IntroductionFootnote 1
Standard English need not be a matter of prescriptivism or any attempt to ‘create’ a particular standard, but, rather, can be a matter of observation of actual linguistic behaviour. For Hudson (2000), standard English is the kind of English which is written in published work, which is spoken in situations where published writing is most influential – especially in university level education and so in post-university professions – and which is spoken ‘natively’ at home by the ‘professional class’, i.e. people who are most influenced by published writing. In the papers in Bex and Watts (eds, 1999), it is recurrently claimed that, when speaking English, what the ‘social group with highest degree of power, wealth or prestige’ or more neutrally ‘educated people’ or ‘socially admired people’ speak is the variety known as ‘standard English’. However, ‘standard English’ may also mean that shared aspect of English which makes global communication possible. This latter perspective allows for two meanings of ‘standard’: it may refer both to an idealised set of shared features, and also to different sets of national features, reflecting different demographic and political histories and language influences. The methodology adopted in the International Corpus of English (henceforth ICE – cf. Greenbaum, Reference Greenbaum1996) enables us to observe and investigate each set of features, showing what everybody shares and also what makes each national variety of English different.
In the case of Ireland, it is mostly educated or professional people, including students and many with power and/or prestige who, we surmise, have been influenced by the norms of written language, whose speech is transcribed in the ICE-Ireland corpus (Kirk et al., Reference Kirk, Kallen, Lowry, Rooney and Mannion2007; Kallen & Kirk, Reference Kirk, Kallen and Filppula2008). The spoken language of this group may be regarded as those speakers' standard English, i.e. their own Irish standard English.
In this paper, I will be particularly concerned with three of those elements: ‘standard’, ‘spoken’, and ‘Irish’; with one exception, this paper will not deal with Irish written standard English.
For some, ‘standard’ English is first and foremost an ideology (Milroy and Milroy, Reference Milroy and Milroy1999) concerning the importance of unflinching conformity to an invariant set of forms or realisations which are codified in grammar books. Anecdotally, I have observed successive cohorts of students from Ireland attesting their subscription to this ideology, without really knowing what it is that they claim to be affirming. The reality is that English has various standards – different sets of forms in different countries or communities – and ICE-Ireland shows that Irish Standard English is one of those multiple standards, increasingly referred to as ‘World Englishes’ (cf. Kortmann and Szmrecsanyi, Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004).
Standard English in Ireland has much in common with the standard varieties in Britain, North America, and elsewhere. Our research shows that both exo- and endo-normative forces combine to create the standard in Ireland too – a distinct blend of international or world English features with many local characteristics unique to Ireland. As a result, speakers of standard English in Ireland reflect variable realisations because not everyone uses consistently or invariably those features which have become fully standardised through selection and codification in other national varieties. Rather, some speakers, as we shall see below, reflect nonstandardised or historical variants, while others reflect vernacular features either transferred from Irish (e.g. the after perfect) or retained from earlier stages of English (e.g. the be perfect) or Scots (e.g. the contracted negative forms –nae and the negative isolate no).
For all its primacy, linguists have only in the last few decades come of age with the description of spoken language, the great diversity within which reflects a great many differences of structure and expression from the written language. The spoken professional registers of educated people are influenced by their written professional register or occupational jargon, including grammar and sentence structure, with little scope for gesturing towards colloquiality or informality except in grammatical items or basic, common-core vocabulary items. ‘Speakers who spend much of their time speaking in formal situations may carry over the linguistic features and constructions typical of formal styles to their conversational speech styles, especially if they are treated with the same respect in their private lives that they command in their professional roles’ (Cheshire, Reference Cheshire1999: 146).
