New Zealand English Transcription Model

In the late 19^th century, James Murray’s inclusion of transcriptions in A New English Dictionary embedded the descriptivist stance that OED’s pronunciations editors emulate and hold dear today. Pronunciations were never intended to show how a word ‘should’ be articulated by prescribing the pronunciation of a particular social group to be the ‘acceptable’ and therefore ‘correct’ form. To Murray, pronunciations were essential as they sought to represent ‘the actual living form or forms of a word, that is, the word itself, of which the current spelling is only a symbolization’ (Vol. I, p.xxiv). Ever since then, pronunciations have retained a special status in the OED as to give the current form or forms of each ‘living word’, or as Murray puts it, ‘the latest fact in the form-history of the word’.

The notation system has changed since then, and has developed to cover an ever-expanding range of English varieties, but our descriptivism has not; we strive to show how speakers of that variety are actually pronouncing each word. Certain social and cultural demographics offer particular challenges in this regard, few more so than when we began drafting New Zealand English pronunciations in 2015. Reflecting the distinctive qualities of New Zealand English (NZE) in words such as agist and rouseabout was relatively straightforward; the qualities given by Bauer & Warren’s (2008) description of New Zealand English phonology were readily adapted into a meaningful user-friendly  transcription system, with Deverson & Kennedy’s The New Zealand Oxford Dictionary (2005) giving a clear steer towards the most salient contrasts.

Compared with Australian English (whose pronunciation model was being drafted at the same time), NZE shares similar start and dress vowels, but quite a few of the other vowel qualities are slightly higher or slightly lower. These subtleties pass by many British and U.S. listeners and are likely the root of many a mistaken national identity. The common ‘flapping’ of /t/ between vowels in unstressed syllables is represented too – but this doesn’t apply to all such situations, as NZE uniquely treats words of Māori origin very differently from words of other roots.

Find a television or radio broadcast in New Zealand English and you will usually hear a step change with any Māori word, whether the speaker is Māori or Pakeha. (This is not the same as ‘Māori English’, a group of distinct varieties in linguistic literature and not the focus of our NZE model – Warren & Bauer explain ‘[t]here are Pakehas who speak Maori English, and Maori people who speak Pakeha English’, 2008:78). In borrowing words from Māori, cultural sensitivity has evolved to reject the most ‘assimilationist’ anglicization, embracing Māori and some of its key features. Māori wh is usually rendered /f/. ‘Flapping’ of /t/ does not occur in words such as matuku in the way it does in bitter, but does crop up as a possible articulation of /r/ in words such as parera. For vowels, the qualities do seem somewhat anglicized by many NZE speakers but they may vary in length from their usual NZE forms (usually indicated by macrons in the Māori spelling) and OED transcriptions account for this. The spelling-to-sound relationship in Māori is far more predictable than in English, but there is a layer of complexity when considering that OED does not represent code-switched ‘Māori’ pronunciations, but the most likely minimally naturalized English forms, and naturalization can be variable. The sequence au, for example, is articulated in Māori close to /ʌu/, which is the vowel symbol usually used for words such as goat. Stronger anglicization brings this to /æu/, the vowel in mouth, but not all such words are equally likely to be anglicized this far. Fortunately, it has never been easier to access and compare pronunciation data of different words and in different varieties.

OED’s evolving patterns for Māori anglicizations are shown in a table on our New Zealand English pronunciation model page : this supports the drafting of new pronunciations as well as revision of older ones. Nevertheless, following Murray, the ‘living form’ of a word is what speakers say, and evidence of what is actually spoken will always trump a model’s predictions.

The opinions and other information contained in the OED blog posts and comments do not necessarily reflect the opinions or positions of Oxford University Press.