The "How do You Pronounce X" Thread

[Zaarin]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I don't drop it, but I do debuccalize it in that'd (consistently) and it'd (often): [ðæʔ(ə)d] [ɪʔəd~ɪɾəd]; I consistently flap the t in that'll and it'll.

[Imralu]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I really must hear a recording of you speaking sometime. I can't even imagine understanding what you're saying.

Nah, I don't drop the /t/, I just flap it. Didn't you at some point say you quite often drop what would be an intervocalic flapped /t/ or /d/? This shouldn't be all that surprising then.

[Imralu]

Pol isn't really a word I'd use. Politician is COT, so...

Never talked about Pol Pot?

Sure, but that has the GOAT vowel, homophonic with pole.

Ah. I just thought of the word "poll" though. I have
/pɒl/ [pʰɔɫ]~[pʰɔʊ̯] - poll, Pol (Pot)
/pəʊl/ [pʰɔʊ̯ɫ]~[pʰɔːʊ̯] - Pole, pole
/pɔːl/ [poːɫ]~[poːʊ̯] - Paul, pall (of smoke)

Do you pronounce asphalt with an ass or ash at the beginning?

It's not a word I use often, since IMD this is largely covered by "pavement". But when I do say it, it's with [s].

And that means the driving surface, not the walking surface?

"Pavement" covers all manner of paved surfaces--asphalt, concrete, pavers, cobblestones, etc. If you say "I tripped on the curb and hit my head on the pavement", I won't immediately know if you were stepping into the street or onto the sidewalk.

Ah, personally, I don't use "pavement" at all but I know Brits use that for what most Americans call the sidewalk and what Australians call the footpath. And I wouldn't call any of those surfaces you listed "paved" except for one with pavers ... a road with asphalt (which we usually refer to as bitumen), we call a "sealed road", not a paved road or anything like that.

And there are some Australians who have the funny thing that the Queen has with "off" and "cloth" (pronouncing them with the CAUGHT vowel)

Ah, thanks for explaining the eye-dialect spelling "orf" to me at long last!

Most Australians I know have a marginal phoneme pronounced like a long version of the LOT vowel which only appears in the word "gone" and interjections such as "awww" (cuteness, disappointment) and "oh" (but only in a quick "oh no", not in "oh, I see"). If I'm using RP as my base for phonemic writing I call it /ɒː/ but its value is actually approximately [ɔː]. In any case, some people pronounce "gone" with the THOUGHT vowel and in eye-dialect that's "gorn". I have one Australian friend who pronounces "gone" with the LOT vowel /ɒ/ [ɔ], rhyming with "on", but both of her parents are British and she also doesn't seem to distinguish between /æ/ and /æː/ and after realising both of these things, I told her "I have more vowels than you" and she was like "you can keep them!" She also pronounces "us" with /z/, so she's not quite a typical Australian.

[Sumelic]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I really must hear a recording of you speaking sometime. I can't even imagine understanding what you're saying.

Nah, I don't drop the /t/, I just flap it. Didn't you at some point say you quite often drop what would be an intervocalic flapped /t/ or /d/? This shouldn't be all that surprising then.

So you can't say "It'd be good" with three syllables? I do this even though I don't normally drop intervocalic /t/ or /d/. I think it's more of an assimilation of /td/ to /d/.

Actually, that made me think of a question for Travis: how do you pronounce "had to"? I have /dt/ here, but it seems conceivable to me that someone might be able to simplify it to just /t/ or /d/. Actually, maybe I can have /d/ in running speech: I just looked up "hadda" and one of the examples "Tell all the guys he had somethin' come up, hadda go outta town for a while and take care of it" sounds pretty normal to me.

[Imralu]

So you can't say "It'd be good" with three syllables? I do this even though I don't normally drop intervocalic /t/ or /d/. I think it's more of an assimilation of /td/ to /d/.

Ah, yes, now that you ask that specifically ... but if I do that, the /d/ also assimilates to the /b/ and it ends up with [bː] I think.

