[Lingtyp] contrast between [ɪ] and [e]

Sun Jul 13 03:42:32 UTC 2025

It seems everyone is trying to force the languages into a rigid phoneme structure, but languages show variation, and even something that can be shown to form a minimal pair can in other words be variable. For example, in Tagalog there are a few words that show a contrast between /i/ and /e/, such as iwan ‘being left behind’ and ewan ‘not knowing’, but in many words the two are interchangeable, e.g. puti ~ pote ‘whiteness’.

Some people now call typology “diversity linguistics”, which is very good, but we also need to pay attention to the diversity language-internally. Weinreich, Labov, and Herzog (1968) discuss how problematic the Structuralist
conception of language as homogeneous is, and argue instead for a view of language as
heterogeneous, and argue that “command of heterogeneous structures is not a matter of
multidialectism or “mere” performance, but is part of unilingual linguistic competence”
(Weinreich et al. 1968: 101).
Hockett (1967[1977]:256), towards the end of his career, said, “Beyond the design implied by the factors and mechanisms that we have discussed, a language has no design. The search for an exact determinate formal system by which a language can be precisely characterized is a wild goose chase, because a language neither is nor reflects any such system. A language is not, as Saussure thought, a system ‘où tout se tient’. Rather, the apt phrase is Sapir’s ‘all grammars leak’.”

Randy

On Sun, Jul 13, 2025 at 1:26 AM, Alex Francois via Lingtyp <[lingtyp at listserv.linguistlist.org](mailto:On Sun, Jul 13, 2025 at 1:26 AM, Alex Francois via Lingtyp <<a href=)> wrote:

