[ðə ‘printsɛs] – prince’s or princess?

Moderator: JackFrost

User avatar
Woods
Posts: 275
Joined: 2007-11-14, 12:43
Gender: male
Location: София
Country: BG Bulgaria (България)

[ðə ‘printsɛs] – prince’s or princess?

Postby Woods » 2017-01-29, 13:51

As I was reading and whispering the words to myself, I got to wonder about the phonology of those two words declined in genitive:

the Prince’s
the princess

Is there any difference? If you hear a story out loud and there are both a prince and a princess involved, and you do not quite know who does a certain thing, can you distinguish from the way the words are pronounced?


As we are on the topic of English phonology, I’m going to ask one more thing I’ve wondered about – how do you pronounce analysis vs. analyses? My main reference dictionary (Cambridge Advanced Learner’s) only shows the phonology for the singular, so I can’t check if there’s any difference. On the other hand, Oxford English Dictionary 1989 suggests that the singular is (əˈnælɪsɪs), whereas for the plural (-iːz) is indicated. But I’m not entirely sure they’re right and up-to-date, so I would like to have a confirmation – do I have to lengthen the last vowel and finish with [-z] instead of [-s] for the plural?

It’ll be interesting to hear both speakers of British and American English, as mine is somewhere in between, on the one hand, and on the other – on the two sides of the ocean things might be different.


 (en) Please correct all mistakes and imperfections you can spot in the way I use English or any other language.
 (fr) Merci de me corriger à chaque fois que vous sentez que ce que j’écris ne sonne pas au mieux ou ne vous paraît pas parfaitement français.
 (dk) I bedes venligst rette alle fejl og ufuldkommenheder I får øje på i det jeg skriver.

User avatar
dEhiN
Posts: 3512
Joined: 2013-08-18, 2:51
Real Name: David
Gender: male
Location: Toronto
Country: CA Canada (Canada)
Contact:

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby dEhiN » 2017-01-29, 14:10

Woods wrote:As I was reading and whispering the words to myself, I got to wonder about the phonology of those two words declined in genitive:

the Prince’s
the princess

I say [pʰɹɪnsɘ/əz] for "prince's", [pʰɹɪnsɛs] for "princess", and [pʰɹɪnsɛsɘ/əz] for "princess'". (I think I alternate between /ɘ/ and /ə/ depending on whether it's regular speech or individual articulation.)

Woods wrote:As we are on the topic of English phonology, I’m going to ask one more thing I’ve wondered about – how do you pronounce analysis vs. analyses? My main reference dictionary (Cambridge Advanced Learner’s) only shows the phonology for the singular, so I can’t check if there’s any difference. On the other hand, Oxford English Dictionary 1989 suggests that the singular is (əˈnælɪsɪs), whereas for the plural (-iːz) is indicated. But I’m not entirely sure they’re right and up-to-date, so I would like to have a confirmation – do I have to lengthen the last vowel and finish with [-z] instead of [-s] for the plural?

Yeah the OED 1989 is correct. The plural takes [-iːz]. Also, the first "a" I pronounce as /æ/ like the second one.
Follow my TAC 2017 here.

(N)  (en-CA) | (B2)  (fr) | (B1)  (pt-BR) | (A2)  (es-CO) | (A1)  (ja) (ko) (sv) (ta-LK)
(A0)  (de) (fy) (haw) (hi) (hu) (id) (it) (oc) (oj) (pl) (ro) (sq) (tl) (tr) (zh)

User avatar
linguoboy
Posts: 19327
Joined: 2009-08-25, 15:11
Real Name: Da
Location: Chicago
Country: US United States (United States)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby linguoboy » 2017-01-29, 16:59

Woods wrote:Is there any difference? If you hear a story out loud and there are both a prince and a princess involved, and you do not quite know who does a certain thing, can you distinguish from the way the words are pronounced?

Absolutely. They are distinguished by vowel quality, consonant voicing, and stress. (Double ss virtually always indicates /s/ in English while single s can be either /s/ or /z/. Even speakers which don't put primary stress on the final syllable of princess stress it more than they do the final syllable of prince's/princes.)