[Travis B.]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I really must hear a recording of you speaking sometime. I can't even imagine understanding what you're saying.

The problem is I find it very hard to speak naturally when recorded or over the phone or like - I normally speak much closer to GA in such contexts than when I speak naturally.

Nah, I don't drop the /t/, I just flap it. Didn't you at some point say you quite often drop what would be an intervocalic flapped /t/ or /d/? This shouldn't be all that surprising then.

Yes, that is one way of seeing it, but another way is cluster reduction of /td/ in the case of that'd and it'd. If that is cluster reduction though it must be phonemically frozen, since the vowel is long, and /td/ predicts a short vowel.

[Travis B.]

Actually, that made me think of a question for Travis: how do you pronounce "had to"? I have /dt/ here, but it seems conceivable to me that someone might be able to simplify it to just /t/ or /d/. Actually, maybe I can have /d/ in running speech: I just looked up "hadda" and one of the examples "Tell all the guys he had somethin' come up, hadda go outta town for a while and take care of it" sounds pretty normal to me.

I pronounce had to with a geminate [tː] or sometimes [ttʰ] but underlyingly it is /dt/, since the preceding vowel is long, no preglottalization takes place, and it may aspirate.

[Sumelic]

So you can't say "It'd be good" with three syllables? I do this even though I don't normally drop intervocalic /t/ or /d/. I think it's more of an assimilation of /td/ to /d/.

Ah, yes, now that you ask that specifically ... but if I do that, the /d/ also assimilates to the /b/ and it ends up with [bː] I think.

Actually, that made me think of a question for Travis: how do you pronounce "had to"? I have /dt/ here, but it seems conceivable to me that someone might be able to simplify it to just /t/ or /d/. Actually, maybe I can have /d/ in running speech: I just looked up "hadda" and one of the examples "Tell all the guys he had somethin' come up, hadda go outta town for a while and take care of it" sounds pretty normal to me.

I pronounce had to with a geminate [tː] or sometimes [ttʰ] but underlyingly it is /dt/, since the preceding vowel is long, no preglottalization takes place, and it may aspirate.

This sounds the same phonologically as in my accent, then... I wrote "it'd" as /ɪd/ and "had to" as /hædtə/ but of course phonetically the processes of place and voicing assimilation would affect these /d/s as usual in my pronunciation, so "it'd be" would be something like [ɪbbi] and "had to" would be something like [hæːd̥tə] or [hæːd̥tʰə].

[Travis B.]

I have generalized place of articulation assimilation of /t d/ before plosives and full assimilation of /d/ before /v/ and various sonorants. Note that coda /d/ is already voiceless for me. However, vowel length and preglottalization contrasts are preserved. As a result I could analyze, say, vodka as /ˈvɑgkə/, corresponding to [ˈvaːkːə] or sometimes [ˈvaːkkʰə]. And yes, I readily pronounce it'd be as [ˈɘːpˈpi(ː)].

[Travis B.]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I really must hear a recording of you speaking sometime. I can't even imagine understanding what you're saying.

I do not think it is actually that weird, at least in this case, though; it's just initial ð-stopping combined with prevocalic and postvocalic devoicing of lenis plosives combined with preglottalization/glottalization of postvocalic fortis plosives combined with the NCVS combined with l-vocalization combined with allophonic vowel length.

Weird is when I do things like pronounce sister as [ˈsɘɕtɕʁ̩(ː)] IMO. (Yes, that is particularly weird; people even around here do not normally do this.) (Properly this is probably an example of me analogically spreading a dialectal phonological rule further than it is actually applicable; e.g. [ˈʁʷɜɕtɕʁ̩ːˌʁɑ̃ʔ] for restaurant and [ˈfɛʔktɕʁ̩ːʁi(ː)] for factory are typical dialectal pronunciations here, but I spread the rule from unstressed /tərV/ where /t/ is not intervocalic to all cases of unstressed /tər/ where /t/ is not intervocalic as a kid, and then lost said innovation in all but a few words.)