The earliest accounts of spoken syntax, such as those in the Quirk grammars (e.g. Reference Quirk, Greenbaum, Leech and Svartvik1985), tended to be along such formal lines. Only in Biber et al. (Reference Biber, Johansson, Leech, Conrad and Finegan1999) and Carter and McCarthy (Reference Carter and McCarthy2006) has the spoken standard language come to be described in terms of its own functional and pragmatic uses. However, informal styles of the spoken language have their own constructions which are used just as much by educated people as by anyone else because of their communicative usefulness. They, too, are part of the repertoire of spoken standard language – and speakers included in the ICE-Ireland corpus show that their own repertoire of those colloquial features has shared exo-normative as well as unique, local endo-normative features. Indeed, corpus texts in the same corpus category can display a full range of styles. For instance, the category of ‘unscripted speeches’ contains lectures some of which are marked by rhetorical eloquence (e.g. S2A-048), others by directive chattiness (e.g. S2A-047).
The colloquiality of Irish Standard English is striking because of those features transferred from Irish,Footnote 2 which no other national or world variety has (cf. Kirk and Kallen Reference Kirk, Kallen and Tristram2007) – and in addition there is a limited set of dialect features from England and Scotland (Kirk and Kallen Reference Kirk, Kallen and Millar2010). Moreover, fluent or confident speakers of Irish code-switch between English and Irish.Footnote 3 But there is also a cultural context for Irish speakers, reflected in lexical choice, which is considerably different in referencing and in usage between Northern Ireland and the Republic of Ireland (Kirk and Kallen in press). Together, those are the features which accumulatively constitute the language's Irishness, which no other standard variety has, despite the huge amount of other features which are shared universally among national, world varieties.
Our approach
The approach to ICE-Ireland adopted by Jeff Kallen and myself (see references) is to consider Irish Standard English as an empirical standard. We describe what we find in each particular speech situation in ICE-Ireland, paying particular regard to the North-South zone, medium, and register (and within that to issues of formality, style and speaker identity). We demonstrate what speakers of the Irish version of international standard English say and, in so doing, are able to demonstrate that very variability within Standard English which enables flexibility in language use – a flexibility which would be impossible if the standard were an idealised, invariant, static, change-and-variation-inhibiting form. People presume themselves to be speaking standard(ised) language although, as suggested above, they are attempting to conform to idealisations which they might envisage in any underlying belief-system about a fixed, invariant norm.
ICE-Ireland shows abundant evidence for all major shared syntactic structures: NP structure and all types of pre- and post-modification, embedded clauses of all kinds, functions of adverbials, etc. Some items are refunctionalised (such as them as a determiner) or redistributed (present tense –s inflection occurs with third person plural subjects in certain environments – the so-called Northern Subject Rule). Discourse markers used pragmatically would seem on the whole to be no different from their use elsewhere (cf. Kirk and Kallen on just (Reference Kirk, Kallen, Bowen, Mobärg and Ohlander2009) and but (in press) and Kallen (2006) on like). On many core features, the default hypothesis just has to be similarity because there is no reason to expect their structural, semantic or pragmatic functionality to be at all different.