I feel like a lot of English instruction does a terrible job of teaching vowel reduction. It seems fairly non-intuitive for a lot of learners but you really can't speak intelligible English without it.
"Richmond is a real scholar; Owen just learns languages because he can't bear not to know what other people are saying."--Margaret Lattimore on her two sons

User avatar
Woods
Posts: 275
Joined: 2007-11-14, 12:43
Gender: male
Location: София
Country: BG Bulgaria (България)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby Woods » 2017-01-29, 23:07

linguoboy wrote:Even speakers which don't put primary stress on the final syllable…

This is indeed the first time I realise it is an option.

Is it common to pronounce it this way in America? The only way I’ve ever heard it or used it is with the stress on the first syllable. But I think I’ll start saying [prin’tsɛs] from now on!


linguoboy wrote:They are distinguished by vowel quality

How exactly? I mean if the stress falls on the same syllable, otherwise my question is no longer valid.


linguoboy wrote: …consonant voicing…

So, prince’s is [‘printsəz]?

How about the prince’s castle? Does the ‘s remain [z]?

The prince’s absence. – I guess here we’ll have a [z], but not if the next word starts with a consonant. What do you think? I’ve never thought of whether ‘s would be pronounced [s] or [z]. If I try to come up with some other examples:

the house’s roof = [ðə ’hawsɪs ru:f]
the house’s artificial appearance = [ðə ’hawsɪz ’artɪfiʃəl əp’iərənts]

Am I right or just not used to discerning the [z]?

And, to finish with the houses, we could add:

the houses’ roofs
the houses’ artificial appearance

I guess that would be:

[ðə ’hawzɪs ru:fs]
[ðə ’hawzɪz ’artɪfiʃəl əp’iərənts]

I’m pretty sure I’ve gotten it wrong up until here, so I hope you will fix my phonology a little bit.

Indeed, I don't think I pronounce the genitive suffix as [z] very often, if ever – I just say [s] almost all the time. So the question is, where should I replace the [s]'s with [z]'s instead. The strange thing is, natives say I speak great English and nobody points these things out to me, and then all of a sudden I'm surprised to find I haven't spoken right. Maybe they can't hear the difference either?


Linguoboy, I’ve also got one question about your grammar:

Even speakers which don't put primary stress on the final syllable of princess stress it more than they do the final syllable of prince's/princes.

Can you really use which in this sentence? I would either use who or that, otherwise it sounds to me like we’re talking about an object. Sorry for deviating into your grammar, which is without a doubt much better than mine, but that also grabbed my attention.


I’ll try to read the article about vowel reduction tomorrow or asap – I’ve got to get some sleep now. Many thanks for the explanations!


dEhiN, honestly I have no idea what the difference between ə and ɘ is. If you have time, you can explain :)

User avatar
linguoboy
Posts: 19327
Joined: 2009-08-25, 15:11
Real Name: Da
Location: Chicago
Country: US United States (United States)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby linguoboy » 2017-01-30, 0:40

Woods wrote:
linguoboy wrote:Even speakers which don't put primary stress on the final syllable…

This is indeed the first time I realise that is an option.

Is it common to pronounce it this way in America? The only way I’ve ever heard it or used it is with the stress on the first syllable. But I think I’ll start saying [prin’tsɛs] from now on!

I'm not sure how common it is. I know I've heard it and I know it doesn't strike me as particularly noteworthy to pronounce it in this way.

Woods wrote:
linguoboy wrote:They are distinguished by vowel quality

How exactly? I mean if the stress falls on the same syllable, otherwise my question is no longer valid.

As dEhiN points out, only princess has /ɛ/.

Woods wrote:
linguoboy wrote: …consonant voicing…

So, prince’s is [‘printsəz]?

The excrescent [t] is optional. (I have it in my speech.) I don't think many people have a true [ə] there. Despite what dEhiN says, I think in General American [ɪ̈] is much more common. (This sound is sometimes referred to as "s(c)hwi" by analogy with s(c)hwa.)

Woods wrote:How about the prince’s castle? Does the ‘s remain [z]?

In rapid speech, it might be devoiced to [z̥], but I don't feel this is mandatory.

