Rhotic speakers pronounce ‘r’ after vowels in all positions, even in words such as ‘cart’, ‘horse’ and ‘far’. It has long been assumed that American English is r-full, in that speakers of this variety always pronounce ‘r’, while English English is r-less, as those speakers drop ‘r’ after vowels. Non-prevocalic ‘r’ is a part of most people’s stereotype of a North American accent, and its use there could be the result of immigrants from the British Isles at a time when the non-prevocalic ‘r’ was more common in the UK. In her article ‘English dialects transported’, Kniezsa claims that ‘what was transported of the language was preserved, in which case the language of the new country shows a variety of archaic forms not to be found any longer in the speech of the mother country’ (Kniezsa quoted in Ureland 1996:351) However, to say that rhotic ‘r’ is a variant of English that was exported to America and has since become extinct in British English is misleading. There are certain areas of the UK where post-vocalic ‘r’ prevails, and on the other hand, there are regions in the USA where this feature has never occurred. Some interesting, yet complex comparisons can be drawn between the historical developments of rhoticity in both countries.

1.0 The development of Rhoticity in the UK

In the UK, the most revered form of English is Received Pronunciation. The fact that RP is r-less suggests that when British English started to lose post-vocalic ‘r’ in the 18th Century, the reason behind it was prestige. Dialect maps from the Survey of English Dialects show the South West, North West and Tyneside and Scotland to be the conservative regions of this feature in the UK. This suggests that, like the /æ/ vs. /a:/ distinction before it, the change from rhoticity originated in the South East and diffused outwards from here. However, unlike the former distinction, the change from rhoticity spread to the north of England but not to the West Country. The reason for this pattern of diffusion becomes more apparent when one considers where rhotic ‘r’ occurs in urban speech. There are very few cities, in England at least, where post-vocalic ‘r’ is a feature. Even in those cities located in ‘r’ conservative areas, rhoticity is not a feature. Clearly, urban areas were significant in the process of this particular linguistic change, which is not surprising given the period during which this change started. The 18th Century saw a rapid growth in cities, particularly in the north of England, as a result of the Industrial revolution. Communication and travel between the South East and the ‘power house’ cities of the North was more essential and available than ever before and facilitated the spread of non-rhotic speech northwards. This would also account for the continued use of ‘r’ in the urban speech of Bristol and the South West. Not only did the West Country have little part in the revolution, but also, as a city, Bristol did not grow much during the 18th century. The docks here were limited in their ability to expand to deal with the new industries, and were subsequently overtaken in importance by Liverpool.
The differentiation does not just occur on a geographic scale, but on a social scale too. Not only is rhoticity in British English no longer the norm, it is now stigmatized as a non-prestigious archaic form and a sign of rusticity, but at the same time of a kind nature. In cities such as Bristol and Exeter, rhoticity occurs in speech a long way up the social scale. However, in nearby Bournemouth, which traditionally has had closer links with London and has a far more varied class population, rhoticity is much less common. Bournemouth was also a ‘new’ city that did not exist before the coming of the railways. It was populated by ‘incomers’ from London, which gives a further reason for its lack of rhoticity.

