English questions

[Travis B.]

In this case I think a better way to treat "furry" is as /ˈfɜri/, with /ɜ/ being a vowel phoneme that only exists before /r/ in rhotic varieties and with an intervocalic glide /r/ rather than having a rhotic vowel existing in hiatus.

However, I just heard a US pronunciation of rural as something like [rɜ˞əl] on Episode 4 of Resident Alien; the lack of an intervocalic consonant did surprise me.

I pronounce rural as a monosyllable [ʁˤʁ̩ˤːːɯ̯]; note that this contrasts with earl [(ʔ)ʁ̩ˤ(ː)ɯ̯] not just in vowel length but also because in isolation, as it begins with a vowel, I precede earl with a glottal stop (even though I typically don't transcribe this).

[Nortaneous]

TBH these alternations, aside from the allophony between initial [ə(ː)ʁˤ] and [ʁ̩ˤ] are probably best classed as allomorphy and are not productive. The affected words include your, you're, pure, cure, sure and like.

I assume your and you're have FORCE-NURSE variation due to re-stressing of reduced forms, but maybe there was an earlier form with CURE? It's the same pattern...

I have [ˈtʰokiːo(ː)]~[ˈtʰokjo(ː)] but [kʰoːˈwaːɤ̯ə(ː)]~[kʰəːˈwaːɤ̯ə(ː)].

The vowel in the first syllable of "koala" is longer than the vowel in the first syllable of "Tokyo"?

I didn't give phonetic transcripts because I'm not actually sure what the vowel of "koala" is - it's some kind of actual back vowel, but I'm not sure about its height, or whether there's even a contrast between word-internal /ɨw əw/ - [kʰŭwɑɫə] doesn't sound obviously unreasonable.

This is complicated in the speech here because of intervocalic elision before /ər/ resulting in non-r-colored long or overlong vowels preceding /r/.

But is it obligatory? It seems reasonable to model rapid speech effects as a layer on top of the usual phonetic output - this could also resolve the "Saturday" issue. [sæɾɹ̩deː] is the input to the rapid speech layer, and there, you have ɾ > 0 followed by æɹ̩ > æːɹ. Or something like that.

Syllables provide good reason for effects such as atlas [ˈɛʔɰɘs] versus metro [ˈmɜtʃɻ͡ʁo(ː)]; the former has a glottal stop because /tl/ cannot form an onset outside of some loanwords, whereas the latter does not because /tr/ is a valid onset.

What about football vs. footsie? In both cases, /t/ must be in coda position - neither /tb-/ nor /ts-/ are valid onsets - but t-glottalization is only permissible in football. There's a reasonable mechanical explanation for that - /t/ and /s/ are homorganic, so there's no articulatory pressure toward debuccalization or anything like that - but if mechanical explanations are acceptable for /ts/, why not for /tr/?

(Didn't John Wells or someone decide to analyze /tr dr/ as affricates? This is obviously wrong, but I'm pretty sure I've seen it proposed.)

I have TRAP > FACE before some but not all /ŋ/ (e.g. I have it in banker but not Lancaster). and I have DRESS > FACE in length and strength but not in Genghis or penguin.

I have no idea what vowel length and strength have. Could be any of /æ e ej/. Probably /æ/, at least in length.

I pronounce rural as a monosyllable [ʁˤʁ̩ˤːːɯ̯]; note that this contrasts with earl [(ʔ)ʁ̩ˤ(ː)ɯ̯] not just in vowel length but also because in isolation, as it begins with a vowel, I precede earl with a glottal stop (even though I typically don't transcribe this).

/rɝəl/ and /ɝl/ for me - "rural" is clearly a disyllable.

(In principle I should write /ɨ˞/ instead of /ɝ/, but there's no precomposed character for that, and the rhotic hook is one of IPA's most typographically offensive mistakes.)

[anteallach]

(Didn't John Wells or someone decide to analyze /tr dr/ as affricates? This is obviously wrong, but I'm pretty sure I've seen it proposed.)

Descriptions of traditional RP (e.g. Daniel Jones's) generally treat them as affricates on the phonetic level.

In Wells's paper on his syllabification algorithm, which generally makes a lot of sense to me, he treats them as indivisible units, leading to syllabifications such as petr.ol and cauldr.on, which I find very strange. My conclusion is that my accent must treat them differently from his, and indeed however fast I say matter of fact it's not going to contain the same cluster as petrol. I perceive these clusters as actually being /tʃr/ and /dʒr/, which supports the idea that what he says here may not apply to my speech. But I'd be interested to know if it makes sense to anyone else.

[Qwynegold]

Wiktionary says it's /ˈfɝi/. But phonetically it would be something different?

