2. Material and Method
The method used to collect linguistic material depends to a large extent on the research objectives. Lexical and phonological studies, such as the Leeds Survey, often use questionnaires to collect data. However, syntactic studies such as the present one require longer stretches of speech, and thus the material was collected using tape-recorded interviews. In order to guarantee long stretches of speech, the interviewer tried to avoid interrupting the informant. Even words such as ‘yes’, ‘yah’ and ‘ah’ were avoided whenever possible. Instead, approval or agreement had to be shown through gestures, such as nodding. This principle had proved successful in recordings of Finnish folk speech in the 1950s and 1960s (Yli-Paavola 1970: 44). Minimal interruption allows the informant to finish their sentences and thoughts, and provides the highest possible number of “clear” or non-doubtful items and constructions.
During the first weeks of the first data-collecting period, I was fortunate to meet a local person who was genuinely interested in local history, who knew the inhabitants of both his own and the neighbouring villages, who had drawn family trees of villagers, and who was bi-dialectal, that is, he was able to use the local dialect with dialect speakers and the standard with standard speakers. He introduced me to local persons in various villages. [1] Local ‘assistants’ were of great importance in locating suitable informants, especially at the beginning of the fieldwork period. They knew the other villagers and their social, educational and family backgrounds, and introduced me to the dialect speakers of their own and neighbouring villages.
In each village, the village post office was worth visiting, since it had lists of pensioners, who were potential informants. The village shop was another information centre, as the shopkeeper usually knew most villagers.
The original purpose of the Helsinki Dialect Syntax Group (see Background: I) was to supplement the Survey of English Dialects material. With this in mind, the principles for the sampling process were to some extent based on the Survey criteria for the selection of localities and informants.
Not everyone speaks the dialect of the area they belong to. To avoid misrepresentations, the following criteria were set for the selection of informants:
(1) First, a consistent regional dialect is most likely to be used by speakers who have always lived in the same regionally limited area, with minimum contact with people of other speech habits. Hence, in selecting informants for the Cambridgeshire dialect it was essential to make sure that they had spent all their lives in that county, preferably in the same village. Special attention was paid to the ancestors of the informant. The parents of the informants had to be natives of Cambridgeshire. The spouses of the informants were usually natives or long-time residents (with a few exceptions in the case of spouses who were born in villages very near the Cambridgeshire border and who had moved to Cambridgeshire at a very early age).
The informants were also selected from various parts of Cambridgeshire in an attempt to discover whether there was any noticeable difference in dialect across the region. Attention was focused on the rural areas, where the dialect has been less exposed to outside influences. The localities chosen are indicated with a dot on Map 2.1. The city of Cambridge is indicated with a square.
(1) Over
(2) Willingham
(3) Rampton
(4) Waterbeach
(5) Burwell |
(6) Swaffham Prior
(7) Fulbourn
(8) West Wickham
(9) Castle Camps
(10) Bartlow |
(11) Bassingbourn
(12) Newton
(13) Harston
(14) Harlton
(15) Little Eversden |
Map 2.1. Localities in Cambridgeshire chosen for the present study.
(2) Age was the second consideration. The informants were selected from the oldest section of society, since older speakers are generally considered to be more conservative in their speech. [2] An elderly speaker is thus more likely to remember and use the older forms and constructions of the local dialect, which are often interesting from the point of view of the historical development of the language. I therefore chose informants whose age at the time of the interviews was between seventy and ninety-four, with the exception of one informant who was sixty-one.
(3) In addition to area and age, social class was taken into consideration in selecting the informants. Although in England today sharp divisions between the social classes are disappearing, the fact is that Standard English (StE) is in general use among the upper and upper middle classes, while the regional dialects are still very largely used by the working and lower middle classes (e.g., Orton 1962; Trudgill 1974a). We can further say that the higher a person's social position is, the less regionally marked his speech is likely to be. Thus, in choosing the informants I concentrated on working-class people.