ICE-Ireland also shows the following generally well-known non-standard features of Irish English (cf. Filppula, Reference Filppula1999; Hickey, Reference Hickey2007): (a) past tense done (<ICE-NI TEC S1A-093$B>Footnote 4 <#> It 's a rough job actually <#> I done that before) and seen (<S1A-023$A> <#> I seen him off last night with that woman Coffey); (b) third person singular don't (<ICE-ROI FTF S1B-077$B> <#> Yeah what she done is <unclear> several sylls </unclear> she don't take Rent Allowance at all); (c) punctual bes (<ICE-NI FTF S1A-032$A> <#> He just stands there and <{> <[> bes Frankenstein); (d) habitual DO (<ICE-ROI FTF S1A-087$B> <#> <[> You know </[></{> who he looks like <#> That <.> m </.> that buck that does be on the television on the video); (e) don't + progressive imperatives (<ICE-NI FTF S1A-007$E> <#> Doreen don't be saying that to her); (f) second-person pronouns ye (<ICE-ROI SOC W1B-014$A> … <#> Where did ye eat?) and youse (<ICE-NI FTF S1A-031$B> <#> Were they not walking up the road with youse or behind youse then); (g) personal determiner me (<ICE-ROI BRI S1B-047$B> <#> That's right <#> It was me brother Brendan actually said); (h) ‘hot-news’ after perfect (<ICE-ROI FTF S1A-046$A> <#> Yeah <#> Lads <#> A new fella is after taking over uhm one of the pubs at home <#> And he 's after coming back from England you <{> <[> know </[>); (i) past-tense collocation of temporal adverbs already, ever, since and yet <ICE-NI FTF S1A-042$B> <#> Oh well you already lent them <&> laughter </& >; (<ICE-ROI FTF S1A-090$B> <#> It was the only sewing he ever did in his life was it; <ICE-ROI FTF S1A-062$B> <#> I was only at the cinema once since I came back from America); and <ICE-ROI FTF S1A-070$A> <#> Talking of which my boyfriend was supposed to ring me <,> but he didn't yet); (j) for to infinitives (<ICE-NI BUT S1B-075$B> <#> That 's I suppose passion for the subjects and passion for to change <{1> <[1> things </[1>); (k) ‘pseudo-perfects’ (cf. Kirk & Kallen Reference Kirk, Kallen and Tristram2007) (<ICE-ROI BRD S1B-035$E> <#> And my daughters <,> I 've two daughters married today); (l) epistemic mustn't (<ICE-NI FTF S1A-009$B> <#> Well Esther Dunne mustn't have been there at it then); (m) may used directively (<ICE-ROI SPC S2A-016$A> <#> Now if I can cut away from that theme for the moment Brian <,> you may see from the television pictures that the crew appear to be heading for the plane now) and intentionally (<ICE-NI STE W1A-007$A> <#> Body posture also functions in conversational strategies for the above reasons and for its use in turn taking, e.g. we may sit forward to indicate we want to take the next turn or sit back to let someone else have their turn.); (n) multiple negation (<ICE-NI FTF S1A-023$D> <#> He says sure it 's about fifty foot deep <#> You couldn't do nothing with it); (o) punctual never and punctual whenever, (<ICE-NI FTF S1A-026$A> <#> Like I mean she sang so many songs but her voice never went <#>) (<ICE-NIFTF S1A-005$B> <#> <[> But whenever </[></{> Granda died he changed and he <,> he joined just to sort of respect him or whatever); (p) special collocations of the definite article (the day, the night, the wife, the mother, etc.); (q) empathetic that (<ICE-NI BUT S1B-072$B> <#> Not exercising <,> not do anything with that leg after a while <{ >); (r) non-anaphoric reflexives <ICE-NI UNS S2A-028$A> <#> Again it 's up to yourself which type of pricing policy you use) (cf. Kirk & Kallen Reference Kirk, Kallen and Tristram2007); (r) there's + plural NP (<ICE-ROI FTF S1A-055$A> <#> <[> I 'd </[></{> say that <#> There 's great characters there); (s) omission of subject relative pronouns (<ICE-ROI FTF S1A-063$B> <#> And this time it was the parents got on to the teachers because they wanted a private teacher); (t) indicative apodoses in conditional sentences (<ICE-ROI CLD S1B-019$B> <#> If it was going the other way <,> it would be would be better I suppose); (u) unmarked inverted structures (<ICE-ROI FTF S1A-076$A> Tuesday morning it was); and (v) it-clefts (<ICE-NI BRD S1B-021$D> <#> How how come <,> uh how come that <.> i </.> in in a situation of of jobs and and that it 's the people from that part of the province that have to emigrate), among others. The list of speech features (cf. O'Keeffe and Adolphs, Reference O'Keeffe, Adolphs, Schneider and Barron2008; Murphy, Reference Murphy2010) include (a) left dislocation (<ICE-ROI FTF S1A-082$E> <#> <[> Yeah </[></{> <#> This guy it was his birthday); (b) the topicaliser see (<ICE-ROI BUT S1B-077$A> <#> No it 's not <#> See Wellington Road <,> is all flatland so there 's always loads of people around); (c) declarative tags (<S1A-047$A> <#> Uhm it was in the Guardian so it was); (d) clause-final like (cf. Kallen, 2006) (<ICE-NI FTF S1A-015$A> He got on he got on with the Black and Decker business like); (e) sentence-final but (<ICE-ROI BRD S1B-036$F> <#> But I I mean you 're asking the others what they would do but), and many more.
