Holmes, J. An Introduction to Sociolinguistics. (Harlow: Pearson Education Limited, 2008) third edition [ISBN 9781405821315] Chapter 6: ‘Regional and social dialects’; Chapter 7: ‘Gender and age’; Chapter 8: 'Ethnicity and social networks’; Chapter 10: ‘Style, context and register’.
Eppler, E. ‘Language and social class’ in Mooney et al. Language, Society and Power. (Abingdon: Routledge, 2011) [ISBN 9780415576598] pp.154–72.
Coates, J. Women, Men and Language. (Harlow: Pearson Education Limited, 2004) third edition [ISBN 0582771862].
Crystal, D. The Cambridge Encyclopedia of the English Language. (Cambridge: Cambridge University Press, 2003) second edition [ISBN 0521530334].
Hughes, A. and P. Trudgill English Accents and Dialects: an introduction to social and regional varieties of English in the British Isles. (London: Arnold, 2005) fourth edition [ISBN 9780340887189].
Holmes, J. An Introduction to Sociolinguistics. (Harlow: Pearson Education Limited, 2008) third edition [ISBN 9781405821315] Chapter 9: ‘Language change’.
Labov, W. Sociolinguistic Patterns. (Oxford: Blackwell, 1972)
[ISBN 9780812210521].
Milroy, L. Language and Social Networks. (Oxford: Blackwell, 1987) second edition [ISBN 9780631153146].
Wardhaugh, R. An Introduction to Sociolinguistics. (Oxford: Wiley Blackwell, 2009) sixth edition [ISBN 9781405186681].
Coupland, N. and A. Jaworksi ‘Introduction: social worlds through language’ in Coupland, N. and A. Jaworksi (eds) The New Sociolinguistics Reader. (Basingstoke: Palgrave Macmillan, 2009) [ISBN 9781403944153]
pp.1–21.
Eckert, P. ‘Gender and sociolinguistic variation’ in Coates, J. and P. Pichler (eds) Language and Gender: A Reader. (Oxford: Wiley-Blackwell, 2011) second edition [ISBN 1405191279] pp.57–66.
Eckert, P. and S. McConnell-Ginet ‘Communities of practice: where language, gender, and power all live’ in Coates, J. and P. Pichler (eds) Language and Gender: A Reader. (Oxford: Wiley-Blackwell, 2011) second edition [ISBN 1405191279] pp.573–82.
Freeborn, D., P. French and D. Langford Varieties of English: An Introduction to the Study of Language. (Basingstoke: Macmillan, 1993)
[ISBN 9780333589175].
Mesthrie, R., J. Swann, A. Deumert and W. Leap Introducing Sociolinguistics. (Edinburgh: Edinburgh University Press, 2009) second edition
[ISBN 9780748638444].
Mitford, N. (ed.) Noblesse Oblige: An Inquiry into the Identifiable Characteristics of the English Aristocracy (London: Futura Publications, 1980) [ISBN 9780708817681].
Trudgill, P. The Social Differentiation of English in Norwich (Cambridge: Cambridge University Press, 1974) [ISBN 9780521202640].
In this chapter we start by considering what sociolinguistics is and what sociolinguists study. Next, we move on to think about language variation and linguistic varieties and what we mean by these terms. Then we outline some definitions, starting with accent, dialect and language varieties, before moving on to consider what grammatical, phonological and lexical variables are. The last section in the chapter deals with some classic sociolinguistic studies, which as a student of sociolinguistics you are expected to be familiar with. We draw on the groundbreaking work of William Labov, Peter Trudgill, and James and Lesley Milroy.
Whereas linguistics is, broadly speaking, the study of language (rather than the study of a particular language), sociolinguistics is concerned with the relationship between different social and cultural contexts and identities on the one hand, and the way that people speak on the other. Sociolinguistics is a descriptive discipline, and this means that sociolinguists are concerned with describing language variety rather than prescribing how language should be used. So as a student of sociolinguistics, you are going to be interested in the relationship between language and society. One of the core texts (Holmes, 2008, Chapter 1) explains sociolinguistics with examples of language use and variation, which you may want to reflect on before you continue with your reading. It is as well to remember, however, as Coupland and Jaworski (2009, p.2) point out, that ‘sociolinguistics is now a broad and vibrant interdisciplinary project…’. This means that the field cuts across academic disciplines, using theories and methodologies from several disciplines, for example, anthropology, education, social psychology, and media and cultural studies.