(4) The interviewee's sex was the final consideration. During the field work I noticed that, as a rule, the speech of working-class men and women differed. The differences were not only in accent, but also in vocabulary and grammar. I noticed that, in general, women tended to use words and constructions which occurred in colloquial Standard English, but also those which occurred in rather formal Standard English. After transcribing the speech of several women, I became convinced that the speech of working-class women included fewer non-standard elements than the speech of working-class men. (However, see Vasko 2010a Male and female language in Cambridgeshire: differences and similarities.) For this reason, of the nineteen informants selected for this study, only one was female. She was selected because her speech seemed to reflect to a great extent the characteristic features of the local dialect.
In summary, the interviewees were people who could be presumed to have best preserved the features of oral language inherited from earlier generations of speakers of their variety of regional English, being particularly isolated from modern linguistic standards. They were people who had stayed in their villages most of their lives, whose educational training had had a minimum effect on their speech, and who were too old at the time when the cinema, radio, and television made their appearance to allow these standardizing influences to alter their inherited habits of speech.
More detailed information on the nineteen informants who provided the material, and on the recording procedures used, are presented in Appendix A.
The recording of spontaneous speech is the best means of collecting speech units large enough for a thorough investigation of syntax. A conversation can be recorded in its entirety, thus guaranteeing that all essential syntactic aspects of a dialect can be taken into consideration. Furthermore, it is possible to record the informant's speech with great accuracy, since the recordings can be transcribed later over a longer period of time.
It is essential that the speech recorded is natural. Before conversation sessions involving a recording, I usually went to see the chosen speakers to give them the chance to get to know me. I also interviewed them to make certain that they met the selection criteria. These preliminary interviews were not recorded. In the recorded conversation sessions, I tried to make the informant comfortable by creating an atmosphere normally associated with spontaneous and free conversation, in the hope that they would pay minimal attention to the way they were speaking. The less attention paid to speech, the more informal and natural it becomes. Thus, there were occasions when the informant became excited, turned away from the microphone, was interrupted by other speakers, lit a pipe, etc.
The informants were encouraged to talk about any topics they pleased at the time of the recording. [3] The topics covered in my recordings ranged from personal reminiscences from the speaker’s working life through village activities to opinions on today’s lifestyle. Like the recordings for the SED, “the material was never rehearsed, and, of course, never recited” (Orton 1962: 19). [4]
The recordings usually contain the speech of one informant; however, friends or spouses of the informants were often present to make the interview situation feel more natural. The recordings also include an interview in which two informants have a conversation with each other. I myself usually acted as the interviewer. However, six interviews were conducted mainly by a person who had known the informants in question all their lives and spoke the local dialect in the company of the informants, and in six interviews I was also assisted by a native speaker who helped me in locating suitable informants. When I was not performing the interview, my role was mostly that of a listener and recorder.
The interviews were conducted in the informants’ homes. This domestic environment was considered conducive to the natural use of the dialect. See the discussion about Collecting spontaneous speech data (forthcoming Additional information).
The recordings were made between 1974 and 1977. The total length of the recorded interviews with the nineteen informants selected for this study is some 24 hours. This adds up to some 160,000 words.
Even when studying speech, it is necessary to record the language in some written form, i.e. to make a transcription. In Ojanen (1982), spontaneous dialect speech was transcribed orthographically. Orthographic transcription was considered sufficient, since the material was intended mainly for the investigation of morphological and syntactic features. The interviews for this study were transcribed in their entirety by the present writer. Some typical pronunciation features of Cambridgeshire dialect speech were indicated in the transcriptions. For instance, an apostrophe was used to indicate the dropping of a word-initial syllable or sound (as in ’tatoes for ‘potatoes’, ’at for ‘that’and ’ere for ‘here’) and the simplification of the word-final consonant cluster (as in sittin’ for ‘sitting’). Forms and structures which would otherwise remain ambiguous were also given in the broad International Phonetic Alphabet (IPA).
