7.1.1 The Demolinguistics of Contact in Ireland Then and Now

Since the Anglo-Irish Treaty of 1922, the island of Ireland has consisted of two discrete nation states. The Republic of Ireland (RoI) (in black on Figure 7.1) is an independent country extending across twenty-six of the original thirty-two counties of the island. It consists of four provinces – Ulster in the far north, Leinster in the east, Munster in the south, and Connaught in the west. Northern Ireland (NI) occupies the other six counties in the historical province of Ulster but remains part of the United Kingdom. All of these sub-regions boast distinctive varieties of English, Scots and Irish Gaelic. Moreover, they have been and are currently also being influenced by exogenous minority languages to greater or lesser degrees. In addition to generic volumes on IE such as Amador-Moreno (Reference Amador-Moreno2010) or Hickey (Reference Hickey2007a, Reference Hickey2016), there is an extensive body of focused research on specific rural and urban dialects within this geographical region, such as Corrigan (Reference Corrigan, Filppula, Klemola and Paulasto2009), Filppula (Reference Filppula1999), Hickey (Reference Hickey2005), Henry (Reference Henry1957), Maguire (Reference Maguire2020a), McCafferty (Reference McCafferty2001), Peters (Reference Peters2016). Companion volumes on the language history of both nation states and the traditional dialects which typify them are published in Corrigan (Reference Corrigan2010a) and Kallen (Reference Kallen2013), respectively. More recently, a new research orientation has emerged addressing the impact on linguistic diversity of historical and contemporary population movements into and out of the island. These include Amador-Moreno (Reference Amador-Moreno2019), Corrigan (Reference Corrigan2020a, Reference Corrigan, Beaman, Buchstaller, Fox and Walkerb), Corrigan and Diskin (Reference Corrigan and Diskin2020), Diskin et al. (Reference Diskin and Levey2019), Diskin and Regan (Reference Diskin, Regan, Forsberg Lundell and Bartning2015), Diskin and Levey (Reference Diskin and Levey2019) and Nestor (Reference Nestor, Singleton, Regan and Debaene2013). Furthermore, the region is extremely well represented in previous editions of this collection (Bliss Reference Bliss and Trudgill1984; Harris Reference Harris and Trudgill1984; Hickey Reference Hickey and Britain2007b; McCafferty Reference McCafferty and Britain2007). It similarly features prominently in world Englishes compendia such as Ball (Reference Ball2010), Cheshire (Reference Cheshire1991) and Kachru, Kachru and Neilson (Reference Kachru, Kachru and Neilson2009), not to mention the more recent Filppula, Klemola and Sharma (Reference Filppula, Klemola and Sharma2017), Hopkins, Decker and McKenny (Reference Hopkins, Decker and McKenny2017) and Schreier, Hundt and Schneider (Reference Schreier, Hundt and Schneider2020).

Figure 7.1 Location of the island of Ireland in Europe.

Source: https://commons.wikimedia.org/w/index.php?curid=14630791, CC BY-SA 3.0 (last accessed 11 April 2023).

As Figure 7.1 indicates, the entire region is geographically situated on the extreme western fringes of Europe. Unsurprisingly therefore it is frequently described as ‘peripheral’ within a European Union context (Nitzsche Reference Nitzsche2013). However, its north-eastern tip (Torr Head, County Antrim), as this map likewise illustrates, lies just 23 kilometres (14 miles) from the Mull of Kintyre in Scotland across the Northern Channel. Its proximity to the island of Britain has thus led to sustained contact between the two regions since the earliest times.

English, however, did not permeate the island of Ireland’s language ecology (in the sense of Mufwene Reference Mufwene2001:21–4) until the Middle Ages. This new state of affairs arose from a late-twelfth-century military expansion by Anglo-Norman leaders such as John de Courcy and Richard Fitz Godbert de Roche, which subsequently led to the colonisation of lands formerly occupied by indigenous Gaelic chieftains (principally in Leinster and Ulster, as Figure 7.2 demonstrates) (Flanagan Reference Flanagan and Smith1999).

