In the weeks following the publication, of our article ‘Do we need a symbol for a central open vowel?’ (Barry & Trouvain Reference Barry2008), we received 12 personal communications commenting on and discussing the points we had made. Martin Ball has since published his discussion points in JIPA (Ball Reference Ball2009), as has Daniel Recasens (Recasens Reference Recasens2009), who was not among the 12 who contacted us directly. Thus it would be fitting for us to summarize the points made by the 11 membersFootnote ¹ whose opinions have not yet been made known to the rest of the IPA community and to refine our standpoint in the light of Recasens’ and Ball's comments.

All eleven argued in favour of a symbol for an open central vowel, with different slants and emphases. The main recurring argument is that, when comparing languages or dealing with varieties within a language, there is a very real need to simply and clearly differentiate front, central and back qualities of open vowel, a point that we made in section 3.1 of our article. There were two acknowledgements that a separate symbol is not justified if the principle of a phonemic contrast is taken as a prerequisite for the adoption of two symbols, but it was also pointed out that such principles (‘simplicity’ being another one) are certainly not rigorously observed in the present system. The arguments can be summarized as: whatever theoretical objections there might be, a separate symbol is needed in practice.

The choice of which symbol to use was almost evenly divided between the small capital A [ᴀ] and the present Cardinal Vowel 4 symbol [a]: Six (ten, if the members of the team represented by one correspondent are considered separately) wanted to keep the system as defined at present, but add [ᴀ] for the central open vowel (our option (a)). However, one of the six only accepted the logic of that choice while actually using [a] in his research and teaching, without any redefinition of the other symbols. Four argued to redefine [æ] as front open and [a] as central open (our option (c)). One chose option (d), i.e., redefining [a] as central open and adding small caps [ᴀ] as front open.

Assuming that those who contacted us directly represent a larger silent constituency, we feel it is significant that nobody felt strongly enough moved to argue against the central-open vowel symbol. We therefore claim that phoneticians and phonologists do feel the need for a separate symbol for a central open vowel and many have made their own modifications to the IPA system for their personal use. De facto a multiform vowel symbol is in use. If the IPA wishes to maintain an internationally recognized transcription norm, it should not ignore the need but prune the rank growth of individual solutions and provide an officially recognized option.

For that very practical reason we must continue to argue against Daniel Recasens’ wish to keep the bottom line as it is, despite his cogent articulatory and acoustic arguments that the vowel space is more restricted along the open back-to-front dimension. Ultimately, the differences in the auditory domain are binding and those who mailed us were practically unanimous in their acceptance of a need to represent the central open quality. The cross-language comparisons which supply part of the supporting argument for such a symbol can in fact be presented in terms of the oft-cited phonemic basis of the IPA symbol inventory: why should languages with a ‘triangular’ system be forced to select from an inventory which is based on the assumption of a front-back open-vowel opposition? The question must be asked yet more emphatically in a contrastive context, where a ‘triangular’ system may be compared with a ‘quadrilateral’ one.

Admittedly, that argument opens the door to a demand for mid front and mid back vowel symbols since, as Recasens points out, systems with three degrees of opening are badly served by the present four-degree IPA framework. Also, it is unfair to point out (reapplying Recasens’ argument when it suits us, after refuting it when it didn't) that the acoustic, articulatory and (we would stress) the auditory space between [e] and [ɛ] or [o] and [ɔ] are considerably smaller than between [a] and [ɑ]. However, the mid front and back vowels are a separate issue, and before potential additions along the close-to-open dimension are considered the status of [æ] in its present position on the chart would need to be reviewed, since it contravenes the principle of phonemic contrast and, together with [ɐ] overloads the chart at least as much as an additional central-open vowel symbol would.

Martin Ball expressly avoids taking sides on the basic question we asked, but we agree with the stance underlying his discussion of the different options if a symbol for the central open vowel were to be introduced, and we have no problems with the conclusion he comes to.

Having received considerable feedback on our article, much of which supported our call for a central open vowel symbol on practical grounds at least, some of which argued theoretically against its introduction, we would like to conclude with concrete suggestions for the Council to consider and vote on in two stages:

1. 1. The central open vowel quality should be represented on the IPA vowel chart by the [ᴀ] symbol.

If this proposal is not accepted, we suggest the following compromise, which provides both a practical solution and brings the vowel chart closer to the theoretical principles that have been invoked as arguments against the central open vowel in this discussion:

1. 2. The ‘Other Symbols’ category should be expanded to include optional vowel symbols which do not fulfill the phonemic contrast principle and overload the vowel chart but are considered essential in practice, namely:

   1. [ᴀ] for the open central vowel between [ä] and []

   2. [æ] for a vowel quality between [] and []

   3. [ɐ] for a vowel quality between [] and []

This would provide an officially sanctioned symbol for a widely and strongly felt need and would theoretically ‘purify’ the Vowel Chart to some extent.