Discussion
The morpho-syntactic and discourse features listed above occur throughout the ICE-Ireland corpus – those which do not occur in Face-to-face conversations are in situations where a standard and relatively formal variety of the language is to be expected. Overall, however, these features are not frequent. Nevertheless, an empirical observation of the ‘standard’ inevitably reveals features which are not considered part of the codified standard – and our examples reflect features which are local to Ireland through demographic history and language shift as well as features which have ‘universal’ distribution. This last point will be developed further.
Further to Kallen & Kirk (Reference Kallen, Kirk, Kirk and Ó Baoill2001: 65), we can now confirm that, in the absence of a specifically codified Irish standard English on either side of the political border, linguistic features associated with traditional dialects and vernacular Irish English are to be found in standard English in Ireland. However, these features are muted relative to the material found in a dialect or sociolinguistic study: muted both quantitatively in low frequencies of occurrence, and qualitatively in the sense that those features which have crossed into the standard language constitute a strong marker of Irishness in a standard English corpus.
Corpus methodology allows the creation of what may be thought of as an empirical standard. The basis of selection for an ICE-corpus is the situation (such as broadcasting, business, education, parliament, the courtroom, etc.) where it may reasonably be presumed that what is spoken is done so in a standard or standard-like way. In ICE-Ireland, there is to be found a high degree of uniformity with descriptions of the ideal variety and with other national varieties – which may be viewed as linguistic variation, not hegemonic dominance. Varieties of a given language exhibit the same linguistic features, but each national variety has its own peculiarities – in the present case: Irishness, which is exhibited, not by the selection of system sentences to demonstrate the potential for Celtic influence, but in the overtly and covertly structural and also indexical differences which emerge from comparison with similar data in other English-speaking countries. Irish speakers of English use Irish and Irish-derived features as a resource to point to a social, cultural, and historical fact about their Irishness – and that Irishness is inextricably bound up with the Irish language. The Celtic element shows that standardisation has not made Irish Standard English identical to standard Englishes found elsewhere. At the same time, a selection of morpho-syntactic and discourse features, particularly those which have attention in the recent scholarly literature, shows evidence of Scottishness in ICE-Ireland, especially in Northern Ireland. This aspect, too, reflects specific historical and social developments.Footnote 5
Our approach to Irishness is also conditioned by insights from the world-wide survey of variation in English morpho-syntactic features carried out by Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004). This investigation, which compares some 76 features across 46 ‘non-standard varieties of English’ (p. 1144), helps us to see many features of Irish English in the context of world English more generally. For instance, accounts of Irish English regularly draw attention to the use of reflexive pronoun forms as subject or other non-reflexive pronouns, as in <ICE-NI FTF S1A-070$A> <#> And myself and Angela talked for the night. Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004: 1158), however, show that such uses with first person reflexives are found in 30 of the varieties surveyed, including both first-language and second-language Englishes. On the other hand, the after-perfect construction, exemplified by <ICE-ROI FTF S1A-046$A> A new fella is after taking over, is shown to be truly limited in its distribution: it arises in the Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004) survey only in Ireland, Newfoundland (where historical links to Irish English based on settlement are well-documented, e.g. by Clarke, Reference Clarke2010 and Hickey, Reference Hickey2002), and Cameroon (no doubt through immigrant missionaries from Ireland and Scotland). The use of never as a punctual negative particle (or ‘preverbal past tense negator’ in Kortmann and Szmrecsanyi's terms), which is sometimes thought to be particularly associated with Irish English, is actually among the most widespread variable features of English, occurring in 40 of the varieties surveyed (p. 1158). The challenge for the study of dialect syntax identified in Kirk (Reference Kirk, Kirk, Sanderson and Widdowson1985), whereby dialectological study of a single place or area tells us little about where else any given feature might occur (see also Kallen, Reference Kallen and Ó Baoill1985), has at least a partial answer from the Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004) survey: future comparative work of this kind may shed further light on traces of Irishness – or Scottishness – in world English.