Figure 7.2 Map of land distribution on the island of Ireland c. 1450.

This map is taken from K. P. Corrigan, Linguistic Communities and Migratory Processes: Newcomers Acquiring Sociolinguistic Variation in Northern Ireland. Berlin: De Gruyter, 2020:26, fig.2 and is used with permission.

This turn of events introduced not only English, Welsh and Flemish to the population of the time, which was predominantly Irish-speaking, but also Norman French because Henry II, under whose reign these incursions took place, led the so-called Angevin Empire (Winkler Reference Winkler, Rouse, Echard, Fulton, Rector and Fay2017).

The next phase of medieval incursions was Scots and therefore rather more Celtic than Anglo in character. This is because the Scots Gaelic used by this aspiring group of colonisers in the early fourteenth century is a descendant of the Irish language brought to the western seaboard of Scotland when the sixth-century kingdom of Dál Riada was established. Between 1315 and 1318, Edward Bruce (bilingual in Scots Gaelic and Anglo-Norman) led campaigns to further Scottish political interests in Ireland.

However, Irish monoglottism prevailed on account of the extent to which such erstwhile Hiberno-Norman colonisers had become culturally and linguistically assimilated to Gaelic norms. This is a key reason why materials representing the IE of the early and later Middle Ages are rare, including, for instance, the Kildare Poems found in BM Harley 913 (Hickey Reference Hickey1993, Reference Hickey2007a, b; Kallen Reference Kallen2013; Lucas Reference Lucas1995). There are also some medieval inscriptions and governmental and administrative documents as well as ecclesiastical manuscripts which are not written in either Latin or Norman French but in English (Bliss and Long Reference Bliss, Long and Cosgrove1987). As Corrigan (Reference Corrigan2020a:32–3) notes, these include, for instance, minor records such as that illustrated in Figure 7.3 from the Armagh Diocesan Registry Archive, which includes text in both Latin and English. They relate to the primacy, the province and the archdiocese of Armagh (the ecclesiastical capital of Ireland during the Middle Ages and subsequently).

Figure 7.3 Sample page from a seventeenth-century transcript of the original Armagh Diocesan Registry, 1428–1441 (PRONI: DIO/4/2/4/286/A).

Permission to reproduce PRONI catalogue reference number DIO/4/2/4/286/A has been granted by the Deputy Keeper of the Records, Public Record Office of Northern Ireland and the Registrar of the Diocese of Armagh.

The minoritisation of English customs, language and legal processes across much of the island well before this letter ever saw the light of day resulted in various attempts to de-Gaelicise the Anglo-Norman dominant regions where Anglo power brokers like Swayne and Boteler associated with this document retained a foothold. These would have included parts of Armagh (where Swayne resided) and Carrickfergus and Newry in Ulster as well as The Pale region of eastern Leinster. The shoring up of their dominance included the passing of laws such as the Statutes of Kilkenny in 1366. These sought to suppress Ireland’s majority language and customs amongst the emerging Hiberno-Norman community who were becoming rather more acculturated to Irish life than was considered conducive to maintaining the classic feudal system. The rationale for such statutes stemmed from the Irish language’s role as a marker of Gaelic ethnic and national identity in the same manner that English was coming to be viewed in relation to Norman French following its amelioration after the Hundred Years’ War (Green Reference Green2014:245). The efficacy of minoritising acts of this type in Ireland that were aimed at the Hiberno-Normans during this period has long been called into question. Indeed, as Casey (Reference Casey2012:74) notes, it was in fact the early seventeenth century which truly brought about ‘the passing of the old order’ in Ireland and which sowed the seeds for the wholescale language shift to English typifying the nineteenth century. This is because events such as the Flight of the Earls in 1607, in which Gaelic nobles decamped to Continental Europe never to return, left a political vacuum. Their lands were declared forfeit to the Crown and cleared for the Plantation of Ulster that followed.

