The Ulster-Scots Dialect Boundaries in the Province of Ulster*

Robert J. Gregg

Introduction: General Background to the Research Project

A comprehensive and systematic survey of any aspect of the Ulster-Scots (US) dialects in Ulster would be a linguistic study interesting and valuable in itself. The survey, whose results are presented here and for which the culminating fieldwork was completed between 1960 and 1963, becomes much more meaningful, however, when related to the wider background of research in English dialectology in different parts of the British Isles and North America prior to 1963.

In Ulster itself, a programme of dialectal research had been initiated in 1951 — almost ten years before the present study was begun — by the Folklore and Dialect Section of the Belfast Naturalists' Field Club, which in addition to its own efforts undertook in 1953 to work in close co-operation with the Linguistic Survey of Scotland (LSS), in particular by circulating the Scottish postal questionnaires to suitable Ulster informants and excerpting the information they provided before returning them to the Edinburgh headquarters. From this activity much valuable material was collected, not only in Edinburgh, but in Belfast where files were established, maps drawn, articles written, talks given, and seminars and exhibitions held.

For Ireland as a whole, a carefully planned preliminary linguistic survey had been carried out by Dr. Patrick L. Henry during the years 1953-55 and the results published in Lochlann.(1)

The LSS, mentioned above, from its beginnings in 1949 had accumulated an impressive archive in Edinburgh, resulting from two widely plied lexical questionnaires compiled by Angus McIntosh, H. J. Uldall, and Kenneth Jackson and circulated by mail. The resultant information had been tabulated and some items tentatively mapped. The data provided by a later phonological questionnaire devised by J. C. Catford for the Lowland Scots dialects were also being analysed linguistically. The Scottish surveys, incidentally, were planned to include the two northernmost counties of England and the whole province of Ulster.

From Leeds as centre, an extensive survey of the dialects of England had been in progress since 1950, organized by the late Harold Orton and Eugen Dieth and using a questionnaire which included phonological, morphological and syntactic, as well as purely lexical, items. A preliminary explanatory volume — including the questionnaire — had been published in 1952, and ten years later a series of detailed reports began to appear, the first volume of which was entitled The Six Northern Counties and the Isle of Man (1962).(2)

The fieldwork for the Ulster survey, whose ultimate object was the mapping of the distribution and boundaries of the US dialects throughout the province, was started early in 1960 and completed during the summer of 1963.

It was obvious that the delineation of boundaries between the two major varieties of dialectal English spoken in Ireland would be of importance internally to any Ulster or general Irish linguistic survey and would be equally the concern of the Edinburgh investigation, which had always been intent on tracing the external expansion of Lowland Scots speech. The dialects of US, of course, march — on almost all their limits —with Ulster Hiberno-English (UHE). The latter proved to have characteristic features that tied in with — frequently archaic — North and West Midland English dialectal forms, thus creating a natural link with the Leeds survey.

These bonds with British dialectology seemed easy enough to establish, but important also — although rather less obvious — were the links with North American linguistic geography. Dr. Hans Kurath and his fellow dialectologists in the United States had found traces of the Ulster-Scots in parts of New England, in Pennsylvania, the Ohio Valley, the Southern Highlands and, at a later stage, points west of these areas. In fact, the Scotch-Irish (so styled) immigration to North America was so intense in the early 18th century that, just after the War of Independence, estimates show that out of a total population of three millions, between half a million and a million were of Ulster descent.(3)

Later, in the 19th and 20th centuries, eastern Canada had received (and is still receiving) a large influx of immigrants from Ulster, who had left their linguistic mark in various parts of Ontario: in particular, the Ottawa Valley, Essex County, and the area around Peterborough and Parry Sound.(4)

It must be borne in mind, however, that when North American historians or linguists use the term 'Scotch-Irish' it is in a rather wide sense, covering things or persons of Ulster origin in general and, with reference to speech in particular, covering all types of Ulster dialect, which are perceived as clearly distinct from the typical and well-known southern Irish 'brogue'.

