The Celtic Englishes

Abstracts of Papers

On 27 December 1997 Ned Maddrell, the last reputed native speaker of Manx Gaelic, died and was buried shortly after in Rushen Parish churchyard. His death marks the end of some 1500 years of Gaelic speech in Man. However, more than 100 years before his death the process of language death had begun its campaign of attrition against the fabric of Manx Gaelic, and particularly wreaked havoc in the phonology (in the form of wild allophonic variation) and in the morphosyntax and syntax, lexicon and phraseology, where the influences of English idiom and expression began to take hold. However, the process was not one-way. English began to establish itself in Man ca. 1800 with the arrival of lower middle class small business operators, mainly (but not exclusively) from Lancashire. In the process of English Manx-Manx English contact the English in Man, particularly (but not exclusively) among those who spoke Manx traditionally, began to take on a Gaelic hue, with idiom and calques as well as a substantial number of lexemes transferred from the Abandoned Language (Manx) into the Target Language (English), so much so that Manx English ceased to be comprehensible to a speaker of Standard English. This paper seeks to look at the contact of both languages on each other and to see whether any pattern is discernible.

In this paper I propose to discuss the way that I set about compiling a dictionary of Hiberno-English for the publisher Gill and Macmillan.
The prospect of a Dictionary of Hiberno-English, whether it be exclusive or inclusive, raises expectations in several different quarters - the writer (e.g., Seamus Heaney), the linguistic scholar (e.g., Manfred Görlach), the literary critic (e.g., Tom Paulin), the general Irish and non-Irish readership, and so forth, each of whom will be looking for different types of lexicographical information, with some degree of unanimity all the same.
The primary concern is with the identification and collection of Hiberno-English words - by direct communication with users of the dialect, by way of newspapers, interviews, radio programmes, etc., and also by analysing entries in comparable published lexicons, word-lists, and glossaries. In this respect, the identification of what constitutes 'Stage-Irish' will aid the decision-making process of what to include.
The lexicographer's task is a series of confrontations with decisions, basically what to include in each entry, having identified what constitutes admissible Hiberno-English material for inclusion (e.g., should 'boycott', 'galore', etc. be excluded on grounds of general absorption into Standard English?). Such decisions should be based on objective criteria.
For each entry, the compiler decides on the inclusion (or not) of alternative spellings, I.P.A. Transcription (which dialect to use), alternative meanings, source (entry in the English Dialect Dictionary; Old English, Middle English, medieval Latin, French, Middle High German, etc.); exemplary quotations submitted from oral and written (i.e. letters) sources; locations of sources; supportive citations from Anglo-Irish Literature; explanatory notes (linguistic (e.g., 'and', 'after', etc.)), cultural (e.g., 'blarney', 'donnybrook', 'Pro-', 'shillelagh', etc.), and historical (e.g., 'Arus', 'Free State', etc.).
These decisions are made with the proposed readership of the dictionary (academic and general, inside and outside Ireland) in mind.

Received scholarly opinion has it that old people speak Hiberno-English more 'genuinely' or 'originally' than younger people (de-creolisation hypothesis). This view can be both read in the handbooks and heard in personal interviews in Ireland. Is this a preconceived idea or is it true? I would like to investigate this problem by means of a case study and present my findings for discussion.
I carried out fieldwork in rural East Galway (Ireland) in the summer of 1995. My eight informants were members of two different families in the age-range between 14 and 101 (at the time). My questionnaire consisted of three parts. The first part elicited answers concerning social background and personal biography. The second part consisted of reading aloud and the third of questions concerning the informants' personal attitudes towards their own language.
I propose to select from the data gathered those which are of relevance to the investigation of variants between speakers of different age-groups in grammar and lexis. My approach will be socio-linguistic and basically in the line of Peter Trudgill's work.

I will argue in this paper that the kind of 'stratum approach' implicit in the notion of the 'Layer Cake Model' does not do justice to the ontology of grammar of contact vernaculars. Based on the untenable supposition that elements drawn from either the substratum or superstratum form neat layers piled on top of each other, this model completely overlooks the intricate patterns of variation which exist in contact vernaculars both at the inter- and intraindividual levels.
I will use evidence from spoken Hiberno-English to illustrate some patterns of variation between syntactic features which reflect the input from Irish, on the one hand, and from standard English, on the other. It will be argued that the choice of variants is determined by a number of factors such as the syntactic environment and lexical or pragmatic considerations, but significantly, there also appears to be a certain amount of random variation.
The overall picture of the grammar emerging from this kind of examination is a far cry from a layer cake of any description: it shows features which are grammaticalised to varying degrees, variants (or 'quasi-variants') in competition with each other, features evidently on the rise or fading away, features belonging to the 'core grammar', or alternatively, to the 'periphery', etc. In short, while the concepts of substratum and superstratum can in my view still serve a useful purpose in the study of 'Celtic Englishes', the variationist approach proposed in this paper is essential to an understanding of the inner dynamics of the grammar of contact vernaculars.