[Qxentio]

Does anyone else consistently drop the /t/ in that'd, that'll, it'd, and it'll? I have [tɛːt tɛːɤ̯ ɘːt ɘːɯ̯] respectively, and should note that that and it are, in contrast, [tɛʔ ɘʔ].

I really must hear a recording of you speaking sometime. I can't even imagine understanding what you're saying.

sister as [ˈsɘɕtɕʁ̩(ː)]
[ˈʁʷɜɕtɕʁ̩ːˌʁɑ̃ʔ] for restaurant
[ˈfɛʔktɕʁ̩ːʁi(ː)] for factory

Maybe I am missing something here, but you're not actually pronouncing you <r>s as voiced uvular fricatives, are you?

[Travis B.]

Maybe I am missing something here, but you're not actually pronouncing you <r>s as voiced uvular fricatives, are you?

They are uvular approximants. And yes, it may seem weird to think of English varieties as having them, but another way of seeing it is as a bunched /r/ where there is no coronal articulation (except after another coronal).

Of course there is little audible difference from any other English /r/ except when a bit of epiglottalization is added (but I normally pronounce it without epiglottalization).

[opipik]

recipe

[Travis B.]

recipe: [ˈʁʷɜsəˌpʰi(ː)]

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

bonus vs. boners
dace vs. days
Joyce vs. joys
dice vs. dies
house (n) vs. house (v)

I want to prove my friend plural (n) and 3rd person present (v) marker is /z/, not /s/. I would also like to see how does pre-lenis lengthening in case of diphthongs (Travis),

[alynnidalar]

Well, the plural marker is /s/ after unvoiced consonants, and /z/ after voiced consonants and vowels.

e.g. <cats> /kæts/ vs. <cads> /kædz/

[linguoboy]

bonus vs. boners
dace vs. days
Joyce vs. joys
dice vs. dies
house (n) vs. house (v)

I want to prove my friend plural (n) and 3rd person present (v) marker is /z/, not /s/. I would also like to see how does pre-lenis lengthening in case of diphthongs (Travis),

I'm not really sure how showing them our transcriptions would help with that. Are you saying you'd like us to record our pronunciations so you could give them acoustic evidence?

[Imralu]

bonus vs. boners
dace vs. days
Joyce vs. joys
dice vs. dies
house (n) vs. house (v)

[ˈbɐʉ̯nəs] [ˈbɐʉ̯nəz]
[dæɪ̯s] [dæˑɪ̯z]
[d͡ʒoɪ̯s] [d͡ʒoˑɪ̯z]
[dɑe̯s] [dɑˑe̯z]
[hæɯ̞̯̈s] [hæˑɯ̞̯̈z]

I'm not in the habit of talking about daces ([ˈdæɪ̯səz]) though, but I'm confident of how I'd pronounce it.

If your friend does not trust the standard descriptions of English phonology, I don't know why he's going to trust some random Australian dude on the internet. Especially one who thinks he pronounces [æɯ̞̯̈] for the standard /aʊ/ diphthong whereas the standard description of that sound in Australian is [æo̯]. Like, I've just noticed that I round the end of that diphthong in "how" and "now" but not in "house" or "how've", so the consonant seems to affect it.

What I can say, is that I'm sometimes thrown off hearing Germans speaking English when they otherwise have goot* pronunciation but fail to voice final consonants. For me, that makes, for example, "surfers" become "surface" and "servers" become "service".

In any case, after doing this, I am really starting to become sure of my theory that allophonic vowel lengthening before voiced consonants is something that's only present in my speech in vowels that do not have a phonemic length distinction. Like, it definitely happens with all of these diphthongs, and with /iː/ and /uː/ ... which don't have short counterpart phonemes, but it doesn't appear in anything like "bus" vs "buzz" because pronouncing the latter with too long a vowel would turn it into "bars". (Maybe there'd be some allophonic length distinction if I had my speech scientifically analysed, but it's not distinct enough for me to pick up on.)

