Linguistic change in a nonstandard dialect: phonological studies in the history
\nof English in Ireland

It is not my aim to write a unified history of the English
\nlanguage an Ireland, since this has already been undertaken elsewhere
\n(e.g. Hogan 1927; Bliss 1977,1979; Barry 1982). Rather I wish to
\nconcentrate on several specifically phonological developments which
\nallow us to disentangle the competing but sometimes complementary
\ninfluences exerted by Irish, English and Scots on Hiberno-English
\n(henceforth HE). This emphasis reflects an attempt not only to chart
\nthe historical developments in greater detail than has been done
\nhitherto but also to contribute to our theoretical understanding of
\nphonological change. For various reasons the focus is for the most
\npart on northern HE. For one thing, a detailed history of southern
\nHE is already available (Bliss 1979). Secondly, much of my own
\nresearch has been undertaken in the north, particularly while I was
\nworking on the project Sociolinguistic variation and linguistic change
\nin Belfast M Milroy et al 1983). Most importantly, however, northern
\nHE with its mixed linguistic heritage offers a rich and relatively
\nuntapped source of data for investigating the extent to which dialect
\ncontact may be implicated in phonological change. Despite the emphasis
\non northern HE, I draw on comparative material from southern dialects
\nthroughout the following pages.
\n
\nIt is often remarked that, in contrast to the relative
\nhomogeneity of southern BE, the linguistic situation in the north of
\nIreland is quite varied (e. g. Adams 1977: 56). In fact Adam (1973)
\nfinds it convenient to recognise at least seven basic northern varieties
\nspoken in an area with a population of just under two million. This
\ndiversity is in part a reflection of the complex interaction of Scots
\nand English influences in the north, in addition to the contribution
\nof Irish which has left its mark to varying degrees on all types of HE.
\nIt is the Scots element in particular that distinguishes much of
\nnorthern from southern BE.
\n
\nAlmost everything that has ever been written on HE stresses the
\nsupposed effects of Irish Gaelic contact with English on its development.
\nI take up this issue in a treatment of HE consonant phonology in Chapter 3
\nand give it more detailed attention elsewhere in a discussion of the
\ngrowth and structure of the HE verb phrase (Harris 1982). Although
\nthis is clearly an important area, it has been much discussed and it
\nis not my intention to go over the sane ground here. Rather the focus
\nin this thesis is on two other aspects of language contact in the north
\nof Ireland. Firstly, I examine the linguistic developments that have
\narisen from contact between the typologically divergent phonological
\nsystems of English and Scots dialects. Secondly, I attempt to isolate
\nseveral changes which reflect contact between nonstandard HE and modern
\nstandard British varieties.
\n
\nIn Chapter 1 I describe the main types of northern HE that can
\nbe identified according to the different ways in which the tensions
\nbetween English and Scots influences have been resolved. The conditions
\nof contact are in many ways similar to those that obtained in the early
\nstages of British settlement in North America. It therefore comes as
\nno surprise to discover close linguistic parallels between certain United
\nStates and Canadian dialects on the one hand and northern HE on the other.
\nThese similarities also reflect the fact that the major British
\ncolonisation of Ireland was roughly contemporary with that of. North
\nAmerica. Hibernian and American dialects of English display many
\ncommon seventeenth-century features which have since been lost from
\nstandard British varieties.
\n
\nThe tension between Scots and English influences in northern HE
\nmanifests itself most clearly in the area of vowel phonology. Some
\ndialects display a typically English system in which vowel length is
\nphonemic. That is, in these varieties it is possible to identify
\none subsystem of inherently long vowels and another of inherently
\nshort vowels. In characteristically Scots dialects, on the other
\nhand, vowel length is to a large extent phonetically conditioned.
\nBetween these two types lies a range of 'mixed' dialects which show
\nvarying degrees of compromise between phonemic and positional length.
\nThe diffusion of the English and Scots length patterns across different
\ndialects, vowels and phonological environments can be expressed in
\nterm of implicational hierarchies which I set up in Chapter 2.I
\nalso examine the phonetic facts which can plausibly be said to deternine
\nthe order of segment-types on the hierarchies.
\n
\nIn Chapter 3, Iattempt a partial. reconstruction of the internal
\nhistory of the urban HE vernacular spoken in Belfast. By inspecting
\npresent-day sociolinguistic variation for signs of change in progress
\nand checking the results against historical records, it is possible to
\nidentify the main phonological developments that have occurred over the
\nlast century or so. Comparative material from the city's rural
\nhinterland dialects and from the descendants of the original British
\nsource dialects allows us to chart the continuing competition between
\nEnglish and Scots linguistic features. It is also possible to offer
\na fairly clear picture of the sorts of adaptive change that have been
\ntaking place in the vernacular as a result of contact with external
\nstandard norms.
\n
\nThanks to its conservative nature, HE provides the historical
\nphonologist with an invaluable store of archaic patterns of distribution
\nwhich were once current in Early Modern English but which have since
\ndisappeared from standard varieties. Through direct observation of
\nthis material it is possible to gain new insights into some of the
\nwell-known problematical issues of English historical phonology. One
\nof these, which I take up in Chapter 4, concerns the fate of Middle
\nEnglish (ME) /6: / (as in meat) in Southern Standard English. According
\nto some interpretations, it merged with ME /a: / (as in mate), only to
\nreseparate and undergo merger with ME /e: / (as in meet). Belfast
\nVernacular is one of several modern dialects in which these vowels
\nremain three-way distinct. Comparative reconstruction of the changes
\nthat have produced the current reflexes in these dialects contributes
\nto our understanding of what might have happened to M /ɛ:/ in the
\nSouthern Standard. In addition, the results have a bearing on the
\nwider issue of the sorts of strategy that can be : implemented to avoid
\nmerger during chain-shifting. In Chapter 5,1 examine the other side
\nof the coin. I identify different ways in which phonological merger
\nis achieved and suggest how these might be modelled in terms of rules
\nand representations. I take up the issue of falsely reported mergers
\nand discuss some of the theoretical and methodological implications.
\n
\nIt is a commonly held belief that there is a general trend
\ntowards dialect convergence in English as a result of the standardising
\npressures exerted by universal education and the media. However,
\nrecent sociolinguistic work suggests that, while old rural dialects
\nmay be in decline, diversification is continuing in recently evolved
\nurban vernaculars (see Labov 1972a: 324; 1980a: 252). A survey of
\nthe changes that have affected Belfast Vernacular over the last 120
\nyears or so does indeed confirm that a degree of standardisation has
\ntaken place. However, this has been restricted for the most part to
\nthe lexical incidence of phonemes. At the phonological level, almost
\nno major structural alignment with Southern Standard English has taken
\nplace. Indeed, evidence from change in progress suggests that sane
\ninternal innovations are actually moving in directions which run
\ncounter to standard norm.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.