1.2 The development of Rhoticity in America

The situation in America is far more complicated than in the UK, as there are two possible ways in which a distinction between rhotic and non-rhotic accents in America came about. The first theory claims that the original settlers in the r-less regions of America were non-rhotic before they arrived. Wyld, for example, proposed that the weakening of /r/ began as early as the mid 15th Century, allowing it to have been transported to America as a prestige feature. However, a second suggestion is that the r-less variety may have become prestigious in the UK a lot later. Most linguists believe it to have only become prestigious as late as the 18th Century. In this case, the weakening of /r/ would have become prestigious in America at the same time as in the UK as a result of the strong bonds between the two countries.
In giving evidence for first theory, Wolfram & Schilling-Estes comment that the ‘configuration of the dialects of the Eastern United States still reflect the distribution of early British habitation of the New World.’ (1998:25) According to them, the distribution of ‘r’ pronunciation was explained geographically; the speech variants could be mapped to the areas settled by various immigrants from different speech communities in Britain. For example, mainly immigrants from rhotic Northern Ireland populated the mountains of South Carolina, while immigrants from supposedly already non-rhotic southern England settled the coastal plains. This is one explanation for the marked difference in ‘r’ pronunciation between these two areas today.
However, some linguists believe all British English was rhotic at the time of settlement. In their article ‘Dialects in the US: Past, Present and Future’ Wolfram & Schilling-Estes claim that ‘the speech of the Jamestown colonists more closely resembled today’s American English than today’s standard British speech.’ (1998:93) This statement presumes that at the time of colonization, rhoticity was a feature of all British Englishes. If this were true, the loss of rhoticity would have become a prestige feature in London at the end of the 18th Century and been transferred to America via contact with the East Coast ports. In support of this theory, a linguistic map of North-Eastern USA from the 1930s shows r-lessness radiating from the cultural centre of Boston. However, there are ‘pockets’ of rhoticity that appear to have maintained r-full pronunciation. It is these so called ‘relic’ areas that suggest diffusion of non-rhotic pronunciation has occurred, and the fact that younger speakers in these communities preferred r-less pronunciation implied that the variant was still spreading. Either way, both rhotic and non-rhotic accents can be found in some areas, which suggests a complex linguistic development has occurred.
In his chapter on rhoticity, Downes suggests that ‘both forms crossed the Atlantic as folk pronunciations’ (1998:156). That is to say both forms reached America before either was the norm or a prestige variety, and as such, r-less speech only acquired prestige once the social elite on the eastern seaboard adopted it from Britain. This argument highlights the very close ancestral ties between southern England, New England and the aristocracy of the southern states. Alternatively, Lass (1987) argued that the first settlers were non-rhotic speakers from southern England, and subsequent waves of further immigrants from Ireland and Scotland imported rhoticity at a later date.
The debate as to whether r-less speech gained prestige from England, or on the other hand whether settlers were already non-rhotic by the time they had arrived can be resolved by a compromise proposed by Downes. He suggests that r-less speech became prestigious well after the colonization of America. However, this would have ‘confirmed’ any ‘r’ variability in the eastern ports and encouraged continued ‘r’ loss. Downes added that ‘there is no doubt that non-rhoticity had general prestige across the coastal south’ (1998:159). In support of this, we have evidence of hypercorrected r-less speech in some southern US accents where even today, ‘r’ is dropped altogether, giving us pronunciations such as ‘Pais, Texas’, meaning ‘Paris, Texas.’

