The Ulster Dialect Survey*

Robert J. Gregg

The primary purpose of the project — under the direction of the well-known folklorist Dr. Richard Hayward and Mr. G. B. Adams of the Belfast Naturalists' Field Club — is to collect material for an Ulster Dialect Dictionary. Over 12,000 words have poured in to headquarters in Belfast, including useful contributions from Canada, where in Toronto, for example, many well-known Ulster dialect words such as dunt 'to nudge', clabber 'mud', gully 'large knife', spraghal 'sprawl', stiaghy 'an unappetising mixture of food', switherin' 'hesitating', and the poetic dailigon ('twilight', i.e. 'daylight gone') are still in use, if not publicly, at least within the family circle, among those whose ancestors came out from 'the ould sod' even as much as seven or eight generations ago. Because a great deal of attention has been given to the exact pronunciation and distribution of the dialect words within the northern province of Ireland, it is hoped that the information accumulated will be a help to the linguistic geographers of Canada and the U.S. who are trying to trace certain North American usages to a precise spot in the British Isles.


Since the scope of [the Canadian Linguistic] Association's interests was officially widened some time ago, I do not need to apologize for choosing a topic which seems to be so far over the horizon from a Canadian point of view. It may be worth emphasizing, however, that the Survey I am discussing, which deals with the dialects of English spoken in Ulster, is not without relevance for other similar surveys both in Great Britain and in North America. From a diachronic viewpoint Ireland in general and Ulster in particular is in some ways a linguistic museum where we find preserved many features apparently characteristic of the speech of both England and Scotland during the 16th and 17th centuries, which is of course also the period during which English moved across the Atlantic to gain a foothold in North America. A further trans-Atlantic link arises from the mass migrations during the 18th century of the 'so-called' Ulster Scots (or Scotch-Irish) to the U.S., as well as from their movement in more recent times to Canada, especially Ontario, where their influence seems to be most clearly discernible in the Ottawa Valley region. North American linguistic geographers thus still find traces of Ulster speech in various parts of Canada and the U.S., just as English and Scottish dialectologists find remnants of their own dialects in Ulster.

Apart from this historical and linguistic-geographical relevance, the Ulster dialects have features of interest to general linguistics, especially those which arise from the creative interaction of two widely divergent language types, viz., Celtic and Germanic, which started on the east coast of Great Britain some 1500 years ago and which is still a vital issue in the Gaeltacht of county Donegal to-day.(1) The vagueness which usually attends substratum, adstratum and superstratum discussions elsewhere does not apply to the situation in Ulster, for we may still refer to the source dialects in England or Scotland to the east, and to the living Gaelic dialects of the west in our attempts to disentangle the constituent strands — phonological, lexical, syntactical —which make up the present-day Northern Irish dialects.


Although the importance of Ulster dialects has long been recognized by outside linguists, the detailed appreciation of their influence has been hampered by a lack of scientific descriptions comparable to those available for England, Scotland and North America in the works of men like Orton, Dieth, Kurath, etc. Thus an important gap was filled when the Ulster Dialect Survey was started in 1951, and the task of collecting dialect material was undertaken by a specially created section of the Belfast Naturalists' Field Club, an old-established and well-connected learned society with wide-ranging regional interests. It was decided that the area under consideration should include not only the six counties of the political unit known as Northern Ireland, but also three other counties — Donegal, Cavan and Monaghan — belonging to the historic province of Ulster, and two other border counties —Louth and Leitrim — as well. The whole organization was set up on a voluntary basis by the prime movers — the well-known folklorist Mr. Richard Hayward and his nephew Mr. G. Brendan Adams, a brilliant philologist and gifted linguist. Six other collector-editors were co-opted, including myself, and the whole area was divided into eight zones with a collector-editor in charge of each. I had been collecting dialect material independently for over twenty years at the time, especially in county Antrim. One of the most enthusiastic sponsors of the Survey was Professor E. Estyn Evans, Dean of the Arts Faculty at Queen's University, Belfast, professor of Geography, archeologist, and keen supporter of regional studies. Reliable informants were found by means of radio-talks, articles in local newspapers and periodicals, and questionnaires went out to teachers, lighthouse-keepers, fishery inspectors, etc. I had replies with useful contributions from places as far afield as Toronto and Philadelphia. A team of a dozen writers or recorders was recruited whose job it was not only to sort out and classify all the in-coming words but also to extract and file material from the old printed glossaries dating chiefly from the late-19th century.