I would not necessarily trust dictionaries; after all I've seen supposedly reliable dictionaries give cot-caught-merged "American" pronunciations despite the fact that a majority of Americans are cot-caught unmerged.

In this case I think a better way to treat "furry" is as /ˈfɜri/, with /ɜ/ being a vowel phoneme that only exists before /r/ in rhotic varieties and with an intervocalic glide /r/ rather than having a rhotic vowel existing in hiatus.

Okay, thanks! I'll do some changes to my text and then post it here to get it corrected.

[Travis B.]

TBH these alternations, aside from the allophony between initial [ə(ː)ʁˤ] and [ʁ̩ˤ] are probably best classed as allomorphy and are not productive. The affected words include your, you're, pure, cure, sure and like.

I assume your and you're have FORCE-NURSE variation due to re-stressing of reduced forms, but maybe there was an earlier form with CURE? It's the same pattern...

I presume your and you're originally pattered with CURE, but underwent the same vowel lowering that /ur/ did in much of English but which only barely happened here.

I have [ˈtʰokiːo(ː)]~[ˈtʰokjo(ː)] but [kʰoːˈwaːɤ̯ə(ː)]~[kʰəːˈwaːɤ̯ə(ː)].

The vowel in the first syllable of "koala" is longer than the vowel in the first syllable of "Tokyo"?

I didn't give phonetic transcripts because I'm not actually sure what the vowel of "koala" is - it's some kind of actual back vowel, but I'm not sure about its height, or whether there's even a contrast between word-internal /ɨw əw/ - [kʰŭwɑɫə] doesn't sound obviously unreasonable.

For me an unstressed long vowel is about the same in length or a bit shorter than a stressed short vowel.

This is complicated in the speech here because of intervocalic elision before /ər/ resulting in non-r-colored long or overlong vowels preceding /r/.

But is it obligatory? It seems reasonable to model rapid speech effects as a layer on top of the usual phonetic output - this could also resolve the "Saturday" issue. [sæɾɹ̩deː] is the input to the rapid speech layer, and there, you have ɾ > 0 followed by æɹ̩ > æːɹ. Or something like that.

The thing is, in the case of particular words like Saturday and twenty, I perceive the forms with elision as the normal forms, and the elision-less forms as specifically careful.

Syllables provide good reason for effects such as atlas [ˈɛʔɰɘs] versus metro [ˈmɜtʃɻ͡ʁo(ː)]; the former has a glottal stop because /tl/ cannot form an onset outside of some loanwords, whereas the latter does not because /tr/ is a valid onset.

What about football vs. footsie? In both cases, /t/ must be in coda position - neither /tb-/ nor /ts-/ are valid onsets - but t-glottalization is only permissible in football. There's a reasonable mechanical explanation for that - /t/ and /s/ are homorganic, so there's no articulatory pressure toward debuccalization or anything like that - but if mechanical explanations are acceptable for /ts/, why not for /tr/?

I have t-preglottalization but not full t-glottalization in footsie [ˈfʊʔtsi(ː)], but this is also found in coda /ts/ in words such as cats [ˈkʰɛʔts].

(Didn't John Wells or someone decide to analyze /tr dr/ as affricates? This is obviously wrong, but I'm pretty sure I've seen it proposed.)

I would be open to treating them as affricates, with the caveat that they cannot appear in codas.

[Kuchigakatai]

Okay, thanks! I'll do some changes to my text and then post it here to get it corrected.

What text are you talking about here, by the way? Is it a post or page elsewhere?

[Qwynegold]

A text I'm writing that's going to be included in the dictionary for my IAL. So it needs to be written for people with no linguistic awareness. (Not that it's actually going to be read by anyone, but it needs to look legit.)

[ratammer]

In this case I think a better way to treat "furry" is as /ˈfɜri/, with /ɜ/ being a vowel phoneme that only exists before /r/ in rhotic varieties and with an intervocalic glide /r/ rather than having a rhotic vowel existing in hiatus.

However, I just heard a US pronunciation of rural as something like [rɜ˞əl] on Episode 4 of Resident Alien; the lack of an intervocalic consonant did surprise me.

There's a bit in an episode(s?) of 30 Rock about a TV show called "The Rural Juror", which keeps causing everyone pronunciation difficulties and turning into something like "Rurrjurr".

[bradrn]

Recently, I’ve noticed that some people consistently write an historical … where I’d write a historical …. For me, only the second variant is acceptable, as it reflects how I say /ə hɪsˈtɔɹikɫ̩/; where does the first variant come from? (I’d expect it if the initial ⟨h⟩ was silent, but I don’t think I’ve ever heard it pronounced that way.)

[KathTheDragon]

It's a rule in formal (British?) English that h- words take "an" rather than "a", precisely because of historical h-dropping, despite most modern lects having restored h-.

