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

[Mark Donohue]

Hello Christian, and all,

What a nice question.
Interrogating a range of languages, the evidence for an /e/ ≠ /ɪ/ contrast
is robust, but highly restrained.
Out of 8600 (ish) languages/dialects, the figures for front, non-low,
unrounded contrasts (namely, languages with all or any of ɛ e ɪ i i̝)[1]
are as follows (not controlling for area or genus):

1 Ø 63
2 ɛ 28
3 e 28
4 ɛ e 12
5 ɪ 41
6 ɛ ɪ 110
7 e ɪ 5
8 ɛ e ɪ 11
9 i 904
10 ɛ i 4723
11 e i 302
12 ɛ e i 1857
13 ɪ i 12
14 ɛ ɪ i 186
15 e ɪ i 20
16 ɛ e ɪ i 347
17 ɛ i i̝ 7
18 ɛ ɪ i i̝ 1

The systems of interest are definitely 16, possibly 15, and perhaps 7 and
8. 7 and 8 are suspect, since without an actual i, it's moot as to
what ɪ represents.
(System 10 is perhaps more accurately listed as ɛ~e i.)

Now the fun begins when we map out systems 16 and 15 (and 12):
[image: Vowel systems.png]
Green = 12, Yellow = 15, Red = 16, grey = other.

Clearly very geographically skewed, very much a west-of-the-Old World
phenomenon. The numbers are (n = 8657):

ɛ e i ɛ ɪ i ɛ e ɪ i other
Europe 38% 3% 3% 56%
Mainland SE Asia 42% 1% 2% 55%
Rest of Eurasia 15% 1% 1% 83%
Northern Africa 0% 0% 0% 100%
Sudan-Sahel Africa 43% 3% 19% 34%
Southern Africa 28% 5% 2% 65%
Pacific 11% 1% 0% 88%
Australia 0% 0% 0% 100%
Americas 5% 0% 0% 95%

So, as Larry wrote, this is not trivial, and in ~5% of languages it is
certainly contrastive, and it's no wonder an Africanist with a broad
perspective on how things work should point this out.

-Mark

[1] Thus ignoring æ, which isn't really justified, given variation between æ
and ɛ. Sorry, Lemerig.


On Sat, 12 Jul 2025 at 20:53, Christian Lehmann via Lingtyp <
lingtyp at listserv.linguistlist.org> wrote:

> Here is a trivial little problem for the phoneticians and phonologists:
>
> IPA (
> https://www.internationalphoneticalphabet.org/ipa-sounds/ipa-chart-with-sounds/#ipachartstart)
> says that [ɪ] and [e] contrast in two features, height and
> frontness/backness.
>
> Being a speaker of a language whose phonetic transcription has involved
> both of the above symbols for generations of phoneticians, and the symbols
> represent different phonemes, I have always taken this for granted.
> However, this pair of phones does not constitute clean minimal pairs in
> German because [e] is long, [ɪ] is short.
>
> Describing now the Cabecar phonetics and phonology, there is a front
> mid-high (IPA says 'near-close' or 'close-mid') vowel phoneme which
> contrasts with both /i/ and /ɛ/, and there is no length. Chibchanist
> tradition transcribes it by [ɪ]. (There is an analogous configuration for
> /u/, /ʊ/ and /ɔ/.) I have two innocent questions here:
>
>    1. Do [ɪ] and [e] actually sound differently? If I click them on the
>    IPA webpage indicated, they sound identical to my ears. Same if I stretch
>    the [ɪ] in my own pronunciation of *bitte*.
>    2. Even supposing that these are two different phones, should the
>    (Cabecar) phoneme covering them not be taken to be /e/, rather than /ɪ/
>    (and likewise for /o/ rather than /ʊ/)?
>
> My (less innocent) suspicion is (but please correct me) that transcribing
> German words like *bitte* with [ɪ] instead of the [e] of *bete* is due to
> a phonological or even orthographic bias.
>
> Curiously, if you ask Google "Is there a phonological contrast between [ɪ]
> and [e]?", its KI cheats you, adducing English examples spelled with <e>
> which represents an [ɛ].
> --
>
> 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/20250716/1ac6f9df/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Vowel systems.png
Type: image/png
Size: 484731 bytes
Desc: not available
URL: <http://listserv.linguistlist.org/pipermail/lingtyp/attachments/20250716/1ac6f9df/attachment-0001.png>