Sum: symbols for unaspirated

Alexandre Arkhipov sarkipo at RAMBLER.RU
Wed Apr 9 21:49:16 UTC 2008

Dear colleagues,

I would like to thank all those who replied to my query, both on-list and off-list (in timeline order of responses): 
Greville Corbett, Andrew Spencer, Alice Harris, Geoffrey Haig, and Martin Haspelmath via LINGTYP;
Terry J. Klokeid, Joseph F. Foster, Jakob Dempsey, Nora Wiedenmann, Bruce Moren, Peter T. Daniels, Michael Job, John E. McLaughlin, Geoff Hooker, Paul D. Fallon, Mark J. Jones, Steven B. Chin, Mark A. Mandel, Elizabeth J. Pyatt, and James L. Fidelholtz via LINGUIST List.

I. Choice of symbol 'in principle'

Although several possibilities for marking unaspirated have been proposed, the most familiar and widespread of them seems to be the superscript 'equals' sign. It is used e.g. in Spencer (1996), cited in Pullum & Ladusaw (1996), being traced back to Gleason (1961) and Wells (1982). The superscript 'equals' is currently used, among others, in studies of Southern Wakashan languages; it is recognized by the IPA and is listed in the Extended IPA charts for disordered speech ( ; Handbook 1999: p. 190, symbol 680).

The other known options include: 
-- tilde above, U+0303. Used in the comparative Daghestanian dictionary by Kibrik & Kodzasov (1990), and also e.g. in the grammar of Lezgian (Haspelmath 1993). This was our starting point, but finally the tilde had to be dismissed as (potentially) a *third* diacritic on top of a consonant (together with macron and hacek/caron).
-- the (right upper) corner, U+02FA, for 'no audible release'.
-- using voiced (b,d,g...) as unaspirated, contrasting with voiceless (p,t,k,...) as aspirated. Not possible in our case since we also have voiced series. 
-- voiced series + 'devoicing' ring below (b,d) or above (g) -- U+0325 and U+030A correspondingly. Problems with uvulars as we don't have the voiced uvular stop or affricate.
-- variations of apostrophe/quotation mark placed before, after or above the character.
-- colon. Confusion with length.
-- underline
-- superscript 'minus', as reported by Joseph F. Foster. To this one we will return below.

All in all, we accept the superscript 'equals' as the most established means.

II. Technical issues

...However, the practical realization of our decision turned out to be far from straightforward. Having to use a good deal of diacritics, we stick to the recent SIL fonts which have all the imaginable stuff and even more, namely Charis SIL and Doulos SIL. These are Unicode fonts with lots of excellent features. Unfortunately, in both of them the unaspirated 'equals'  (U+02ED 'modifier letter unaspirated') is *enormous* in width -- just unusable inline. (This character is not to be confused with U+033F, double overline, which is still bigger). One more analogous symbol is U+207C 'superscript equals sign' from the maths section, but it is spaced and as well much too wide for a tiny modifier. 
So in fact, there is no possibility to have the symbol of acceptable dimensions in our font without some per-character tricks with font properties. We will try to contact people from SIL who are making these fonts, and maybe we'll find a way to settle this question, but this is the present state of affairs.

As a temporary technical solution, we will thus be using the superscript minus, U+207B, which is reported to also have been used for unaspirated.

III. Khinalug alphabet(s)

The symbol in question will be used in our project of documentation of Khinalug, a North Caucasian language. We're using (alternatively, for different purposes) a 'scientific' orthography (with superscript equals/minus), an IPA narrow transcription (with aspiration marked instead), and a 'public' alphabet (to be finalized during this summer's fieldwork). The 'public' alphabet making extensive use of digraphs (vs. diacritics), unaspirated are rendered with double consonants.
Those of you who are interested are invited to visit the project's website at for details. Pages will be updated as available.

Kind regards to all,
Alexandre Arkhipov


Ball, MJ. 1988. The contribution of speec pathology to the development of phonetic description. In MJ Ball (Ed.), Theoretical linguistics and disordered language (pp. 168-188). San Diego CA: College-Hill Press.
Ball, MJ. 1991. Recent developments in the transcription of nonnormal speech. Journal of Communication Disorders, 24, 59-78.
Ball MJ, Rahilly J. 2002. Transcribing disordered speech: the segmental and prosodic layers. Clinical Linguistics & Phonetics, 16, 329-344.
Duckwork M, Allen G, Hardcastle W, Ball M. (1990). Extensions to the International Phonetic Alphabet for the transcription of atypical speech. Clinical Linguistics & Phonetics, 4, 273-280.
Gleason, Henry A., Jr. 1961. An introduction to descriptive linguistics. New York: Holt, Rinehart and Winston.
Handbook 1999: Handbook of the International Phonetic Association: A Guide to the Use of the International Phonetic Alphabet. Cambridge: CUP, 1999.
Haspelmath, Martin. 1993. A grammar of Lezgian. (Mouton grammar library; 9). Berlin & New York: Mouton de Gruyter.
Pullum, Geoffrey K., and William A. Ladusaw.  1996.  Phonetic Symbol Guide.  Second edition.  Chicago:  University of Chicago Press.
Spencer, A. 1996. Phonology. Theory and description. Oxford: Blackwell.
Wells, J. C. 1982. Accents of English. Cambridge: Cambridge University Press.

(Cross-posted to LinguistList)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the Lingtyp mailing list