The question was prompted by trying to find an English analogue to the many words in Welsh that end with /ɒ/ (today's example: crwydro = wander). This sound is after all common in English at the beginning or in the middle of a word (ox, not).

I can't think of any common word in reasonably standard English that ends this way. As a Londoner originally I'm prone to t-glottalisation, which leads to something pretty close. Words spelt –o generally seem to end with the highly variable (so I haven't tried to give IPA) ɢᴏᴀᴛ vowel.

I'm approaching this from a Southern British perspective, but other common accents/dialects are also interesting. The original version of this question was confused by using IPA examples that were based on pronouncing cot to match caught but didn't say so explicitly.

So do any English words end with this sound?

  • 2
    The word law (as well as claw, floor, and numerous others) ends with /ɔ/, at least in some British dialects. I assume you're worried about /ɒ/. Oct 9 '18 at 14:18
  • @PeterShor you may well be right, I can never remember my IPA vowels so have to look them up and may have hit an American source
    – Chris H
    Oct 9 '18 at 14:21
  • 1
    Looking online, the British vowels are currently moving around, law going from /ɔ/ to /o/, and pot going from /ɒ/ to /ɔ/, so it seems you're not actually wrong. But using /ɔ/ to represent this vowel is confusing. Oct 9 '18 at 14:26
  • I would say no. /ɒ/ and /ɔː/ are often considered to be the checked and unchecked counterparts of each other (like /ɪ/ and /iː/). As a checked vowel, /ɒ/ can occur only in closed syllables, so you’ll never get a word that ends in it. There are lots of words that end in /ɔː/, but (and this relates to what @Peter wrote too) /ɔː/ is always realised long in English, whereas Welsh /ɔ/ is short. English has no short [ɔ] at all, so no words to end in it either. English /ɔ/ is also further back than Welsh /ɔ/, but that’s perhaps going into too much phonetic detail. Oct 9 '18 at 14:27
  • 1
    @tchrist That’s why I specifically said that /ɔː/ (I always write the length marker in the phonemes for consistency and maximum cross-dialect clarity) is always realised long. It may not be long phonemically, but it is invariably long phonetically, which the Welsh vowel isn’t. If we ignore that, then yes, there are many words that end in this sound; but acoustically, Welsh /ɔ/ and English /ɔː/ are very different. The final vowel in saw, etc., is acoustically as far from that in crwydro as the final vowel in know is from that in Spanish no. Oct 9 '18 at 14:36

In general, English words do not end with any of the stressed "short" vowel sounds (/ɒ/, /æ/, /ɛ/, /ʌ/, /ɪ/, /ʊ/). This is not an absolutely exceptionless rule: interjections may not follow it (for example, I have /æ/ in "yeah" and /ʌ/ in "duh"), and I don't find it particularly difficult to pronounce nonsense words ending in stressed /ɪ/, for example.

In American English, historical "short o" has been merged into the originally "long" vowel sound /ɑ/, so the original restriction on the distribution of the "short o" sound no longer applies, at least not on the surface level (the word spa ends in the same sound /ɑ/ that is used for the "short o" sound in pod, so pod and "spa'd" rhyme).

Furthermore, as far as I know there is no dialect of English where the sound /ɒ/ is in common use in fully unstressed open syllables. (Unlike /ɪ/ and /ʊ/, which some accents use in words like ready or gradual.)

So I don't think you'll be able to find any word ending in /ɒ/.

The symbol /ɔ/ is most commonly used in transcriptions of English to represent a vowel sound distinct from the sound of "short o". It is also transcribed /ɔː/ (with the IPA length marker "ː") in the context of British English to indicate that it functions as a "long vowel" in the British English vowel system. This vowel sound does occur word-finally, in various words spelled with -aw (law, claw, raw, straw), and in "non-rhotic" accents also in words spelled with -oar, -ore such as roar, more, tore, bore (and some words spelled with -oor such as door and floor).