Almost immediately, contact was established with Professor Angus Mcintosh and his colleagues on the Linguistic Survey of Scotland, a relationship which has, I think, proved very profitable to both sides. The two bulky Edinburgh questionnaire books have been circulated to informants throughout our Ulster area, and our own recorders have entered all the information thus gathered on to our cards before returning the books to Scotland. During the summer of 1953 I worked with Mr. J. S. Woolley, who had come over to do some exploratory field-work, and in 1957 and 1959 Mr. J. Y. Mather spent part of the summer doing a more detailed investigation of word-distributions and phonology. Until his departure for India close contact was also kept with Mr. J. C. Catford on phonological matters.


The original purpose of the Survey was to produce an Ulster Dialect Dictionary comprehensively documented with phonological, distributional, etymological and semantic data for each entry. To date some 12,000 words have been collected but the final total may be around 15,000, and the present plan is to publish this material in fascicules corresponding to the letters of the alphabet. Before my departure for Canada about five years ago Mr. G. B. Adams and myself went through the entire file and with the OED and Henry Cecil Wyld as our guides critically examined the entries and selected the genuine dialectal items. Occasional entries from old glossaries prove to be standard language and not dialectal, and some informants are liable to confuse dialect with slang. Our preliminary inquiries had put us in touch with reliable informants in every corner of our region, and we are now going back to these with our edited lists of, for example, the A-words, the B-words, etc., and checking with them the current use, non-use or obsolescence, the distribution, meaning(s), and pronunciation of each item. Since 1957, when the Ordnance Survey published its new ¼" map covering Northern Ireland and adjacent areas to the south and west, we have scrapped our old reference system in favour of the Irish grid system, which the new map employs. We use the grid reference letters and numbers to refer to the places from which we have received information, and this is now immediately intelligible to our Scottish colleagues, who have a similar arrangement. We have also for our own purposes divided our area into the larger baronies and groups of smaller baronies. The barony was an ancient administrative unit in Ireland, and its boundaries often mark the limit in the distribution of speech features and other cultural traits.

Most of our original informants belong to the older generation and to counteract this bias we have recently made a check with the co-operation of the Stranmillis and St. Mary's Teacher Training Colleges on the speech of the younger folk. The students in these establishments come from all parts of the province and range in age from seventeen to about twenty-one. This investigation has given us valuable information about the obsolescence and the distribution of words and the results are now being utilized in a series of maps by Mr. G. B. Adams.

From the beginning it was decided that the fullest possible phonological information should be sought from all parts of our territory. Mr. G. B. Adams has already published An Introduction to the Study of Ulster Dialects (which includes a description of Standard English as spoken in Belfast) and two further studies dealing with Donegal and Antrim phonology. I have personally made a special investigation of the pronunciation of parts of county Derry, county Down and county Cavan as well as a study in depth of specific county Antrim urban and rural sound systems. On the basis of information collected we have been working on the complex problems of phonemic analysis and an orthography for dialect words which would meet the conflicting claims of the various regional speech subdivisions. The only previously existing phonological data published are the entries under Antrim and Ulster in J. S. Wright's English Dialect Grammar, which unfortunately turn out to be unreliable in many particulars. A preliminary critical account of Wright's material has been written by G. B. Adams in his paper on Antrim phonology, and I intend to follow this up at a later stage with a 'correction' of Wright's forms so that investigators using the EDD may no longer be misled by erroneous statements.

Further aims include an investigation of the etymological and semantic aspects of each entry. Work on this has begun but will not be fully developed until the collection of material is completed. Special lists of dialectal plant and animal names are being drawn up by our botany and zoology experts, which will be useful, it is hoped, to their colleagues in biology and others unfamiliar with our dialects.

So much for the original purpose. In May 1959, however, it was decided to postpone indefinitely the Dictionary project and instead to follow the Edinburgh example by setting up an Ulster Dialect Archive to be housed probably in the new Ulster Folk Museum building under the care of a Board of Trustees and available to those engaged in research and, under supervision, to the general public. At the same time a guide will be published containing the material which was to have been included in the Introduction to the Dictionary of the original scheme: (1) a preface, (2) an historical account of dialect studies in Ulster, (3) a list of informants with grid reference, (4) an index of the main items in the records, (5) a register of all phonological work published or in process of collation, and (6) some special articles by various contributors, which would be varied in successive editions.


When it comes to assessing the results of our eight years' work we can claim, I think, that apart from the ever-growing lexical archives we have now a much clearer picture of the nature and limits of the various dialects and sub-dialects spoken in Ulster. Our impressions have been confirmed by the recent map-work done by G. B. Adams and further by the publication of 'A Linguistic Survey of Ireland: Preliminary Report'(2) by Dr. R L. Henry of University College, Dublin, whose entirely independent researches have given similar results. He recognises in Ireland four types of dialect:

(1) Ulster Scots

(2) Mid-Ulster English

(3) Irish (i.e. Gaelic)

(4) Blended dialects of British English with Irish.