The transcription principles used in Vasko (2010) differ from those used for Ojanen (1982). For a discussion of transcription principles in Vasko (2010), see Transcribing Cambridgeshire dialect speech: problems and solutions.
In studies based on fieldwork, this fieldwork may be considered to be part of the method, since the method of collecting material has considerable influence on the type of linguistic material which will be available for analysis. Similarly, the transcription process may be regarded as part of the method.
The analysis of the 160,000-word body of material, collected from 15 locations in Cambridgeshire, was carried out within the framework of Standard English. Standard English is considered the best comparative basis, since it is the most widely known variety. It also allows the study to concentrate on those aspects which distinguish Cambridgeshire dialect from Standard English.
When I started to analyse the Cambridgeshire dialect speech data, it became obvious that the tools of analysis had to be different from those used in analysing Standard English. The differences between dialectal (and colloquial) grammar and written standard grammar can be striking. Complex, hierarchical syntactic constructions are systematically shunned in relaxed, informal face-to-face communication. In dialect speech, short, often paratactic, structures are preferred. For this reason, identifying a sentence, which is a basic unit in the analysis of the standard language, is often problematic. In fact, we have to accept that in dialect speech (as in all informal conversational language) we may rarely come across stretches of language that constitute ‘sentences’ in the strict sense of the term. For the analysis of dialect speech in this study I used a stretch of speech not limited to a ‘sentence’ or to the words of one speaker. Even this stretch of speech was not always sufficient. Personal knowledge connected with the concrete speech situation was necessary for the interpretation.
Since there is no universally accepted grammar of English – and especially no full and authoritative description of spoken English grammar – no single grammar book was used as a comparative framework. However, various grammars were consulted: for instance, Quirk et al.’s Grammar of Contemporary English, a ‘common-core’ grammar of “educated English current in the second half of the twentieth century in the world’s major English-speaking communities” (1979: v), because it is concerned (almost exclusively) with syntax; I also consulted such ‘traditional grammars’ as Jespersen’s Modern English Grammar on Historical Principles (1909-49) and Curme’s Grammar of the English Language (1935), both of which are major descriptive accounts of English grammar. However, the study essentially deals with the distinctive morphological and syntactic ‘facts’ of the Cambridgeshire dialect (in comparison with the standard language), these being the aspects which I found of particular interest. Some forms and constructions which are perhaps borderline cases are included because they seem to be used more frequently in the dialect than in the standard, and because the boundary between standard and non-standard is in any case fuzzy.
The analysis covers the use of verbs, nouns, pronouns, adjectives, adverbs, and prepositions typical of Cambridgeshire dialect speech as it was recorded in the 1970s. References are made to the earlier stages of English, especially in the course of mapping certain salient features in the material. Findings from the Cambridgeshire material are related to those from other regional dialects of the same and earlier periods.
The tape-recorded speech material – though comparatively extensive at 160,000 words – is less adequate for morphosyntactic studies than it would be for phonological studies. From this point of view, one might wish to supplement the recordings with other material from the same regional dialect. However, in earlier dialectal studies, Cambridgeshire has been totally neglected or, at most, very poorly covered (see Vasko 2010b Scarcity of information on Cambridgeshire speech up until the 1970s).
[1] Using an ‘insider’ as an intermediary, in this case a close friend, is not exceptional (e.g. Labov in Harlem), although this was not generally done in the early 1970s. In subsequent dialect studies, this method has become more popular.
[2] Indeed, it is assumed that throughout their lives older speakers continue to use basically the same language they learned in their youth (Francis 1983: 72).
[3] In this sense, my recordings differ from the SED recordings, which consisted mainly of connected discourses of the ‘directed conversation’ type, these being used to gather Incidental Material, and to check impressionistic notes on pronunciation (Francis 1983: 95).
[4] However, my talk sessions were recorded in their entirety, unlike the SED recordings, none of which consisted of complete interviews (Francis 1983: 95).
|
