Scottish English: phonology*

Defining the term ‘Scottish English’ is difficult. There is considerable debate about the position and appropriate terminology for the varieties which are spoken in Scotland and which ultimately share a common historical derivation from Old English. Here I follow Aitken (e.g. 1979, 1984) and describe Scottish English as a bipolar linguistic continuum, with broad Scots at one end and Scottish Standard English at the other. Scots is generally, but not always, spoken by the working classes, while Scottish Standard English is typical of educated middle class speakers. Following Aitken’s model, speakers of Scottish English either switch discretely between points on the continuum (style/dialect-switching), which is more common in rural varieties, or drift up and down the continuum (style/dialect-drifting), which is more characteristic of the urban dialects of cities such as Edinburgh and Glasgow. Throughout Scotland, Scots is increasingly becoming limited to certain domains, for example, amongst family and friends, while more formal occasions tend to invoke Scottish Standard English. Of course the boundaries between Scots and Scottish Standard English, and English English, spoken by a small percentage of the population, are not discrete, but fuzzy and overlapping.

Scottish Standard English, taken here as Standard English spoken with a Scottish accent, is a possible variety for many speakers across Scotland, depending on social context. There are only slight regional differences in Scottish Standard English across the country. Scots is also widely available to speakers in the appropriate context. The Scottish National Dictionary recognizes four main dialect divisions of Scots whose names reflect their geographical distribution across Scotland: Mid or Central Scots, Southern or Border Scots, Northern Scots, and Insular Scots. Alongside spoken Scots, there also exists a literary variety, Lallans (literally ‘Lowlands’), but this is rarely spoken and thus not discussed here. Northern Scots, particularly the variety spoken in the North East, is often called the Doric. Urban Scots spoken in the cities of Glasgow and Edinburgh and across the Central Belt, is historically derived from forms of Central Scots.

The Scottish English continuum is the result of dialect contact and language change over many centuries. A brief account follows. Before the Anglian invasions during the seventh century AD, Scotland was predominately Celtic-speaking. The invaders introduced a northern variety of Anglo-Saxon (‘Anglian’) into south-east Scotland. A century and a half later, the southern borders of Scotland were invaded again by the Vikings, who also separately reached the far north of the country. At the time of the Norman Conquest, most people in Scotland spoke a form of Celtic. Anglian was spoken in the south-east, and Norse was used in the far north and possibly in the western borders. Political developments in England and Scotland during the twelfth century led to an influx of northern English speakers into Scotland. The twelfth to the fourteenth centuries saw the gradual development of a particular variety of English in Lowland Scotland which we recognize as Scots, but which was known as ‘Inglis’ (Gaelic was called ‘Erse’ or ‘Irish’). By the fifteenth century Scots was noted as distinct from contemporary forms of southern English English. Despite the early Anglian settlement, the main historical basis of Scots was probably the language of northern English settlers from 1100 onwards, which was considerably influenced by Norse after the long period of Scandinavian occupation of the north of England. Prolonged contact with Norman French also contributed to its distinct character. Before the first large-scale literary work in Scots, Barbour’s Brus (1375), preliterary Scots is only scantily attested, e.g. in place names and glosses. In 1398 the Scottish Parliament moved from Latin to Scots as the language of record, and until the Union of the Crowns (1603), Scots flourished as a literary and spoken language. Thereafter, with increasing English influence, particularly after the Act of Union of the English and Scottish parliaments in 1707, the use of literary Scots declined beyond specific literary genres (e.g. comedy, satire) and gave way to Standard Southern English, which is today the written standard. The eighteenth century also saw the development of Scottish Standard English in the emergence of a variety of Standard English spoken with a refined Scottish accent, typically by the middle classes whose reference for prestige were Southern English accents of England. While literary Scots declined, spoken Scots remained vigorous, at least in rural areas and among the burgeoning working classes. Despite ongoing dialect change and levelling of Scots towards Scottish Standard English, this linguistic situation still persists, although with the additional qualification of Scots as either ‘good’, i.e. traditional and rural, or ‘bad’, i.e. degenerate and urban (cf. Aitken 1984: 529).