2.0 Comparison of development of Rhoticity in UK and USA

Once we have understood how the two variants came to co-exist, it is necessary to understand how the non-rhotic pronunciation came to spread unevenly across the east of America. The explanation for this distribution mirrors the social diffusion that occurred in the UK during the 18th Century. American linguist Raven McDavid claimed that ‘rhoticity was associated with lower levels of education, older speakers and rural as opposed to urban speech.’ (Downes 1998:155) This suggests a similar scenario to what was happening in the UK at the time. Like the Industrial Revolution, the expansion of the plantations saw the social elite spread outwards from the coastal ports, but certain areas were left culturally isolated. The mountains, inhabited by rhotic speaking Ulster immigrants, were not suitable for plantations and were left undeveloped. The consequence was that the non-rhotic plantation owners diffused towards the hinterland taking their prestigious r-less pronunciation with them.
Perhaps the most significant difference between the development of rhoticity in American and British English, is that in the last century, there has been something of a revival in the non-prevocalic ‘r’ in the American English up and down the Eastern Seaboard. In Britain, or rather England, on the other hand rhotic ‘r’ is very much an archaic variable that is unlikely to become prestigious and is most likely to eventually become extinct. However, it is important to point out that in some parts of the UK, rhoticity is still in use and much debated. In her study ‘Postvocalic /r/ in Scottish English: Sound Change in progress?’ Romaine discusses some of the ambivalent attitudes expressed towards this feature in Scottish English. Grant (1914) is quoted as labeling it a ‘idiosyncratic speech defect’, (Romaine quoted in Trudgill 1978:145) while Spietel (1969) found it to be a marker of ‘polite Edinburgh speech.’ (Romaine quoted in Trudgill 1978:145). At the same time, other types of non-prevocalic ‘r’ are becoming widespread in the UK. The so-called ‘intrusive ‘r’’ is a special form of juncture between a word with a spelling that suggests a final ‘r’ and a word that starts with a vowel. For example, ‘four eggs’ may be pronounced ‘four reggs.’
However, in most of the UK, r-less pronunciation has diffused and replaced r-full pronunciation, while in the USA both types were spreading vigorously from opposite centres. In 1966 Levine and Crocket performed a linguistic investigation in a town called Hillsboro, which sat on the border of Midland r-full pronunciation and Southern r-less pronunciation. Speakers with high status were perceived to use both rhotic and non-rhotic pronunciation, while in more formal speech, r-full pronunciation was more common. In complete contrast to non-rhotic use in the UK, r-less speech was the variant used by the old and working class members of society. People in white-collar jobs, and the middle classes preferred r-full speech and thought it to be correct. Consequently, this community appeared to be transitional, and, under the pressure of outside norms, saw rhoticity become the prestige norm.
While the socially mobile, upper classes were responsible for the original transition to r-lessness in both America and Britain, some investigators have questioned who is responsible for this reversal of a historic pattern. Labov performed an investigation into the use of rhotic ‘r’ in New York in 1962. He observed that traditionally, New York was an r-less city during the 19th Century; however, like most of the eastern seaboard, the use of rhoticity had become more variable. In his investigation he looked at the use of rhotic ‘r’ amongst several class types in New York on the basis that ‘the linguistic variable ‘r’ is a social differentiator in all levels of New York city speech.’(Labov 1972:169). He did this by grading the class of shop assistants according to the prestige of the department store they worked in. His method was simple; by asking the assistants loaded questions that required the answer, ‘on the fourth floor’, he could record the use of rhotic ‘r’ in correlation with class. The results showed that the assistants in the more prestigious department stores were more likely to use rhoticity than those in the less prestigious stores. In later sociolinguistic interviews, the amount of r-fullness increased with the formality across all classes, however the middle class informants appeared to overtake the upper class informants in more formal contexts. This hypercorrection suggests that it is the linguistically insecure lower middle classes, and particularly women, who are the agents of this change. The curvilinear hypothesis supports these findings. It claims that linguistic change does not occur at extreme ends of a social continuum, rather, those in the upper working and lower middle classes are at the forefront of linguistic change.

3.0 Conclusion

In conclusion, the east coast of the USA, like the south east of England, has for generations been characterized by non-rhotic speech. However, rhotic speech has been the norm in the American mid-west for just as long. Those living on the east coast have looked to the historic cultural centres of Boston and Philadelphia for linguistic guidance, eschewing rhotic pronunciation as rustic. However for many reasons, there appears to be a growing perception that a rhotic ‘General American’ accent is the norm for American speakers. This may be because the connotation of British English as the most prestigious form has ceased since Britain is no longer the global super-power it once was. At the same time, according to Trask ‘radio, films and television are dominated by rhotic speakers from the Midwest.’(Trask 1996:281) In either case, the studies of Labov, Levine and Crocket serve to demonstrate that there has been a change in the social significance of rhotic ‘r’. Many linguists are unable to provide an explanation for this apparently whimsical change. Trask, however, claims ‘the solution to the problem of the apparent (linguistic) reversal is simple, almost banal. Indeed, it was hardly a linguistic change at all, but rather a social one.’ (Trask 1996:284)