As Corrigan (Reference Corrigan2010a:112–21) documents, the success or otherwise of such plantation or colonisation schemes whereby tracts of land were appropriated by various monarchs who settled new English and Scottish migrants on territories formerly occupied by Gaelic chieftains ebbed and flowed over time and geographical space. Thus, the Ulster plantation scheme under James I was considerably more successful than those attempted decades earlier by Henry VIII in Munster. The colonising population movements which ensued exposed Hiberno-Norman and Gaelic populations alike to new varieties of English and Scots depending on where they were situated. For instance, the early-medieval Anglo-Norman incursions into what is now the Republic of Ireland had brought South West and south-west Midlands English to eastern regions within The Pale. By contrast, the later seventeenth-century settlers, who fanned out southwards and westwards beyond this originally Anglo territory, hailed primarily from the North/West Midlands (Hickey Reference Hickey and Britain2007b:137; Kallen Reference Kallen2013:213). These two regions of England have long been recognised as showing dialectologically distinct boundaries since at least the Anglo-Saxon period (Gneuss Reference Gneuss1972). This fact is an important reason why the English now spoken in these sub-regions of the Republic of Ireland maintains a range of linguistic differences between them. This premise is readily testified to by comparing Henry’s (Reference Henry1957) account of North Roscommon English in Connacht with that of Hickey (Reference Hickey2005) for Dublin, situated in Leinster to the far east and originally a heartland of the medieval Anglo-Norman Pale settlements. It is contemporary NI, however, which – despite its relatively limited geographical extent by comparison to the RoI – boasts the widest array of diverse dialect zones. Figure 7.4 is based on data calculating the number of Irish speakers returned in the 1911 census in addition to research on the phonologies of Ulster English and Scots by Adams (Reference Adams1958), Barry (Reference Barry and Barry1981), Braidwood (Reference Braidwood and Adams1964), Gregg (Reference Gregg and Wakelin1972), Henry (Reference Henry1958), Harris (Reference Harris1985) and Robinson (Reference Robinson, Smyth, Montgomery and Robinson2006), inter alia. These works established the dialect boundaries of the historical nine-county province of Ulster for the twentieth century. Corrigan’s later dialectological survey based on new interview and reading task data confirmed that these historical divisions have, in fact, largely persisted into the twenty-first century (see Corrigan Reference Corrigan2010a:29–49 and also Maguire Reference Maguire2020a). There are, in fact, three major dialect zones for the northern counties. Mid-Ulster English (M-UE) (the white areas on this map) is spoken by the vast majority of the population, and this zone is sandwiched between South Ulster English (SUE) and Ulster Scots (US) in the north and east. The former describes the varieties found in the Armagh and Fermanagh border counties and is a type that is shared with that spoken in the northernmost counties of the present RoI (Barry Reference Barry and Barry1981). This dialect zone remains quite distinctive – in broad terms – from either M-UE/US or the ‘Southern Hiberno-English’ type illustrated in Figure 7.4 that dominates elsewhere in the Republic. For instance, the typical northern fronted allophones of /u:/ and /u/ are prevalent in SUE but not in other southern IE dialects. Moreover, speakers in this zone share common features of the latter (such as the substitution of alveolar or dental stops in place of what would instead be realised as dental or labio-velar fricativesFootnote ⁴ in the M-UE and US regions). Ulster Scots is, as Figure 7.4 indicates, not confined to contemporary northern and eastern NI but is also spoken across the border in County Donegal.

Figure 7.4 Map of Ireland indicating Irish-speaking districts in 1911 and the English/Scots isoglosses of historical Ulster.

This map is reproduced from K. P. Corrigan, Linguistic Communities and Migratory Processes: Newcomers Acquiring Sociolinguistic Variation in Northern Ireland. Berlin: De Gruyter, 2020:79, fig.22 and is used with permission.