Language varies according to who is using it and in what situation or context it is being used. The significant factors which may cause language to vary can be geographical or social and, moreover, languages also change or vary over time. In addition, you should note that sociolinguists often use the term ‘variety’ to describe the linguistic features used in particular geographical regions or in other social or situational contexts. This is because it is a somewhat more neutral and broader sociolinguistic term than others that could be used. In fact, we can use the term linguistic variety to refer to a language, a dialect, an accent, to different languages and even to language styles. Here are some examples of linguistic varieties:
Activity
Can you think of some geographical varieties of English? There may be those associated with different countries and those associated with different regions within a particular country (for example, the UK).
Activity
The variety you use will also depend on your age, sex, ethnicity or social group. In addition, your use of language will also vary according to the situation or context you find yourself in (for example, in a lecture or seminar, at a family dinner, with friends at a social gathering, at a job interview). Think about these different factors and make a list of how these affect the way that you speak.
At this point, it is useful to think more closely about two terms listed above: accent and dialect. As Freeborn et al. (1993, p.59) observe, they are often used as if they mean the same thing. As they point out, you often hear people say things like:
‘She has a broad northern accent’ or ‘She speaks a strong northern dialect’.
In sociolinguistic terms, however, accent is a (regionally or socially distinctive) variety of language identified by particular features of pronunciation (phonology). It is a defining feature of a dialect but may be separated from it. The other point to remember is that everyone speaks with an accent: it is impossible to speak without one, even though a lot of people think they don’t have one.
Dialect, on the other hand, is a (regionally or socially distinctive) variety of language identified by a particular set of words (lexis), grammatical structures (syntax and morphology) and pronunciation (phonology). It can also mean any subdivision of a language associated with a particular geographical area and/or an identifiable group of people. Dialects tend to be mutually intelligible with other dialects of the same language. Here are some examples of variation in different dialects of English:
| Negation: | Nobody wants any sweets/Nobody don't want no sweets |
| I haven't done anything/I haven't done nothing | |
| I don't want any/I don't want none | |
| Present tense: | Jane runs every morning/Jane run every morning |
| Past tense: | I did it last night/I done it last night |
| I gave it to him yesterday/I give it to him yesterday | |
| We were hot/we was hot | |
| Pronouns: | Can I have those books?/Can I have them books? |
It is important to bear in mind that while there may be sociolinguistic definitions of the terms accent and dialect, there are also non-linguistic criteria. These include political, cultural and religious reasons as well as issues of identity, national or otherwise. Languages and dialects can be markers of similarity between people or markers of difference. Consider, for example, groups of speakers in the UK who speak with Scottish accents (whether those associated with Edinburgh or Glasgow) and use certain words such as bairn (to mean a young child). Another example might be an American accent, perhaps the one associated with educated New York speakers that is rhotic (r pronouncing) or the lexical differences associated with American English; for example, sidewalk is used where British English speakers would say pavement.
Activity
List a number of English language varieties, highlight some of their features and think about whether you would classify them as accents, dialects or according to some other criterion. In addition, think about general perceptions towards those varieties. For example, are they highly valued or are they stigmatised varieties? You may need to do some research and reading into these varieties to help you.
If a language variety is highly valued within a society and is generally admired or evaluated highly, it is said to have overt prestige. The standard variety in a society has overt prestige. For example, the list providing examples from different dialects of English that we presented in the previous section contains some forms from what sociolinguists call the ‘Standard English dialect’, which has overt prestige. There are, however, large groups of speakers (in the UK for example) who speak non-standard varieties (that is, they use many non-standard or vernacular forms of language). They are not publicly prestigious and are often stigmatised varieties. An example of this might be Cockney, a variety associated with working class London English speakers. But even though these varieties are not publicly prestigious and are often negatively evaluated, even by their speakers, they continue to be used. There is a prestige value attached to them which is labelled covert (or hidden) prestige.
Activity
Why do you think low or ‘covert’ prestige varieties of language continue to exist? What value do you think they have to their speakers?
You will find out more about the debate around standard and non-standard English in Chapter 3 of this guide.
Sociolinguists investigate linguistic variation by analysing the linguistic variable. As Coates (2004, p.48) puts it: ‘A variable... is something which varies in a socially significant way’. So a linguistic variable is a linguistic feature with various realisations which are known as variants. These variants (look at the examples below) are linguistically equivalent but socially different ways of saying something.
Example of phonological variables:
Activity
One of the variants associated with the variable (u) is associated with Southern British English speakers and the other is associated with Northern British English speakers. Research the historical development of English so that you can see how this difference came about. You may want to begin your research with Chapter 9 in Holmes (2008) (see Essential reading).