[bradrn]

It's a rule in formal (British?) English that h- words take "an" rather than "a", precisely because of historical h-dropping, despite most modern lects having restored h-.

Huh, I’ve never heard of that rule — thanks for clarifying! I’ve always just written it the way I say it, so an hour and an honour, but a history and a hand.

[Linguoboy]

It's a rule in formal (British?) English that h- words take "an" rather than "a", precisely because of historical h-dropping, despite most modern lects having restored h-.

This is the only word I can think of which has [h] in contemporary usage and yet still follows this obsolete rule. I'd always assumed that it survived due to the stress pattern, the first syllable in historical being very weakly stressed compared to history. I don't really notice whether [h] is present or absent in an historical whereas if you said *an history it would leap out at me.

You occasionally see this in older North American English, where I assume it's a slavish imitation of the British usage (and thus considered pretentious).

[Richard W]

This is the only word I can think of which has [h] in contemporary usage and yet still follows this obsolete rule. I'd always assumed that it survived due to the stress pattern, the first syllable in historical being very weakly stressed compared to history..

The prescriptive rule says the syllable should be unstressed for /h/ to be treated like a vowel. There's also an hotel, though I think the /h/ is dropped in that combination, and in a related collocation, the /h/ can be dropped in at home.

[anteallach]

It's a rule in formal (British?) English that h- words take "an" rather than "a", precisely because of historical h-dropping, despite most modern lects having restored h-.

Huh, I’ve never heard of that rule — thanks for clarifying! I’ve always just written it the way I say it, so an hour and an honour, but a history and a hand.

I think that's the rule that most people, including in BrE, actually follow (see for example the Guardian style guide, which says an before silent h but a before /h/) but some people do write an before a word beginning with an unstressed syllable starting with /h/, with hotel and various derivatives of history being the usual examples.

It's definitely a hotel and a historical novel for me, with [h] and no [n].

[ratammer]

I've heard that the rule is that you use "an" if the h- word came from French, on the logic that "h" is silent in French. So "an hotel" and "an historical" but "a hill". (Prescriptivists, eh?)

[Linguoboy]

I've heard that the rule is that you use "an" if the h- word came from French, on the logic that "h" is silent in French. So "an hotel" and "an historical" but "a hill". (Prescriptivists, eh?)

That can't be right. Historical is an English coinage based on Latin historicus. The corresponding French is historique. (But history is from Old French and was originally /h/-less. So it's complicated.)

[ratammer]

Blame that on whoever came up with the rule...

[Nortaneous]

Has anyone tried to use word truncation to establish rules for English syllabification?

e.g. vulnerability > vuln implies /vʌln.ər-/ rather than /vʌl.nər-/, but captain > cap (Wikipedia alleges that this is real, at least) implies /kæp.tən/, etc.

(but also coronavirus > rona implies that maybe it's actually /vʌln.ə.r-/, which is a little weird because it's not pronounced like that)

Seems like that'd be a source for natural data, but I haven't seen anything on it (tho I also haven't gone looking) - and it'd be a little tricky to deal with loans between dialects with different syllabification rules ("rona" seems a little weird to me)

[zompist]

Has anyone tried to use word truncation to establish rules for English syllabification?

e.g. vulnerability > vuln implies /vʌln.ər-/ rather than /vʌl.nər-/, but captain > cap (Wikipedia alleges that this is real, at least) implies /kæp.tən/, etc.

In general abbreviations should fit English phonotactics, but they might not tell us about the syllabification of the original word.

Abbreviations may be modified for recognizability. E.g. merchandise > merch. It's quite possible that "merch" is the first syllable. But "mer" would be a much worse abbreviation, being much less guessable. Another example might be ultimate > ult. I think the syllabification is ul-tim-ate, but "ul" as an abbreviation would seem really weird.

Abbrevations can also follow other patterns, e.g obviously > obvs, association > soccer.

[anteallach]

Has anyone tried to use word truncation to establish rules for English syllabification?

e.g. vulnerability > vuln implies /vʌln.ər-/ rather than /vʌl.nər-/, but captain > cap (Wikipedia alleges that this is real, at least) implies /kæp.tən/, etc.

(but also coronavirus > rona implies that maybe it's actually /vʌln.ə.r-/, which is a little weird because it's not pronounced like that)

Seems like that'd be a source for natural data, but I haven't seen anything on it (tho I also haven't gone looking) - and it'd be a little tricky to deal with loans between dialects with different syllabification rules ("rona" seems a little weird to me)

For me it's /ˈvʌlnrəbəl/ and /vʌlnrəˈbɪlɪtɪ/, with the second syllable you're showing gone.

Wells's rules would give co.ron.a.vir.us, which fits with rona (and makes sense to me), but capt.ain. For Zompist's examples, they'd give merch.an.dise and ult.im.ate.