Interestingly, these present-day dialect divisions align with the patterns of English and Scots plantation settlement across the whole of historical Ulster which began in earnest after the Flight of the Earls, as already noted (see Corrigan Reference Corrigan2010a, Reference Corrigan2020a; Maguire Reference Maguire2020a; Robinson Reference Robinson1994).

Only by documenting the distinctive periods of migratory movements to the island of Ireland and exploring not just the regional origins of its founder populations but their destinations post-arrival, can we arrive at a clear understanding of why this ‘wee’ place boasts the extent of dialectological diversity further articulated in Section 7.2.

7.2.1 Lexical Variation

As Hickey (Reference Hickey and Britain2007b:149) also notes, scholarly interest in the lexicon which characterises English and Scots on the island of Ireland is a long-standing tradition. He cites Vallancey’s 1787–88 collection of word lists pertaining to the relic dialect known as ‘Yola’ originally spoken in the Forth and Bargy baronies of Wexford in the south-east of the contemporary Republic. Dolan (Reference Dolan2012:xxii), in examining the decline of kiver (from Middle English kever ‘to cover’) there, remarks that the increased social and geographical mobility of the twentieth century is ‘tending to diminish or even destroy’ the stock of traditional dialect vocabulary. Documenting IE lexis is thus considered to be an important scholarly enterprise which flourished particularly in the twentieth century when eclectic word lists of the type that Vallancey Reference Vallancey1787–88, Patterson (Reference Patterson1880), Marshall (Reference Marshall1904) and Bigger (Reference Bigger1924) collected were augmented with more systematic word geography surveys. On account of its close connection with Scotland, certain regions of NI and parts of the RoI were surveyed via questionnaires in the 1950s, for instance, as part of the Linguistic Atlas of Scotland (LAS) (Macafee Reference Macafee, Smyth, Montgomery and Robinson2006; Maguire Reference Maguire, Kopaczyk and McColl Millar2020b; Mather and Speitel Reference Mather and Speitel1986). The late twentieth and early twenty-first centuries proved to be an especially prolific period, with the publication of Macafee’s dictionary in 1996 followed by Fenton (Reference Fenton2006), both of which focused on Ulster dialects (especially US). Ulster lexis was likewise represented in Dolan (Reference Dolan2012) and Ó Muirithe (Reference Ó Muirithe2000), though these dictionaries included materials from written sources and amateur correspondents across the entire island.

Not surprisingly, perhaps, given the long gestation period that the English language in Ireland has had, these works demonstrate that this variety is distinctive lexicographically from a world Englishes perspective in the number of lexical items that it retains which are of early English (like kiver in Forth and Bargy) or Norman origin. Thus, scallion is typical across the island to describe a ‘shallot’ or ‘spring onion’ and it ultimately derives from Anglo-French scaloun (Old French escalogne) (Corrigan Reference Corrigan2010a:88; Dolan Reference Dolan2012:ix). The lexeme is obsolete in Standard English and indeed in most contemporary dialects of English globally (though Wright’s English Dialect Dictionary ascribed it to Scotland, Ireland, England and America in the late eighteenth/early twentieth century).Footnote ⁷

Amongst the vocabulary which remains current in IE and also adds to the distinctiveness of this variety more broadly, are items derived from the Irish language.Footnote ⁸ These include examples like barmbrack (a type of fruit-enriched baked dough), which is derived from bairín breac ‘little speckled loaf’. This collocate also incorporates the Irish diminutive suffix {‑ín}, which is readily applied to nouns in southern IE – particularly amongst older speakers – to produce forms such as girl{-een}, man-{een} and so on (Dolan Reference Dolan2012:xx). There are other lexemes, of course, which are now so embedded in English on account of the long relationship between the islands of Ireland and Britain that they no longer seem connected in any way with Ireland or even the Irish language. These include Standard English galore, which actually derives from Gaelic go leor (Dolan Reference Dolan2012:109). Interestingly, the meaning of the latter is ‘enough’, so there has been a semantic extension at work during its evolution in mainstream Englishes into the sense of ‘plentiful’ which it is generally nowadays taken to mean. Other contact-induced lexis of this type in English retains the original meaning and includes items such as bog, which in Irish is an adjective meaning ‘soft’ and is thus often used to refer to ‘wet place where peat or turf is cut’ (Ó Muirithe Reference Ó Muirithe2000:40).