Activity
Carry out some reading and research into the other phonological variables mentioned to see if you can ascertain which groups of speakers the variants of these variables are associated with.
Examples of grammatical variables:
Activity
Try to identify what the grammatical variables in each example vary according to. Is it according to region or to some other factors?
Examples of lexical variables:
Activity
Carry out some reading and research into the types of variables mentioned above and try to determine the reason for the variation. Begin your research with the essential reading indicated at the beginning of this chapter.
We have made the point that variation may occur in any speech community depending on regional or geographical factors or due to social factors; for example, class, ethnicity, gender or age. You will need to carry out in-depth and extensive reading on various other types of language variation because it is not possible to deal with all of these within this subject guide. For example, both Holmes (2008) and Mooney et al. (2011) have dedicated chapters and/or sections on each of these, including one concerned with language and age.
Activity
Read Holmes (2008, pp.173–81) and Peccei's chapter 'Language and Age' in Mooney et al. (2011, pp.146–53) to consider linguistic theory and research on age-graded features of speech as well as on the similarities between Child Directed Language (CDL) and Elder Directed Language (EDL).
This subject guide also gives several examples of language variation that are relevant to age as well as to other socio-cultural variables in later chapters. See, for example, the section on crossing or London Jamaican in Chapter 4. In the remainder of this chapter, however, we will concentrate on some studies which have looked at the effect of class. Two other chapters within this subject guide consider the co-variation of language and ethnicity and language and gender.
Let us start by considering some populist work on language and social class, followed by some of the early, classic sociolinguistic studies where researchers considered language variation by looking at linguistic variables and how these varied according to social class.
The first example concerns the British linguist Alan Ross, writing in the 1950s. He wrote an article which provoked a lot of public reaction in which he classified what he described as ‘U’ and ‘Non-U’ usage in relation to language forms, including vocabulary. ‘U’ terms were those used by the higher social classes (upper classes) and ‘Non-U’ terms were used by the lower social classes. After reading Ross’s article, Nancy Mitford wrote a response which was published in Stephen Spender’s magazine Encounter in 1954. The response was also published, along with contributions from others, including a version of Ross’s article, in a book edited by Mitford in 1956. We reproduce some of the items which upper class speakers apparently had different names for in comparison with the names other speakers used:
| 'U' Terms | 'Non-U' terms |
| rich | wealthy |
| luncheon | dinner (also U-children and U-dogs!) |
| writing paper | note paper |
| spectacles | glasses |
| lavatory | toilet |
| false teeth | dentures|
| die | pass on |
| sofa | settee/couch |
| napkin | serviette |
| vegetables | greens |
| mad | mental |
Activity
Which of these words do you use? Do you think usage of any of these would be a reliable indicator of social status today?
Many of the examples seem quite dated some 60 years later. Moreover, it is questionable whether using a particular term instead of another would be a reliable indicator of social status today. The other point to remember is that although lexical items, such as those above, can be a potent marker of social class for older speakers, lexical items go in and out of fashion, and often quite quickly. As we will see in the next chapter, the English language has undergone a process of language standardisation, which has had several consequences. Since around 1800 the structure of the language has changed very little. In the main, changes have been in the area of vocabulary. For example, one item which has been omitted from the list above is ’wireless’ (U term) and ’radio’ (Non-U term). Nobody uses the term ’wireless’ to mean a ’radio’ these days, but the term has come back into usage in the last few years to describe a concept related to internet technology. It can be more productive, therefore, to look at grammatical features and pronunciation features, as these tend to be more stable, reliable indicators of social class: there are features which repeat themselves across the generations. So let us turn to some studies which have considered these types of features.
Labov carried out an extremely simple but successful pilot study in the 1960s where he researched the use of postvocalic (r) in New York. Postvocalic (r) is the sound that occurs after vowels in words like car, floor and heart. We summarise here drawing on Mesthrie et al (2009, pp.82–83) and Labov's (1972) work itself. He carried out his study in three department stores in New York which were quite different in terms of status and prestige: Saks, Macy’s and Klein’s. Saks, the upper class store, was roughly equivalent to Harrods in London; Macy’s, the middle class store, is similar in status to John Lewis, and Klein’s, the least prestigious store, to Woolworths (interestingly, neither Klein’s nor Woolworths are trading today). Labov determined the prestige and status of the store by the newspapers they advertised in and how they priced and advertised their products. The pronunciation of /r/ after vowels is the prestige pronunciation and Labov hypothesised that employees’ pronunciation would reflect the social status of the store:
… our population is well defined as the sales people (or more generally, any employee whose speech might be heard by a customer) in three specific stores at a specific time.
