A.2 ACCENT AND DIALECT In common perception, accent is something other people have: ‘He’s very broad’, ‘She has a lovely speaking voice’, ‘That’s a thick accent’, can be heard on innumerable occasions. Many of my students deny that they have an accent at all. These common judgements are interesting for what they tell us about people’s attitudes to language, but as sociolinguists we must not share their evaluative biases. Apart from sign-language, it is as impossible to speak without an accent as it is to speak without making any sound. A.2.1 Describing accents Phonology (the study of speech-sound and articulation) provides us with a scientific and objective means of discussing accent, in the form of the International Phonetic Alphabet (IPA). Unlike the standard spelling alphabet, the IPA sets out symbols which have a fixed value. For example, though ‘s’ is a normal alphabetical letter (a grapheme), it can sound differently in ‘six’, ‘lies’, ‘sugar’ and ‘leisure’. In the IPA we can distinguish these sounds: /s/, /z/, /Σ/ and /Ζ/ (these are phonemes, written within slashed lines as shown). The IPA allows us to describe, compare and contrast accents in a systematic way. There are many IPA symbols to cover all the various sounds that can be meaningfully produced in the world’s languages. The following is a selection of those that will be useful in this book. Assume a British ‘BBC English’ accent in the example words, unless stated otherwise. Selected IPA Symbols Consonants (including glides/liquids) p - pip b - bib t - ten d - den k - cat g - get f - fish v - van Π - thigh ∆ - thy s - set z - zen Σ - ship Ζ - leisure h - hen tΣ - church

1

dΖ - judge m - man n - man Ν - sing l - let (‘light l’, at the front of the mouth) λò - pull (‘dark l’, at the back of the mouth) r - ride, parrot Ρ - rubbish (Scots) (‘tapped r’) w - wet ã - which (aspirated, with breath) j - yet / - bu’er (glottal stop) x - loch (Scots) Vowels (Monophthongs) (Diphthongs) Ι Ε Θ ℘ Υ ↔ o a y Ο i: ∈: Α: : u: e:

-

pit pet pat pot putt put patter eau (French), low (Northern England) calm (Scouse), farm (Teesside) tu (French), school (Scouse) peu (French), boat (Geordie) bean burn barn born boon bait (Northern England)

aΙ ↔Ι ΕΙ Ι ↔Υ aΥ ↔Υ Υ↔ u↔ Ι↔ i↔ Ε↔

2

bite, night night (Scots, Canadian) bait boy house poor (Northern England) ear ear (Northern England) -

roe house (Scots, sewer, poor

air

You will have noticed some familiar symbols, as well as some unusual ones (the ‘tapped’ /Ρ/, for example), especially among the ‘pure’ vowels (monophthongs) and complex vocalic elements (diphthongs). In all the example words given, I assume what used to be called a ‘BBC English’ accent (known properly as Received Pronunciation – RP). Sometimes I have indicated a different regional accent, where RP does not use the sound. The IPA includes many notational marks – known as diacritics – to modify the sound represented. In the chart above, the colons (:) indicate that the vowel is lengthened. Accent variation is often most noticeably carried in the vocalic elements of pronunciation, and in the glides (/j/, /w/) and liquids (/r/, /l/) that are sort of ‘semivowels’. For example, a speaker who says [fΑ:rm] rather than [fΑ:m] is likely to be American or Irish rather than English: that is, their accent is said to be rhotic if they pronounce this sort of ‘non-prevocalic /r/’ (/r/ when it is not before a vowel, as in ‘farm’ or ‘car’). Americans and most Irish people have a ‘retroflex’ /r/. By contrast, if they ‘tap’ the ‘r’ (by flicking the tip of their tongue against the ridge behind their front teeth), the vowel quality is likely to change slightly and they are likely to introduce a vowel between the ‘r’ and ‘m’ to make the last two letters syllabic: [fΘΡ↔m]. This is more likely to be a Scottish speaker, or someone influenced by Scots, such as speakers in Ulster (and the square brackets are used to write down actual realisations in speech). Phonetic details like these can help you pinpoint the differences between accents. The crucial factor for sociolinguistics is that accent variation tends not to happen just randomly, but in relation to observable social patterns. Accent can often tell us where someone comes from, their age, gender, level of education, social class, wealth, how well-travelled they are, and whether they are emotionally attached to their home-town, job or political party. All of these factors can also be carried in someone’s dialect. B.2 ATTITUDES TO ACCENT VARIATION The reason why accent variation is so important in sociolinguistics is because of the significance people attach to different accents. The case-study below is partly a replication study of original work in informants’ subjective evaluations of accent. B.2.1 Evaluative reactions to accents Sarah Wood researched original work done by Giles and Powesland (1975) and others which used a method of data elicitation known as the matched-guise technique. Briefly, this involves playing a recording of the same speaker imitating a variety of different accents, and then asking listeners to rate each ‘speaker’ on a range of different dimensions. These might include their sense of the attractiveness of the speaker, how communicative they were, what their social status seems to be, and so on. In this way, a pattern of common stereotypical associations in attitude to accents is built up. The original studies used a range of British accents (northern, southern, rural, urban and RP) and some foreign-accented English (American, Italian, Indian, German, French, and so on), and used informants from south Wales and the south-west of England.

