English World-Wide - Volume 27, Issue 3, 2006

The division of the world’s Englishes into rhotic and non-rhotic types is clearly due to the fact that the former are conservative in not having undergone loss of non-prevocalic /r/, whereas the latter have. The beginnings of the loss of non-prevocalic /r/ in English have generally been dated by historians of the language to the 18th century. It is therefore obvious, and has been widely accepted, that Irish English, Canadian English, and American English are predominantly rhotic because the English language was exported to these colonial areas before the loss of rhoticity in England began; and that the Southern Hemisphere Englishes are non-rhotic because English was exported to these areas in the 19th century after the loss of rhoticity. Analysing newly-discovered data from Australia, we present some surprising evidence that shows that this obvious conclusion is incorrect.

This paper provides an acoustic description of the vowels of Brunei English (BrunE). Ten female BrunE speakers were recorded reading The North Wind and the Sun (NWS) passage. The formant values of the eleven monophthong vowels and the rate of change (ROC) of the diphthong /eI/ were measured and compared with the data of seven British English (BrE) speakers and also the results of similar studies on Singapore English (SgE). It was found that BrunE shares some common features with SgE as both groups do not distinguish between /i˜/ and /I/, /e/ and /æ/, and /f˜/ and /#/. The high back vowels of BrunE, however, are unlike the SgE vowels. Whereas in SgE /u˜/ and /~/ are fully back, in BrunE these two vowels are fronted, so they are similar to the vowels of the BrE speakers. The data also shows that BrunE /f˜/ is more open and less back than BrE /f˜/. For /eI/, the average ROC for Bruneian speakers is considerably less negative than that for British speakers, which indicates that in BrunE, just as in SgE, this vowel is less diphthongal than its counterpart in BrE.

A common feature of Singaporean English, and found to a lesser extent in British and US English, is the “conventionalised scenario” (Goldberg 1995); i.e. a causative construction in which an intermediate causee is neither expressed nor necessarily recoverable from context and common ground, e.g. You cut your hair, in which the action is normally attributed to another, unexpressed participant. The present study provides written data on the use of conventionalised scenarios in Singapore English and explains their link with competing resultative constructions (e.g. You had/got your hair cut) in terms of an ACTION FOR RESULT grammatical metonymy. In this metonymy, the passival resultative construction is substituted with an active‑voice construction and the causer now stands for both the causer and causee together. Contact features in the Singaporean dialect, relating particularly to local Chinese languages and/or Malay, may influence the distributional extent of conventionalised scenarios, as may the overgeneralisation of the semantic constraints on its usage.

Ulster differs from the other three historical provinces of Ireland in the presence of Ulster Scots, an off-shoot of Lowland Scots brought principally from the Western and Central Lowlands of Scotland in the 17th century through a plantation established by King James I and through periodic migrations, especially in times of economic duress in Scotland. Since that time Ulster Scots has been spoken in rural parts of Counties Antrim, Donegal, Down, and Londonderry/Derry, where it was mapped by Robert Gregg in the 1960s mainly on the basis of phonological features. The present article, based on eight years of fieldwork with native speakers in Antrim, analyzes a range of pronominal, verbal, and syntactic features, seeking to identify general patterns as well as variation within Ulster Scots. When possible, comparisons are made to Lowland Scots and Irish English in order to situate structural features of Ulster Scots within the larger linguistic landscape of the British Isles.

Prosodic rhythm was measured for a sample of 20 African American and 20 European American speakers from North Carolina using the metric devised by Low, Grabe and Nolan (2000), which involves comparisons of the durations of vowels in adjacent syllables. In order to gain historical perspective, the same technique was applied to the ex-slave recordings described in Bailey, Maynor and Cukor-Avila (1991) and to recordings of five Southern European Americans born before the Civil War. In addition, Jamaicans, Hispanics of Mexican origin who spoke English as their L2, and Hispanics speaking Spanish served as control groups. Results showed that the North Carolina African Americans and European Americans were both quite stress-timed overall, with no significant difference between them. Spanish emerged as solidly syllable-timed, while Jamaican English and Hispanic English were intermediate. The ex-slaves were significantly less stress-timed than either younger African Americans or European Americans born before the Civil War. This finding suggests that African American English was once similar to Jamaican English in prosodic rhythm.