As Hickey (Reference Hickey and Britain2007b:149) points out, echoing Dolan’s (Reference Dolan2012) comments regarding Yola already referred to, the missing link here is establishing the extent to which vocabulary such as scallion remains current. Has it perhaps been replaced amongst younger speakers by Standard English ‘spring onion’ because this is how the vegetable is packaged in urban grocery chains originating in Britain such as Tesco or Marks and Spencer? Much remains to be investigated in this regard with respect to rural and urban dialects alike both north and south of the Irish border. Corrigan’s small-scale lexicographical survey between 2008 and 2009, which considered the retention of relic features and production of new vocabulary, found that children aged 5–11 in Belfast, the capital of NI, had lexical systems in which early English lexemes like uxter ‘armpit’ and dander ‘stroll’ remained very productive indeed – despite their antiquity. It also uncovered a word stock of new items that do not feature in any of the dictionaries mentioned thus far – especially in semantic fields associated with the era of sectarian violence during the later twentieth century known as ‘The Troubles’. These included Coca-Colas, Provos and Stickies, which are all terms representing different factions of the Irish Republican Army that itself is commonly referred to locally as the Rah (see Corrigan Reference Corrigan2010a:84). No doubt other major cultural and social changes such as the Covid-19 pandemic (which started in 2020 and appears to have generated new lexemes and semantic extensions galore in a range of world EnglishesFootnote ⁹) likewise have the capacity to increase the word stock of IE. Coveejit, for instance, was a trending hashtag that appeared on Irish-generated social media (https://twitter.com/hashtag/coveejit?lang=en-gb). It refers to anti-vaccine/anti-mask protestors or more generally to members of the public who otherwise do not comply with official directives issued to reduce the spread of the virus. By 16 March 2020, the Urban Dictionary had already defined Covidiot as ‘someone who ignores the warnings regarding public health or safety’ or ‘a person who hoards goods denying them from their neighbours’ (www.urbandictionary.com/define.php?term=Covidiot). Eejit is, of course, a well-known, traditional IE naive spelling for Standard English idiot that preserves the original vowel of French idiot (Macafee Reference Macafee1996:115). What is needed, of course, is a new, large-scale systematic survey of the whole island to fully document the currency of words amongst diverse social groups, to find other new uses of traditional lexemes and to explore semantic fields that take account of social and demographic changes since the word lists of the twentieth century were collected by scholars like Patterson and Bigger. As such, the new project led by Robert McColl Millar at the University of Aberdeen, which seeks to update the LAS already mentioned, is much anticipated because it will produce a digital lexicography of contemporary Scots that includes the US regions of NI and RoI indicated in Figure 7.4. There is also the recent expansion of the Speak for Yersel project at the University of Glasgow which is worth mentioning in this context. This new survey invites responses from the general public across the entire island of Ireland so as to document current lexis in addition to vocabulary derived from the indigenous language as well as earlier forms of English and Scots.Footnote ¹⁰ Such initiatives are crucial given views expressed in Maguire (Reference Maguire2018:488), for instance, that ‘borrowings from Scots’ at the lexical level are more prevalent in parts of Ulster than those with an undoubtedly Irish contact origin.Footnote ¹¹