It is probably fair to say that a good proportion of the population of Scotland, now estimated at 5,062,011 according to the 2001 census (GROS 2003), are potential speakers of Scottish Standard English. There are no official estimates or census statistics for the number of Scots speakers in Scotland, although Scots is now counted as a ‘language’ by the European Bureau for Lesser-Used Languages. Defining the number of speakers of Scots in Scotland is extremely difficult, and cannot be easily resolved by asking speakers. The problem is created and exacerbated by a number of interrelated factors:

1. the difficulty of recognizing Scots as a variety which is linguistically distinct from Scottish Standard English (for both linguists and native speakers);

2. the broad range of communicative competence in Scots found in speakers across Scotland;

3. the unresolved difficulty of determining whether Scots is an autonomous language;

4. the negative attitudes held towards Urban Scots, which is often regarded as a degenerate form of speech synonymous with slang (e.g. Macafee 1994);

5. the ongoing process of dialect levelling towards English throughout Scotland.

Two recent studies (Murdoch 1995; Maté 1996) have attempted to survey the number of Scots speakers, and at the same time (Maté 1996) to evaluate the feasibility of assessing Scots-speaking population through a survey tool such as a Census question. The number of self-professed Scots speakers was relatively low in both sample surveys (57% in Murdoch, 30% in Maté). In both cases, older working-class speakers were more likely to classify themselves as speaking Scots. The conclusions of Maté’s research, sponsored by the General Register Office For Scotland, state that the “inclusion of such a Census question would undoubtedly raise the profile of Scots” (1996: 2), but at the same time do not argue strongly for the Census as the optimal tool for estimating Scots speakers:

Adequate estimates of the numbers of people who assess themselves as Scots speakers can be obtained from sample surveys much more cheaply than from a Census [...] A more precise assessment of genuine Scots language ability would require a more in-depth interview survey and may involve asking various questions about the language used in different situations. Such an approach would be inappropriate for a Census. (Maté 1996: 2)

The 2001 Census did not include a Scots language question.

It is possible to provide a very gross estimate of Urban Scots speakers by using Census data which refer to the population of the Central Belt. Of the total population of the Central Belt, 3,088,938, 66% are assigned to classes 3–8 of the socio-economic classification index used to compile the Census. If we guess that people assigned to these classes may in some domains and to differing degrees be more likely to use Scots than those in classes 1–2, on the grounds that Scots is likely to be continued in the lower middle and working classes (including those who have never worked or who are long-term unemployed), we could suggest that potentially this proportion has access to Urban Scots in some form. The population of the Central Belt makes up approximately two-thirds of the population of Scotland, and hence those classified as class 3–8 in that area make up 40% of the total population. People assigned to classes 3–8 in Scotland as a whole gives 67%, which might be very roughly indicative of a potential for Scots across the country, though this is much less certain.

Beside the varieties of English origin which make up the Scottish English continuum, there are also other languages spoken in Scotland whose influence on Scottish English is known to a greater or lesser extent. Scottish Gaelic was once widespread across Scotland, particularly in the Western Isles and Highlands of Scotland. The proportion of speakers bilingual in Gaelic and English living in Scotland is now estimated at 58,652 (1.2% of the population; a slightly higher figure reported comprehension of Gaelic: 93,282, 1.84%). These figures are a slight reduction from those registered in 1991 (65,978, 1.3%). The English spoken in these areas, and also in small Gaelic/English bilingual enclaves in the cities, such as in Partick in Glasgow, has particular phonetic and phonological characteristics, for example the realization of /l/ as clear in all environments (e.g. Johnston 1997: 510), or the use of voiceless /s, ʃ, tʃ/ where voiced  are expected, or the retroflex fricative  as the outcome of /rs/ in words like force, some of which are due to Gaelic influence.