(Labov, 1972, p.49)
Labov asked the sales staff where to find an item which he knew was sold on the fourth floor (postvocalic (r) occurs in both these words). So, for example, he would ask: ‘Excuse me, where are the women’s shoes?’ Next, he pretended he hadn’t heard and asked people to repeat what they had said. In this way, he was able to elicit four examples from each person of the postvocalic (r). Moreover, the second elicitation would be a more careful pronunciation, as this is what people do when they are asked to repeat something someone appears not to have understood the first time around. Next, he moved out of sight of the informant to make a written record of the data. This included details of pronunciation, the estimated age of the informant, their gender, their ethnicity and their occupation within the store. He also went up to the fourth floor and asked the staff there what floor it was. In six- and-a-half hours he managed to ask 264 people which gave him over 1,000 examples of the postvocalic (r).
Labov found that the higher the class of the store, the more people pronounced the /r/. The percentages were approximately 62 per cent for Saks, 51 per cent for Macy’s and 20 per cent for Klein’s. Moreover, in the second elicitation, which he gained by pretending to the staff he asked that he had not heard the first time, in all the stores there was an increase in the use of /r/. The greatest increase, however, was to be found at the middle class store, Macy’s, even more so than at Saks, the more prestigious store. When Labov carried out his larger study, which involved detailed linguistic interviews involving a variety of tasks, a similar phenomenon occurred with his lower middle class informants. They used a higher percentage of prestigious variants than the classes above them when they were asked to do formal tasks as part of the interview.
The use of prestige variants among the upper working classes and lower middle classes which is greater than the percentage used by the groups above them can be described as hypercorrection. Social insecurity is a characteristic of the lower middle classes, who can be described as being on the borderline between the middle classes and the working classes. They are concerned about others’ judgments and perceptions of them, including evaluations about their speech. Hypercorrection involves using variants that a speaker would be unlikely to use if they did not feel they were being evaluated, but that they are using because of the situation they find themselves in (for example, in a linguistic interview). Peter Trudgill (1974), a sociolinguist in Britain, found something similar, although he focused on different features in his work in Norwich. You should note that in Norwich, Trudgill found that upper working class speakers also hypercorrect in careful (that is, self-conscious) speech. This suggests that the upper working class/lower middle class boundary is very sensitive and speakers are more anxious about their social position and how they are perceived. This makes them more self-conscious when they feel they are being judged on their speech. Some of Trudgill’s findings in Norwich are reproduced in Figure 8.4 of Eppler’s chapter ‘Language and social class’ in Mooney et al. (2011, p.167.)
Activity
Note the absence of the upper classes in Trudgill’s study. Can you think of some reasons why this might be the case?
Activity
Both Labov’s and Trudgill’s studies do not only reveal interesting patterns of variation with regards to speakers’ social class background, but they also show how speakers have a range of different styles, from very informal to very formal. Labov’s observation was that the more speakers pay attention to their speech, for example, when they have to read out passages aloud, the more standard pronunciation features they use. There are, however, also other explanations for why and when speakers’ styles vary. Read Chapter 10 ‘Style, context and register’ in Holmes (2008) and consider the range of explanations given for speakers’ variation of style.
The Milroys approached the study of language variation differently from the way in which Labov and Trudgill studied it, making use of the concept of social networks. A social network is like a map depicting the relationships between individuals in a social group. A person whose personal contacts all know each other is said to belong to a closed network, whereas someone whose personal contacts tend not to know each other is said to belong to an open network. Closed networks are described as being of high density; open networks as being of low density. People can be linked in various ways: through family relationships, through friends or acquaintances, through their workplace and through other social activities. Links between people may be multiplex, meaning they are linked in several ways, or uniplex, meaning they are linked in only one way (for example, this describes people in many work places in Western societies).
There is often a social class dimension attached to the types of network as this table indicates:
| Middle Class | Lower working class |
| most open | most closed |
| loose | dense |
| uniplex | multiplex |
| professionals, socially mobile | working class, often rural |
| higher education | less education, less mobility, but can also characterise upper class ties |
The Milroys carried out fieldwork in Belfast in 1975–76 where they studied three lower working class inner-city communities, all of which were characterised by unemployment, crime and illness (‘social malaise‘). However, there was increased solidarity between members of the communities:
Good summaries of the Milroy's work can be found in many of the introductory sociolinguistic textbooks such as Mesthrie et al. (2009). In addition, you can also read about it in Milroy (1987).