7.2.2 Phonological Variation

Barry (Reference Barry and Barry1981) plotted forty-five different phonological forms across the island, drawing on data from the TRSHES. These were considered to have the potential to discriminate between the dialects of the northern (N) and southern (S) regions as well as those preserving archaic pronunciations that have largely become obsolete in other British Isles Englishes (e.g. COLUMN as [lj] (Reference Barry and Barry1981: 72)). Thirteen of these features, summarised in Tables 7.1 and 7.2, revealed ‘a significant N/S distribution pattern’ (Barry Reference Barry and Barry1981:69).Footnote ¹²

As McCafferty (Reference McCafferty and Britain2007:123–5) also notes, Harris (Reference Harris1985) makes a convincing case for further discriminating between the northern and southern dialects by attending to whether varieties within the region conform to the historical phonemic vowel length system of West Germanic or instead use an option whereby vowel length is determined phonetically. The former vowel quantity principle applies throughout much of the English-speaking world whereas the latter is an innovation associated particularly with contemporary dialects of Scots. The phonemic type is more typical of dialects in the south of Ireland whereas the phonetic option is commoner in the northern varieties (to varying degrees depending on the extent of their influence from Scots, as Corrigan Reference Corrigan2010a:17 and Maguire Reference Maguire2020a:400–1 argue).

Another area of phonological difference between varieties across the island of Ireland which, in this case, appears to be used to index social and stylistic traits, is the extent to which utterance-level contrasts are distinguished by intonation. Most varieties of English use pitch to discriminate between statements and questions. Intonational falls accompany the former while question intonation is usually achieved by adding a pitch rise to an utterance. By and large, this state of affairs also obtains in the dialects spoken in the Republic of Ireland. However, there is a significant body of research evidencing the fact that in more northerly dialects the regular neutral intonation observed for statements is actually one in which there is a rising rather than a falling pitch contour (Corrigan Reference Corrigan2010a:48; Kallen and Kirk Reference Kallen and Kirk2012; Lowry Reference Lowry2002; McCafferty Reference McCafferty and Britain2007:127; Rahilly Reference Rahilly and Kallen1997; Sullivan Reference Sullivan, Migge and Ní Chiosáin2012). In an analysis of urban teenagers in Ulster, Lowry (Reference Lowry2002) documents a preference for rising intonation when the level of formality in the conversational exchange is low. In such contexts, she also finds that both genders prefer rising nuclei in statements but observes that young women favour falling nuclei when operating in more formal settings. Acoustic analyses described in Sullivan (Reference Sullivan, Migge and Ní Chiosáin2012), also based on Belfast data, has considerably extended our understanding of the dynamics of this phenomenon. In particular, it has called into question the widely held view (based only on auditory analyses) of how NIE intonational rises are actually shaped. In fact, there is now good evidence to suggest that instead of the contour being ‘rise-plateau-(slump)’ in type (Sullivan Reference Sullivan, Migge and Ní Chiosáin2012:78), the rise does not end in a plateau at all but instead continues until the final syllable. As such, it resembles the contour typically associated with the ‘Uptalk’ or High Rising Terminal option in some English dialects that has also become closely linked to the indexing of social meaning, as argued in Podesva (Reference Podesva2011), inter alia.

Rahilly (Reference Rahilly and Kallen1997) contends that the source of the differences in tune shape between dialects of English north and south of the Irish border is difficult to understand, not least because of the lack of either auditory or acoustic analysis of intonation patterns in different varieties of Celtic (including Ulster Irish and Scots Gaelic, which may be implicated here). However, Sullivan’s (Reference Sullivan, Migge and Ní Chiosáin2012) work has been able to consider new research on the phenomenon in Gaoth Dobhair Irish (County Donegal) by Dalton and Ní Chasaide (Reference Dalton and Ní Chasaide2005). It also identifies statement rises in that dialect that could offer insights of the type that Rahilly is arguing for and may well be worth examining from a linguistic contact perspective, especially since Corrigan (Reference Corrigan2010a:48) identifies other intonational phenomena shared by both the Celtic languages and Englishes influenced by them.

