Can you please clarify whether the phonemic symbols /ɪ/ and /i/ can be used interchangeably to represent the sound at the end of such words as coffee, taxi, happy, easy, ready?

  • 2
    Fair question. Don't know why you got the downvotes!? Mar 26 '18 at 15:14
  • 2
    @curiousdannii Well things get sticky with vowels don't they? So, for example a a typical realisation of SSBE /e/ could be represented as either a lowered cardinal vowel 2, or a raised 3 (so either [ e̞ ] or [ ɛ̝ ] ). And it gets more problematic with phonemic representations, because these are language specific and the choice between one symbol and another may be affected by many things, for example the orthography used in the target language. So one load of people might use /e/ for the English SET vowel and others / ɛ/. Of course once we choose a symbol, we need to stick with it. ... Mar 26 '18 at 16:37
  • 2
    @curiousdannii, IPA is unique and unambiguous only in that a given letter has a single articulatory description. The mapping from phonetic fact to articulatory description is non-unique: articulatory descriptions are partially overlapping ranges. Also, "a sound" is not really a countable unit.
    – user6726
    Mar 26 '18 at 17:01
  • 2
    Phonemic symbols are arbitrary and vary from language to language and from linguist to linguist. Consequently they can represent anything at all. In English, for instance, /e/ represents [ei] in some transcriptions and [ɛ] in others.
    – jlawler
    Mar 26 '18 at 19:43
  • 1
    @jlawler Yes, exactly so. We could use numbers or smileys or anything to represent a phoneme. This should be unsurprising because, of course, an individual phoneme may have several completely different phonetic realisations. Mar 27 '18 at 8:04

Yes and no.

There are some, mostly older, speakers of Southern Standard British English (SSBE or RP) who consistently use [ɪ] at the ends of the words happy and so forth. However, a majority of younger speakers use something closer to [ i̞ ]. And other speakers use a vowel where the difference between the two is neutralised.

In older transcriptions of SSBE, /ɪ/ was consistently used to transcribe the vowel at the end of the word happy in phonemic transcriptions, and this reflected the two facts that a) the sound most speakers used at that time was [ɪ] and b) they considered this vowel to be the same phoneme that they used in the word KIT. However, as SSBE English evolved and the pronunciation changed in the manner described in the paragraph above, people started to use the symbol /i/ in broad, phonemic transcriptions to represent this vowel.

This instance of the use of /i/ in SSBE broad transcriptions was an innovation used because for some speakers the vowel is the same phoneme as the KIT vowel and for others it is the same phoneme as the FLEECE vowel. Notice that in traditional transcriptions of SSBE, the FLEECE vowel is represented by the symbol /i:/, which uses two dots after the i. The /i/ symbol therefore represents a vowel which may be /ɪ/ for some speakers and /i:/ for others, and indistinguishable between the two for others still.

Some SSBE dictionaries and other resources still use /ɪ/ for HAPPY vowels, but most now use /i/. So, in the sense that you may decide to fall in with one group or the other by using either /ɪ/ or /i/, you can make that choice. However, of course once you have made that choice you need to stick with it.

As John Wells remarked in one of his blogs about /i/ - 'it seemed like a good idea at the time!'


To quote the Wikipedia article on ɪ,

Sometimes, especially in broad transcription, this vowel is transcribed with a simpler symbol ⟨i⟩, which technically represents the close front unrounded vowel.

The question is really, "how specific do you want to be". In a language where there is no contrast between these two sounds, you would typically only use /i/ but in some contexts, you might want to consider the difference as marked, and thus highlight the contrast, even though the language doesn't generally see the two sounds as phonologically distinct.

Wiktionary lists British pronunciation of coffee as /ˈkɔː.fɪ/ as an alternative with the caveats (conservative RP, dated) and prefers a final /i/ for both British and American English. To my untreaned eye and ear, an /ɪ/ pronunciation would vaguely suggest an Australian accent, not a completely free (i.e. unmarked) variation in English.

  • Yes. With the proviso that you can't ever be completely specific in any alphabetic/phonemic system, because it's digital and sound is continuous. Emic units are statistical generalizations of actual speech that are specific enough to be useful in some context, though usually not in all.
    – jlawler
    Mar 27 '18 at 15:25

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.