This paper tries to establish theoretical aspects of linguistic accomodation processes through looking at language data culled from 19th century Irish-Australian letters.
First a general theoretical frame of language and language accomodation and change will be presented. Notions of 'Core' and 'Periphery' and their implications for language change will receive prominence in this.
Then a possible mode of change will be discussed where the system of determinancy analysis provides a mathematical basis for the investigation of change.
After that, changes in the expression of perfective and habitual aspects, as evidenced in letters written to and from Australia by Irish-Australians in the 19th century, will be presented, discussed and evaluated.

The Romantic period has often been characterised by the search for primeval origins. Indeed, the late 18th and early 19th century witnessed major strides forward in the field of linguistics, the discovery of the bond between the Indo-European family of languages being perhaps the most far-reaching in its implications. It goes without saying that the contribution made by German philologists was of critical importance. The latter part of the century saw history established as a major academic discipline in European universities. The fascination by scholars in the ethnic and linguistic origins of the European peoples was one of the consequences. In its most extreme form, this fascination eventually led to the development of an ideology based on the concept of racial and linguistic purity.
This paper attempts to show how such notions conditioned the thinking of 19th century scholars and, consequently their views on the ethnogenesis of their own nations. More specifically, we concentrate our attention on the work of the 19th century English historians in an attempt to trace the line of thought that led to the conclusion that the Brittonic-speaking population in what is today England was exterminated or driven out by the conquering Anglo-Saxons from the 5th to the 8th centuries. Although this hypothesis is still widely disseminated, it should be noted that most modern historians no longer place much stock in the idea. The extent to which the British population survived is a matter of current debate but, recent archaeological studies show that their numbers were superior to those of the incoming Anglo-Saxons, even in the most anglicised areas of eastern England. We do not wish to go so far as to say there was harmony between the two populations, however. On the contrary, there can be little doubt that the relations between Britons and Anglo-Saxons were often conflictual and marked by prolonged periods of bloodletting.
Nevertheless, of special importance to scholars of Celtic English is the fact that the linguistic pendant of the argument, namely that there is no Celtic substratum in English, has rarely ever been challenged, the logic being, of course, that if the British population was wiped out how could Brittonic have influenced the English language? Another common argument against a Brittonic substratum is that there is virtually no Celtic vocabulary in English thus providing that the Britons must have been exterminated or driven out. The argument is, of course, a circular one.
The paper concluedes with the proposal of a sociolinguistic model showing how the shift from Brittonic to English may have occurred and suggests the possibility of such a substratum as well as the form it may have taken.

Description of aspectual types in Irish English (northern and southern) have not been wanting with authors like Harris and Kallen devoting much attention to the matter. A major difficulty in the linguistic treatment of aspect is determining how many types exist in Irish English. But this question can only be answered by being explicit about what descriptive model one uses for dealing with the issue.
A number of models can be recognised in the literature, especially for dealing with the two main aspectual types, perfective and habitual. The number of subdivisions arrived at, especially for the perfective, depends on the model used.
1) The grammaticalisation hypothesis maintains that Irish speakers in the historical shift to English grammaticalised all the semantic/pragmatic categories which they could distinguish by using means from Irish and the input varieties of English (Kallen's initial hypothesis in Kallen, 1990).
2) The speech act view is not so much a hypothesis about how aspectual categories arose in Irish English as a model for describing the distinctions which can be recognised today and is that applied by Kallen (1990) in the remainder of his paper.
3) The most flexible approach in the present writer's opinion is that which avails of prototype theory (Hickey 1997) as it allows one to postulate a small number of central aspectual types and various subtypes which are located on the periphery of each major one. The purpose of this paper will be to present a critical discussion of the prototype approach and to illustrate the advantage of analyses within this framework.

The study of Irish English in the crucial 16th and 17th centuries is hampered by the paucity of available data. Bliss's 1979 volume 'Spoken English in Ireland: 1600-1740' has been heavily criticised, both for the choice of texts and for the methodology of his analysis. Pending the publication of other sources, however, this textual material deserves reconsideration.
Of the texts in this collection, that which contains the most substantial body of allegedly Irish English is 'Purgatorium Hibernicum', written c. 1670. It consists of a parody of Book VI of the Aeneid, in an Irish setting, and includes some 1400 lines of direct speech purporting to represent the archaic English dialect of Fingal, in north Co. Dublin. The sample is large enough to permit an examination of its internal consistency and also its similarity to the more reliably attested dialect of Forth and Bargy, with which contemporary observers compare it. An investigation of the morphology and syntax of these 1400 lines shows that this 'Fingallian' dialect can be neither a descendant of medieval Irish English, nor an ancestor of Modern Hiberno-English. On the other hand, few of its features can be attributed to the influence of an Irish substratum.
The paper nevertheless argues for the validity of the language variety reflected in this text, and locates its origin in the short-lived and poorly documented linguistic phenomena resulting from the language contact processes which must have been at work in the colonial and post-colonial Ireland of the Early Modern Period.