*Holy crap! That was a completely unintentional typo. I sometimes wonder what's going on in my head sometimes because I fairly often make "devoicing" type typos, which is just really weird when you think about what that means ... and funny that it was relevant to the topic here.

[Travis B.]

bonus: [ˈpõːnɘs]
boners: [ˈpõːnʁ̩ːsʲ] (when followed by a vowel, [ˈpõːnʁ̩ːzʲ])
dace: [tes]
days: [teːs] (when followed by a vowel, [teːz])
Joyce: [tʃɔɪ̯s]
joys: [tʃɔːɪ̯s] (when followed by a vowel, [tʃɔːɪ̯z])
dice: [təe̯s]
dies: [taːe̯s] (when followed by a vowel, [taːe̯z])
house (n.): [hʌo̯s]
house (v.): [hɑːɔ̯s] (when followed by a vowel, [hɑːɔ̯z])

Underlyingly, I have /s/ and /z/ where just about every native English-speaker has them; it is just that I devoice final /z/ more frequently than not.

[Travis B.]

The thing about Germans that gets me is not the devoicing but rather the lack of proper vowel length allophony - except in environments where final devoicing does not take place. But, of course, it seems like my number one complaint about non-native speakers of English in general is they don't do vowel length allophony properly, and they don't get the proper distribution of realizations of fortis and lenis obstruents right - even though over time I have gotten used to people who don't do vowel length allophony and who just use tenuis and voiced plosives without proper allophony.

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

I'm not really sure how showing them our transcriptions would help with that. Are you saying you'd like us to record our pronunciations so you could give them acoustic evidence?

If your friend does not trust the standard descriptions of English phonology, I don't know why he's going to trust some random Australian dude on the internet. Especially one who thinks he pronounces [æɯ̞̯̈] for the standard /aʊ/ diphthong whereas the standard description of that sound in Australian is [æo̯].

I told her about this forum and then about the fact -s is /z/ after voiced/lenis consonants and vowels, so she wanted an example from probably real and trustworthy English native speakers.

Also, I wonder if there's a dialect or maybe an idiolect where people make a distinction between e.g. "dace" and "days" with different onsets of the diphthongs (applying something similar to Canadian raising to other diphthongs).

[Imralu]

Also, I wonder if there's a dialect or maybe an idiolect where people make a distinction between e.g. "dace" and "days" with different onsets of the diphthongs (applying something similar to Canadian raising to other diphthongs).

dice: [təe̯s]
dies: [taːe̯s] (when followed by a vowel, [taːe̯z])
house (n.): [hʌo̯s]
house (v.): [hɑːɔ̯s] (when followed by a vowel, [hɑːɔ̯z])

Emphasis mine.

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

Also, I wonder if there's a dialect or maybe an idiolect where people make a distinction between e.g. "dace" and "days" with different onsets of the diphthongs (applying something similar to Canadian raising to other diphthongs).

dice: [təe̯s]
dies: [taːe̯s] (when followed by a vowel, [taːe̯z])
house (n.): [hʌo̯s]
house (v.): [hɑːɔ̯s] (when followed by a vowel, [hɑːɔ̯z])

Emphasis mine.

I saw it immediately after submitting my post, but thanks .

[Travis B.]

Also, for me, the following are often more like:

dace: [te̞s]
days: [tɛ̝ːs] (when followed by a vowel, [tɛ̝ːz])

(Most of my vowels are more open when long than when short, but it is more noticeable with /eɪ/ than the other vowels that are normally monophthongs.)

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

days: (when followed by a vowel, [tɛ̝ːz])
boners: [ˈpõːnʁ̩ːsʲ] (when followed by a vowel, [ˈpõːnʁ̩ːzʲ])
days: [teːs] (when followed by a vowel, [teːz])
joys: [tʃɔːɪ̯s] (when followed by a vowel, [tʃɔːɪ̯z])
dies: [taːe̯s] (when followed by a vowel, [taːe̯z])
house (v.): [hɑːɔ̯s] (when followed by a vowel, [hɑːɔ̯z])

What about semivowels?