There are several important factors about the Milroys’ study which you should be aware of and which we outline here. The political and social situation in Belfast was very difficult at this time. It was the height of the ‘Troubles’ between Protestant unionist and Catholic nationalist communities in Northern Ireland. (If you do not know anything about this period it would be a good idea to do some research to find out more about this period and the reasons for the’ Troubles’.) Outsiders were viewed suspiciously. The research methodology involved Lesley Milroy being a participant observer. The researcher in this study had to be a woman and she had to go into the communities alone. This was because of the tense political situation in Belfast at the time: a male researcher would not have been able to gain access to the communities under study. Milroy gained access to the communities as ‘neither an insider nor an outsider’, but as a ‘friend of a friend’ who, in Ballymacarret, for example, was a student who had lived there. The initial contact allowed her to make further contacts. People felt obligated to help her because she was introduced as a ‘friend of a friend’ and also because she assisted people (helping people to fill in official forms and giving lifts to people and their work equipment in her van). Moreover, her small children were invited to parties and other events and were looked after by local people.
In the three communities under study, ‘extended visiting’ was normal. If the front door of a house was open, neighbours and friends were free to walk into the house and join the others. On the basis of her contacts, Lesley Milroy was able to participate, which allowed her to gather tape-recorded data consisting of formal interviews and spontaneous conversations. In addition, she was able to observe the interactional norms of these communities (see Milroy, 1987, Chapters 3 and 4).
The Milroys devised a six-point scale which allowed them to determine how strongly an individual was integrated into their social networks and how this correlated with their linguistic behaviour. They researched how vowels were pronounced in certain words and demonstrated that, particularly in Ballymacarrett, there was a clear relationship between the variables and network strength. They demonstrated that in close-knit working class communities such as those they studied in Belfast, certain vernacular (that is, non-standard) features are maintained and this is related to the cohesive ties within the networks.
Activity
We have referred to the term vernacular. Sometimes non-standard language varieties are called vernacular varieties but the term has several, related meanings in sociolinguistics. See if you can research and ascertain what these are.
Activity
Research and evaluate the Milroys’ work. Do you think that the concept of social networks is important in establishing and maintaining norms of behaviour including linguistic behaviour?
Activity
After researching and studying some examples of social network diagrams, draw a diagram of a social network you are involved in. You will have to decide how to limit it in some way. When you have drawn it, describe and analyse it, considering what makes it a network and how it affects the members’ behaviour, linguistically and otherwise.
Whereas most of the studies introduced above argue that speakers’ social class background is reflected in their speech, more recently, so-called constructionist theories argue that speakers contribute to the construction of their social class, gender and other identity. This means that speakers use language as a resource to present themselves in certain ways, either as posh, or as tough and cool, and these stances are linked in interesting ways to social class (or gender, or any other aspect of identity). This does not mean that speakers are fully aware of their linguistic ‘choices’ or that there are no constraints on those choices, but speakers are not seen as being pre-determined to speak in a specific way.
A lot of the research drawing on social constructionist theory investigates language use in specific, local contexts as part of an ethnographic project, which is based on long-term fieldwork and observation. One famous study was conducted by Penelope Eckert in a US high school context in the 1980s and 1990s. She observed that in this context two main groups, or ‘Communities of Practice’, as she refers to them, stood out: the Jocks and the Burnouts. The Jocks identified with school and middle class values and aimed to be seen as popular whereas the Burnouts resisted school/middle-class norms and presented themselves as tough in many ways. Eckert argued that this differentiation between Jocks and Burnouts, which was affected by wider social class norms but did not reflect them exactly, was much more relevant to the speakers’ identities than social class per se. Most importantly, in order to present their membership in the Jock or Burnout groups, boys and girls would use a range of linguistic and non-linguistic markers and practices (hence, ‘Community of Practice’).
Activity
We return to Eckert’s research in Chapter 6 of this guide, but for now read Holmes (2008, pp.198–200) and answer the following questions. Which Communities of Practice are described and how do their members signal their group membership and identity?
In this chapter we have introduced you to many terms and concepts which are central to the field of sociolinguistics. In addition, we have also summarised several ‘classic’ sociolinguistic studies where variation has been researched, making use of variables and their variants. Variation may be according to class, ethnicity, gender or age. While we have focused mainly on social class in this chapter, gender and ethnicity are covered in Chapters 4 and 5 and we have drawn your attention to the relevant chapters in Holmes (2008) and in Mooney et al. (2011) so that you can research age as a social variable.
Having worked through this chapter, and done a substantial amount of reading on the topic as well as the activities, you should: