Contact-induced grammar formation: A model from a study on Hiberno-English

Introduction

In Ireland, like many other places, various forms of English have penetrated the everyday lives of the population via the media. Within this state of affairs, Hiberno-English (HE), the English spoken by the Irish, is on a path toward convergence with other varieties, but it still preserves some of its Irish heritage. This article addresses this heritage in its expressions for time and its manner of expressing information structure, two elements which form part of the core of the grammar. It discusses how HE has been formed and developed up until today by an examination of three salient grammatical features. The article focuses on varieties of Southwest Hiberno-English (SwHE), spoken in the counties of Cork and Kerry where a language shift from Irish to English has not progressed to the same extent as in other regions of Ireland. In the investigation of contact-induced properties of grammar, SwHE is a prime subject due to its robust connection with the Irish language.

The article investigates the process behind the establishment of constructions in present-day HE. One of our concerns is the process with which the be after perfect, as exemplified in (1), has gained its present form and usage.

(1) We are after missing the bus.

  “We have just missed the bus”. [Cork City]

(1) is an example of be after V-ing from contemporary varieties of SwHE. The speaker said this when “we” were running to catch a bus leaving from the bus stop in front of us. The basic property of be afterV-ing is to denote that, as of the speech time, a certain activity or event has been completed. The be after perfect is often labeled as the “hot-news perfect”. Using be after, the speaker presents the event as “hot-news” and signals their emotional attachment to it.

Besides the be after perfect, this article highlights key distinctive constructions (vis-à-vis Standard forms of English) including the do be habitual and the 'tis….. pattern. (2) is an example of the do be form in SwHE. The do be form, generally speaking, seems to be falling out of use if we look at the contemporary situation and speakers' attitudes toward the form. However, the majority of HE speakers, both urban and rural, even including the younger-middle generations, have certain knowledge of this construction (Shimada, 2016a).

(2) He does be eating in John B's.

“He usually eats in John B's (pub)”. [Co. Kerry, elicitation]

In SwHE, do be (V-ing, AdjP) is used to describe an inherent property or a habitual behavior of the agent, which is expressed in the -ing form after do be. Thus, it is called habitual do be, with the do being unemphatic. Formal realization of the habitual categories varies in dialects of HE; northern varieties tend to have be(z)~do(se) + V, while speakers of southern varieties use do be (Henry, 1957; Harris, 1986).

The 'tis….. construction is the third salient characteristic of HE that we highlight in our investigation of contact-induced grammar formation. Although this construction displays similarities in appearance to the it-cleft in standard varieties of English (StE), clearly, important differences have been pointed out (Filppula, 1999; Shimada, 2018).

The 'tis…. pattern in SwHE is characterized by its high frequency of use, the wider variety of phrasal categories allowed in the supposed “focus” position, and the primacy of that-lessness (see The 'tis… construction). (3) is an example of the 'tis….. construction.

(3) I suppose 'tis boozing on brandy you are with McFillen.

“I suppose you are boozing on brandy with McFillen”. (prosodic prominence on boozing) [John B. Keane STD56, = (29)]

(3) admits a different context of the use of 'tis…in HE from that of the it-cleft in StE. In (3), 'tis presents a proposition that is characterized as a supposition by suppose. In the clause of 'tis, a marked constituent order results; the fronted constituent boozing on brandy is salient in the clause you are boozing on brandy with McFillen. The sentence expresses that this particular component of the state of affairs is highlighted in the speaker's mind. The 'tis… pattern in HE, despite an apparent resemblance to the it-cleft in general English, exhibits a different information structure decoded in the sentence.

(1)-(3) are present-day forms in SwHE. The be after perfect and the do be habitual are expressions of time, while the 'tis….. construction is concerned with information structure. Using these salient characteristics as the basis of discussion¹, we will investigate the central part of contact-induced grammatical formation.

Studies on features of HE that are distinctive from StE including the seminal studies of Bliss (1979), Harris (1986, 1993), Filppula (1999), Siemund (2004), Hickey (2007), and Kallen (1997, 2013) and (Kallen and Kirk, 2007) have made valuable contributions to the description of HE and the superstratal-vs.-substratal debate. Recently, development of corpora such as ICE-Ireland (Kallen and Kirk, 2008) and CORIECOR (McCafferty and Amador-Moreno, 2012), which contains letters of immigrants, improve our understanding of HE, both its current and earlier forms. Today, research on Irish English particularly flourishes in areas of discourse markers, as represented by the volumes edited by Barron and Schneider (2005), Amador-Moreno et al. (2015), and in sociolinguistic interests (Hickey, 2016; Hickey and Amador-Moreno, 2020). Another important research stand has been on the connection of the grammar of Irish English with other Celtic Englishes, which has been elaborated by the authors of The Celtic Englishes, edited by Tristram (1997). Also important to note is the pursuit for how Celtic languages have influenced English (Filppula et al., 2008).

Many past studies concerning the grammar of Irish English have taken a form-to-form comparative approach, focusing on the way in which forms in HE are deviant from StE and which forms in Irish or earlier English they most closely correspond to. This study, taking a different angle, examines the systematicity of the connection between HE and Irish. It seeks to find general principles from which features of HE result naturally. Taking the three distinctive features above as windows into the underlying mechanisms of the formation of HE, this study examines how their current forms and meanings are the result of a dynamic process of language formation guided by language contact. It will survey earlier forms in the rise of HE (section Earlier examples and the form-function establishment) and describe some details of contemporary HE (section Irishness realized in English). This article will finally introduce an integrated perspective across the characteristics to discuss how elements from Irish have been inherited and realized in English, along with how these realizations have been updated until today².

Data and methodology

For the investigation of present-day SwHE, this study employs a fieldwork-based qualitative methodology, which includes elicitation techniques, participant observation, and interviews, both linguistic and sociolinguistic. The majority of the author's data come from regular stays in Cork since 2002 and Listowel since 2003. The author has also visited Dublin and an Irish-speaking district called Dingle Peninsula. The examples of forms used in present-day SwHE include both the ones the author encountered in natural speech and email texts and the ones that were elicited during linguistic and sociolinguistic interviews. Furthermore, as part of the analysis of the process of the formation of HE, the author draws on examples and studies concerning earlier varieties in the literature review.

The elicited data used in this article come from five consultants living in Cork City and County Kerry whose birthdates range from the 1920s to the 1970s. The linguistic interviews were semi-structured with questions inspired by hypotheses the author drew based on a corpus of John B Keane's playscripts and letter series written mainly in the 1960s and 1970s (refer also to note 8 for the “Keane corpus”). The author had compiled this corpus for previous research and had selected sentences in his studies, which included target features including the three addressed in this article: do be, be after, and the 'tis… construction. Analyzing this corpus with guidance from the pioneering descriptions of HE made by Filppula (1999), among others, led to construction of hypotheses concerning morphosyntactic environments and contexts in which the features appear. The HE variety reflected there is most likely a “stative HE” used during the 1960s and 1970s, which had not yet been exposed to the dominant varieties of English made commonly available in Ireland by TV and the Internet (refer to Two different contacts in the formation of present-day HE and Figure 1 for the transition of HE in timeline). This study initially started with the author's observations of contemporary HE in use in terms of Filppula's (1999) “distinctive features”, examining what led to the robustness of some features and the decline of others.

FIGURE 1

Figure 1. Timeline of the sociohistorical background of HE.

Formation of southwest Hiberno-English in a sociohistorical timeline

A language is naturally affected by the situation of its community. Thus, HE has changed along with the changing times of its community. In the following, after first covering the flow of history with a focus on language, we will consider the formation of HE and its changes over time.

Formation and the development of SwHE

In the southwest of Ireland, including the counties Cork and Kerry, statistics of Irish and bilingual speakers in the 18th and 19th centuries show that Irish had not succumbed to English before the Famine in 1840s, but that it was not inherited by the generation of speakers who were born post-Famine. According to FitzGerald (1984), the 1881 census indicates that 41% of the population aged 70–79 in 1881 had been Irish-speaking in youth, and 77% in Munster, the southern province (ibid: 125). From his examination, it is concluded that Irish remained vigorously spoken in South Cork and North Kerry in 1841 and was more prominent in southeast and southwest Cork and Kerry from Tralee southwards (ibid: 128). Nevertheless, a large proportion of the Irish-speaking population may have already been bilingual; data illustrating that “Irish and English” speakers were outnumbered four times by “Irish-only” speakers in 1851 corroborate this (ibid: 140).

After the Great Famine, there was little room for doubt that the language shift progressed rapidly. In southwest Ireland, the shift started at a comparatively late date, understood to be around the mid to late 19th century. However, the earlier English-lexifier varieties emerged in most cases as the result of natural or untutored L2 acquisition. At the beginning of the 19th century, the only schooling available for the population was obtained from “hedge” schools (Edwards, 1981, p. 241), where local masters were paid a small sum by parents, and Irish was likely to be used as the language of instruction (Ó Cuív, 1986, p. 380, 381). High illiteracy figures show the scarcity of schooling in the mid to late 19th century. The illiteracy figure for the whole country was 53% in 1841 and 47% in 1851, but it decreased to 18% in 1891. For the counties Kerry and Cork, the focus of the author's investigation, it was 70 and 66%, respectively, in 1841 (Edwards, 1981, p. 242, 243). Odlin (1997, p. 12), using demographic data to assess the effect of seasonal migration, states that “most workers from Ulster went to Scotland, whereas the southwestern counties of Kerry, Cork, and Limerick sent few migrants at all to Britain; most workers from these areas sought work in more prosperous Irish counties such as Wexford”. Also, it is assumed that before the 19th century, the acquisition of English by many Irish speakers in southwest counties did not result directly from schooling.

Supposing that the formation of HE accelerated in some south-western communities in the latter half of the 19th century, we can estimate that sometime between 1930 and 1950 a largely stable system for HE had taken root in southwest Ireland. However, that stable system, amid the changes in environment brought by TV broadcasting, widespread education, and entry into the EU (European Union), along with emigration and migration, came into contact with other major forms of English, and it underwent standardization and leveling. Following this, the social situation brought by the rapid economic growth of the latter half of 1990s, called the Celtic Tiger, brought even further changes to the English spoken in Ireland. As many varieties of English, including StE, flowed into Ireland, the once stable system began to falter. HE is now experiencing what we might call a “second contact”, this time with other forms of English.

Two different contacts in the formation of present-day HE

The linguistic history leading up to the formation of HE and subsequent changes to it can be roughly summarized as shown in Figure 1. Figure 1 is the author's summary of the historical background of the development of HE.

On the timeline of the development or transition of HE, two contacts in different times seem to be recognized as prominent milestones. The first contact is that of Irish and English. The second is that of HE and major varieties of English or what can be described as mainstream English. Thus, in a simple way, three stages can be distinguished: (I) English enters the Irish-speaking community, (II) HE forms and gains relative stability, and (III) HE undergoes contact and convergence with mainstream English. The dynamics of contemporary HE can be regarded as the outcome of two different occurrences of contacts, one completed and the other still in progress.

During the period in which English, in some form, came into common usage in Ireland, stage (I) in Figure 1, Irish monoglots became bilingual mostly through contact and interaction with other Irish people who had already acquired English. This English is most likely the prototype of HE. According to Bliss (1972, p. 63), in the mid-17th century, English was “acquired, gradually and with difficulty, by speakers of Irish; and in the process of their acquisition of it they modified it, both in pronunciation and in syntax, toward conformity with their own linguistic habits”. There were, of course, different processes working at the same time in different regions and spurred by different contact situations, all of which fostered language shift and the formation of HE. If we see this phenomenon from a micro-perspective, that is, focusing on a speech community in one particular area and not on the whole country of Ireland, HE has been shaped in a somewhat similar way to other contact varieties which came into being as a result of the two different languages. Although the spread of the English language in the whole of Ireland has taken a long time, language shift at the level of each community is, in general, accomplished in three generations over approximately 100 years.

Earlier examples and the form-function establishment

This section addresses examples of the three constructions found in the texts written in the 16th-19th centuries to see initial forms of HE. In terms of the timeline given in Figure 1, stages (I) and (II) are discussed. The focus is on the do be and be after V-ing forms, concerning the expression of tense/aspect, and the 'tis… construction, concerning the expression of information structure. For the do be and be after constructions, there is already a wealth of research, so the main points will be summarized as part of the literature review. For the 'tis… construction, there seems to be a room for further research. Examples of the 'tis… construction, mainly from Bliss (1979), will be analyzed to examine how the Irish language was mixed into English speech/writing in the early stage of HE.

Be after perfect

Examples of be after V-ing emerged from an earlier stage of the development in HE. The be after form is traced back to the 1680s, as reported by Filppula (1999, p. 103) and Kallen (1994, p. 173). The following is an example of be after V-ing cited from Bliss (1979, p. 133).

(4) Deare Catolicks, you shee here de cause dat is after bringing you to dis plaace: 'tis come bourying you are de crop, de cadaver, of a verie good woman, God knows!… [Report of a Sermon (1698)].

A common view in the literature on HE is that the be after construction is calqued on the Irish tréis (~tar éis) “after” construction. This view is based on the obvious parallelism between Irish and HE, shown in (5) below, and the fact that no recoded form of English offers an alternative model (Greene, 1979, p. 126). The be after NP form is given in (5a) and be after V-ing in (5b).

(5) a. Tom is after his supper. [HE]

Tá Tomás tréis a shuipéar. [Irish]

be.pres. Tomás after his supper

“Tom is in the state where he has had supper”.

b. I am after taking three plates from the cupboard. [HE]

Tá mé tréis trí phláta a thógáint ón gcófra. [Irish]

be.pres. I after three plates PRT take._VN from-the cupboard.

“I've just taken three plates from the cupboard”.

Having seen the parallelism between the two languages, one may presuppose that the Irish tar éis pattern furnishes the model for the be after construction in HE; however it is noted that the existence of such a parallelism does not promise identical semantic and pragmatic properties between the parallel forms in the two languages. Also, the establishment of the be after construction in HE will be re-examined in Where today's form came from.

Besides the view that studies on HE have offered, Heine and Kuteva (2005, p. 93–94) present the be after perfect as an example of “replica grammaticalization”, by which they mean the case where the grammaticalization process itself is transferred from the “model” to the “replica” language. According to them, the conceptual source of the spatial/temporal schema [X is after Y] is common in HE, the replica language, and Irish, the model languages; this after schema is cross-linguistically rare. This may support the idea that HE speakers borrowed this construction from Irish. Heine and Kuteva suggest that it is likely that speakers of HE have adopted the Irish grammaticalization process to HE, perhaps by means of the same cognitive path that links the notion of completion with the spatial/temporal prepositional schema of after. In the view of the author, however, a more significant factor in the formation of the be after perfect is the Irish bí construction for expressing aspect. This is discussed in Where today's form came from.

Do be habitual

There are several examples of periphrastic do found in earlier texts, i.e., in John Michelburne's Ireland Preserved, which dates back to 1705 (Filppula, 1999, p. 138). Bliss (1979) notes the “consuetudinal” usage of do and do be in the text, which includes (6).

(6) Why Neighour, you do be mauke de Rauvish upon de young Womans, and when…. (Bliss, 1979, p. 147)

In a later stage, in the 1860s, emigrants' letters show that the do be form occurs in HE at the time (Filppula, 1999). The following examples were written by an Irish mother and daughter, Nancy and Bridget Oldham, from Rossmore, Co. Cork (TCD MS 10435: Oldham Papers, Department of Manuscripts Trinity College Dublin).

(7) i. I do be disputing with my mother… [1857/TCD 10435-15]

ii. I do be sick every year at this time but I was not prepared anytime until now. [1863/TCD 10435-21].

In (6), the verb mauke (“make”) follows do be, while in the later stage V-ing/AdjP follows do be as in (7). McCafferty (2017) reports the following example from an emigrant letter in the 1860s.

(8) Indeed I do be thinking of ye when ye least suspect it (Schrier, 1997, p. 36; B. Colgan, 13.06.1862).

The do be construction in (7) and (8) is still seen in present-day SwHE.

The 'tis… construction

The word 'tis and clauses headed by it can be observed in Early Modern English, i.e., in the studies of William Shakespeare. The Oxford English Dictionary (OED) cites an example of 'tis from John David (1569–1618), a renowned English poet, in which 'tis heads a negated clause.

(9) 'Tis not that she would renovate her affection with this Prince. [1656 J. Davies tr. M. de Scudéry Clelia II. ii. 51]

The 'tis in (9) is a type of 'tis taking a that-clause, as in “it is not the case that…”, while the HE construction is taken as a translation of the Irish copula Is (the “i” in Irish “is” is capitalized to distinguish it from is in English). (10) is the entry of the OED for 'tis.

(10) ME tys, ME– tis, 15 t'is, 15–'tis (now poetic, archaic, regional, and colloquial), 16 t'is, 16 ti's, 16 tish (apparently only in representations of Irish English), 16 ty's, 19–'tes. [OED 3]

Variants of 'tis appeared in English in the 15th and 16th centuries. Tish is especially noted as Irish in the OED. It is also noted that 'tis is seen in the literature of Anglo-Irish and Irish writers including Laurence Sterne and James Joyce. Besides that, it is also common in Irish writing, including famous examples such as Frank McCourt “'Tis”, published in 1999 by Flamingo in New York City, although 'tis itself is not an Irish invention. (11) is an example of the 'tis… construction from an earlier stage of development in HE.

(11) Deare Catolicks, you shee here de cause dat is after bringing you to dis place: 'tis come bourying you are de corp, de cadaver, of a verie good woman, God knows!, fwom cruel deat hate devoure. (Bliss, 1979, p. 133, John Duntion, Report of a Sermon 1698).

Importantly, Bliss (1979, p. 296) notes: “Here, 'tis corresponds to the Irish copula Is, the element come burying is emphasized, and the rest of the clause is recast as a relative clause, although the relative pronoun is omitted, as it usually is at the present day”. In Bliss's collection of texts from the 1600s to 1740s, we can find examples of the variants of 'tis, including 'tish and tish, aligning with the Irish sentence-initial copula Is.

The Irish copula Is appears during the initial rise of HE in parallel contexts and settings to those where 'tis and the variants appear. This shows us the mixture of Irish and English in the un-institutional language contact. In the 'tis sentence in (11), the verb in infinitive form comes after 'tis. (12) parallels this with Is followed by an infinitive verb.

(12) Commanded bee superior powers,

Is make me h[a]unt dese donny bowers;

And fate!, and be!, I never thought

(Bliss, 1979, p. 117; Purgatorium Hibernicum, 1670–75)

The Irish Is, usually located in the initial position, functions the same as 'tis. (11) is an example where a verb comes just after Is, while in (13) Is is followed by the subject-NP and an adjective.

(13) …; is none of you strong enough, or stout enough, to overcome him, or wise enough or cunning enough, to sheet him—no, no, 'tis a shad ting, not won! Well, den, fwat's to be done?

(Bliss, 1979, p. 134–5; John Duntion, Report of a Sermon, 1698).

(13) uses is same as the Irish copula form, which would likely be replaced by 'tis in present-day HE. It is important to note that Irish would not use the Is copula. (14) is another pattern in which a subject noun and a finite verb follow Is.

(14) I know de[e] vell enough, and bee!

Is de old hawke have de old eye

(Bliss, 1979, p. 119; Purgatorium Hibernicum, 1670–75)

According to Bliss (1979: 297), Is in earlier HE is used to mean “I am”, as in Is thanke my mester (“I thank my master”) and as in (14), and ishto means “it is”, as in What ish my Nation? in William Shakespeare's Henry V in 1599 and 1623. [Bliss (1979), p. 35] notes the reference to the Earl of Essex's campaign in Ireland in 1599 and deduces the year of the text]. Bliss also mentions omission of the personal pronoun in those sentences.

In addition, Is takes the auxiliary may in the clause.

(15) Is may as velkisse my breesh

“You may as well…”

(Bliss, 1979, p. 119, 298; Purgatorium Hibernicum, 1670–75)

In (16), there is an example of the tensed copula, namely, vas, the past form meaning “was”.

(16) Vas carry it on his Shoul-deer

“He carried me…”

(Bliss, 1979, p. 129, 298; The Irish Hudibras, 1689).

This use of vas is inherited from the usage of Irish ba, the past form of the copula Is. The examples (12)-(16) suggest that the Is pattern modeled after Irish is often used to construct sentences using English words. Over time, Is was replaced by the English equivalent/correspondent 'tis or 'ish with the information structure to be expressed held in the construction. In addition, this pattern came to be used to express information saliency, as the Irish Is construction does. This will be discussed in more detail in The 'tis… construction and Expressions of information structure.

Functional properties of the forms in contemporary HE

In this section, our discussion turns to HE after it reached a certain stage of maturity. We thus highlight facets of grammar concerning the three features in focus in stages II and III in Figure 1. The following addresses the three illustrations: (i) in present-day HE, the have and be after perfects exhibit their separable functional distributions (Be after perfect vis-à-vis have perfect in HE today), (ii) the do be form has two discrete usages, “habitual do be ₁” and “inherent property do be ₂” in the text published by a playwright of SwHE in 1950s–1970s (Two meanings of do be in a stabilized SwHE), and (iii) the 'tis… construction inherits the Irish strategies of expressing informational saliency (The 'tis… construction). The examples in this section are from linguistic interviews, from natural speech or from the John B. Keane corpus. Examples from natural speech and from John B. Keane are followed by the source in brackets. Those that do not mention the source are from linguistic interviews with speakers.

Be after perfect vis-à-vis have perfect in HE today

The be after perfect is a well-known feature of HE; however, the have perfect also exists in HE. The segregation in usage has resulted in a contemporary phase in which two perfects coexist. Note that the have perfect in HE behaves differently from StE and exhibits a much lower frequency than StE, which Kirk (2017) reveals in a comparison between London-Lund Corpus and International Corpus of English (ICE)-Grate Britain Corpus on one hand and ICE-Ireland on the other hand. In HE, the simple past tense form is often the first choice in natural speech, although the have perfect can be elicited for most sentences with the simple past.

A contrast between the be after and have perfects can be seen in (17) and (18) according to their different contexts.³

(17) {Context: Mary has cleaned her son Brian's room before he comes home from school. When he gets home, Brian throws his uniform, shoes, and school bag on the floor, and then he puts on his casual clothes and is about to go out to meet his friends. Mary sees the mess in the room. She says to Brian:}

a. # I've cleaned your room. (normal pitch)

b. Ok I'm after cleaning your room. [elicitation] (Shimada, 2010, p. 205)

(18) {Context: Ger comes home from work. He is in a hurry and leaves his working clothes and shoes on the bed. He says to his wife, Mary, will you put away my clothes? and goes out. He returns home again, and Mary says:}

a. Ok I've tidied your clothes.

b. # I'm after tidying your clothes. [elicitation] (Shimada, 2010, p. 205)

The be after example (17b) expresses the speaker's intention to act upon the listener, in addition to expressing the proposition that the activity of “cleaning your room” has been done. For example, the speaker may want to imply that “so you should not make a mess” or “how dare you have said that and make me wash your T-shirts,” or “please do not bother me so that I can relax a little”⁴. The be after perfect reflects a focus on the present state after the action or event is completed rather than the action or event itself. This, in turn, invites the hearer to shift their attention to the utterance-time state, and conditions the hearer to act in such a way as to address the state. Having pragmatic oppositions to the have perfect as such, this perlocutionary effect of the be after form is triggered by the grammatical form itself.

The pair of (19) and (20) is an example of a “having a guest” situation.

(19) A (host): D'you want some tea?

B (guest): I'm after having tea.

(20) A: D'you want some tea?

B: I've had tea. (Shimada, 2010, p. 187)

In (19), the speaker (guest) asks the listener (host) not to get him tea, simultaneously encoding the fact that “I have already had tea”. On the other hand, in (20), the SwHE speaker's relatively neutral attitude toward the proposition of having had tea is observed.

There is another example that highlights the pragmatic difference between the have and be after perfects in SwHE. The following (21) is cited from an email text.

(21) I just started working on Monday– am working in a center for adults with learning disabilities as a psychologist – i've just started so idon't know exactly what they want me to do –but am enjoying the work so far! [email text, Cork City]

In (21), the italicized sentence cannot be interchanged with the be after perfect primarily because of the absence of the interlocutor. The have perfect is used for a neutral description of the situation rather than affecting the reader of the mail. (22) is the author's recent encounter of the be after perfect.

(22) {Context: A taxi driver picked me up on the road. She contacted the company to report this, while listening to me, and said to me after the contact:}

I was after being called, you know. [Dublin]

In (22), the utterance of be after was intended to encourage the listener to resume the conversation. This suggests that the “hot-news perfect” is an appropriate tag for the be after perfect in that the speaker of be after intends to share the content of his utterance as if it was a piece of news. The focus is laid on the state of the utterance time, implying something like “I am no longer talking on the phone, so please go on with what you were saying”.

The preceding illustration of the pragmatic contrast between two perfects in SwHE suggests that the be after form and the have forms have developed so that speakers who use both forms assign them different pragmatic functions. This is an autonomous process in HE, apart from the approximation to the “target language” of English.

The be after form exhibits high productivity; it co-occurs with a wide range of verbs and nouns.⁵ Its use is not restricted to any social group (Kallen, 1991). The be after perfect is usually labeled as “hot-news perfect.”⁶ In the author's observation, the label, referring to the status as news of the thing being reported (McCawley, 1971) portrays the property of be after V-ing/NP not so much because of the indication of recency that some studies (Harris, 1984; Kallen, 1989, 1990) discuss but rather because of its expressive connotations. Using be after, the speaker exhibits their emotional attachment to the given event completed at the time of speaking while presupposing that their addressee does not yet know the news that they are reporting (Shimada, 2010).⁷

In summary, the have perfect denotes completion of an action that was initiated in a prior point in time, while the be after construction highlights the relevance of its completion to the context of speaking. By virtue of be, the be after construction relates the fact that some activity or event is in the status of completion in the speech context, where not only the speaker but also the listener/interlocutor are involved. This difference in focus, along with the perlocutionary effect, may furthermore be supported by Kallen's survey of what type of social interaction in which be after tokens occur. The be after form is present more frequently in the conversational domain of “Friendly (work)”/“Friendly (general)”/“Family”/“Shops” than in public domains.

Two meanings of do be in a stabilized SwHE

The do be form is generally described as a habitual-aspect marker. Examples from John B. Keane's (1928–2002) literature basically confirm this, but further examination of settings where do be is used has allowed for a more explicit description of the meaning and usage of the do be form⁸.

(23) and (24) are examples of the do be form. The form expresses the recurring activity of the agent of the verb.

(23) 'Tis not aisy, a-girl, to kill you! You have the appearance of a small one, a young one. We do be praying for you in our prayers, whenever we get the notion to kneel. [SIV 21]

(24) Why do you be always singin' that oul song? Where did you pick it up, anyway? [MYM 1]

This usage is now called “habituality do be ₁”. Importantly, besides do be ₁, there are a number of examples where the do be form appears in a predicate phrase expressing an inherent property of the subject. This “inherent property” use of do be, henceforth labeled as do be ₂, refers to a stated quality that belongs to the object by nature. Significantly, the do be in SwHE occurs in relative clauses, often following an NP + where, the way (including how). This kind of clause, appended to a head NP, defines an inherent property, offering a complementation of the antecedent.

(25) Will you open it or you'll drive me to Gleannnan Gealt where your own equals do be. [SIV 39]

“Gleannnan Gealt where your own equals are (Gleannnan Gealt is known to be the place where your own equals are)”

(26) 'Tis a wonder you took your backside from the table where people do be eating. [SIV 3]

(27) What would be in it but thoughts to disturb her young head the night before her marriage. Have you no knowledge of the way a woman do be the night before? [SIV 38]

In (25), the place name Gleannnan Gealt is specified in the relative clause; similarly, in (26), the inherent property of “the table” is specified in the clause. (27) is an example of dobe₂ occurring in a clause followed by the antecedent “the way”.

The do be habitual is considered to have developed elaborate usages until it constitutes a part of the stable system of SwHE. Shimada (2013) reports that while there are a number of examples of dobe₂ found in Keane's play scripts in 1950–1970s, this usage is not attested by present SwHE speakers. Uses of the dobe₂ type have largely converged with those of the unmarked present tense in contemporary HE, and the convergence, combined with the speakers' awareness toward standard norms, may lead to a relative decline in usage; this will be further addressed in Norm shift by the second contact. The presence of two types of do be in the Keane corpus and the probable decline in stage III may foreshadow its change and relocations between form and function in the system of SwHE.

The 'tis… construction

(28) and (29) are examples of SwHE cited from Filppula (1999), whose informants were born sometime in 1900–1910, with the recordings conducted in 1970s. (30) is an example from the same county (Co. Kerry) of the author's Keane corpus.

(28) a: And did they speak English and Irish?

I: There's more spoke Irish one time. It have died away. Our language is dying away, since we got our own, independence. It is more English they are speaking now. [Kerry] (Filppula, 1986, p. 93)

(29) …and I here in bed with my nerves in a bad state and my left breast sore. Maybe 'tis cancer I'm getting or maybe 'tis something wrong altogether. [STD 56]

(30) How do we know but maybe 'tis dead you are, or worse. [STD 11]

These types of sentences are often described as it-clefts in the literature because of similarities in appearance to the it-cleft in British English. However, because of some stark differences with BrE it-clefts, the author gives them their own label. Since, in SwHE, a majority of the examples have procliticized forms beginning with 'tis (e.g., 'tis, 'twas, 'tisn't) and separable properties from it-clefts, the author refers to them as the 'tis… construction.

Studies have noted that in HE, “clefts,” the 'tis… construction in the author's terminology, occur highly frequently (Curme, 1931; Taniguchi, 1972; Filppula, 1999) and allow for wider ranges of phrasal categories in the focus position than they do in British English (e.g., Guilfoyle, 1985; Filppula, 1986, 1999; Henry, 1995). For example, the phrasal categories AdjP and non-finite VP can occur in the focus position in HE. Besides the phrasal categories in the focus position, there are important differences in form and function of the 'tis… construction. Examining the characteristics of HE and how they contrast with StE will allow for us to begin to see the sentence construction and the way of expressing information structure in HE.

(i) 'Tis

'Tis is prominent in SwHE. 'Tis (pronounced as /tiz/) is a procliticized form whose major tensed/negated representations are 'tis, 'twas, and 'tisn't. It is recognized by SwHE speakers as a single unit. 'Tis is frequently used in SwHE daily conversation, both in cleft-like patterns and apart from them. In particular, 'tis forms are often used as answers to yes-no and tag questions, and as affirmative responses to sections of dialogues. (31)–(32) illustrate these usages.

(31) A:'Tis a grand day thank God!

B: 'Tis! (with a nod) [Cork]

(32) Ellen: But Glory be to God is anyone safe? That's a dead loss.

Nora: Tis.

Ellen: That's a dead loss that place.

Nora: Tis tis. [cited from (Murphy, 2006)]

This usage of 'tis as a single-word utterance may gain the idiomatical status of a response token. It is common among sentences containing the sentence-initial 'tis in HE. In HE discourse for giving responses, 'tis is often used as observed in (33), cited from the International Corpus of English (ICE)-Ireland (Kallen and Kirk, 2008).

(33) B: A lot of, a lot of them now we get, come from Beechwood or, some of these […] yeah yeah. And they can't understand, that the measure of care they get, by comparison with paying a wad of money.

A: Mm

B: For nothing.'Tis only… basically… get them up and feed them [ICE-Republic of Ireland, S1A055]

(33) is a conspicuous example that reveals continuity of the function of 'tis ('twas) between its use as an independent lexeme, seen in (31) and (32), and its usage in the 'tis… construction. Both usages appear in (34).

(34) I: But 'twas the tenants put up that, like, his tenants. I believe, that put up that.

b: I didn't know that.

I: 'Twas, 'twas, I don't know, but it is written on it, that it is his tenants put up that, like. (Filppula, 1986, p. 171)

The first sentence, in which 'twas is used, is likely to have been spoken to connect things in mind with the preceding utterance. In the third line, 'twas is used to recall things in mind. The following it is may reveal the speaker's affirmation that the event (i.e., his tenants put up that) is true.

Furthermore, the continuity of the discourse function of the 'tis… construction can be observed in the usage of 'tis in (35), which the author happened to hear in a conversation during his fieldwork in Kerry. B is explaining why A should read the book.

(35) A: Why do you say so?

B: 'Tis everything in that book happened in this area.

The 'tis… construction is used to give a response to the question in (35). The responding speaker indicates her strong commitment to the truth of the thought.

The 'tis in HE can be described as a discourse marker for affirmation, showing subjective commitment to the statement. The word “subjective” is used here as an adjective derivation of subjectivity, which means the “expression of self and the representation of a speaker's perspective or point of view in discourse” (Finegan, 1995, p. 1), in other words, a “speaker's imprint” (ibid.). This article, for the current purpose, does not pursue the issue of subjectivity any further; however, it suggests that 'tis is used as a marker of the subjective commitment to the content to be expressed by the utterance that the said marker introduces.

(ii) “Salience sensitivity”

This article maintains a “non-cleft” analysis suggested by Shimada (2018) for the 'tis… construction in present-day SwHE. The “non-cleft” means that the analysis does not depend on it-clefts in StE. The description was made not on the norm of StE but on the norm of SwHE on its own to allow for a more economical description of SwHE. This permits a uniform explanation for the HE data on its own. “[T]he initial assumption is that if a linguist who has never spoken or learned any variety of English encounters HE for the first time for linguistic description, (s)he could find a better, or at least different, way of describing 'tis(~it is)… and the things involved in light of the data from this language (Shimada, 2018, p.249).” According to Shimada (ibid.), the 'tis… pattern can be accounted for as the combination of the clause-initial 'tis and a finite clause, in which salience marking may operate. She defines “salience” as having a heavier informational load than other elements in the state of affairs that is being expressed in a sentence. Salience can be marked in speech syntactically and prosodically. HE is salience-sensitive; informational saliency determines not only prosodic presentations but also syntactic forms.

To make the concept of salience more explicit, Shimada (2018) compares salience and focus using pitch accent and the it-cleft in StE. In general English, prosodic prominence is used to mark saliency of the information on a particular constituent in a sentence. For example, a speaker may say “I saw MARY in the park” with prosodic prominence on Mary when they place more significance on the Mary constituent than the other syntagmatically related constituents. That is, Mary is salient. On the other hand, the syntactic strategy of using an it-cleft, as in “it is Mary that I saw in the park”, expresses focus. Focus, more precisely identificational focus, is defined as a new piece of information vis-à-vis the presupposition made by the sentence, while salience is, by definition, independent from the presupposition and addresses the syntagmatic relation of constituents. There is, of course, overlap between focus and salience, but salience has proven to be a more economical way to precisely describe the function(s) of the 'tis… construction in SwHE. This “salience sensitivity” comes from the Irish language, as will be further considered in Expressions of information structure.

While the it-cleft is a syntactic expression of focus, the HE pattern that resembles it-clefts on the surface is the combined outcome of the 'tis(~it is) clause and the fronting of a salient constituent. HE is salience-sensitive; informational saliency determines not only prosodic presentations but also syntactic forms.

(iii) The sentence construction

'Tis… clauses the following: giving the reason or adding to the topic in the previous utterance, i.e., (36); assuring one's self or someone else, (37); recollecting an event in the past or opening a discourse, (38).

(36) I do like to be beside the seaside. 'Tis far away from the seaside we are, God bless us and save us in the warm weather. Far away indeed from the time I was a slip of a girl walking the streets of Ballybunion in my figure and you winking at me. [STD 56]

(37) …and I here in bed with my nerves in a bad state and my left breast sore. Maybe 'tis cancer I'm getting or maybe 'tis something wrong altogether. [= (29)]

(38) (In the beginning of a column) 'Twas at the river bank I met {X=name}. She stopped to talk to my dog. I told her it wasn't my dog and that I was walking Dougal for a friend. [Cork]

'Tis takes a clause in which a salient constituent, in syntagmatic terms (i.e., compared to other constituents of the sentence), is fronted, if there is a one as such.

In HE, salience is marked by fronting a salient constituent in cooperation with prosodic prominence, sometimes with the supplement of that placed after the salient constituent of NP-subj, PP, and AdvP. In addition, the insertion of that significantly lessens acceptability and the context that the construction appears different from that of the it-cleft in StE as discussed.

(39) a.'Tis dead you are.

b. ^*'Tis dead that you are.

(40) a. 'Tis grumbling he is.

b. ^*'Tis grumbling that he is.

The use of that in the case of AdjP and non-finite VP salience is ungrammatical or infelicitous, as noted in (39) and (40). A data survey that the author conducted confirmed that that is present when the chunk of the fronted constituent (NP-subj, AdvP, and PP) is large relative to the other constituents, as in (41).⁹

(41) He's doing well for himself when you consider he left the national school from the fourth class. 'Twas from studying the television programmes in the papers that he learned how to read. [STD 10]

The boundary of the salient constituent from studying the television programmes in the papers is clearly expressed with the support of that. Another notable feature is that that occurring in this pattern is spoken with or without a break after the salient constituent on some occasions (e.g., in reading a text aloud), and it is also true that SwHE speakers find the that-less sentence 'Twas from studying the television programmes in the papers he learned how to read preferable in spoken language.¹⁰ That occurs when informational saliency is not entirely expressed with the fronting of a salient constituent in the 'tis complement and prosodic prominence of the fronted constituent. It is used to supplement salience marking.

Irishness realized in English

Expressions of tense/aspect

The be after perfect is generally supposed to have calqued on tar éis (~tréis) in Irish. Although the be after construction seems to be a straightforward translation of the tar éis construction in Irish, the low frequency of tar éis in Irish (Greene, 1979; Ó Sé, 2004) makes it a poor candidate for calquing. Greene (1979) reports that there was only 1 example of tréis in 19,753 verbal forms in Buntús Gaelige of the 100,000-word corpus, in addition to his reference to Hartmann (1974) who reveals that there was no example in a corpus of conversations totalling about 117,000 words recorded in Connemara. Ó Sé (2004) notes that there are only four tréis examples of the 360 of verbal adjective constructions in AntOileánach.

Greene (1979) also reports that use of the construction using tréis as in the example (42) is rapidly extending its field in Irish spoken outside the Gaeltacht. This suggests influence on Irish from HE.

(42) Tá sé tréis leitir a scríobh.

“He's after writing a letter.”[HE] (Greene, 1979, p. 122)

Tréis used in a substantive verb clause had not always been an idiomatic or grammaticalized way of expressing aspect, evinced by its low frequency in corpora. The sudden increase in the use of tréis outside the Gaeltacht, in addition to the evidence of low frequency in corpora, implies that the expression of perfect using tréis in Irish is gaining in use now, so it is more of a modern expression. This, by extension, gives support to the idea that the be after perfect is an invention of HE on its own. If not an idiomatic use of tréis in Irish, what was involved in forming the be after construction? The author located the answer in the Irish method of expressing aspect. Irish has verb-initial construction, and this is likely to have provided the schema for HE's sentence production. Rather than being a calque from tréis, it is more likely that the general characteristics of sentence construction in Irish have influenced the formation of HE. In other words, as the author will argue below, the be after construction is construed in the stability of the subjunctive verb construction in Irish, which allows for aspectual expressions.

In Irish, aspect is denoted by constructions with the substantive verb bí (tá in the present tense) in the sentence-initial position, one of which is the passive construction (Stenson, 1981; Ó Siadhail, 1989; Russell, 1995; Doyle, 2001; Nolan, 2006). Note that the passive used here is particular to languages with substantive verbs and refers to the sense where “a noun phrase, which does not present the agent, appears as the subject of the substantive verb in the first argument slot following the substantive verb in the position reserved for the grammatical subject” (Nolan, 2006, p. 140). The term “passive” is used in the literature (Ó Siadhail, 1989; McCloskey, 1996; Ó Sé, 2004) in reference to many European languages as a syntactic term concerning the realization of arguments; in the passive in Irish, the argument corresponding to the subject of an active verb appears as the object of the preposition ag “at” (McCloskey, 1996).¹¹ “Perfect” has been generally adopted by grammarians of Irish (Ó Siadhail, 1989; Ó Sé, 2004; Stenson, 2020). There are some dialectal variations over the choice of active and passive variants. “The passive variant has a far higher frequency in Munster” (Russell, 1995, p. 102), the area where SwHE is spoken. In Munster, the aspects of the progressive, the prospective, and the perfect are expressed in the passive construction. (43a, b) and (44) are cited from Nolan (2006, p. 141–142), and (43 c, d) from Ó Sé (2004, p. 181); the glosses are the author's. English translations are given based on Greene (1979), Ó Siadhail (1989) Russell (1995), Doyle (2001), Nolan (2006), and Stenson (2020). (43) shows perfect forms, (44) prospective forms, and (45) progressive forms.

(43) a. Tá an leabhar léite agam. [perfect, Irish]¹²

be.pres. the book read.VAat-me

“The book is read by me./ I have the book read [HE]”.

b. Tá an leabhar léite.

be.pres. the book read.VA

“The book is read”.

c. Tá sé tagtha

be.pres. he come.VA

“He has arrived”.

d. Tá sé feicthe ag Máire.

be.pres. it seen.VA at Máire

“Mary has seen it”.

Ó Sé (2004, p.186) has left a note that “evident increase in frequency [of the perfect constructions, especially the past participial types in southern Irish, as in (43 c, d) of our examples] in the twentieth century is doubtless due to bilingualism with English”. Ó Sé (2004, p. 186) also informs us of the fact that “until quite recently, the perfect was considerably less frequent in Gaelic dialects than in English or other languages of western Europe”. Given these insights, we shall extend our view to other aspectual constructions in Irish that are similar in the use of the peripheral ag (=at) + agent (or actor).

(44) a. Tá an leabhar leléamh agam. [prospective, Irish]

be.pres. the bookwith reading.VNat-me

“The book is to be read by me”.

b. Tá mé le/chun léamh an leabhar.¹³

be.pres I with/toward reading.VNthe book

“I am to read the book”.

(45) Tá an leabhar (dh)á léamh agam. [progressive, Irish]

be.pres the book at reading at-me

“The book is being read by me/ I'm reading the book”.

These periphrastic sentence constructions seem to have provided Irish speakers who were shifting to HE with the sentence-construction schema for expressing aspectual meanings. Considering the deep-rooted use of the bí construction and its multiplicity of uses in Irish, a plausible scenario is that it provided a schema in the incipient days of HE for creating an additional aspect using “after” in the expansion of the pattern [bí (= substantive verb) NP_undergoer PP ag NP_actor]. Also noted is that aspectual categories that are expressed in the bí construction are transferred from Irish to HE. Thus, in HE, Irish semantic divisions are realized using the forms provided by English, conforming to the English morphosyntax in an SVO language where the agent becomes the subject of the sentence.

Since the form of tar éis is not used as an aspect grammatical marker in Irish, the bí constructions, which served as the core of aspect-marking in Irish, are the most likely candidates for producing HE-specific expressions for aspect. Specifically, bí constructions inspired the innovated use of be after in HE to decode a completive state after the action that the verb expresses at the time of utterance. Although the use of passive constructions for denoting perfect, prospective, and progressive aspects in Irish could not transfer directly to HE where the agent or actor who does the action that the verb expresses usually becomes the subject of the sentence, it has had some influence by negotiation with the expression frame of English.

Explaining the habitual do be of HE also requires reference to the Irish language. In HE, variations are seen in realization of the habitual aspect, with, roughly speaking, be(es) in the north and do be in the south. The existence of the habitual forms in the different dialects suggests that the habitual category in Irish paved the way for its formation. There are numerous descriptions concerning the origins of this feature in HE (Joyce, 1910; van Hamel, 1912; Bliss, 1972; Harris, 1986; Kallen, 1986; Filppula, 1999; Hickey, 2000). The substratal presence of the category created a need by the speakers for its linguistic expression. This conditioned their reanalysis of the periphrastic do in primary linguistic data of Early Modern English as marking habitual aspect. In addition, Kallen (1985) has postulated that the do be form results from re-interpretation of do of Early Modern English, which was then juxtaposed with be in HE to mark a habitual, durative, or generic aspect. A similar analysis is given later by Hickey (2000) under the term “usurpation” (ibid: 113). According to him, periphrastic do forms that were semantically empty in the input variety of English were functionalized in HE because of a habitual grammatical category in the Irish substratum.

In Modern Irish, habituality is expressed as in bíonn, the inflectional form of bí. The present habitual bíonn indicates a repeated or regular state of affairs.

(46) Bíonn sé tinn.

“He is (regularly) sick; he is sickly” (Stenson, 2020: 59).

The existence of a morphological category for the habitual in the form of the substantive verb would have been important, as it would inspire the need for the equivalent category in the newly adopted language. Also, Bliss (1972, 1979) argues that Irish speakers acquiring English associated periphrastic do with the dependent form ending in Early Modern Irish -(e)ann (present indicative) on the basis of their contextual parallels; the use of this ending, which originally functioned as a dependent form, was extended into the absolute environments over the course of the 17th century. “[T]he ending -(e)ann is found only in the consuetudinal present bídheann”; the auxiliary do “would therefore come to be associated with a consuetudinal meaning” (Bliss, 1979, p. 293). Harris (1986) criticizes this hypothesis, pointing out discrepancies between syntactic contexts in which the Irish-dependent form and StE periphrastic do occurs and explains that the dependent form in Modern Irish occurs not only after the particles ni (negative), an (interrogative), and nach (negative-interrogative), as the expanded use in the contexts where do-support is required but also in other contexts such as after the conjunctions mura (“unless”), dá (“if”), go/gur (“that”), and the relative particle a. The characteristic he points out actually shows a striking parallel with dobe₂ of SwHE from the Keane corpus, which only appears in relative clauses, as we observed in the section Do be habitual.

Concerning the use of do in English in Ireland, the periphrastic do remained in the nonstandard daily spoken English vernaculars in south Ireland, whose use of do was strengthened under the linguistic conditions of Gaelic/English bilingual speakers¹⁴ despite the assumed decline until the 18th century in south England. According to Filppula (1999, p. 140–142), based on a thorough survey of studies on the use of periphrastic do in BrE, “this construction [periphrastic] reached its peak in the middle of the 16th century, after which it started to decline very rapidly and became quite rare by the early 1600s” (Ellgård, 1953; Filppula, 1999, p. 140). Rissanen (1991), in a study based on the Helsinki Corpus, “notes the significant drop in frequencies of use of periphrastic do in the records of trials (which can be considered to be closest to the spoken mode) as early as the period 1570–1640, whereas the pattern retained a relatively high frequency of use in official letters even in the last EModE [Early Modern English] period, i.e., between 1640 and 1710” (Filppula, 1999, p. 141). There is also a note in Wright (1900) that periphrastic do “became obsolete after about 1700 (apart from archaic and poetic uses), except in the south-western dialects where it survives as the normal form up to the present day”.

In Irish grammar, the opposition between permanency and temporality is primarily significant; this is most evident in two types of verbs for “to be”, namely, copula and substantive verbs. (47) illustrates the fundamental distinction.

(47) a. Is bainisteoir mé.

“I am a manager”.

b. Tá mé i mo bhainsteoir.

“I am a manager (lit. I am in my manager.)” (Stenson, 2020, p. 51).

In (47b), a construction consisting of the substantive verb bí (tá in the present form) and a prepositional phrase are used to indicate the temporal status of being a manager. Most relevantly is that the substantive verb is used when temporal reference other than to the speech time is needed. This property of bí provides the basic outline for the expression of various kinds of aspect.

Aspectual distinctions such as perfect, progressive, prospective, and conditional are subsequently and periphrastically expressed by a combination of the substantive verb bí and a preposition with a verbal noun phrase (Stenson, 1981; Ó Siadhail, 1989; Russell, 1995). It is noteworthy that habituality is, on the other hand, morphologically expressed with the inflected form of bí (i.e., bíonn in Modern Irish). In addition to analyzing the fit of do be in the overall aspectual system of HE, which is the perspective taken by this article, a specific look at the historical formation of the do be form in HE provides insight into the rationality of the grammar system of HE. Importantly, English did not use the auxiliary do with be, and there was no parallel in Irish for the HE do + V pattern (Filppula, 1999, p. 137). Filppula (1999) discusses the pattern involving be in the do be form, referring to Bliss (1972) argument: Phonetic resemblance between Irish bí/bídh and English be would facilitate transfer of the use from Irish bí/bídh to English be by Irish learners of English and thus provide the basis for the adoption of be as a consuetudinal aspect marker in HE. The dependent form in Irish had a syntactic distribution very similar to the uses of the auxiliary do in English, i.e., in early Modern Irish “be”, the ending (e)ann was only found in the consuetudinal present bídheann.

Furthermore, by taking into account the oppositions in the Irish verbal system, we could seek the answer to the question “why do be V-ing, not do + V, was innovated and has survived in SwHE? The difference between tá and bíonn is most likely to be displayed in the semantic distinction between “I am writing a letter” (but not “I write a letter.”) and “I do be writing a letter” in SwHE. Owing to the substantive construction expressing both continuous (bí) and habitual (bíonn), the do be V-ing form, having be within, is more stable than do + V when expressing habituality and has thus become an established part in contemporary varieties of SwHE.

Expressions of information structure

We are now in a position to consider why the 'tis…. construction has the form it does today. As discussed in the sections of 'Tis and “Salience sensitivity”, it holds distinct properties from the it-cleft of English, which it resembles on the surface. We examine Irish traits and their integration into forms of English via the 'tis… construction in SwHE. The claim that HE has inherited its way of expressing information structure from Irish. In other words, in both languages, salience affects the form of sentences. Recall that salience refers to the syntagmatic relationship between constituents. In a salience-sensitive language, a constituent having more saliency, in theory, is marked prosodically and/or syntactically. The following gives evidence that the salience dependency of SwHE is traced to the Irish language.

We have adopted “salience” rather than “focus” to describe the information structure of 'tis…. In describing salience, the expression of salience in StE via prosodic accent and the tendency of HE to express this syntactically were discussed. This discussion is now extended to the Irish language. Greene (1966, p. 42) states that “Irish expresses emphasis by grammatical means rather than by intonation, and any stressed word can be brought to the head of a sentence, with is [Is] before it”¹⁵. Prosodic marking is not a general strategy for denoting information structure in Irish, but it adopts the syntactic device using Is.

Let us have a closer look at examples of HE and Irish. In HE, salience can be syntactically expressed even when not accompanied by 'tis. (48) is an example of fronting in SwHE.

(48) Pats: I saw the young girl, Sive, and the other one going the road to town airly [early] in the day.

Nanna: Gone to buy the wedding clothes they are. Fifty pounds Dota gave to buy the clothes and the drink for the wedding.

Pats: 'Tis about the wedding I came. Last night we made a plan in the caravan. [SIV 34]

Marked constituent order, underlined in the conversation above, is often used for expressing informational saliency in SwHE,¹⁶ and this strategy is one of the elements manifested in the 'tis… pattern. In StE, on the other hand, this type of informational saliency is likely to be prosodically encoded.

The analysis of the 'tis… pattern as a combination of 'tis and a clause in which salience marking may operate is similar to the behavior of the corresponding Irish constructions. 'Tis functions as a discourse marker to mark the commitment of the speakers' subjectivity. Omission of 'tis does not affect grammaticality in case of a that-less construction. (49a) resembles the 'tis… pattern in HE; (49b) shows fronting of the salient constituent without a copula.

(49) a. Is lúchorpán a chuartaíonn Seán. [Irish]

COP leprechaun REL seeks John

b. Lúchorpán a chuartaíonn Seán.

leprechaun REL seeks John (McCloskey, 1979, p. 116)

In Irish, the clause initial Is, a copula, can be omitted, as in (49a). The translation is given in general English by McCloskey as “It's a leprechaun that John seeks” for both (49a) and (49b). The information structures are, however, more apparent in their HE translations in (50).

(50) a. Is lúchorpán a chuartaíonn Seán. [Irish]

“Tis a leprechaun John seeks” [HE]

b. Lúchorpán a chuartaíonn Seán. [Irish]

“A leprechaun John seeks” [HE]

The Irish Is construction is straightforward to be translated into HE, unlike StE, thanks to the similarity between HE and Irish in the syntactic expression of salience. Further examples are shown in (51) and (52); the translations of which the author has provided in HE and StE.

(51) Is ag caint a bhíodar. [Irish]

Is at talk.VN PRT were.they (Doyle, 2001, p. 89)

“Tis talking they were” [HE].

“They were TALKING./(^*It is talking that they are.)” [StE].

(52) Is é an t-arbhar a bhaineann m'athair le speal [Irish].

Is it the grain REL reaps my father with scythe (Cotter, 1994, p. 136, 139)

“Tis the grain my father reaps with a scythe“ [HE].

“My father reaps the GRAIN with a scythe/(It is the grain that my father reaps with a scythe)” [StE].

The correspondence between 'tis and Is is supported by an independent observation of Greene (1966, p. 40–42) concerning Irish. He observes that Irish, from its oldest form¹⁷, always uses Is in a way which resembles it's and c'est, as opposed to is and est, in modern English and French, respectively. Greene (1966, p. 40) states: “Even in the oldest Irish it is already used in much the same way in which it's and c'est are used in modern English and French; these constructions, like Is, represent a departure from the old rule that the subject and predicate of the verb ‘to be' must be in the same person, as we can see clearly from the sentence It is I in the English Bible (Matt. xiv 27), where the Greek original has ego eimi and the Vulgate ego sum”.

In other words, Is is used impersonally. The literal English translations of Greene's examples in (53) below capture this property of the Irish Is, where he gives it's for the translation of Is.

(53) a. Is múinteoir é. [Irish]

“It's a teacher him/He is a teacher”.

b. Is é Seán an múinteoir [Irish].

“It's him, Seán, the teacher/Seán is the teacher”.

Greene also points out that in the identification sentence (54a), the words Is múinteoir were translated as “it's a teacher”, and they can also be used in emphatic sentences including (54b).

(54) a. Is múinteoir Seán.

“It's a teacher, Seán”.

b. Is múinteoir atá ina chónaí anso.

“It's a teacher who lives there”.

The Is used in identification sentences such as (53a,b) and (54a) is likely to be the same as the Is used in the emphatic sentence in (54b), which would mean that Is is always used in an auxiliary role in salience-marked sentences. The shared impersonalness of Is in Irish and it is in English allowed for HE to develop its own uses of 'tis and its variants it's and it is for the salience-aiding function.

In this way, the 'tis of HE and the Is of Irish show significant similarity in the expression of information structure. In addition, they allow a similar range of phrasal categories in the position of a salient constituent. As in HE, in the Irish pattern (Is + X + relative particle a + Y), NP-Subj, NP-Obj, AdjP, PP, and AdvP can occur in the X position (McCloskey, 1979, 2005; Stenson, 1981; Ó Siadhail, 1989; Doyle, 2001). It is also noteworthy that the sentence-initial usage of 'twas, 'tisn't, 'twasn't, isn't it, and is it, respectively corresponds to the Irish copula forms of past ba, non-past negative ní, past negative níor, negative interrogative nach, and the interrogative an. The Irish copula construction was transferred to HE and has developed into the 'tis construction used to mark speaker's commitment to the proposition the clause expresses. As a whole, 'tis marks affirmation, 'twas recollection and 'tisn't negative confirmation and so on. Having the copula-derived marker in the sentence-initial position, HE has inherited the strategy of mapping information structure onto sentence forms from Irish and realized it using English morphosyntax.

Contact-induced grammar formation

HE has been referred to as a “contact vernacular”, and the resemblance of its historical background to that of creoles has been discussed by some linguists of HE (Harris, 1984, 1990; Filppula, 1990, 1999; Odlin, 1997; Todd, 1999). Harris (1991, p. 314), for example, states that, “HE is not typical of English vernaculars in that it has a recent history of language contact which at least partially resembles those of creoles”. Although the majority of writers on HE “fight shy of labeling HE as a creole” (Filppula, 1999, p. 15) and thus designating it as a distinct language, it is also a fact that HE cannot be fully understood if it is described only as a local dialect of English. In the initial development of HE, it becomes necessary to regard HE as an independent product resulting from the synthesis of English and Irish. Moreover, the claim that HE is not a mere dialect of English is sustained by empirical evidence in linguistic and sociolinguistic data of HE. Many of the characteristics would be difficult to explain without the influence of an Irish substratum, as already argued in pioneering studies, especially Filppula (1999) among others. The attempt in this section is to visualize the process of formation of HE as a contact vernacular, a process that continues to the present day as discussed in prior sections.

Studies of other scholars from contact-linguistic perspectives also support this analysis of HE, leading us to consider the link between so-called creole and non-creole languages like HE and focus on language phenomena. DeGraff (2005), i.e., disaffirms the distinction between creole and non-creole languages. Similarly, Winford (2001, 2009) discusses basic similarities between English-based creoles and indigenised varieties of English in the characteristics of their paths of development, both of which, he claims, were shaped by the interaction of L1 knowledge and universals of language creation. Inspired by these discussions, this article proposes a model of contact-induced grammar formation based on HE, which has obvious involvement of the two source languages, Irish and English. The following model is suggested based on the analyses presented in this article, which were greatly helped by the findings and analyses of HE in major studies by Harris (1984) and Filppula (1999). The approach of the author hereafter highlights changing linguistic aspects of HE as they relate to the likewise changing socio-historic situation, and makes no commitment as to whether HE should be deemed a creole or a dialect.

Where today's form came from

So far, we have observed some earlier examples in the rise of HE and some of the contemporary facets of the system behind it. We have discussed Irish language traits and their realization in HE morphosyntax. A common factor among do be, be after, and 'tis… is that they reflect grammatical categories inherited from Irish. Irish offers a base for grammatical oppositions, which can be described as conceptual constructions dividing the world up in order to express it with language. This base is involved in the formation of a sentence by forming the basis for grammatical categories, with which states of affairs are rendered.

For example, we have seen that habituality is grammaticalized both in Irish and HE, and that the informational saliency of constituents is represented syntactically in both Irish and HE. The above examinations on the aspectual expressions of HE suggested the robust involvement of the sentence construction pattern in Irish. The verb-initial property of Irish has contributed to the formation of HE, inducing features apart from StE. The examination in this article specifically suggests that the bí construction in Irish has influenced the formation of aspectual expressions in HE, and that the Is construction has triggered the formation of expressions for information structure.

Based on the discussions so far, the following working hypothesis 1 is suggested: The main contributor of vocabulary (the lexifier) is English, and the basic morphosyntax also comes from English. Irish provides grammatical oppositions, thus serving as a base for expression formation.

In Section Irishness realized in English, we have examined the development of original forms in the system of HE. The distribution of be after was established in competition with other aspectual forms, one of which is the have perfect. In do be V-ing, habituality is marked as opposed to the be stative and the be V-ing continuous; furthermore, the Keane corpus has confirmed two types of do be, including dobe₂ in a relative clause. HE, as a matter of course, not only shows distinctive characteristics vis-à-vis StE, such as the three characteristics that the present study has highlighted, but also as a language, has an autonomous grammar formation system in force, which produces expressions of tense/aspect and information structure.

With hypothesis 1, we may find some grammatical states of the supposedly stable HE (remember stage II in Figure 1) explicable, whereas the latest states of HE, which are relentlessly exposed to other varieties of English, remains unexplained. We must also take this exposure to other varieties and its effects into account. In fact, we have already come across some aspects in which the exposure seems to have influenced HE in prior sections of the article, including the extra use of that to mark informational saliency (The 'tis… construction). Besides, Shimada (2013, 2016a) reports social connotations of do be among speakers of SwHE and illustrates speakers' awareness toward “Standard” based on her field research. Speakers of HE, in exposure to standard norms, find the form of do be bad grammar, sometimes with the stigma of “poorness” and “uneducatedness”. The linguistic form has, in a way, served as a criterion of education and socioeconomic status.

Norm shift by the second contact

The do be form is an overt example of the second contact, where the first was between English and Irish to form HE (refer to Figure 1). The second contact is concerned with HE as a stable variety and other major varieties of English resulting from the spread of mass media, mainly television broadcasting, joining the EC, and increased access to education after the 1970s. In other words, in the second contact, HE went from endo-normative to exo-normative as interacting with main standard varieties of English. This norm shift is diagnosed by the acute rise of awareness of a “Standard” among speakers of SwHE.

Note that not all the Irish characteristics of HE are judged as bad grammar or taken negatively by the speakers. See details in (Shimada, 2010, 2015, 2016b), who has conducted a survey of speakers' awareness toward 26 HE sentences in 11 feature categories. The contrast in speaker perception of do be with be after is telling. While both do be V-ing and be after V-ing are equally regarded as showing Irishness, the do be form is judged as bad grammar and is avoided in speech while the be after form has a significantly more positive judgment. Shimada (2016a) explains the gap in their awareness with the term of morphosyntactic conformity. The sequence of two auxiliaries, do and be, does not comply with a morphosyntactic constraint of “Standard”, which is an abstract construct in the linguistic knowledge of the speakers¹⁸. The be after form, on the other hand, having two types of the complement V-ing and NP, does not violate English morphosyntax, although it is distinct from StE. The noun categories including gerund follow a preposition; this conforms precisely to English morphosyntax. Speakers of SwHE are highly conscious of morphosyntactic conformity but not semantic congruity. A form that satisfies morphosyntactic conformity is generally accepted.

The norm adopted in the primary and stable stages of contact-induced language formation, stages I and II in Figure 1, is not equivalent to the one that the lexifier language adopts for itself. HE was generated in the indigenously established norm by untutored adult L2 acquisition. The majority of Irish adult monoglots became bilingual through contact and interaction with other Irish people who had already acquired English in a context where English was a significant means to promote socioeconomic security. In such non-institutional contact situations, HE, whose speakers were in a process of shift from Irish to English, would be likely to exhibit transfer of Irish-driven sentence construction as well as culturally essential Irish lexicon. This state if agreed finds some similarities to the “endonormative stabilization” described by Schneider (2007) as one of the five characteristic stages in his Dynamic Model to capture general features of postcolonial English.¹⁹ However, in a later stage, speakers of HE are assumed to have put into use a model of StE and have also acquired this through a written medium. The shift from a norm indigenously established on the community basis to an exo-normative model can be referred to as “norm shift”.

Working hypothesis 2: In the process of HE formation to date, a norm shift occurred during the second contact (stage III in Figure 2). HE shifted its norm from the indigenously established one to a perceived Standard English. This has been triggered by contact with other major varieties of English since the 1970s (refer to Figure 1).

FIGURE 2

Figure 2. Grammar formation of HE.

Based on speakers' statements from sessions with them during fieldwork (e.g., Shimada, 2013), it seems that the overall degree of speaker awareness of “Standard” grows when the variety they speak is undergoing contact with major or standardized varieties of the lexifier language. This awareness may affect the linguistic repertories that speakers use in their everyday linguistic exchanges. The increased degree of awareness of "Standard" among speakers of HE can be an outcome of the norm shift. Contact with globally dominant or internationally standardized varieties is not special in the case of HE but is true of all languages that are currently exposed to a dominant variety of their lexifier language. Potential examples include African American Vernacular English (AAVE), Jamaican English, and TokPisin. Although of course careful research is needed before any definite claim can be made, AAVE and Jamaican English are exposed to mainstream American English and are clear cases of decreolization. In this sense, there seems to be similarities with HE in terms of the contact phenomenon. Norm shift may be another aspect that HE shares with other contact languages, including creoles exposed to the lexifier language in its matured stage.

A model to be suggested

HE provides an example of what happens when two different languages come into contact within a community. HE, unlike creole languages, has a single substrate language. This has an advantage in isolating a specific area of substrate influence and assessing contact-induced grammatical formation. Harris (1984, p. 191) states: “the task of isolating specific areas of substratal influence would appear to be much easier in the case of contact vernaculars for which a single substratal language can be unambiguously identified. Varieties of this type would thus seem to constitute one of the most fruitful research sites for those interested in testing substratum hypotheses. One such vernacular is Irish English”. HE thus provides a valuable material for investigating how the grammar of a new language is formed when two languages (we will call them “Language A” and “Language B” for convenience) meet, and for investigating contact-induced language change. Based on the examination of HE, this article concludes by proposing a model of the way grammar comes to be, stabilizes, and then undergoes further changes.

In Figure 2, A is given for Irish and B for English. In the first contact between A and B in a speech community (stage I), new language formation is facilitated, with the native speakers of A acquiring and accommodating B. The newly developed language is tagged with A'+ B' in the figure (stage II); this language forms a grammatical system and attains what can be described as “stable HE,” which is thought to have been spoken by a few generations after the first contact between Irish and English. Stable HE is exemplified in varieties of southwest HE in the early to mid-20th century. The formation and development of HE is contingent upon the sociolinguistic circumstance that the history of Ireland has laid, sketched in Figure 1. As the working hypotheses 1 and 2 suggested in Where today's form came from and Norm shift by the second contact, in the case of HE, Irish offers a base for grammatical oppositions, while English provides the basic vocabulary and morphosyntax. The formation toward the stability of the system then proceeds.

Going further down the timeline, HE is then in the second contact phase and exposed to the major varieties of English. This is indicated by (III), where (A' + B')₁, the progressed version of A' + B', faces intense contact with B₁ in the given community. The contact fosters convergence; the renewed language is tagged with (A' + B')${}_{1^{\text{′}}}+$ B₁. This is the result of the second contact with another English variant B₁, experiencing a norm shift as suggested in the working hypothesis 2. Note that following the process outlined in Figure 2, the element of A may be diluted (i.e., buried under new elements) but will never completely disappear even if further intensive contact with other varieties of English occurs, because it was part of the formation. To give an extreme example, in the distant future, we may end up with ((((A' + B') + B”) + C) + D), and some of the new elements may compete with A', producing complicated results that are exceedingly difficult to trace back to A', but A' is still there at the base, unless a drastic demographic change overtakes the community and a language replacement occurs. The heritage from Language A gradually fades through a process of convergence with Language B, but even so, grammatical oppositions of Language A are retained in some way or another.

Conclusion

This article, taking salient characteristics of HE, including the be-after perfect, the do be habitual, and the 'tis….. construction as windows to its underlying properties, has examined how the current forms and meanings of these characteristics are the result of a dynamic process of language formation guided by language contact. The article surveyed earlier forms in the rise of HE, described some facets of contemporary HE, and finally introduced an integrated perspective across the characteristics to discuss how elements from Irish have been inherited and realized in English and updated over time until the present.

One of the main issues was Irish language traits and their realization in HE morphosyntax. Examinations of expressions of tense/aspect and information structure in HE have led to the conclusion that it has inherited grammatical oppositions from Irish. The division of grammatical categories, with which states of affairs are rendered, has been realized using English morphosyntax. This article illustrated cases of habituality marking and saliency marking, in which HE adopts the oppositions from Irish. Also pointed out was the high degree of involvement of the Irish sentence construction pattern in the aspectual expressions of HE, wherein the verb-initial property of Irish has contributed to the formation of HE. Specifically, this article argues that the bí construction of Irish has influenced the formation of aspectual expressions in HE, and that the Is construction of Irish has triggered the formation of expressions for information structure. These findings suggest that further investigation of the Irish morphosyntax is likely to show even more involvement of Irish in the formation of HE.

From a sociolinguistic perspective, norm shift, which occurs in the second contact with other varieties of English (refer to Figure 2), presumably over the past 50 years or so, is key in the explanation why HE is as it is today. Speakers' awareness toward the standard norm brings them to prune HE so that it conforms to the morphosyntactic constraints of StE. Importantly, the model presumes that grammatical oppositions of Irish remain even after such pruning, although the heritage from Irish is gradually diluted through a process of convergence with other varieties of English. This presumption and the supposed model should be tested against many different grammatical forms of phenomena in HE in the future both to assume this model's variability and to uncover the nature of the forms and phenomena.

This article has confined itself to the description of HE; thus, the model suggested within addresses the grammar formation of HE, considering settings and language contact situations particular to HE. However, the model, along with the two working hypotheses of the formation of grammar in a contact situation and of norm shift, can be applied to find similarities and contrasts with other language-contact phenomena. This, in turn, can contribute to various studies in a variety of languages, helping to enrich our understanding of contact-induced language change and the mechanism of grammar formation of a language in contact.

Data availability statement

The original contributions presented in the study are included in the article/supplementary materials, further inquiries can be directed to the corresponding author.

Author contributions

TS: design of the study, collection and analysis of the data, and preparation of the manuscript.

Funding

This study was supported by JSPS Grant-in-Aid for Scientific Research (C) Knowledge of Language and Constant Modification: A Theory and its Verification based on Contemporary Hiberno-English (KAKENHI Grant Number 20K00611) and JSPS Grant-in-Aid for Scientific Research on Innovative Areas Linguistic Chronogenesis (KAKENHI Grant Number 18H05521).

Acknowledgments

The author would like to express her gratitude to all those who have supported her fieldwork. Particularly, she would like to note her friends who have been key consultants: Elsie Harris, Emmet Stones, and the Keane, Clifford, and O'Connor families. The comments and questions she has receiving at various occasions in conferences since 2005 and those from discussions with her teachers and colleagues have also been of great help to her investigation. In addition, the author is very thankful to reviewers and editors whose comments and advice have been of great help for the current study.

Conflict of interest

The author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher's note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

Footnotes

1. ^The phrase “salient characteristics” is used when comparing with other varieties of English including British Standard English. The idea of the “salient features”/“salient characteristics” comes from Filppula (1999), who uses the phrase to refer to features or characteristics of HE vis-à-vis Standard English. Such salient characteristics help to uncover influence from the Irish language and, as in the current study, can serve as “windows” to investigate contact-induced properties of grammar.

2. ^The article is based on the author's prior studies and descriptions, which include Shimada (2018) for the analysis of the 'tis….. construction, Shimada (2013) for do be habitual, and Shimada (2008, text in Japanese) for the be-after perfect. The aim of this study is to integrate properties of these features and their formation into a theory of contact-induced grammar formation.

3. ^The examples (17)-(22) are discussed in detail in Shimada (2008, 2010).

4. ^These implications are not made by the have perfect. Of course, such conversational implicatures could be conveyed in an utterance containing the have perfect by virtue of intonation and shared background by the speaker and listener.

5. ^As to the wide range of verbs, Kallen (1991) reports that his Dublin corpus, which contains 114 tokens of after obtained from 74 speakers in Dublin, includes 56 verbs, with the following frequencies: being (10 tokens of copula and seven tokens of auxiliary/passive) and getting (12 tokens).

6. ^This form has been labeled as “hot news perfect” Harris (1984), the “after perfect” (Filppula, 1999), and “immediate perfective” (Hickey, 2000).

7. ^This is based on the author's sessions with speakers in years 2004-2007. In the impression the author has formed from visiting Ireland regularly, the usage of be after V-ing has expanded and generalized to include a use in non-expressive contexts (Shimada, 2016b, p. 163–164). Ronan (2005) notes that, the range of use of the after perfect is speaker-specific.

8. ^Keane is a playwright from Listowel, County Kerry, where the author has been undertaking fieldwork since 2003. Examples from John B. Keane's plays and letter series written from the 1950s to1980s are cited with their abbreviations; the collection of examples is referred to as the “Keane corpus”. The examples are sorted by grammatical features and categories; one of the features is' Tis(~it is)… constructions. The Keane corpus, as a whole, comprises over 18,000 words. Keane is a local writer and his work, cherished by local people, is an invaluable source of the local language from the 1950s-1970s. He is known as a major Irish writer with many successful plays and books (Smith and Hickey, 2002). The following is a list of his woks cited in this article, headed with their abbreviations: SIV, Sive (Keane, 1959); HHM (Keane, 1961); STD, Letters of a Successful TD (Keane, 1967). The spellings used in the citations are those used in Keane's plays. This article also includes examples that the author happened to record and encounter during his fieldwork.

9. ^The author's data looked at the constituent in the position following 'tis, represented as X in the following. In the author's data, there are respectively in [X (supposed focus position) = NP-Subj] 48 examples of the that-less type and 60 tokens with that; in [X = PP/AdvP] 60 and 21, in [X = NP-Obj] 43 and 1; in [X = AdjP/Non-finite VP] 18 and 0. Data presented in other studies, although they do not give numerical information, also turned out to this distributional tendency (Filppula, 1986, 1999; Ó hÚrdail, 1997).

10. ^Some SwHE speakers who pronounce that rather clearly in AdvP-saliency sentences report that such examples are “modern” [Co. Kerry]. This may suggest the influence of their awareness of normative grammar. It is assumed that this awareness motivates the use of that in HE.

11. ^The passive status of this construction may be questioned in terms of how it is formed and the fit of the passive in the system of Irish. Dillon (1941), claiming that Modern Irish aa sa daantaagam (“I have done it”) is not passive, points out that the substantive verb was construed with the past participle already in the Old Irish period. Dillon further argues that “the participle is an adjective and the verb “to be” is the main verb, not an auxiliary” (1941, p. 59). In contrast, McCloskey (1996, p. 254–255) argues that “in a full clause, the participial form is supported by the verb be used as an auxiliary; in a small clause, the participle appears without any auxiliary”. Ó Sé (2004), to whom this article adheres, basically agrees with McCloskey (1996) except that he adopts “perfect passive” instead of “perfective passive” to describe this construction.

12. ^This aspect has been referred to in the literature with various terminologies such as “perfect,” “perfective”, and “completive” (Ó Siadhail, 1989; Ó Dochartaigh, 1992; Mac Eoin, 1993; Russell, 1995; McCloskey, 1996; Ó Sé, 2004; Nolan, 2006; Stenson, 2020). In light of the fact that the state-of-affairs that the verb phrase expresses is a completed state with current relevance, the author chooses perfect, not perfective. His use of “perfect” and “perfective” is based on Comrie (1976, p. 12): “perfect” refers to a past situation that has present relevance, for instance, the present result of a past event; whereas “perfective”, contrasting with “imperfective”, denotes a situation viewed in its entirety without regard to internal temporal constituency. Following Comrie, to avoid ambiguity between the two meanings, the author avoided using “perfective” as the adjectival form of “perfect”. Note that while “perfect” focuses on the state resulting from the completion of an action, the “completive” focuses on the completion itself, and so it can be applied to, i.e., (43).

13. ^There are dialectal variations for the use of the preposition le. In Munster, “there is a distinction between obligation using le and intention using chu(i)n ‘toward”' (Russell, 1995, p. 100).

14. ^The author's assertion here is bolstered by Harris' insights. He notes (Harris, 1986, p. 193): “it can be argued that the distinctive habitual markers in both the substrate and nonstandard varieties of the superstrate had a mutually reinforcing effect on the development of a similar category in the new contact vernaculars”.

15. ^Likewise, Cotter (1994, p. 134), contrasting the Irish and English systems, notes that, “Much of what is signaled by intonation in English is encoded grammatically in Irish. In particular, the most “important” aspect of an utterance will be moved toward the front of a sentence in Irish or be marked morphologically”. Cotter, for the purpose of illustrating the “focus-marking” system in Irish, defines focus as a “highlighting or emphasis of a particular constituent in an utterance” but “not in relation to presuppositions or background”. This means that in order to illustrate the Irish “focusing” constructions, the primary definition of focus must be discarded. In the description of information structuring in HE, a similar change or adjustment compromise is necessary. Rather than redefining focus, this article addresses salience.

16. ^The basic word order of HE, including SwHE, is SVO. It is often said, however, that the order of HE is not so rigid because “topicalisation is more frequently used in the rural varieties of HE” (Filppula, 1990:44). “Topicalisation”, in Filppula's words, refers to the movement of a constituent from an unmarked to a marked, i.e., the clause initial, position. According to his data on the frequencies of topicalisation per 1,000 words (1986: 190-94), Kerry and Clare, where SwHE is spoken, show the highest level (1.4) compared to HE dialects and British English (BrE) [cf. Dublin 0.9, BrE 0.4].

17. ^Thurneysen (1946, p. 327), a grammarian of Old Irish, also describes Is as “it is”, corroborating the similarity.

18. ^As an anonymous reviewer has pointed out, do plus a non-finite verb does not generally violate morphosyntactic conformity when do is used emphatically. However, emphatic do does not cooccur with be, hence, i.e., ^*I DO be writing is not registered by native speakers of English. As such, the use of do be in HE clearly digresses from StE. Note that in habitual do be seen in HE, do is not accented, in contrast to emphatic do.

19. ^Schneider (2007), in his research on the evolution of postcolonial forms of English, proposes a five-stage process of emergence: foundation, exonormative stabilization, nativization, endonormative stabilization, and differentiation. Taking a general approach, Schneider (2007) does not focus on the linguistic process alone, instead describing general features of sociopolitical background, identity constructions, sociolinguistic conditions, and linguistic effects. Schneider (2007) considers HE a dialect of English, not a postcolonial English.

References

Amador-Moreno, C., McCafferty, K., and Vaughan, E. (2015). Pragmatic Markers in Irish English. Amsterdam: John Benjamins. doi: 10.1075/pbns.258

PubMed Abstract | CrossRef Full Text | Google Scholar

Barron, A., and Schneider, K. P. (2005). The Pragmatics of Irish English. Berlin: Mouton de Gruyter. doi: 10.1515/9783110898934

PubMed Abstract | CrossRef Full Text | Google Scholar

Bliss, A. J. (1972). Languages in contact: some problems of Hiberno-English. Proc. R. Ir. Acad. Sci. 72, 63–82.

Google Scholar

Bliss, A. J. (1979). Spoken English in Ireland 1600-1740. Dublin: The Dolmen Press.

Comrie, B. (1976). Aspect. Cambridge: Cambridge University Press.

Google Scholar

Cotter, C. (1994). “Focus in Irish and English: contrast and contact,” in Proceedings of the Twentieth Annual Meeting of the Berkeley Linguistic Society. p. 134–144. doi: 10.3765/bls.v20i1.1450

PubMed Abstract | CrossRef Full Text | Google Scholar

Curme, G. O. (1931). A Grammar of the English Language, Volume II syntax. Essex: Verbatim.

DeGraff, M. (2005). Morphology and word order in “creolization” and beyond. In The Oxford Handbook of Comparative Syntax, Cinque, G., and Kayne, R. S., (eds). Oxford: Oxford University Press. p. 293–372.

Google Scholar

Dillon, M. (1941). Modern Irish atá sé déanta agam ‘I Have Done It'. Language. 17, 49–50. doi: 10.2307/409460

CrossRef Full Text

Doyle, A. (2001). Irish. Muenchen: Lincom Europa.

Google Scholar

Edwards, R. D. (1981). An Atlas of Irish History, Second edition. London: Routledge.

Google Scholar

Ellgård, A. (1953). The Auxiliary do: The Establishment and Regulation of its Use in English. Stockholm: Almqvist and Wiksell.

Google Scholar

Filppula, M. (1986). Some Aspects of Hiberno-English in a Functional Sentence Perspective, University of Joensuu Publications in the Humanities 7. Joensuu: University of Joensuu.

Google Scholar

Filppula, M. (1990). Substratum, superstratum, and universals in the genesis of Hiberno- English. Irish Uni. Rev. 20, 41–54.

Google Scholar

Filppula, M. (1999). The Grammar of Irish English: Language in Hibernian Style. London: Routledge. doi: 10.2307/29742738

CrossRef Full Text | Google Scholar

Filppula, M., Klemola, J., and Paulasto, H. (2008). English and Celtic in Contact. London: Routledge. doi: 10.4324/9780203895009

CrossRef Full Text | Google Scholar

Finegan, E. (1995). “Subjectivity and subjectivisation: An introduction,” in: Subjectivity and Subjectivisation, Stein, D., and Wright, S,. (eds). Cambridge: Cambridge University Press 1–10. doi: 10.1017/CBO9780511554469.001

CrossRef Full Text | Google Scholar

FitzGerald, G. (1984). Estimates for baronies of minimal level of Irish speaking amongst successive decennial cohorts: 1771-1781 to 1861-1871. Proc. R. Ir. Acad. Sci. 84, 117–155.

Google Scholar

Greene, D. (1966). The Irish Language. Dublin: The Cultural Relations Committee of Ireland.

Google Scholar

Greene, D. (1979). Perfect and perfectives in Modern Irish. ÉRIU. 30, 122–141.

Google Scholar

Guilfoyle, E. (1985). Hiberno-English: A parametric approach. In Perspectives on the English Language in Ireland: Proceedings of the First Symposium on Hiberno-English, Harris, J., Little, D., and Singleton, D. Dublin: Centre for Language and Communication Studies, Trinity College Dublin. P. 121–132.

Google Scholar

Harris, J. (1984). Syntactic variation and dialect divergence. J. Linguistics. 20, 303–327. doi: 10.1017/S0022226700013918

CrossRef Full Text | Google Scholar

Harris, J. (1986). Expanding the superstrate: Habitual aspect markers in Atlantic Englishes. English World-Wide. 7, 171–199. doi: 10.1075/eww.7.2.02har

CrossRef Full Text | Google Scholar

Harris, J. (1990). More on brogues and creoles: What's been happening to English short u? Irish Univ. Rev. 20, 73–90.

Google Scholar

Harris, J. (1991). “Conservatism versus Substratal Transfer in Irish English,” in Dialects of English: Studies in Grammatical Variation, Trudgill, P., and Chambers, J. K,. (edds). London: Longman. p. 191–212. doi: 10.4324/9781315505459-14

CrossRef Full Text | Google Scholar

Harris, J. (1993). “The grammar of Irish English,” in Real English: The Grammar of English Dialects in British Isles, Milroy, J., and Milroy, L,. London: Longman. p. 139–186.

Google Scholar

Hartmann, H. (1974). Distribution und Funktion der Expanded Form in einigen Dialekten der irischen Sprache von Co. Galway. Zeitschrift für celtische Philologie. 33, 140–284. doi: 10.1515/zcph.1974.33.1.140

CrossRef Full Text | Google Scholar

Heine, B., and Kuteva, T. (2005). Language Contact and Grammatical Change. Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511614132

CrossRef Full Text | Google Scholar

Henry, A. (1995). Belfast English and Standard English: Dialect Variation and Parameter Setting. Oxford: Oxford University Press.

Google Scholar

Henry, P. L. (1957). An Anglo-Irish Dialect of North Roscommon. Dublin: University College Dublin.

Google Scholar

Hickey, R. (2000). “Models for describing aspect in Irish English,” in The Celtic Englishes II, Tristram, H. L. C,. (ed). Heidelberg: Carl Winter. p. 97–116.

Google Scholar

Hickey, R. (2007). Irish English: History and Present-day Forms. Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511551048

CrossRef Full Text | Google Scholar

Hickey, R. (2016). Sociolinguistics in Ireland. Basingstoke: Palgrave Macmillan. doi: 10.1057/9781137453471

CrossRef Full Text | Google Scholar

Hickey, R., and Amador-Moreno, C. P. (2020). Irish Identities: Sociolinguistic Perspectives. Berlin, New York: De Gruyter. doi: 10.1515/9781501507687

CrossRef Full Text | Google Scholar

Joyce, P. W. (1910). English as We Speak It in Ireland. Dublin: Wolfhound Press.

Google Scholar

Kallen, J. L. (1985). Bilingualism and the genesis of Hiberno-English syntax. TEANGA. 5, 98–113.

Google Scholar

Kallen, J. L. (1986). “The co-occurrence of do and be in Hiberno-English syntax,” in Perspectives on the English Language in Ireland: Proceedings of the First Symposium on Hiberno-English, Harris, J., Little, D., and Singleton, D,. Dublin: Centre for Language and Communication Studies Trinity College Dublin. p. 133–147.

Google Scholar

Kallen, J. L. (1989). Tense and aspect categories in Irish English. English World-Wide. 10, 1–39. doi: 10.1075/eww.10.1.02kal

CrossRef Full Text | Google Scholar

Kallen, J. L. (1990). The Hiberno-English perfect: grammaticalisation revisited. Irish University Rev. 20, 120–136.

Google Scholar

Kallen, J. L. (1991). “Sociolinguistic variation and methodology: after as a Dublin variable,” in English Around the World: Sociolinguistic Perspectives, Cheshire, J,. Cambridge: Cambridge University Press 61–74. doi: 10.1017/CBO9780511611889.005

CrossRef Full Text | Google Scholar

Kallen, J. L. (1994). “English in Ireland,” in English in Britain and Overseas: Origins and Development, vol. V, The Cambridge History of the English Language, Robert, W,. (ed). Burchfield, Cambridge: Cambridge University Press. p. 148–196. doi: 10.1017/CHOL9780521264785.005

CrossRef Full Text | Google Scholar

Kallen, J. L. (1997). Focus on Ireland. Amsterdam: John Benjamins. doi: 10.1075/veaw.g21

CrossRef Full Text | Google Scholar

Kallen, J. L. (2013). Irish English Volume 2: The Republic of Ireland. Berlin: De Gruyter. doi: 10.1515/9781614511298

CrossRef Full Text | Google Scholar

Kallen, J. L., and Kirk, J. M. (2007). “Accessing Celticity in a corpus of Irish standard English,” in The Celtic Languages in Contact, Tristram, H. L. C,. (ed). Potsdam: Potsdam University Press 270–288.

Google Scholar

Kallen, J. L., and Kirk, J. M. (2008). The International Corpus of English. Ireland: CD-ROM.

Google Scholar

Keane, J. B. (1959). Sive. Dublin: Progress House.

Google Scholar

Keane, J. B. (1961). Highest House on the Mountain. Cork: Mercier Press.

Google Scholar

Keane, J. B. (1967). “Letters of a Successful TD,” in The Celebrated Letters of John B. Keane. Cork: Mercier Press.

Google Scholar

Kirk, J. M. (2017). The present perfect in Irish English. World Englishes. 36, 239–253. doi: 10.1111/weng.12256

PubMed Abstract | CrossRef Full Text | Google Scholar

Mac Eoin, G. (1993). “Irish,” in The Celtic Languages, Martin, J,. (ed). Ball with James Fife, London: Routledge. p. 101–144.

Google Scholar

McCafferty, K. (2017). Irish English in emigrant letters. World Englishes. 36, 176–190. doi: 10.1111/weng.12252

CrossRef Full Text | Google Scholar

McCafferty, K., and Amador-Moreno, C. P. (2012). “CORIECOR: A corpus of Irish English correspondence: Compiling and using a diachronic corpus to study the evolution of Irish English,” in New Perspectives on Irish English, Migge, B., and Ní Chiosá, N., Amsterdam: John Benjamins. p. 265–287. doi: 10.1075/veaw.g44.13mcc

CrossRef Full Text | Google Scholar

McCawley, J. (1971). “Tense and time reference in English,” in Grammar and Meaning, McCawley, J., London: Academic Press. p. 257–272.

Google Scholar

McCloskey, J. (1979). Transformational Syntax and Model Theoretic Semantics: A Case Study in Modern Irish. Dordrecht: D. Reidel. doi: 10.1007/978-94-009-9495-9

CrossRef Full Text | Google Scholar

McCloskey, J. (1996). “Subjects and subject positions in Irish,” in The Syntax of the Celtic Languages: A Comparative Perspective, Borsley, R. D., and Roberts, I,. (eds). Cambridge: Cambridge University Press. 241–283. doi: 10.1017/CBO9780511586279.009

PubMed Abstract | CrossRef Full Text | Google Scholar

McCloskey, J. (2005). “A note on predicates and heads in Irish clausal syntax,” in Verb first: On the Syntax of Verb-initial Languages, Carnie, A., Harley, H., and Dooley, S. A,. Amsterdam: John Benjamins. p. 155–174. doi: 10.1075/la.73.12mcc

CrossRef Full Text | Google Scholar

Murphy, B. (2006). Gender and its influence on linguistic production in responding across different age groups. Presented at Sociolinguistic Symposium 16, University of Limerick.

Google Scholar

Nolan, B. (2006). The passives of modern Irish. in Passivization and Typology: Form and Function, eds. A. Werner and L. Leisiö, Amsterdam: John Benjamins 132–164. doi: 10.1075/tsl.68.10nol

CrossRef Full Text | Google Scholar

Ó Cuív, B. (1986). “Irish language and literature, 1691-1845,” in A New History of Ireland: IV Eighteenth-century Ireland 1691-1800, Moody, T. W., and Vaughn, W. E., (eds). London: Clarendon Press. p. 374–423.

Google Scholar

Ó Dochartaigh, C. (1992). “The Irish language,” in The Celtic Languages, MacAulay, D,. (ed). Cambridge: Cambridge University Press. p. 11–99. doi: 10.1017/CBO9780511551871.003

CrossRef Full Text | Google Scholar

Ó hÚrdail, R. (1997). “Hiberno-English: Historical background and synchronic features and variation”, in The Celtic Englishes, Tristram, H. L. C,. (ed). Heidelberg: Universitätsverlag C. Winter 180–200.

Google Scholar

Ó Sé, D. (2004). The ‘after' perfect and related constructions in Gaelic dialects. ÉRIU. 54 179-251. doi: 10.1353/eri.2004.0015

CrossRef Full Text | Google Scholar

Ó Siadhail, M. (1989). Modern Irish: Grammatical Structure and Dialectal Variation. Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511620621

CrossRef Full Text | Google Scholar

Odlin, T. (1997). Hiberno-English: pidgin, creole, or neither? CLCS Occasional Paper. 49, 1-39. Dublin: Trinity College Dublin.

Google Scholar

Rissanen, M. (1991). “Spoken language and the history of do-periphrasis,” in Historical English Syntax, Kastovsky, D,. (ed). Berlin: Mouton de Gruyter.

Google Scholar

Ronan, P. (2005). “The after-perfect in Irish English,” in Dialects Across Borders, Filppula, M., Klemola, J., Palander, M., and Penttilä, E, Amsterdam: John Benjamins. p. 253–270. doi: 10.1075/cilt.273.16ron

CrossRef Full Text | Google Scholar

Russell, P. (1995). An Introduction to the Celtic Languages. London: Longman.

Google Scholar

Schneider, E. W. (2007). Postcolonial English: Varieties around the World. Cambridge: Cambridge University Press. doi: 10.1017/CBO9780511618901

CrossRef Full Text | Google Scholar

Schrier, A. (1997). Ireland and the American Emigration 1850–1900, Second edition. Chester Springs PA: Dufour Editions.

Google Scholar

Shimada, T. (2008). Airurando eigo be after V-ing no hyougen kouka: have kanryou to no tairitsu wo chuushin ni [On the Pragmatics of the be after Construction in Hiberno-English: a Contrast with the have Perfect]. Tokyo University Linguistic Papers. 27, 187–206.

Google Scholar

Shimada, T. (2010). English in Ireland: Beyond Similarities. Hiroshima: Keisuisha.

Google Scholar

Shimada, T. (2013). The do be form in southwest Hiberno-English and its linguistic enquiries. Festschrift for Professor Hiroshi Kumamoto, Tokyo University Linguistic Papers. 33, 255–271. doi: 10.15083/00027516

CrossRef Full Text | Google Scholar

Shimada, T. (2015). Morphosyntactic features in flux: awareness of ‘Irishnessness' and ‘Standard' in Hiberno-English speakers. Arbbeiten aus Anglistik und Amerikanistik. 40, 41–65. Available online at: https://www.jstor.org/stable/24922039

Google Scholar

Shimada, T. (2016a). Speakers' awareness and the use of do be vs. be after in Hiberno-English. World Englishes. 35, 310–323. doi: 10.1111/weng.12197

CrossRef Full Text | Google Scholar

Shimada, T. (2016b). Eigo to iu Sentaku: Airurando no Ima [lit. The Choice of English: Ireland Today]. Tokyo: Iwanami.

Google Scholar

Shimada, T. (2018). 'Tis…. pattern in Hiberno-English as a grammatical innovation. Festschrift for Professor Tooru Hayasi, Tokyo University Linguistic Papers. 39, 243–263. doi: 10.15083/00074543

CrossRef Full Text | Google Scholar

Siemund, P. (2004). “Substrate, superstrate and universal. Perfect constructions in Irish English,” in Dialectology Meets Typology, Kortmann, B,. (ed). Berlin: Mouton de Gruyter. 401–434.

Google Scholar

Smith, G., and Hickey, D. (2002). John B. Cork: The Mercier Press.

Google Scholar

Stenson, N. (1981). Studies in Irish Syntax. Tübingen: Gunter Narr.

Google Scholar

Stenson, N. (2020). Modern Irish: A Comprehensive Grammar. London: Routledge. doi: 10.4324/9781315302034

CrossRef Full Text | Google Scholar

Taniguchi, J. (1972). A Grammatical Analysis of Artistic Representation of Irish English: With a Brief Discussion of Sounds and Spelling. Tokyo: Shinozaki Shorin.

Thurneysen, R. (1946). A Grammar of Old Irish, Translated From the German by D. A. Binchy and O. Bergin. Dublin: School of Celtic Studies, Dublin Institute for Advanced Studies.

Todd, L. (1999). Green English. Dublin: The O'Brien Press.

Tristram, H. L. C. (1997). The Celtic Englishes. Heidelberg: Carl Winter Universitätsverlag.

van Hamel, A. G. (1912). On Anglo-Irish syntax. Englische Studien. 45, 272–292.

Winford, D. (2001). “ “Intermediate” creoles and degrees of change in creole formation: The case of Bajan,” in Degrees of Restructuring in Creole Languages, Neumann-Holzschuh, I., and Schneider, E. W,. Amsterdam: John Benjamins. 215–246. doi: 10.1075/cll.22.13win

CrossRef Full Text | Google Scholar

Winford, D. (2009). “The interplay of ‘Universals' and contact-induced change in the emergence of New Englishes,” in Vernacular Universals and Language Contacts: Evidence from Varieties of English and Beyond, Filppula, M., Klemola, J., and Paulasto, H,. London: Routledge. p. 206–230.

Google Scholar

Wright, J. (1900). English Dialect Dictionary. Oxford: Oxford University Press.

Google Scholar