Chapter 2 therefore provides some background information on the phenomenon of /r/-sandhi and provides an overview of the theoretical literature. Contemporary varieties of English are traditionally classified into two different types based on the distribution of /r/. The most famous non-rhotic accent is undoubtedly 'Received Pronunciation (RP)' which has traditionally been associated with educated speakers of higher authority.
4 There was little agreement on the precise ordering, dating and interrelationship of the historical processes leading to the loss of rhoticity. Keep in mind that one of the distinguishing features of non-rhotic dialects of English is the aforementioned link /r/ alternation. To account for the motivation behind /r/ intrusion, or why non-rhotic speakers insert an /r/ when there is no /r/ in the spelling, it has often been assumed that the phenomenon of intrusive /r/ can be interpreted as an overgeneralization or the lexical extension of the surface alternation to link /r/ to words ending in the same set of vowels.
12 Carr adopts the term 'London English' to refer to the traditional working-class speech of the East End of London, commonly known as Cockney. Returning to the role of non-high vowels in the appearance of intrusive /r/, Crystal (1984:42) suggests that /r/ intrusion is particularly noticeable after /ɔː/ given the openness of the vowel and thus in most cases, it is advisable to pronounce law and order without /r/- insertion. Crystal mentions the quality of the vowel, but Gimson refers to the low frequency of words ending in /ɔː/ which can make /r/ intrusion into this context particularly noticeable and shameful.
For detailed information on hiatus resolution techniques, see e.g. 9) The influence of the quality of the preceding vowel on the probability of r intrusion.
Regional variation 1 Introduction
- New Zealand English
- The emergence of /r/-sandhi in New Zealand English
- Newcastle upon Tyne and Derby
- Received Pronunciation & BBC English
- Dialectal distribution of /r/-sandhi
However, differences in pronunciation were observed between different age groups, with younger speakers adopting an 'advanced' pronunciation of final diphthongs. In order to investigate the possible /r/-intrusion of /au/, one of the base words chosen to elicit the intrusive /r/ was the base plow to which various suffixes (-y, -ese) were added and word boundaries (-ing). attached in target sentences, e.g. 'A farmer plows a field' or 'A plow company plans to completely plow up New Zealand's farms' (Hay & Maclagan, 2006, p. 70). Regarding the interpretation of the results, Hay & Maclagan (2006) documented 5 different factors that slightly influenced the production of intrusive /r/, namely gender, age, class, base words and affix identity.
As a result, one of the weaknesses of this study is the inability to demonstrate the effect of gender on /r/ sandhi production. Further observations of their experiment revealed that the presence or absence of intrusive /r/ was influenced by suffix identity, in a way that there was a sharp increase in the occurrence of /r/ insertion after the strong border suffixes (claw #ing, oprah#ish for example). While the study clearly demonstrates both social and phonetic variation in the pronunciation of intrusive /r/, Hay & Maclagan made no attempt to measure the effects of speech style and speed, speech awareness, or potential self-control on the likelihood of / insertion. R/.
As current New Zealand English exhibits non-rhoticity, the main aim of the research was to synthesize data from the available audio recordings and to find information on how linking and intrusive /r/ arose in this dialect in the nineteenth century. Hay & Sudbury are the first to provide concrete evidence of the 'coexistence' of partial rhoticity and intrusive /r/ production, that is, partially rhotic speakers can equally use /r/ insertion in traditional sandhi contexts18 . While it is true that the empirical study of regional accents has so far been assigned a relatively low priority, extensive studies of Newcastle upon Tyne and Derby have shed new light on our understanding of the sociolinguistic pattern of /r/-sandhi in English.
Based on the results, a clear distinction can be drawn between Newcastle and Derby English speakers in terms of /r/-sandhi use. It should be mentioned that one of the limitations of that study is the absence of explanation for the almost categorical appearance of /r/-sandhi in Derby. However, one of the most striking features of Foulkes' findings is that in Newcastle English there are sociolinguistic constraints on the pronunciation of linking and intrusive /r/.
Foulkes noted a clearly recognizable phonological change in relation to the basic representation of the phoneme /r/. claiming that younger people have "erased the historical divide between r-words and non-r-words". The main focus of this study was to document new evidence on how the formation of /r/-sandhi in RP could be constrained by a number of sociolinguistic and/or phonetic factors, the presence of a second /r/ in the spelling (e.g. , criterion(r) of) or for example, the identity of the preceding vowel. However, one major limitation of the study is the inadequate examination of the frequency of intrusive /r/ in compounds and collocations.
Given the wide variety of /r/-sandhi, the following table compares and contrasts different varieties of English on the basis of the behavior of linking and intrusive /r/. Despite the fact that Ghanaian English is non-rhotic, speakers lack both linking and intrusive /r/.
Conclusion
Despite the fact that /r/ intrusion is phonologically conditioned in non-rhotic English dialects, recent research has consistently shown that the likelihood of intrusive /r/ production may also be socially and stylistically determined. Therefore, the occurrence of intrusive /r/ has been shown to be largely influenced by a wide range of linguistic and non-linguistic factors in several non-rhotic varieties of English. After giving some general information about /r/-sandhi in Chapter 2, I outlined the main issues of descriptive descriptions (for example, the rate of /r/ intrusion according to the quality of the preceding vowel).
Then, in Chapter 3, I have provided a detailed overview of empirical research conducted on /r/-sandhi by focusing on certain linguistic and non-linguistic factors related to variation in intrusive /r/ production. Apart from the main drawbacks associated with corpus-driven approaches (i.e., small data sets), these studies have made notable contributions to our understanding of the social and regional pattern of /r/ intrusion. Given the limits of the present thesis, I have concentrated on describing three dialects/accents of English, namely New Zealand English, Newcastle upon Tyne and Derby, and Received Pronunciation.
Finally, I finished this thesis with a table showing the distribution of /r/-sandhi in different varieties of English. Variability and change in received pronunciation: A study of six phonological variables in television newsreaders' speech. Social and phonetic conditioners on the frequency and degree of 'intrusive /r/' in New Zealand English.
Shields (red.) Proceedings of the Seventeenth Annual Meeting of the Berkeley Linguistics Society, Berkeley Linguistics Society, Berkeley, CA.