The British English /ɒ/ phoneme ("short o") may be realized as the IPA phonetic vowel [ɔ], and the British English /ɔː/ phoneme may be realized as the IPA phonetic vowel [oː], but purely phonetic transcriptions are not very commonly encountered, particularly not when discussing restrictions on the distribution of sounds in a language's sound system. (Contrariwise, in American English the phoneme transcribed /ɔ/ may be realized phonetically as something like [ɒ]—even in accents where it is not merged with the open unrounded /ɑ/ sound, it often is relatively close to it phonetically.) This mainly comes up as an issue when people are trying to compare vowel sounds between different languages: for example, comparing English "o" sounds to those of Italian, French, or German.

  • Duh I’ll grant you, but surely your yeah is realized with a centralizing diphthong finish in [jæə̯], isn’t it? Mine certainly is.
    – tchrist
    Oct 9 '18 at 14:34
  • @tchrist: Different speakers evidently perceive different phonemes in the word yeah. I've seen it transcribed as /jɛ/ and as /jɛə/, in addition to the transcription that you propose and my transcription of /jæ/. To me, I think there isn't a phonemic distinction between /æ/ and /æə/, although [æə̯] might be used in my speech as a realization of the /æ/ phoneme.
    – herisson
    Oct 9 '18 at 14:38
  • @sumelic: possibly /æ/, /ɛ/, and all those other variants are merged in English when they're at the end of a word. Of course, we have so few examples that maybe it's impossible to tell. Oct 9 '18 at 14:47
  • What you’re describing seems similar to how phonetic off-glides in (esp. word-final) closed vowels like phonemic /e/ and /o/ are sometimes perceived by ESL learners as phonetic diphthongs [eɪ] and [oʊ]. But native speakers think of them as simple vowels. I bet that that’s what you are doing here with your /jæ/ with its two phonemes: the [jæə̯] version is just an automatic compensation that (some? many?) native speakers make without thinking of it as a separate phoneme. Just as you will find no minimal pair in English for /o/ vs /oʊ/, I bet you will also find none for /æ/ vs /æə/
    – tchrist
    Oct 9 '18 at 14:49
  • 2
    @PeterShor No, when tchrist writes /o/, he’s referring to the phoneme found in the word know as usually transcribed for AmE. Shoring and show ring would be phonetically /ˈʃɔrɪŋ/ and /ˈʃo.ˌrɪŋ/ in AmE. Phonemic /o/ is phonetically [oʊ] (or something along those lines—the details differ, of course) in AmE. Oct 9 '18 at 15:00

Yes, a few do, but not very many. Common examples are saw, law, claw, draw, flaw, jaw, raw, thaw, or a crow’s caw.

This is because for the most part, /ɔ/ patterns like a “checked” vowel (meaning a lax vowel like the ones that also prototypically occur in DRESS, KIT, HAM, PUT) in that it doesn’t like to end a syllable without a consonant or glide following it. So words like soft and loft and coughed, or sawyer and lawyer, are more likely than just plain saw and law type words.

Normally a word-final tense vowel like phonemic /o/ that isn’t reduced will take a terminal glide phonetically, so an extra /w/ and sometimes written phonetically as [oʊ].

Sometimes words spelled with ‹a› are pronounced /ɔ/, as in one regional pronunciation of the stressed syllable in the city of Chicago [ʃɨ ˈkʰɔ goʷ], or even grandma in the eye-dialect spelling of grandmaw.

This ends up being reflected in spellings like fellow. Notice though how when that gets reduced in the unstressed position, it gets spelled fella, reflecting that it has become an open back unrounded /ɑ/, phonetically reduced further even to schwa.

The other kind of reduction we see in words adapted into English from other languages where they had originally ended in /o/ is for them to go to /u/. This happens in words like buckaroo from Spanish vaquero or vindaloo probably from Portuguese vin(ho) d’alho, or even in lasso from Spanish lazo but pronounced /ˈlæsu/ with /u/ in English rather than /ˈlɑso/ as though it were Spanish.

  • It seems like I got my IPA in a muddle when I first wrote the question, because the distinction between the examples in your first paragraph and the sound that prompted the question is clear to my (untrained) ear, but not in the question
    – Chris H
    Oct 9 '18 at 14:32

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.