The northern subject rules states essentially that plural present tense verbs take the -s ending unless they are adjacent to a personal pronoun subject (as in They peel them); if the subject is a full noun phrase (Birds sings), or a personal pronoun not adjacent to the verb (They peel them and boils them), -s ending is not used.
In my paper I will discuss the history and geographical distribution of the northern subject rule. Special attention will be paid to the typological rarity of this type of constraint on the marking of the verb. It will be shown that the language which shows the closest parallel to this type of constraint is Welsh, where a similar constraint on agreement is well-documented from the earliest times. The theoretical considerations that arise from the discussion of the northern subject marking include a discussion of the time-depth of contact phenomena and a discussion of the role that typological rarity may play in arguing for contact phenomena.

This paper will examine the role of toponymy within Ireland's opposing language shifts, Irish > English (1500 - present) and English > Irish (1893 - present), seeking to identify any patterns which would establish whether place-names and toponymic elements are the harbingers or the pawns of change.
The paper will consider, firstly, how the indigenous system of Gaelic place-names was moulded to the requirements of a foreign, English-medium administration, and secondly, how the independent Irish state has set about the re-establishment of an Irish-language nomenclature. It will be seen that concerns as diverse as the legal (i.e. the identifying and delimiting of land ownership) and the symbolic (i.e. the rhetorical resonance of place-name reference) have combined to imbue Ireland's toponymy with linguistic characteristics which distinguish it form general community language. In fact, the structure and forms of the country's place-names can be shown to reflect tension between the largely intuitive societal dynamics, which predominate in cases of language contact and language shift, and certain intrusively reflective concerns with their connotative import.

Millions of Americans had ancestors who spoke either a Celtic language (Welsh, Irish, Gaelic, or Scottish Gaelic) or a variety of English or Scots influenced by a Celtic language. Settlers from Ireland, for instance, formed the third largest emigrant group after the English and the Germans, and the surnames they contributed to the American population are innumerable. However, historical accounts of American English make virtually no mention of linguistic influence from these emigrant streams. The few items of Welsh, Irish, and Scottish background in the first two volumes of the 'Dictionary of American Regional English' have never had more than marginal currency, differing little from those cited as Celtic contributions to British English, whose small number is supposedly attributable to the cultural dominance of the English. It can be shown, however, that the linguistic impact on American English made by emigrants from the Celtic parts of the British Isles, especially those who came from Ulster in the eighteenth century, were both numerous and profound and that they have been routinely overlooked by researchers because they lie not in the lexicon but in the grammar. These are particularly noteworthy in the speech of the Appalachian mountain region.
This paper has three objectives: 1) to discuss the range of grammatical features in American English that are traceable to varieties of English from Ireland and Scotland; 2) to provide explanations for why the ancestry of these features have not been identified before and why they have endured while lexical items have rapidly eroded; and 3) to explore the implications of this ancestry for understanding present-day regional and syntactic variation in American speech.

I would like to consider the general matter of Welsh English by discussing the curious case of Gower English. Tradition and history tell us that the Gower Peninsula, in South Wales, has had an English-speaking population since the time of Norman rule in Britain. The character and origin of this English have been the subjects of discussion and controversy since at least the end of the 17th century. Two possible sources have received particular attention: the settlement of Flemings in the Peninsula in the Middle Ages (for which there seems to be no substantive evidence), the settlement of immigrants from the south-west of England from the Middle Ages onwards (for which there is some evidence). Since the end of the 17th century scholars have been predicting the imminent demise of Gower English; research for my 1994 book Gowerland and its language (Peter Lang) led me to believe that it has not been a living dialect for most of the 20th century.
My paper will have two aims: 1. to look at the interplay of Welsh and non-Welsh culture and language in the case of the Gower Peninsula, and 2. to look at the construction of a dialect, i.e. 'Gower English' (2. having a direct bearing on 1.). John Algeo once described the term variety, when used in the study of language variation, as a "useful fiction". A modern commentator might be tempted to see Gower English as a kind of fiction or mirage, a construction brought about by folk-history and the accidents of geography, and sustained by folk-history and linguistic and historical scholarship. My paper will consider these factors, but will argue also that, curiously, and despite the fact that its origins, lifespan and even its speakers remain indistinct, Gower English might be the nearest thing to a non-fictional variety of Welsh English.