Of these (1) and (2) are of course confined to Ulster, but types (3) and (4) occur in Ulster as well. Type (1) is distributed in a broad arc reaching from the Mourne Mountains in county Down through the counties of Antrim and Derry to East Donegal and coincides basically with the areas heavily settled by Lowland Scots planters in the 17th century. Type (2) corresponds closely to the English plantations of the same period and represents chiefly an extension of W. and N.W. Midland and N. Western English dialects. Owing to their geographical proximity the Scottish planters in the early part of the plantation period tended to predominate. They penetrated to some extent this English-settled area so that even here certain Lowland Scottish speech elements are present. Later, especially in the 19th century, industrial developments favoured the expansion of type (2), which spread down the Lagan Valley, widening the wedge between the southern and the northern branches of type (1) and becoming predominant in particular in the city of Belfast. Type (3) in Ulster is concentrated chiefly in the Donegal Gaeltacht, which has shrunk even since 1925, and perhaps a few pockets of Gaelic speech in the Sperrin Mountains in N. Tyrone. The Gaelic of Rathlin Island and the Glens of Antrim investigated by Nils Holmer of Uppsala in the 1930s must by now have completely died out, as has that of the Mournes and of S. Armagh reported on in 1929 by Professor Alf Sommerfelt of Oslo. Gaelic has also disappeared as a spoken language in the Breifny area in S. W. Ulster. Type (4) has arisen in comparatively modern times in areas such as the Glens of Antrim, which were hitherto Gaelic-speaking — actually since the Great Famine of about 100 years ago, before which over half the population of Ireland (which was c.8,000,000 at the time) was Gaelic-speaking. Consequently the form of English that was adopted tended to be essentially the current standard form of the language rather than any of the neighbouring older-established dialects. In Ulster most of these areas were strung out along the southern border of the province and further north formed a buffer between the Ulster-Scots dialects to the north and the Mid-Ulster English further south. The border areas, although in Ulster, generally adopted with their English a pronunciation strongly influenced by Dublin standards and this has given rise to what may be called a border brogue, typical, for example, of towns like Newry in S. Down or whole counties like Monaghan, Cavan and Fermanagh. These dialects are also characterized by the relatively large number of unaltered Gaelic words and phrases still used in the midst of their English and the frequent use of Gaelic syntactical patterns and translated idioms. It would be a mistake, however, to regard these areas as quite distinctly marked off from each other, for, as P. L. Henry puts it, 'Ulster dialects should rather be visualized as a series of concessions between the four types described above than as a tripartite division with unmixed exponents of each'.

Leaving Gaelic out of account, we may draw up markers of various kinds which help to distinguish the other three dialect types. Even a casual scrutiny of the data will show that the Ulster-Scots dialects are the most clearly differentiated from every point of view - phonetically, lexically, morphologically and syntactically. Educated speakers in the Ulster-Scots areas, however, use when occasion demands it, and especially in the towns and villages, a type of English which approximates much more closely to the norm of the other areas. The division between Mid-Ulster and the Border dialects is much less clearly drawn. In fact, it is rather a question of shading off from one type to the other.

It should be noted first of all that there are certain phonetic features common to almost all Irish dialects of English, notably:

(a) a 'light' (front resonance) lateral which occurs even in word final and pre-consonantal positions (as in hill, feel, field, etc.)

(b) a point-open frictionless continuant r in word-final and pre-consonantal positions (as in far, farm, etc.)

(c) a single-flap r following interdentals, with or without an intervening schwa (as in tree, dry, three, bothering, etc.)

(d) interdentals in close connection or in direct contact with a following r of either type (as in mutter, drink, flattery, banner, balderdash, etc.)

(e) an open front vowel in words like bad, hand, etc., although Ulster-Scots has an open back vowel in such circumstances.

(f) a mid, back-centred, rounded vowel in words like cut, plum, etc., although Ulster-Scots generally has a lowered mid, back, unrounded vowel in such words.

At the same time there are phonetic features which clearly mark off all types of Ulster English from the English spoken in the rest of Ireland, the most conspicuous of which are:

(a) an extremely fronted (actually front-central), somewhat under-rounded vowel in words like too, boot, foot, etc.

(b) a raised and strongly over-rounded o in words like go, hope, boat, etc.

(c) a very narrow descending diphthong, [əi], in words like bite, tide, fine, etc.