Whereas distinctive features of Irish English have been discussed and analysed for some time, what is only now coming to light is the extent to which at least some of those features can be considered universal in the senses explored by, e.g., Chambers (Reference Chambers and Kortmann2004) and Szmrecsanyi and Kortmann (Reference Szmrecsanyi, Kortmann, Filppula, Klemola and Paulasto2009). Because never is the second most widespread feature of the 76 features surveyed by Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004), Kortmann (Reference Kortmann, Kortmann and Upton2008), and Szmrecsanyi and Kortmann (Reference Szmrecsanyi, Kortmann, Filppula, Klemola and Paulasto2009) – occurring in 40 out of 46 varieties of English – Irish ‘ownership’ of the feature is not sustainable. Similarly, whereas it may be empirically demonstrable that progressive aspect occurs relatively more often in Irish English,Footnote 6 there appears to be no functional reason why this should be the case, as all possible uses of the progressive are potentially available to all speakers of English. The inclusion of many features in the literature on Irish English as somehow characteristic of Ireland, North or South, is now re-opened by the very thorough surveys by Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004) and Szmrecsanyi and Kortmann (Reference Szmrecsanyi, Kortmann, Filppula, Klemola and Paulasto2009). Regional features are redesignated vernacular features and appear to occur in very different varieties of English – and no doubt because of their functionality. Because the range of spoken registers in an ICE corpus comprises some 15 different discourse situations, it is striking that the above examples come from across the range of texts in ICE-Ireland and are not simply restricted to informal conversation, as the annotated codings show. The more we try to build a case for Irishness (or indeed Scottishness) on morpho-syntactic grounds, the more we establish that, through observing shared vernacular features, the spoken language of the ICE-Ireland corpus is nothing less than a world English, a variety of world Englishes.
If the standard language is that variety which most strongly suppresses variation, then this paper has shown both how strong that pressure towards standardisation in Ireland is and yet also how resistance to that pressure persists. Superstratal pressure may be due to education, the influence of the standardised written form on individuals represented in those situations under investigation, or the prescriptivist ideology of an invariant standard language. Our present results for ICE-Ireland show that, in all instances, standardisation is almost never quite fully achieved and residual vernacular elements persevere in standard contexts. Although ongoing research shows low occurrences of features or few differences in the frequencies of occurrence of features vis-à-vis standard British English, it is the presence at all of such features – not their absolute or relative quantity – which in the formal contexts under investigation, and in the accumulation of harmonious features at other levels, mark at least part of the speech as Irish. The present investigation has also shown that in some instances, the hypothesis that particular features are uniquely Irish or Scottish is wrong because global perspectives (as shown by Kortmann and Szmrecsanyi (Reference Kortmann, Szmrecsanyi, Kortmann and Schneider2004), Szmrecsanyi and Kortmann (Reference Szmrecsanyi, Kortmann, Filppula, Klemola and Paulasto2009), Chambers (Reference Chambers and Kortmann2004) and others) demonstrate that it cannot be right. Punctual never, epistemic mustn't, and increased use of the progressive, for instance, all appear to be of this type.
