The "How do You Pronounce X" Thread

[Vijay]

Where in the South are you familiar with?

Austin!

[----]

That's interesting. Perhaps just use the normal velarisation and pharyngealisation marks? If it is pharyngealisation in codas.

But is it pharyngealization, that's the question!

[Zaarin]

Where in the South are you familiar with?

Austin!

Yeah, a Texas accent is a completely different animal than a Deep South accent. I don't mind a Texas or Upper South accent; Deep South accents sort of make me want to tear my ears off. :p

[vokzhen]

That's interesting. Perhaps just use the normal velarisation and pharyngealisation marks? If it is pharyngealisation in codas.

But is it pharyngealization, that's the question!

Is it a pharyngealized lateral in the onset and a coarticulated coronal-pharyngeal in the coda? Mine is either uvular or upper pharyngeal, and onset uvularization feels lighter/secondary while coda uvularization feels like it's the primary POA with light, incidental coronal contact.

[Travis B.]

Here /l/ is [ʟ̞] in stressed or initial onsets except after another consonant in an onset where it is [ɰ w] depending on environment and any of [ɯ̞ ʊ ɤ o] depending on environment anywhere else... in careful speech. But in everyday speech it tends to only be [ʟ̞] when geminate (including across word boundaries) and else is things like [ɰ w] in all onsets except that initially it is sometimes something like [ɰʟ̞ wʟ̞].

[----]

That's interesting. Perhaps just use the normal velarisation and pharyngealisation marks? If it is pharyngealisation in codas.

But is it pharyngealization, that's the question!

Is it a pharyngealized lateral in the onset and a coarticulated coronal-pharyngeal in the coda? Mine is either uvular or upper pharyngeal, and onset uvularization feels lighter/secondary while coda uvularization feels like it's the primary POA with light, incidental coronal contact.

My instinct/impression is something like this but with such a complex phone I'm hesitant to make any particularly specific descriptions without actually measuring it. That is, I would agree that in the onset the alveolar contact is primary, while in the coda (and often intervocalically) it's secondary.

[Io]

Where in the South are you familiar with? In my experience there are, broadly speaking, four Southern accents: Texas/Oklahoma, Louisiana*, Upper South (Virginia, Maryland, Tennessee, North Carolina, etc.), and Deep South (Georgia, South Carolina, Alabama, Mississippi, chiefly). What I was describing is absolutely a feature of Deep South and lower registers of Upper South, but not of Texas or Louisiana or of higher registers of Upper South.

*By which I mean non-Cajun Louisiana, which is its own accent...

What about FL then?

[Zaarin]

Where in the South are you familiar with? In my experience there are, broadly speaking, four Southern accents: Texas/Oklahoma, Louisiana*, Upper South (Virginia, Maryland, Tennessee, North Carolina, etc.), and Deep South (Georgia, South Carolina, Alabama, Mississippi, chiefly). What I was describing is absolutely a feature of Deep South and lower registers of Upper South, but not of Texas or Louisiana or of higher registers of Upper South.

*By which I mean non-Cajun Louisiana, which is its own accent...

What about FL then?

7/10 Floridians are from New York, Michigan, or Ontario; another 2/10 are still from somewhere other than Florida. The few native Floridians I know have an accent that's closer to what you'd find in the American Heartland with a few Southernisms mixed in. (Citation: I currently live in Florida.)

[Vlürch]

I just returned from a trip to Louisville, Kentucky, and now I'm wondering if I'll sound pretentious if I pronounce Louisville as natives do, something like /luː.ə.vɚ/ or /luːvɚ/

Wait, are Louisville and Louvre homophones in that dialect?

The only time I've ever heard Americans (or maybe even English-speakers in general) say "Louvre," I swear it sounded exactly the same as "loo."

Un jour dans l'avenir, parce que tous les cabinets sera occupé en un cycle incessant par des immigrants de l'Inde que auront appris les secrets des toilettes, et ainsi comme une solution les Français auront commencé à déféqué sur le sol quand ils auraient dû construit plus de toilettes, en les salles magnifiques du Louvre, la mer de la merde aura inondé à partout les étages.

Sorry ._.

More on topic, how would you guys pronounce "orificialisation"? Something like [oɹɪfɪʃəliseɪ̯ʃən] or [oɹɪfisəlɪʃəɪ̯zən] or...?

[spanick]

I've literally never heard or read this word before but on first reading I'd say /ɔɹɨfɪʃəlɨzeɪ̯ʃən/.

[Travis B.]

I am used to [ʟ̞uːf~ɰuːf] for Louvre.

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

As I can see, many of you, English-speaking people, have at least one of the following: flapping of alveolar stops (with lengthening of the preceding vowel in case of /d/), debuccalization of voiceless stops and affricate (I mean both glottal reinforcement and glottal replacement) and partial or full devoicing of voiced stops. I used to just accept that, but once there appeared a thought: "What if my full voicing of voiced stops and not making use of glottalization and flapping is considered in Anglophone world a very big mistake?".
So, are these three phenomena so common that my pronunciation would be a mistake or not?