When we use the term Ulster-Scots (US) with reference to Ulster itself, it is necessary to define it more closely. We must distinguish two types:

(i) rural US dialects — the subject of the present study — still spoken in the areas that were most intensively peopled by Lowland Scottish settlers during the 17th century.(5)
(ii) urban US speech, which is a regional version of the standard English language, heard in towns situated within the above mentioned areas, and also used by educated rural US bilinguals as their 'other' language. With the extension of education and the development of modern communications, the spread of this second variety of US at the expense of the first is everywhere apparent and in time seems likely to supplant it.(6)

Apart from complementing current surveys in other parts of the English-speaking world, the study of rural US has another aspect which is of particular interest to dialectologists, philologists, lexicographers, etc., namely its preservation of certain apparent archaisms in contrast with the dialects of Ayrshire and SW Scotland in general, from which source it springs. Thus in the phonological changes affecting the vocalic nucleus of words with OE -ald, the Early Scottish transitional diphthongal stage, au, has been preserved in all the Ulster-Scots districts — the US reflex being centered to [əü] — whereas in most Scots dialects the early diphthong was levelled to [ɑ:] or [ɔ:]. This diphthongal reflex survived in Scotland in only a few very marginal places, e.g., in Campbeltown (Argyllshire), Caithness, on Black Isle and Easter Ross, and in parts of Ayrshire.(7) Its very existence in Ulster is evidence of its persistence and widespread — if not universal — use in SW Scotland in the early 17th century and perhaps even later. The competing forms with [ɑ:] crop up in only a few words in a few places in Ulster and always in addition to the [əü]-forms. Mostly there is a semantic difference between the two forms, so that they are not actually in competition with one another.

The [ɑ:]-form of old suggests familiarity and affection,(8) as in

[ˈɑ:l ˈfok]old folk
[ˈɑ:l ˈtəimz]old times
[ðeˈɑ:l ˈsü:]the old sow
[ðeˈɑ:l ˈmi:r]the old mare

On the other hand, the [əü]-form may be used in a derogatory (9) sense:

[ən ˈəül ˈgæ̈rn]

[e ˈgæ̈rne ˈəül ˈɑ:tərˌkɑ:p]

both expressions being used of a person who is continually complaining. Similar semantic oppositions are valid for [kɑ:l] versus [kəül], meaning cold.

There is the further series of archaisms in the vowel of such words as above, done, good (i.e., the reflex of ME ō), which still have [i] in many parts of Ulster (generally remote from the ports of entry of the 17th-century Scots settlers). These [i]-forms were thought to be paralleled only in the NE Scots dialects, but recently the LSS has come upon a SW area at Wanlockhead and Leadhills on the borders of Dumfries and Lanarkshire where the same [i]-forms have survived. Their distribution in Ulster would argue that here again we are dealing with a pronunciation that was widely current, if not universal, in SW Scotland during at least the early 17th century. The forms now current in SW Scotland with [ɪ̈] or [e] instead of [i] are found in Ulster but are restricted to the N Ards and county Antrim, i.e. places close to the Scottish source of diffusion for these innovations.

In the same way archaic or obsolescent pronunciations stemming from N and W Midland English sources are still currently widespread in the UHE areas and are in fact considered so 'normal' that the US speakers who wish to use what they think of as 'standard' have actually adopted these English dialectal archaisms, assuming them to be 'correct' English over against their 'broken Scotch'. Thus, they reply to queries about the local name for a female sheep by saying: 'We ca' her a [jəü], but it should be a [jo:], should it no'?' Similarly with thaw [θo:] and straw [stro:].(10)

From the point of view of general linguistics, US dialectology reveals several interesting phenomena worthy of discussion.

Many linguists are concerned with the general study of what happens when a new language makes a massive incursion into the domain of a language of a very different type. The resultant mutual adjustments which inevitably take place during the period of bilingualism preceding the ultimate and perhaps inevitable disappearance of one of the two languages can be studied at all linguistic levels —phonological, morphological, syntactic, lexical and semantic. For the purposes of the present study the phonological aspect of these adjustments is of the utmost importance, and a detailed investigation of the relationship between the phonology of US urban and rural on the one hand and that of county Antrim Gaelic on the other has shown that there must have been a considerable carry-over from the latter to the former.(11)

Introduction | Historical and Geographical Factors | Lexical Questionnaire | Phonological Questionnaire | Discussion | Notes