3

The findings were that, for many people, standard accents (such as RP) were more likely to be considered as belonging to prestigious, aesthetically pleasing and intelligible articulate speakers. The ‘broadest’ accents and those associated with urban and industrial areas, by contrast, were considered to be used by low-status speakers and were regarded as unattractive. Rural accents were considered aesthetically pleasing, but subordinate to RP on the dimensions of social status and intelligibility. Much of this work was conducted in the 1970s, and Sarah Wood was concerned with discovering the current situation. In general, she replicated Giles’ (1970) study but made a few adjustments to improve the analysis. She restricted the recordings to 8 native British accents (RP, west London, Norwich, north-east England, Nottingham, Cheshire, Burnley, and Sheffield), and she used genuine native speakers of these accents in the recordings. All speakers read a passage which was specially written to contain many accent-variant features (this ‘Goldilocks’ passage appears in C.2). All speakers and informants were female students in their early 20s, to control for gender, age and some class variation, and the informants included 2 northern speakers, 2 southern speakers, and a midlands speaker. Sarah ensured an easy comparability of data by setting a written, multiple-choice questionnaire, as follows: Q1

Please name the accent you have just heard.

Q2

Please circle the description you most agree with based on your view of the pleasantness / unpleasantness of this accent: 1 extremely pleasant 4 unpleasant

Q3

2 prestigious 5 very unprestigious

3 neutral

Please circle the intelligence rating you would give this speaker: 1 very intelligent 4 unintelligent

Q5

3 neutral

Please circle the prestige rating you would give this accent: 1 very prestigious 4 unprestigious

Q4

2 pleasant 5 extremely unpleasant

2 intelligent 5 very unintelligent

3 neutral

Please circle the type of house you would expect this person to live in: 1 homeless 2 council / housing association rented 3 council / housing association owner-occupied 4 rented private housing 5 terraced owner-occupied

4

6 average-sized owner-occupied 7 large owner-occupied 8 ‘mansion’-size owner-occupied Q6

Please circle the type of job you would expect this speaker to have (examples are simply guidelines): 1 unemployed 2 unskilled manual (rubbish collector) 3 semi-skilled manual (factory worker) 4 skilled manual (engineer) 5 routine non-manual (clerical, sales) 6 low professional (civil servant) 7 self-employed (own business) 8 management 9 higher professional (doctor, lawyer).

Sarah presented her detailed results as a table, and then contrasted her findings with those of 30 years previously. She found that the ‘southern’ accents (RP, west London and Norwich) attracted the highest and most prestigious overall ratings in most categories, across all informants. Though the RP speaker was judged more intelligent than the others, they were judged equal in social status, and the RP accent was judged as being less pleasing. The northern accents came out worst in the prestige judgements, with the urban accents more stigmatised than the rural ones. It is interesting that there was largely a consensus across informants, which suggests that language loyalty was a small factor in this study (though the northern informants did rate the northern accents slightly higher). Furthermore, judgements tended to parallel each other across the dimensions: so an accent tended to be judged consistently either prestigious or stigmatised across all the questions. Sarah also conjectured that the status of RP was changing so that it was becoming seen as ‘too posh’ and thus untrustworthy, and she discussed other reading which supported this view. Finally, she discussed the consequences of such accent-stereotypes for non-standard speakers in relation to their social-standing, job opportunities and educational access. There are all sorts of connections to be made from Sarah’s study. First of all, replication studies are a very useful means of investigating language change: in this case, changes in attitude to accents. It is largely as a consequence of such attitudinal changes that the use of RP seems to be diminishing in Britain (it is certainly less commonly heard in universities and on the BBC), and in many cases is being ‘toned down’ by being casualised in the direction of local urban vernaculars like Mancunian, Geordie and Cockney. This process produces the sort of hybrid ‘posh Geordie’, and so on, that can be heard on regional television news programmes, and has been characterised in relation to Cockney as ‘Estuary English’.

5

Secondly, Sarah’s study could lead into a discussion of the prestige and stigmatisation of accents and dialects in general. When people are aware of their own accent and its prestige-value, they will often adjust it either towards a more standardised form (this is hypercorrection if it is over-done) or even towards a more stigmatised form (if they want to sound ‘less posh’, this is covert prestige). Finally, Sarah’s study offers a refinement of sociolinguistic methodology along the lines of using naturalistic elicitation procedures to generate naturalistic and reliable data. It is thus a contribution to the methodological discussion of the field. C.2 DIALECTAL VARIATION C.2.1 An accent-elicitation reading passage The following passage has been specially written to contain many of the linguistic features that have multiple variants in different accents of English. It is, of course, one of the advantages of having a standardised written dialect (Standard English) and a frozen medieval spelling system that written English can be read and written by anyone in their own accent. Once upon a time there were three bears: a Daddy bear, a Mummy bear and a little baby bear. They lived in a cottage deep in the woods. One morning, Mummy bear had made some porridge for breakfast, but it was too hot to eat at once. ‘Let’s go for a walk while it cools down,’ said Daddy bear. ‘What a good idea!’ exclaimed Mummy bear, and, with their bear coats and bear shoes on, they all set off for a short walk in the woods. That morning a little girl called Goldilocks was also walking in the woods. She was picking flowers and had wandered deeper in among the trees than her parents allowed her to go. After a while of being completely lost, she came into a clearing and saw the pretty little cottage. ‘I wonder who lives there?’ she thought to herself, and walked up to the door. When she knocked, there was no answer, so she pushed the door. It swung open, and she went in. Can you identify some of the particular items (letters, words or links between words) in this passage that are likely to elicit a range of variant phonetic features when read aloud in different accents? It might help if you try out different accents while reading the passage: New York, Glasgow, Dublin, Birmingham; or think of ways that men, women, boys, girls, bus-drivers, mechanics, professors, sales assistants and others might pronounce the text.

6