(d) a similar very narrow diphthong with the vowel described above under (a) as second element, [əü], in words like now, house, cloud, blouse, and also old, cold, told, etc.(3)

In spite of the fact that the basic repertoire of speech sounds is similar all over the northern province there are nevertheless a few features which help us to distinguish immediately between the various Ulster dialects. The Ulster-Scots dialects, for example, are contrasted with the others as shown in the following oppositions:

Others Ulster-Scots
(a) vowels ranging from [ι] to [ϊ] (a) an extremely open, slightly centred front vowel in words like big, hill, sit, sieve
(b) front open [a] (b) a long back-open vowel in bad, cat, stamp, hammer, etc.
(c) vowels ranging from [ɒ] to [ɑ] (c) a long, back, rounded, half-open vowel in cot, caught, rock, bottom, etc.
(d) the back but somewhat centred, half-close, rounded vowel [o̤] (d) a lowered version of [ʌ] in cut, mud, supper, etc.
(e) only the close tense front-central slightly rounded vowel [ü] (e) an open, lax version of the front-central [ü] in words like book, room, hoop, etc.
(f) only a 'narrow' type diphthong [əi] (f) a 'broad' diphthong of the type [ɑːe] in words like pie, dial, size, alive, etc., in addition to the 'narrow' diphthong [əi] in other words
(g) short vowels in such words (g) long, half-open and open vowels in words like step, hat, dock
(h) palatalized velars in such words (h) 'normal' velars in words like cat, garden, bang
(i) tendency to voicing these medial plosives, especially with medial t (i) glottalised plosives medially and sometimes finally in words like pepper, butter, baker, what, or even simple glottal stop medially in words like butter

The Border dialects are generally distinguished from both Mid-Ulster and Ulster-Scots by their tendency to use interdental stops in place of [θ] and [ð] in words like three, other, etc., and by the bodily inclusion of Gaelic words and phrases — without any phonological modification towards English — in the middle of English utterances.

At the lexical level there are again many clear-cut phonological oppositions (representing historical divergences of sound) which can be utilized in separating the dialects. Such are:

Mid and South Ulster Ulster-Scots
(a) two, who [tüː], [hüː] [twɑː/twɔː], [ʍɑː/ʍɔː]
(b) alone, home, soap [əˈloːn], [hoːm], [sop] [əleːn], [heːm], [seːp]
(c) blow, crow, sow (verb) [bloː], [kroː], [soː] [blɑː/blɔː], [krɑː/krɔː], [sɑː/sɔː]
(d) ball, fall, wall [bɒːl], [fɒːl], [wɒːl] [bɑː/bɔː], [fɑː/fɔː], [wɑː/wɔː]
(e) father, flat, grass [ˈfaːðər], [flat], [graːs] [ˈfɛːðər], [flɛːt], [grɛːs]
(f) bread, head, mare [brɛːd], [hɛːd], [mɛːr] [brid], [hid], [miːr]
(g) web, wet (adj.), wren [wɛːb], [wɛt], [rɛːn] [wɑːb], [wɑːt], [rɑːn]
(h) clean, lean (vb.), mean [kleːn], [leːn], [meːn] [klin], [lin], [min]
(i) dinner, quit [ˈdιnər], [kwιt] [ˈdɛːnər], [kwɛːt] or [kwɑːt]
(j) twin, witch, whin [twιn], [wιtś], [ʍιn] [twʌn], [wʌtś], [ʍʌn]
(k) blind, find [bləind], [fəind] [blæ̈n], [fæ̈n]
(l) cow, house, mouth [kəü], [həüs], [məüθ] [küː], [hÜs], [mÜθ]
(m) high, sight, fight, laugh, cough, tough [həi], [səit], [fəit], [laːf], [kɒ:f], [to̤f] [hix], [sæ̈xt], [fɛːxt], [lɑːx], [kɔːx], [tśʌx]

Lexical items may be opposed to each other not on the basis of phonological divergence but of entirely separate words:

Mid and S. Ulster Ulster-Scots
eariwig gellick
heifer quey
clootie 'left-handed' fyuggie
pistroag 'object of superstition' freet
bonnive 'smallest pig in the litter' crowl
colcannon 'potatoes mashed with milk, and cabbage, chives, etc. added' champ
omadhaun 'fool' ligg, gype, etc.
There are, of course, some morphological oppositions as follows:
Mid and S. Ulster Ulster-Scots
cows kye
eyes een
shoes shin
cannot cannae
do not dinnae
have not hinnae
have to hittae
from me fimmae
with it wit
There may finally be oppositions of a syntactical or idiomatic nature:
Mid and S. Ulster Ulster-Scots
at all avaw
He'll not be able to go He'll no can go
a quarter till six a quarter tae six
He went down to Belfast He went up to Belfast
He went up to Dublin He went down to Dublin


* Unpublished paper delivered to Canadian Linguistic Association at the University of Saskatchewan, 1959.

(1) For an interesting conception of this interaction see Henry, P. L., An Anglo-Irish Dialect of Roscommon (Dublin, 1957), Introduction.

(2) Henry, P. L., 'A Linguistic Survey of Ireland', Lochlann: A Review of Celtic Studies 1 (1958).

(3) These two associated features are found also in western Lowland Scottish dialects, and, further afield, are a characteristic of most forms of Norwegian.