a companion website which gives full access to the Bolton Worktown Corpus.

Historical Research on Spoken Language is key reading for researchers and students working in relevant areas.

Chapter 1

Introduction

In search of spoken language in the past

Aims

The main aim of this book is to use data from historical spoken corpora as a focal point for the discussion of two related issues: (1) continuity (and change) in spoken grammar, and (2) the factors which appear to sustain certain vernacular grammatical forms for centuries. As the term ‘vernacular’ is central to this study, it is worth defining the sense in which it is used in this book at the outset. The definition of ‘vernacular’ provided by Biber et al. (1999: 1121; my italics) best captures the focus and tenor of this book:

The term vernacular, referring to the popular, untaught variety of a language found in colloquial speech, will serve to cover a range of phenomena in popular speech which may to a greater or lesser extent be felt to lack prestige and to be inappropriate for serious public communication, especially written communication.

It is hoped that by using the term ‘vernacular’ (except where sources cited use ‘non-standard’) at least some of the negative connotations of ‘non-standard’ will be avoided. Two points made by Anderwald (2011) in this regard are relevant to the focus of this book: firstly, in the cases where there has been a focus on vernacular varieties, it has most frequently been on phonology. Secondly, non-standard sources have rarely been used by grammatical theorists. In addressing vernacular grammar through a range of vernacular sources, some of which have received little attention from linguists, this book aims to make a contribution to filling this gap and to add what I will argue is a crucial dimension to the study of the vernacular: an historical perspective.

The study is based mainly, but by no means exclusively, on two historical spoken corpora, the Bolton/Worktown Corpus (BWC) and the Mayhew Corpus (MC). Both these corpora are described in detail in Chapter 2, but to contextualise the discussion of historical linguistics which follows, it is useful to note at this point that the BWC is a corpus of conversations amongst working class people in the town of Bolton, in the North of England, between 1937 and 1940: these conversations, or fragments of them, were written down at the time as part of a sociological study. The MC consists largely of data transcribed at the time from interviews carried out by the social reformer and journalist, Sir Henry Mayhew; these interviews were (mostly) with working class people in London between 1850 and 1860. The claim that these written records of speech in the BWC and the MC present valid spoken data for analysis will be scrutinised in depth in Chapter 2.

The field of historical linguistics

It is clear from the outline of the BWC and the MC that we are in the domain of historical linguistics; it will be clear in subsequent chapters that we need to call on sub-disciplines of this field: historical sociolinguistics and historical pragmatics. Having thus situated this study, it is important to consider the general aims of historical linguistics as outlined by Campbell (2004: 2):

human cognition and the human capacity for language learning are central research interests in linguistics, and historical linguistics contributes significantly to this goal. As we determine more accurately what can change and cannot change in a language, and what the permissible versus impossible ways are in which languages can change, we contribute significantly to the understanding of Universal Grammar, language typology and human cognition in general – fundamental to our understanding of our very humanity.

This book is not so much concerned with the abstract theoretical domains of Universal Grammar and language typology as with the cognitive and affective factors which influence language change and continuity in spoken language. It needs to be acknowledged that in approaching historical linguistics through spoken language, we are taking ‘the road less travelled’. It is far more common to study historical linguistics through written texts, often canonical written texts. However, as Milroy (2002: 7) argues, just as in political and social history there can be alternative histories of the same events, so there can be alternative histories of a language. It follows, then, that just as political or social history can be interpreted from different ideological perspectives, so can language histories. Language histories, for example, can be motivated by a desire to portray a language as ‘pure’ resulting in a focus on the putative invariability of the language in question as a reflection, perhaps, of the stability and longevity of the society which speaks it. With such aims in mind, language histories will typically be based on the written canon rather than the spoken word.