> Dear Christian, dear all,
>
> Among the 17 (Oceanic, Austronesian) languages of northern Vanuatu, one has kept the 5-vowel system of Proto Oceanic {i e a o u}, but the other 16 have increased their inventories, to at least 7 vowel phonemes, and up to 13 or 14 for some languages (François 2005a).
>
> Several languages, like Mwotlap, have 7 vowels, which I analyse like this:
> i u
> ɪ ʊ
> ɛ ɔ
> a
>
> Usually, N Vanuatu languages do not contrast /ɪ/ with /e/, so admittedly someone could propose that the 7-vowel system above could "simply" be a case of cardinal vowels like this:
> i u
> e o
> ɛ ɔ
> a
> And indeed, this analysis is proposed by Malau (2016) for Vurës, one of the languages of the area.
>
> Personally I don't mind, but I'm not convinced that the second analysis would be more simple or economical than the first one. Also, one reason why I analyse the second-degree vowels as /ɪ ʊ/ is a phenomenon (rare in the Pacific) of ATR vowel harmony in Mwotlap , whereby a noun [ i pl u ] 'companion', when taking certain suffixes, involve the lowering of both vowels [ ɪ pl ʊ n] 'his companion', with what looks like some form of ATR vowel harmony  (François 2005b:117) .
> This is one of the reasons why I have been analysing the second degree vowels as /ɪ ʊ/ in that region.
>
> Yet I can sympathise with the hesitation reported by Larry about Bantu languages, on the difficulty to choose between the /ɪ ʊ/ and the /e o/ analyses.
>
> While some systems of northern Vanuatu are richer in vowel phonemes, they usually show the same contrast between /ɪ/ and /ɛ/ (rather than between /ɪ/ and /e/). See for example Lemerig and Mwerlap  (counting diphthongs among vowel phonemes):
> [image.png]
> (vowel charts from François 2011: 195)
> _______
> Now, one special case I encountered was the language of Hiw (Torres islands).
> While its neighbours show systems such as those above, Hiw shows a contrast, unusual in the region, between /ɪ/ and /e/:
> [image.png]
>
> From the phonetic point of view, the system is quite asymmetrical, with the presence of mid-open [ɔ] among back vowels, but no mid-open [ɛ] in the front.
> One might insist that, from some abstract perspective, /i ɪ e a/ is really similar to (or reanalysable as) /i e ɛ a/ (?); but I don't see what we would gain from forcing a symmetric structure onto a vowel system that is simply not symmetrical. When I studied Hiw, I was impressed that [ɛ] never came up as a phone, as /e/ was always pronounced quite high. Thus the verb 'go', which is [vɛn] in neighbouring Lo-Toga, is always realised [ven] in Hiw, with a higher vowel.
>
> My acoustic impressions were confirmed as I calculated the average F1 and F2 for the 9 vowels of Hiw:
>
> [image.png]
>
> As you may imagine from the formant chart, distinguishing between the two phonemes /ɪ/ and /e/ was one of the challenges I faced when learning the language.
> Good speakers insist that they do form minimal pairs, e.g. /ven/ 'go' vs. /vɪn/ 'go up, climb'.
> Some younger speakers also found it difficult to hear the difference between /ɪ/ and /e/, or to reproduce it clearly; but my main teacher Jacob was able to distinguish them with clarity.  (One of the lessons he gave me can be heard  [here](https://doi.org/10.24397/pangloss-0002834) .)
>
> In sum, while there may be hesitation, for other North Vanuatu languages, between analysing the second-degree vowels as /ɪ ʊ/ or /e o/, at least for Hiw there is no such ambiguity that /ɪ/ exists as a phoneme – since it contrasts both with /i/ and with /e/.
> best
> Alex
> __________
>
> PS: References:
>
> - François, Alexandre. [2005a](http://alex.francois.online.fr/AFpub_articles_e.htm#2005b). Unraveling the history of the vowels of seventeen northern Vanuatu languages. Oceanic Linguistics 44 (2): 443-504.
> - —— 2005b. A typological overview of Mwotlap. Linguistic Typology 9-1: 115-146. [→[pp.117-118](https://marama.huma-num.fr/data/AlexFrancois_2005_LingTyp_Mwotlap.pdf#page=3)]
> - —— 2011. Social ecology and language history in the northern Vanuatu linkage: A tale of divergence and convergence. Journal of Historical Linguistics 1 (2). 175-246. [→[pp.194-195](https://marama.huma-num.fr/data/AlexFrancois_2011_JHL1-2_Social-ecology_Vanuatu.pdf#page=20)]
>
> ---------------------------------------------------------------
>
> Alex François
>
> [LaTTiCe](http://www.lattice.cnrs.fr/en/alexandre-francois/) — [CNRS](https://www.cnrs.fr/en)  [—](https://www.cnrs.fr/en)  [ENS](https://www.ens.fr/laboratoire/lattice-langues-textes-traitements-informatiques-et-cognition-umr-8094) – [PSL](https://www.psl.eu/en)  —  [Sorbonne nouvelle](http://www.sorbonne-nouvelle.fr/lattice-langues-textes-traitements-informatiques-cognition-umr-8094-3458.kjsp) [ENS](http://www.sorbonne-nouvelle.fr/lattice-langues-textes-traitements-informatiques-cognition-umr-8094-3458.kjsp) [ENS](http://www.sorbonne-nouvelle.fr/lattice-langues-textes-traitements-informatiques-cognition-umr-8094-3458.kjsp)
>
> [Australian National University](https://researchportalplus.anu.edu.au/en/persons/alex-francois)
> [Personal homepage](http://alex.francois.online.fr/)
>
> ___________________ ___________________ ___
>
> ---------- Forwarded message ---------
> From: Christian Lehmann via Lingtyp <lingtyp at listserv.linguistlist.org>
> Date: Sat, 12 Jul 2025 at 17:28
> Subject: Re: [Lingtyp] contrast between [ɪ] and [e]
> To: < lingtyp at listserv.linguistlist.org>
>
> Thanks to everybody for your helpful hints. I will certainly countercheck my phonetic ear.
> As for my question #2, I will heed Larry's hint "(15c) is the expected 7V system worldwide". Indeed, [e] and [o] figure among the cardinal vowels, while [ɪ] and [ʊ] do not.
> @ Ian and Martin: I assume that the general principles you mention imply an answer to one of my questions. Would you mind giving me a clue in this respect?
>
> --
>
> Prof. em. Dr. Christian Lehmann
> Rudolfstr. 4
> 99092 Erfurt
> Deutschland
>
> Tel.:	+49/361/2113417
> E-Post:	christianw_lehmann at arcor.de
> Web:	https://www.christianlehmann.eu
> _______________________________________________
> Lingtyp mailing list
> Lingtyp at listserv.linguistlist.org
> https://listserv.linguistlist.org/cgi-bin/mailman/listinfo/lingtyp
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250713/70636d02/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 23041 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250713/70636d02/attachment-0003.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 112195 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250713/70636d02/attachment-0004.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image.png
Type: image/png
Size: 11209 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250713/70636d02/attachment-0005.png>