Redefining current symbols

In their suggestions, Barry & Trouvain propose two solutions that would require the redefinition of existing symbols (both with and without the introduction of a new symbol). In one, [æ] becomes the unrounded fully open front vowel (i.e. Cardinal 4), and [a] is changed to represent the unrounded fully open central vowel. In the other, [a] again is used for the unrounded fully open central vowel, and the new symbol [a] is used for Cardinal 4.

I think it would be a major mistake to redefine the values of either [æ] or [a]. It would at a stroke render all previous IPA texts that use these symbols no longer accurate, or no longer readable by anyone who knows only the new values of these symbols. (The alternative, to teach learners both old and new values would seem highly inefficient.) In the past, the Association has removed recognition from symbols (e.g. the fricative trill and the voiceless implosives); allowed alternative shapes (e.g. [ɡ] and [g]); and done both (allowing, then removing, recognition of [ɷ] and [ɩ] as alternatives to [ʊ] and [ɪ]). However, it has rarely – if ever – changed the phonetic value of a symbol,Footnote ¹ especially of such commonly occurring sounds. For the sake of current students and future researchers I would suggest that both of these choices should be rejected by the Association.

Introducing new symbols

In their introductory section, Barry & Trouvain note that the use of the plus and minus diacritics does not unambiguously provide a way of denoting the central open vowel. This is true, but they fail to note that the diaeresis diacritic would do just that; so that [ä] and [] (or, indeed, [] and []) would denote unrounded and rounded central open vowels. They argue plausibly against the use of diacritics, however, and suggest either the use of the barred vowel shapes, or the introduction of a new symbol.

The use of ‘barring’ as in the close vowels [ɨ] and [ʉ] results, unfortunately, in symbols that are difficult to clearly identify. In particular the barred-a [a] is difficult in small type to distinguish from ordinary [a]. Also, [ɒ] is not very different from [ɵ].

Almost by a process of elimination, therefore, my personal preference would be for the use of [a] for the unrounded central open vowel – should it be deemed to be required. As the authors point out, there is a history of usage of this symbol for just such a vowel, albeit outwith the sanction of the IPA.

The suggestion for a rounded counterpart is to use inverted small capital A. This is potentially confusable with [v], however. Are there small capital letter shapes that have a rounded connotation that are currently not in use by the IPA? I can think of two candidates for the rounded central open vowel that fit this criterion (albeit not originally used for open vowels): from Greek [Ω], and from Cyrillic []. The Greek omega is perhaps to be slightly preferred over the Cyrillic, as it represents an opener vowel. Further, the Cyrillic yu symbol does seem to suggest that it represents a sound combination (which of course it does in Russian and other languages).

In conclusion, the authors argue strongly for a means of symbolizing fully open central vowels. Assuming non-diacriticed symbols are desirable, then the route of new symbols should be chosen, rather than re-assigning values to long-established and commonly occurring symbols. Of the new symbols proposed, barring existing symbols leads to lack of clarity in the symbol shapes. My suggestion, therefore, is to choose the small capital A and the small capital omega to represent the unrounded and rounded central open vowels.