The nature of the BWC and the MC, both of which consist largely of data produced by working class speakers in the past, also takes us into the field of historical sociolinguistics, particularly in the comparison of vernacular data with current and past notions of the standard. As we shall see at several points in the book, in attempting to account for the prominence or typical uses of given features in a specific historical period, we will be engaged in historical pragmatics, with a particular focus on the following aspects of pragmatics: the speaker’s intentions; the hearer’s interpretation of the speaker’s utterance and the intention behind it; and the social interaction between the speaker and the hearer (Culpeper and Kytö 2010: 8). As the study of all these aspects has to be rooted in a knowledge of the social and historical circumstances of the time (Rissanen 2008), Chapter 2 sketches some of the sociohistorical background to the BWC and the MC. While this study is not primarily a study in dialectology, it does involve some discussion of dialect, particularly in relation to the BWC in Chapter 5 – as Schneider (2002) points out, dialectology necessarily contributes to historical linguistics.

While historical linguistics, historical sociolinguistics and historical pragmatics may differ somewhat in focus, it is important to establish that, ultimately, from a theoretical standpoint, all three disciplines rest on ‘the principle of uniformitarianism’. The ‘principle of uniformitarianism’, as Romaine (1998: 1454, cited in Nevalainen and Raumolin-Brunberg, 2012: 24) explains, is that the linguistic and sociolinguistic forces which shape language remain fundamentally unchanged over time:

The linguistic forces which operate today and are observable around us are not unlike those which have operated in the past. This principle is basic to purely linguistic reconstruction as well, but sociolinguistically speaking, it means there is no reason for believing that language did not vary in the same patterned ways in the past as it has been observed to do today.

Rationale for a focus on spoken language

In this book, there is an extra dimension to the intrinsic challenges presented by historical linguistics, historical sociolinguistics and historical pragmatics: our concern is with historical spoken language from periods when audio recording was either not available at all (e.g. the MC), or not available in a form to allow discreet recording (e.g. the BWC). The nature of the challenge is well put by Wyld (1920: 1):

It is an unfortunate circumstance for students of the history of a language, but one from which there is no escape, that they are dependent upon written documents for a knowledge of all but the most recent developments, since, in the nature of things, they can gain no direct and personal access to the spoken language earlier than the speech of the oldest living person they may know.

Before we address this formidable challenge, however, we need to ensure that the challenge is worthwhile by answering two related questions:

• Is spoken language significantly different from written language?
• Is there any particular value in studying spoken rather than written language?

It has long been recognised, though perhaps by a minority, that spoken language is sufficiently different from writing to be worthy of descriptive attention in its own right (e.g. Wyld 1920; Palmer and Blandford 1924; Bøgholm 1939). However, it is the development in recent years of large spoken corpora, built to represent the language of specific varieties and/or specific demographic groups and/or spoken genres, which has promoted a focus on spoken language by allowing a more comprehensive and systematic analysis of spoken language than was hitherto possible (e.g. McCarthy 1998; Biber et al. 1999; Carter and McCarthy 2006; Ruehlemann 2007). In general, these sources have suggested that there are indeed important differences between spoken and written language; they have also sought to account for these differences, generally in terms of how the circumstances of production and the goals of spoken language differ from those of written language. Ruehlemann (2007), for example, suggests four factors which determine the nature of spoken language. Adopting the ‘uniformitarian principle’ noted above, we must assume that these circumstances also applied in the periods of interest to us, so it will be useful to consider each of these factors in turn:

• Real-time processing
• Interactive concerns
• Interpersonal concerns
• Shared context.

The role of real-time processing constraints in shaping spoken grammar has been highlighted by Biber et al. (1999: 1048), who draw attention to the need for devices which promote economy of effort: ‘Many conversational traits arise from the fact that conversation is typically spontaneous, so that speakers are continually faced with the need both to plan and to execute their utterances in real time, “online” or “on the fly”.’

To meet the demands of spontaneity, speakers will favour, then, devices which help them to ‘save effort and/or to ease the processing load for speaker and hearer’. Similarly, Salmon (1987: 41) refers to ‘impermanence’ and ‘lack of premeditation’ as two features ‘inherent in speech’, or at least inherent in conversational speech. Amongst specific features of spoken grammar which have been at least partially explained in terms of real-time processing constraints we will consider there’s with plural noun phrase (NP), ‘heads’ and ‘tails’ (also known as left and right dislocation) in Chapter 8.