When trying to unravel what Labov (Reference Labov, Meyerhoff and Nagy2008) terms such ‘mysteries of the substrate’, we also, however, ought to heed the warnings regarding L1 transfer arguments more broadly in Cornips and Corrigan (Reference Cornips, Corrigan, Auer, Hinskens and Kerswill2005) as well as Corrigan (Reference Corrigan1997, Reference Corrigan and Tristram2000a, Reference Corrigan and Hickey2010b) and Maguire (Reference Maguire2018, Reference Maguire2020a). Poplack and Levey (Reference Poplack, Levey, Auer and Schmidt2010:409) put it very succinctly in their suggestion that ‘differences between cohorts or overall rates may be masking other effects’. An excellent case in point with respect to the evidence for contact effects in the phonology of IE is recent research on epenthesis phenomena by Maguire (Reference Maguire2018). Epenthesis in liquid+sonorant clusters (e.g. film [fɪləm] or farm [faɹəm]) is considered to be widely typical of dialects on the island of Ireland whether these are in the north or the south (de Rijke Reference Rijke2016; Hickey Reference Hickey and Britain2007b:145). Its prevalence is generally thought to have arisen as a cross-linguistic transfer phenomenon on account of the dominance of schwa insertion in dialects of Irish (Ó Siadhail Reference Ó Siadhail1989:20–2). In a careful examination of the phonological structure, constraints and historical development of epenthesis cross-linguistically, Maguire concludes, however, that the role of the latter in Irish English ‘has been overstated’ (Reference Maguire2018:505). He marshals evidence from the phonologies of early and dialectal English as well as Scots and of course Irish which demonstrate that the latter may have simply played a reinforcing role in retaining an epenthetic vowel in such clusters. An attenuated version of the Irish epenthetic system was, in fact, already prevalent in the target English and Scots dialects that the first speakers of IE were attempting to acquire, and it is this reduced system – already on the wane in other British varieties – that best reflects the dynamics of how it operates today island-wide. That is not to say, of course, that there are not social or geographical differences observable. Some middle-class speakers, for instance, never epenthesise in these contexts. Moreover, while film [fɪləm] sequences are very typical north and south, segments of the farm [faɹəm] type are more geographically restricted (being especially prominent in the west of Ireland (Peters Reference Peters2016) and rarer in NI). Indeed, Sell (Reference Sell, Migge and Ní Chiosáin2012) found that even in Galway city, which is situated in the heart of Connacht, there is evidence of change in progress. Age and formality have an impact on the frequency of schwa insertion. There is even the possibility that the predominance of the phenomenon in film [fɪləm], as opposed to other liquid+sonorant clusters, may have come to serve as a ‘local (or maybe even national) identity marker’ (Sell Reference Sell, Migge and Ní Chiosáin2012:63).

7.2.3 Grammatical Variation

As noted previously, IE shares a whole raft of linguistic features with other global Englishes and this is especially true of its grammar. All its traditional dialect speakers, for instance, boast ‘vernacular universals’ of the type defined by Chambers (Reference Chambers and Kortmann2004), which include double negation (1) and vernacular verb forms (2).

1. (1)

   They can’t vote for them when they’re not winning nothing (Empire Corpus/DMcC/Y/M/M-UE/Armagh, NI, 2008)

1. (2)

   The ironic thing was they done Shakespearean plays (Empire Corpus/SD/O/M/US/Antrim, NI, 2008)Footnote 13