Another small subset of the population of Scotland are recorded in the Census as belonging to an ethnic minority. The number of people defined as ‘Pakistani/ Indian/Bangladeshi/Chinese/other Asian/Black-African/Black-Caribbean/Black-Other/Mixed/Other’ make up 2% of the total population of the country, and 5.45% of the population of Glasgow (GROS 2003). As Verma (1995: 120) has pointed out, this substantial ethnic minority population also has linguistic implications, leading to “the recent emergence of a bilingual, and culturally and linguistically diverse, population in schools, where for historical reasons monolingualism was the norm”. His analysis of data for ESL provision for the Lothian region reveal 54 languages other than English in primary schools, and 37 in secondary schools, with overall Punjabi and Chinese (Hakka/Cantonese and Mandarin) as most common. The extent of influence of South Asian languages such as Punjabi, on Scottish English and particularly Urban Scots, has not yet been investigated, but my own informal observations suggest that younger members of these communities do show distinctive features, particularly in the realization of FACE and GOAT as closer monophthongs (even with expected breaking), some retraction in the articulation of /t, d/ which are often fronter in Scottish English, and characteristic patterns of intonation (higher nuclear tones) and voice quality (more nasalization and tenser phonation).

Reviews of Scottish English phonology, such as that of Wells (1982: 393), typically concentrate on Scottish Standard English (ScStE), and for good reasons. After all, one could assume that Scots is a language distinct from English and hence not within the scope of any discussion of ‘English’ in Scotland. Certainly, Scots phonology is largely defined through a rather different lexical distribution resulting from differing historical developments in Scots (Wells 1982: 396). However, at the same time, excluding Scots means effectively excluding description of the possible phonological range of a very large number of speakers for whom Scots is a seamless part of their linguistic repertoire. Certainly any sociolinguistic analysis of urban Scottish English which includes phonetic or  phonological variables and which includes working-class or lower middle class (or even middle-middle class) speakers is going to encounter Scots in some form. This will be most overt in lexical alternations such as hame /e/ for home, usually ScStE /o/. It will be less clear for those vowels whose lexical incidence is largely the same, such as Glaswegian KIT/BIT, and where socially-stratified variation occurs along a continuum correlating with social class (e.g. Macaulay 1977). However, close analysis of such data often reveals particular patterns of variants which may occur in working-class speakers that make more sense if we can acknowledge them as ongoing developments within and from Scots. Vowels and consonants may appear to be ‘the same’ in Scottish Standard English and Scots, but the patterns of variation may be rather different, and these differences may correlate with linguistic heritage (Stuart-Smith 2003). Of course this explanation makes it sound as if ScStE and Scots are distinct linguistic entities and the difficulty is that of course they are not. Nevertheless the blurred observable socio-phonetic continua do seem to show focussing about two poles, or at least about one which is ‘ScStE-like’ and another which owes much, but certainly not everything, to what I call Scots here. Another motivation for including some discussion of Scots is provided by recent results of variation and change in Scottish English. For it is the speech of working-class youngsters which is showing the most vigorous innovation and change, and hence it seems that Urban Scots is undergoing the most far-reaching changes.

Thus I take the view here that Scottish English must refer to the entire continuum, not simply to Scottish Standard English, and Scots is therefore included in my discussion. However, I too must choose an uneasy compromise in what material may or may not be included, since there is not space here to outline the phonology of Scottish Standard English and Scots in their entirety. Given that around two-thirds of Scottish English speakers inhabit the ‘Central Belt’, which loosely refers to the cities of Glasgow and Edinburgh and the relatively small strip of land which lies between and about them, and because much recent phonetic and phonological research has been carried out on these accents, the material is biased towards these accents and especially Glaswegian. ‘Scots’ here generally refers to continuations of Central Scots found in contemporary Urban Scots. For an outline of historical developments in Scots, for the most comprehensive discussion of regional differences in Scots phonology. Macafee (1997) provides a full review of sociolinguistic results, many of which are phonological.