[finlay]

Not a mistake - many people don't have these, or don't always have them - but you run the risk of sounding like a newscaster from the 50s.

People don't always devoice their voiced stops, but everyone aspirates their voiceless stops, so if you don't do that, you sound French or Dutch.

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

Not a mistake - many people don't have these, or don't always have them - but you run the risk of sounding like a newscaster from the 50s.

Didn't know it sounds so official and artificial!

People don't always devoice their voiced stops, but everyone aspirates their voiceless stops, so if you don't do that, you sound French or Dutch.

(Scottish?)
The only useful thing school taught me about the pronunciation. No need to mention again.

So, anyone else?

[linguoboy]

So, anyone else?

What finlay said. If your vowels are American but you don't flap, I'll know immediately you're foreign. Even non-native speakers of American English do this. I'm less sensitised to debuccalisation.

[Travis B.]

I assume when you speak of aspirating fortis obstruents, you mean initially or at the start of stressed syllables...

[Travis B.]

So, anyone else?

What finlay said. If your vowels are American but you don't flap, I'll know immediately you're foreign. Even non-native speakers of American English do this. I'm less sensitised to debuccalisation.

Another thing I hear, which I do especially a lot (even though I am rather odd about this), is simply eliding unstressed intervocalic /t d n nt nd/ at every odd distance from a stress (aside for effect on vowel length and nasalization).

[ˈd̪ʲɛ.gɔ kɾuˑl̪]

Thanks for current posts. I'm still worried about glottalization and devoicing since I have difficulty with hearing glottal stop and producing partially voiced consonants. Could native speakers tell more about it?

[linguoboy]

Another thing I hear, which I do especially a lot (even though I am rather odd about this), is simply eliding unstressed intervocalic /t d n nt nd/ at every odd distance from a stress (aside for effect on vowel length and nasalization).

I've seen the claim before that [ɾ] can be freely deleted in casual speech. I'm not convinced that it's always possible, but it's hard to self-test perception in this area.

[Travis B.]

Another thing I hear, which I do especially a lot (even though I am rather odd about this), is simply eliding unstressed intervocalic /t d n nt nd/ at every odd distance from a stress (aside for effect on vowel length and nasalization).

I've seen the claim before that [ɾ] can be freely deleted in casual speech. I'm not convinced that it's always possible, but it's hard to self-test perception in this area.

What I notice that seems to indicate that this really is not me being weird is that no one from around here has trouble understanding me even when I quite aggressively elide in this fashion.

[Sumelic]

Thanks for current posts. I'm still worried about glottalization and devoicing since I have difficulty with hearing glottal stop and producing partially voiced consonants. Could native speakers tell more about it?

I don't think partially voiced consonants are really necessary. It's certainly fine to use fully-voiced consonants intervocalically or in other situations where the consonant is surrounded by voiced segments on both sides (e.g. the "g" of "pilgrim"). It's also fine word-initially--even native speakers have variability here, and not all of them devoice consonants that much in this position. Word-finally, I think devoicing is more common, but not devoicing shouldn't hurt understandability as long as you correctly apply the English rules of vowel-length allophony: use longer allophones before voiced consonants, and shorter allophones before voiceless consonants. The only issue I think that might arise from not devoicing or glottalizing coda consonants as much as a native speaker is that native English speakers might hear the release of the consonant as a schwa, perhaps.

[Travis B.]

The pattern I am used to is basically as follows:

Initial/stressed onset fortis obstruents not preceded by sibilants: aspirated voiceless
Stressed onset fortis obstruents preceded by sibilants: unaspirated voiceless
Initial/stressed onset lenis obstruents: unaspirated voiceless
Unstressed onset fortis obstruents: unaspirated voiceless
Unstressed onset lenis obstruents: unaspirated voiceless
Unstressed intervocalic fortis obstruents: unaspirated voiceless, or flapped or elided if /t/
Unstressed intervocalic lenis obstruents: voiced or half-voiced, occasionally unaspirated voiceless, or flapped or elided if /d/
Coda fortis obstruents: preglottalized unaspirated voiceless, or simply glottalized if /t/
Coda lenis obstruents: unaspirated voiceless

This is all combined with vowel length allophony, which from what I am used to is such that sonorants are transparent to it, and vowels default to long.

Also, /nt nd/ behave like single consonants intervocalically, and /t d/ assimilate to following plosives, being preserved only with regard to vowel length and preglottalization.

Furthermore, if a word is followed by another word in the same utterance which begins with a vowel, a final plosive behaves like an unstressed intervocalic plosive.

One note is that /dʒ/ is always [tʃ], even when unstressed intervocalic, where it is distinguished from /tʃ/ by vowel length alone.

