People without the "horse"/"hoarse" merger?

[skullturf]

I'd like to think that I have a reasonably good ear for accents, and I have some familiarity with the various mergers that are present in some regional varieties of English and absent in others.

I was born and raised in Western Canada, so I personally grew up with the "cot"/"caught" merger and the "Mary"/"marry"/"merry" merger. I also grew up without the "law"/"lore" merger or the "pen"/"pin" merger, but I've certainly heard many people who do have either of those mergers.

I have read that historically, "horse" and "hoarse" had different vowel sounds. I believe a substantial majority of English speakers in both North America and England pronounce "horse" and "hoarse" identically, but there still exist some speakers who maintain the distinction. See, for example, this link.

My question is, can anybody help me find a recording of a native English speaker who pronounces "horse" and "hoarse" differently? I've always been really curious to encounter this distinction "in the wild", but I'm not consciously aware of ever having heard it.

[eSOANEM]

I don't have a recording of it, but one of my maths teachers (Northern Irish) didn't have the for/four merger which is described in that article as the same (I never heard him say horse or hoarse) he pronounced them, as I recall something like:

for = fɔ(ɹ) (possibly an r, can't remember if his accent was rhotic)
four = foə(ɹ) (possibly an r, can't remember if his accent was rhotic)

This seems consistent with the article you linked which said hiberno English was non-merging.

Also, when I (modern RP) speak slowly and deliberately, the [ɔː] in my hoarse starts to centralise a bit towards the end, half way between that and [ɔə̯] but this change is very slight and only when I speak very carefully (and so my accent moves towards more traditional RP which, as the article says, doesn't have a total merge) and I can't be sure that this isn't me trying to make a slight difference (although it feels like I'm not forcing any particular pronunciation).

[skullturf]

Interesting. Perhaps it's time for me to search through YouTube clips of (say) Liam Neeson, and listen closely to the way he says words with an "or" or "oar" sound. (Of course, actors aren't always the best examples, because they sometimes consciously or unconsciously depart from their native accents, but that might be someplace to start.)

Edit: I'm at work now, and generally don't turn on the sound on my computer, but I may look into this later.

[eSOANEM]

Liam Neeson, and listen closely to the way he says words with an "or" or "oar" sound. (Of course, actors aren't always the best examples...

Yeah, Liam Neeson doesn't seem like a great example to me, he sounds 100% American to me (as a Brit so he probably doesn't actually have as American accent as it sounds to me, but he certainly doesn't sound at all Irish) in every clip I've ever seen of him (even out-of-character) interviews. Political speeches might be a better source of natural Irish accents but are still likely to be available on teh intertubes.

[SheffJames]

To confirm, my gf is from the same town as Liam Neeson and she doesn't have the merger.

[iChef]

I'm interested in this as well. In my area of the US we pronounce horse/hoarse/whores all the same, but I've heard English accents that do not.

[eSOANEM]

Whores would end in a /z/ for me whilst the others end in an /s/. There may also be a difference in vowel length between whores and horse/hoarse (with whores being a tad longer on the vowel). Horse/hoarse are merged for me (RP) though.

[gmalivuk]

In my area of the US we pronounce horse/hoarse/whores all the same, but I've heard English accents that do not.

Yeah, I'm not aware of any place that would pronounce "whores" with /s/ instead of /z/ at the end.

[Derek]

It would be a very strange phenomenon indeed: +/z/ is the underlying plural morpheme for regular plural nouns, with a +/s/ allophone* after voiceless consonants and a +/ɨz/ allophone after sibilants. Pronouncing "whores" with a final /s/ would imply that the entire plural morpheme has shifted to +/s/, at least after all consonants, which is an unlikely change.

I suspect he just made a mistake.


*I think I'm using allophone wrong here, but I'm not sure what the right word for a variation of a morpheme that doesn't change it's meaning is, so work with me.

[eSOANEM]

I'd probably have said that the -s morpheme is realised as /s/ after voiceless consonants. The morpheme is -s not a phoneme (because they represent completely different things, -s is a piece of syntax (possibly meaning as well) whilst /s/ is a piece of phonology with no meaning) so really you want to talk about how the morpheme relates to the phonemes and words like pronounced like, realised as etc. are probably more appropriate.

[Makri]

/s/ is not a piece of phonetics. It's a piece of phonology.

The morphological analogue of an allophone is an allomorph; in this case, you would say there are phonologically conditioned allomorphs. (At least the /ɪz/ presumably is one; the [s] case could be considered an allophone of /z/ here.)

The morpheme is, strictly speaking, {plural}. That's the notation that nit-picky linguists use. -/z/ is one allomorph of it, -/ɪz/ is another, and possibly -/s/ is a third. (Actually, some people would go so far as to say they're all only one morpheme /s/ and the rest is allophony. There are issues with the concept of a phoneme.) So phoneme sequences are morphs of morphemes, just as sound sequences are phones of phonemes. (You could invent a system where the morph is a pair of an attachment site specification and a phoneme sequence, of course, so the plural might be <r,/s/> or something. Don't know if anyone does that explicitly, but it's more adequate to our use of -/.../ or /.../-.) For some people, the plural morpheme {plural} has other allomorphs, too; lexically conditioned ones that somehow make you use a different version of the noun stem. Others would like to say that there are several plural morphemes. It's a bit of a conceptual mess when you try to be a hundred percent precise.

However, mostly you don't need that and everyone will understand you anyway, so most people write the plural morpheme as -/s/ because it's the most frequent realization; and in German, they will also talk about distinct plural morphemes being used in different noun classes (-/ər/, -/ən/, -/ə/, -/s/) when they all mean the same.

This slight terminological mess just comes from the fact that people have found the structuralist notions defective for theoretical purposes, but still use them as a convenient descriptive tool to get the facts across. So just be cooperative language users when you talk and hear about morphology.

[eSOANEM]

/s/ is not a piece of phonetics. It's a piece of phonology.

*facepalm* always get those two the wrong way round. Fixed.