At the same time, ICE-Ireland shows that, for some speakers, their choice of forms does indicate an awareness of normative or prescriptive rules which they still operate in their speech (cf. Leech et al., Reference Leech, Hundt, Mair and Smith2009: §11.6.2). ICE-Ireland shows that some speakers still use he/him/his in a gender-neutral way, whereas others use alternative pronouns such as singular they. Examples include both (i) <ICE-NI BRD S1B-025$A > <#> Is that is that the ultimate dilemma then how far the freedom of the individual can be reconciled with the rights of that individual to harm himself or others and (ii) <ICE-ROI SCS S2B-047$A> The patient will survive in spite of such treatment <,> and I believe that the student will also survive to <.> ha </.> <,> to perhaps become as reader in the real sense of the word <,> to perhaps become astonished constantly by what he or she finds between the covers of a novel or a book of poetry). There are also speakers acting to avoid split infinitives (<ICE-NI CLD S1B-009$A> <#> So what I 've done is simply to rank order the scores for the uh proportions for the <.> gi </.> for the boys and for the girls from the lowest to the highest); using wh-forms as relative pronouns (<ICE-NI PAD S1B-052$D> <#> There are some young potato farmers here today whom I have consulted with); and using the subjective forms of pronouns in subject predicate position (it is I) (<ICE-ROI LEH W2A-007$A> <#> Similarly when Buddy Willard visits her at the hospital it is she who digs his car out of the snowdrift.). The conformity to prescriptive norms is also demonstrable on the basis of spelling preferences – in the written component, all of which was published in Ireland except academic texts in the natural sciences and technology. In comparison with ICE-GB, there is, in ICE-Ireland, almost total conformity to British English spelling choices (–ise over –ize, through over thru, etc.), whereas ICE-GB shows a relatively higher tolerance for American preferences.
The final point concerns the methodology of corpus linguistics in the context of ICE-Ireland. In Kirk et al. (Reference Kirk, Kallen, Lowry and Rooney2003), we acknowledge that the end product of a corpus project under-represents the decisions used to make the corpus. Every corpus raises problems and questions — and corpora are not absolutes in any given sense. ICE-Ireland is merely representative of Irish Standard English to the extent that the problem of representativity has been addressed and critically answered before inferences may be drawn. ICE-Ireland acts as mediator between the reality of language usage and the demands of corpus analysis – in the present case, to test for standardisation. The presentation of large amounts of data in a corpus might well create a much more objective picture of language usage than is possible using more traditional methods – yet even a large corpus such as ICE-Ireland can never purport to represent language usage in any ‘objective’ sense. It will always represent the compilers' answer to the kinds of questions such as sampling and representativeness addressed in the compilation (and discussed in Kirk et al. Reference Kirk, Kallen, Lowry and Rooney2003), and the analysis of the corpus. In obliging ourselves to be accountable to the facts of usage, and against the ideology of an invariant standard language, this article has shown that the ICE-Ireland corpus displays inescapable diversity – including the display of Irish and Scottish dialect-indexical features within standardisation. In the particular case, ICE-Ireland offers confirmation of standardisation as a process, an on-going dynamic formation, as distinct from a product, fixed and invariable. In the general case, ICE-corpora have an unrivalled potential to reflect multiple national standards, or multiple world Englishes, as each combines the global with the local and also the ‘glocal’, perhaps documenting some ‘vernacular universal’ (Chambers, Reference Chambers and Kortmann2004) or ‘angloversal’ (Szmrecsanyi & Kortmann, Reference Szmrecsanyi, Kortmann, Filppula, Klemola and Paulasto2009) features.
JOHN M. KIRK is Senior Lecturer in English and Scottish Language at Queen's University Belfast. His main interests are in corpus linguistics and dialectology, especially on Irish English and Scots. With Jeffrey Kallen, he directed the compilation of the Ireland component of the International Corpus of English (ICE-Ireland) and its pragmatically and prosodically-tagged daughter-corpus, SPICE-Ireland. With Dónall Ó Baoill, with whom he arranged ten annual symposia on the language and politics of the Gaeltacht and Scotstacht between 2000–10, he is a general editor of Belfast Studies in Language Culture and Politics. He has edited or co-edited 15 books. Email: j.m.kirk@qub.ac.uk