Within sub-regions of the island, there are very few uniquely IE morphosyntactic characteristics (likely deriving from contact with Irish), such as the unbound reflexives in (3) and the ‘hot-news’ be+after+‑ing perfect in (4), which are not replicated across a majority of speech communities in NI and the RoI (see Bliss Reference Bliss and Trudgill1984; Corrigan Reference Corrigan2010a; Filppula Reference Filppula1999; Harris Reference Harris and Trudgill1984; Hickey Reference Hickey2007a, b; Kallen Reference Kallen2013; McCafferty Reference McCafferty and Britain2007).Footnote ¹⁴ As such, N/S isoglosses of the type outlined in the previous section are harder to motivate, though there are some exceptions. One concerns the use of double modals (5) (Corrigan Reference Corrigan2000b; Montgomery and Nagle Reference Montgomery and Nagle1993; van Hattum Reference van Hattum and Collins2015) and the second relates to the expression of generic-habitual aspect in dialects north and south of the Irish border (6a), (6b).Footnote ¹⁵

1. (3)

   himself would have reminded me of my dad (Corrigan Reference Corrigan2010a:55)

1. (4)

   They were after leaving a christening in Ennis (O’Keeffe and Amador-Moreno Reference O’Keeffe and Amador-Moreno2009:525).

1. (5)

   I’ll can do that the morra (Corrigan Reference Corrigan2010a:59)

1. (6a)

   He bees mad for the bath (Corrigan Reference Corrigan2010a:64)

1. (6b)

   When I do be listenin’ to the Irish here, I do be sorry… (Filppula Reference Filppula1999:59)

The area of modality in IE can indeed be used as a tool for discriminating northern and southern varieties. It is only in the US dialect, for instance, that double modals are attested. Even in these dialects, they are confined to the oldest and most isolated rural speakers – Corrigan’s Empire Corpus survey of the region mentioned previously did not find any US participants for whom the feature remained productive (even in those who were male and aged 40+). Modals of necessity in the dialects of Ulster provide further evidence of differentiation not just between the dialects of NI and the RoI but also between varieties within Ulster. Thus Corrigan (Reference Corrigan2000b) documents the use of a be+to syntagm (7) that marks epistemic modality in an early sub-dialect of SUE. So too does the form maun in the same example. It dominates in the US region but, as (7) shows, the feature can also be found further afield within NI in this attestation from the twentieth-century corpus of South Armagh English analysed in Corrigan (Reference Corrigan1997).

1. (7)

   What bees to be maun be. ‘What must be must be.’ (Corrigan Reference Corrigan2000b:32).

As Filppula (Reference Filppula1999) contends, there is a strong argument for proposing that habitual forms, as in (6a) and (6b), are particularly typical of regional dialects on the island of Ireland on account of their likely origins in the contact situation described in Section 7.1. This is because punctual and habitual are important contrastive categories of the verb in Irish (Corrigan Reference Corrigan2010a:64). What is more, Filppula (Reference Filppula1999) offers strong evidence for the view that the (6b) variant predominates in varieties found in the RoI, such as the sub-regions of Counties Clare, Dublin, Wicklow and Kerry which he examined. The bees variant, by contrast, is more commonly attested in Ulster dialects (Corrigan Reference Corrigan2010a). Moreover, Hickey’s SIEU published in (Reference Hickey2007a:236–7) shows lower rates of do+be acceptance for counties in the north (between 5% and 25%), whereas speakers interviewed in the Republic of Ireland were more readily accepting of this feature (between 36% and 53%).Footnote ¹⁶ While it is indeed possible to see parallels between this construction in IE and the prominence of a habitual-punctual distinction in the substrate, it is important to also note that the building blocks for the transfer are based entirely on early English and Scots verbal forms.

7.2.4 Discourse-Pragmatic Variation

Since the publication of Hickey (Reference Hickey and Britain2007b) and McCafferty (Reference McCafferty and Britain2007) there have been two areas of research on IE which have seen unprecedented growth and will be considered separately under ‘Corpus Pragmatics’ (Section 7.2.4.1) and ‘Newcomers Acquiring Variation’ (Section 7.3).Footnote ¹⁷ These connect partly to the development of the major historical and contemporary corpora outlined in Section 7.2 but they also arise from the radical demographic changes on the island which began in the late twentieth century.