The influence of the interactive nature of conversation on spoken grammar is also noted by Biber et al. (1999), who emphasise the need for spontaneous adjustments as conversation emerges as a joint enterprise: ‘conversation is co-constructed by two or more interlocutors, dynamically adapting their expression to the ongoing exchange’. Amongst the interactive features of conversation they describe are the prevalence in conversation of questions, imperatives and question tags. They also refer to the greater frequency of stance adverbials and linking adverbials in spoken language, as opposed to written language, and to the fact that these adverbials are different in kind in speech. We can also include in interactive features ‘good listenership’ (McCarthy 2003) which involves the use of features such as backchannelling and response tokens to oil the wheels of conversation. In Chapter 4 we consider interjections as devices which are fundamentally interactive in nature.

In addition to processing and interactive concerns, speakers have to take into account interpersonal concerns and, as Carter (2004: 117) observes, the expression of affect: ‘When speakers interact, they do more than transmit information … Speakers also often wish to give a more affective contour to what they or others are saying’. In similar vein, Biber et al. (1999: 1047) remark that, ‘Conversation is expressive of politeness, emotion and attitude’. The importance of interpersonal concerns is also discussed by Salmon (1987: 40) who points out that speakers ‘will choose the exponents of those features which express their attitudes towards all aspects of the situation and to one another’. Such attitudes, as Salmon (1987: 40–41) notes, may be ‘appropriate to permanent relationships such as parent to child or servant to master’ or ‘temporary attitudes such as politeness, formality, anger or dislike’. The form of a given utterance may be influenced by one or more of these attitudinal factors and such factors, Cheshire (2005: 81) argues, ‘influence the extent to which variation arises in syntax’.

The centrality of such aspects of relation management in conversation is underlined by Ruehlemann (2007), who goes so far as to argue that the very goal of conversation can be seen as to establish and maintain relationships. In terms of specific features of conversational grammar which can be attributed to interpersonal concerns, Biber et al. (1999) point, for example, to familiarisers and endearments and politeness markers. In Chapter 4 we consider the interpersonal function of swearing and interjections in conversation and, more generally, the interpersonal function of evaluative language.

An obvious distinguishing feature between written and spoken language is that conversation typically takes place face-to-face such that interlocutors share a context which is often multi-faceted. As Biber et al. (1999: 1042) note, shared context involves ‘not just an immediate physical context of time and space, but a large amount of social, cultural and institutional knowledge’. They also observe that the information which can be retrieved from this shared context and knowledge permits the frequent use of ‘structure-erasing devices’ such as ellipsis, pronouns and pro-forms. Shared context, then, allows conversation to be ‘less dependent on the articulation of overt grammatical structure’ (Biber et al. 1999: 1043). Our main focus on this factor is in Chapter 3 with the focus on homophoric pronouns and vague category markers.

While these four determinants have been defined and exemplified separately above, it needs to be acknowledged (Ruehlemann 2007) that they interact with each other, and that any given feature may be explained in relation to one or more of these determinants. This observation is highly relevant to the study of tails/right dislocation in Chapters 4 and 8. We must also keep in mind our assumption that these determinants are not zeitgebunden.

In addition to a growing awareness of the distinctive nature of spoken language, there has been a growing recognition of the importance of spoken language itself (irrespective of the era). Such a concern is not limited to corpus studies of spoken language; as Leech (2000: 677) argues, there has been a ‘strong convergence of various interests in the topic of spoken language’. Amongst the convergent interests that Leech (2000: 677) notes are ‘the power and influence of spoken media’ and a greater emphasis on oracy in education. For the purposes o...

Table of contents

• Cover
• Title
• Copyright
• Dedication
• Contents
• List of tables
• List of abbreviations
• Acknowledgments
• 1 Introduction: In search of spoken language in the past
• 2 The Bolton/Worktown Corpus (BWC) and the Mayhew Corpus (MC)
• 3 Windows on society: pronouns and vague category markers
• 4 Evaluation, affect and intensity
• 5 Dialect and identity
• 6 Sources and resources for historical spoken language research: beyond the MC and the BWC
• 7 Vernacular grammar: longevity and obsolescence
• 8 Vernacular continuity
• 9 Reflections
• Index