So-called periphrastic tenses occur in the Celtic languages and in English; and it has remained one of the vexed question with regard to the possible Celticity of English, or otherwise, whether these constructions are the result of a languages-in-contact situation, and how this situation could be envisaged. Any attempt to solve this question, or even to approach a solution, requires the detailed consideration of a number of theoretical linguistic issues which are all connected with the data bases researchers use to explore languages in contact. To look at such problems from the specific perspective of the periphrastic tenses in the medieval Celtic languages is therefore relevant not only with regard to this special problem - however important it may be for the notion of Celtic Englishes - but also with regard to all data-based explorations of languages in contact.
Chronology plays an important - but perhaps not decisive - role here, and the medieval Celtic languages have therefore had a special place in the discussion. An adequate grammatical description of the periphrastic tenses in these alleged Celtic source languages is still lacking, and as a result conflicting views about the occurrence of such constructions in their oldest attested stages have been voiced, or their status within the grammatical system - viz. their level of grammaticalisation - has been contested. Corpus-based and systematic analyses are required here, and some of the theoretical issues of the parameters to be applied will form the focus of my paper. Based on Old Irish and Middle Welsh data I shall discuss the notions of 'corpus' and 'grammaticalisation' and the possible consequences of the interaction between textual genre and register on one side and the frequency and distribution of specific grammatical patterns on the other side.
I would like to present this paper together  with a research student if my projected research program on the periphrastic tenses in the medieval Celtic languages at the Phillips-Universität Marburg progresses well enough.

The paper deals with the concept of diglossia which is central to the research of languages in contact and has been used for the description of those specific contact situations which led to the development of the "Celtic Englishes". The paper will discuss the possibilities of operationalising this concept, taking into consideration the history of the concept, so as to better account for the many different types of situations of linguistic contact and their dynamics. The point at issue is that the discussion of diglossia until the 1980s was characterised by a traditional notion of discrete categorisation - the hallmarks of which are feature orientation and binarism. The proposal of a continuum of descriptive parameters, as advanced in Romance Studies in the beginning of the 1990s (Lüdi 1991), offers a new orientation which, among others, corresponds to changes of paradigm in other areas of linguistics.
It remains to be seen whether this proposal and its corresponding instrumentalisations are sufficient for a more adequate and precise description of the contact situations relating to the "Celtic Englishes".

My paper will provide a short introduction to post-colonial theory, focussing on the aspects relevant for actual language use. The main point of interest will be the notion of hybridity, as introduced by Homi K. Bhaba. Bhaba and his followers interpret hybridity as a positive concept, similar to the use of the term in biology, where it denotes a stronger, better adapted specimen derived from the mixture of two different strains. The post-colonial writer (or intellectual) is seen as a wanderer between two cultures, consciously or unconsciously selecting element from both to transcend both in his work. In linguistics, this approach can be linked to the concept of the nativisation of English (cf. e.g. Braj Kachru, E. K. Brathwaite), the 'Acts of Identity'-model (Robert LePage, Andrée Tabouret-Keller) and the whole complex of language attitudes, of covert and overt prestige (cf. e.g. John Rickford/Elisabeth Traugott). I will explore these possible links and address the question, if the meaningful application of the concept of hybridity is restricted to 'clear-cut' cases (e.g. the Caribbean) or if it can also contribute to the study of the Celtic Englishes.

Finally, I shall try to compare conditions of bilingualsim is 20th century Ireland with earlier phases of this same language contact. This might help to clarify to what extent it is justified to reconstruct, at least partially, the origin of Hiberno-English on the basis of later empirical insight.

In common with the other Celtic languages, Welsh is considered to belong - on the syntactic level - to the category VSO, despite the fact that the various spoken dialects diverge somewhat from this basic ordering. In this paper, I will argue in favour of considering Welsh, on the pragmatic/cognitive level, as basically rheme-theme, indefinite-definite orientated. This point of view runs counter to the Prague School tradition, but is largely in line with work carried out on Amerindian languages since the late 70's by researchers based at - for example - Oregon and Santa Barbara. Hopefully, it will help to shed some light on widely-attested phenomena involving predicate-fronting in Welsh English. I will attempt to examine to what extent the various rheme-theme structures in discourse data are neutral, pragmatically unmarked - as they very often are in Welsh; and what role is played by modality. Finally, I will suggest that - paradoxically - monolingual English speakers tend to display more interference of this kind in their spoken English than do bilinguals.