I would take all this with a grain of salt; this is just how I am used to people here speaking.

[Imralu]

"What if my full voicing of voiced stops and not making use of glottalization and flapping is considered in Anglophone world a very big mistake?"

Certainly not a mistake, just a foreign accent and they're lovely. If your consonants sound a bit unusual, it's OK. And for what it's worth, the things that usually sound the most foreign to me in terms of stops are when people don't aspirate voiceless stops in the right places, or when people don't make a distinction between final voiced and voiceless stops (e.g. pick, pig). Also, a dental /t/ and /d/ gives it away for me too. Fully voicing your voiced stops is fine and as far as I can tell, what I usually do. Don't go too far with devoicing. I used to teach a lot of Korean students and when some of them said my name, Ben, it sounded like Pen to me. For me, final voiceless stops are simply unreleased without a following vowel and as far as I can tell, not strongly glottalised, if at all. There are so many possibilities in English - even ejectives as final consonants so if your accent is overall good (vowels. intonation, rhythm, aspiration all correct) then I doubt that a lack of flapping or glottalisation will be the giveaway... maybe if your vowels are really American.

[Imralu]

One note is that /dʒ/ is always [tʃ], even when unstressed intervocalic, where it is distinguished from /tʃ/ by vowel length alone.

I would take all this with a grain of salt; this is just how I am used to people here speaking.

Yeah, I don't think a lot of that about devoicing applies to AusE, certainly not that bit that I've just quoted about /dʒ/. I wonder if it has anything to do with the fact that we have a lot of phonemic length contrasts in our vowels (five long and short vowel pairs, although one has no minimal pairs, and an additional sixth pair which is not exact but pretty close) which leaves less room for vowel length to tell us about consonants. Post-tonic, intervocalic /t/ and /d/ are not distinguished except in pedantically clear speech. There's a pub in a Paddington in Brisbane called the Paddington Tavern and everyone calls it the "Paddo", and before I went there and saw the sign, I had always thought it was the "Patto". I never thought about it being connected to "Paddington" - it's pronounced exactly like the nickname of a guy in my high school, Patto, whose last name was Patterson, so in my head it was a /t/. Latter and ladder are the same. An Aussie friend of mine recently said he didn't know if a short tune was a "ditty" or a "diddy" because he's never seen it written.

In fact, thinking about it now, I can think of a contrast between "beady" as in eyes and "Beattie" as in "Peter", former premier of Queensland... there's a strong allophonic difference between a short vowel in feet and a long vowel in feed, and this is interesting because /iː/ does not have a phonemically short partner (unless you count the HAPPY vowel, but its shortness seems to me to simply be attributable to lack of stress) meaning that my hypothesis about phonemic vowel length prohibiting consonant-conditioned allophonic vowel length may have something to it.

Of course, this is all just from self-observation and I'd like to see it backed up by research. Possibly I was pronouncing the /æ/ in Paddo minutely shorter when I thought it was Patto but no one ever noticed ... and that length distinction is completely overwhelmed by the distinction between short /æ/ and long /æː/, which we distinguish in "Bernard Fanning is fanning the bad lad."

Most description of AusE phonology seems centred on our vowels and when consonant are mentioned, there are a few topics such as our non-rhoticness with a vowel-conditioned epenthetic /r/, our /t/ flapping or our always-dark and sometimes vocalised /l/, preference for /dʒ/ or /ʒ/ in certain words, blending of /j/ with preceding consonants ("There was Jew on the grass on Chewsday morning) and very occasional /h/ dropping but exact values of some consonants, where they differ from other dialects, is something I haven't really seen. I'd be curious to see if my thoughts on our stops are substantiated. I've also never seen it really mentioned that a strong AusE accent pronounces "stupid" and "stew" etc. with an initial [ʃt͡ʃ] (in milder accents, it is [st͡ʃ]) and that word final schwa is greatly lowered towards [ä], which is blocked by anything following it.

[Travis B.]

"What if my full voicing of voiced stops and not making use of glottalization and flapping is considered in Anglophone world a very big mistake?"

I used to have a very hard time with accents that do not roughly fit the phonology I outlined above. A lack of aspiration would wildly throw me off, and a lack of proper vowel length allophony would greatly confuse me, especially without proper preglottalization or non-release. I basically would not hear voiceless-voiced contrasts except in unstressed intervocalic positions. I basically needed aspiration and vowel length allophony to understand people, and preglottalization helped reinforce that. I did not need strong vowel length allophony, and non-release could take the place of preglottalization. So hence while I could understand English English fine, Indian English was essentially unintelligible to me.

However, this has changed since I started working where I work now, where I work with many Indian people. Now I can understand Indian English okay, largely because I have learned to understand distinguishing fortis and lenis pairs by voicing alone and to ignore vowel length.