Woods wrote:The prince’s absence. – I guess here we’ll have a [z], but not if the next word starts with a consonant. What do you think? I’ve never thought of whether ‘s would be pronounced [s] or [z]. If I try to come up with some other examples:

the house’s roof = [ðə ’hawsɪs ru:f]
the house’s artificial appearance = [ðə ’hawsɪz ’artɪfiʃəl əp’iərənts]

Am I right or just not used to discerning the [z]?

Wrong. You'd have [z] in both cases. (And [r] sounds pretty unusual here unless you're Scottish.)

Woods wrote:And, to finish with the houses, we could add:

the houses’ roofs
the houses’ artificial appearance

I guess that would be:

[ðə ’hawzɪs ru:fs]
[ðə ’hawzɪz ’artɪfiʃəl əp’iərənts]

I’m pretty sure I’ve gotten it wrong up until here, so I hope you will fix my phonology a little bit.

Again, it would be [z] in both cases. (And I actually say [ɹu̟ːvz], but that's recognisably nonstandard.)

Woods wrote:Indeed, I don't think I pronounce the genitive suffix as [z] very often, if ever – I just say [s] almost all the time. So the question is, where should I replace the [s]'s with [z]'s instead. The strange thing is, natives say I speak great English and nobody points these things out to me, and then all of a sudden I'm surprised to find I haven't spoken right. Maybe they can't hear the difference either?

You probably do speak great English. It's a very small detail, and even if people did notice, they probably wouldn't consider it worth mentioning.

I have a German friend who speaks excellent English and it took me years to realise that one of the small details which made his English sound non-native was the length of his vowels. There's a subphonemic distinction in English whereby short vowels are slightly longer before voiced consonants than before unvoiced ones. Very view native speakers are aware they do this and it's not taught explicitly in English courses as far as I know, but it can affect comprehension. (It's one of the features which makes Scottish accents difficult to comprehend for other speakers of English.)

Woods wrote:
Even speakers which don't put primary stress on the final syllable of princess stress it more than they do the final syllable of prince's/princes.

Can you really use which in this sentence? I would either use who or that, otherwise it sounds to me like we’re talking about an object. Sorry for deviating into your grammar, which is without a doubt much better than mine, but that also grabbed my attention.

Yes, you really can, as evidenced by the fact that (1) I did and (2) I didn't even notice this until you said something. Stylistically, your suggestions are preferable, but which can and is used here as a quick Googling will tell you. (Apparently, the KJV is shot through with it.)
"Richmond is a real scholar; Owen just learns languages because he can't bear not to know what other people are saying."--Margaret Lattimore on her two sons

User avatar
dEhiN
Posts: 3512
Joined: 2013-08-18, 2:51
Real Name: David
Gender: male
Location: Toronto
Country: CA Canada (Canada)
Contact:

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby dEhiN » 2017-01-30, 3:44

Which, that, and who are all relative pronouns that, in many cases, can be used interchangeably. At least in common speech. In an English course or for formal, proper writing, there are grammar rules specifying when each should be used. But in everyday speech, natives mix them all the time and it's acceptable.

You can hear [ɘ] here. That Wikipedia chart doesn't have a recording for [ə] but it's of course the schwa vowel, so I'll assume you know what it sounds like. Linguoboy is probably right about GA using [ï]; he's the linguist and has studied this stuff, plus I was just writing out what I say. Also, [ə], [ɘ], and [ï] are all very similar sounds for me - you can see that Wikipedia vowel chart that they are all central vowels and ranging from mid to near-close - so I don't really distinguish them well.

Lastly, the ESL rule that I believe is taught regarding the plural/genitive "-s" and pronunciation is this:

1) When following an unvoiced consonant, the "-s" is unvoiced
[kæt] -> [kæts]

2) When following a voiced consonant, the "-s" is voiced
[dɔg] -> [dɔgz]

3) When following a /s,z/, the "-s" is said as [-ɪz]*
[hɔɹs] -> [hɔɹsɪz]

*Note this is the rule in regard to English linguistics; I'm sure in actual speech, the /ɪ/ varies a little.

One thing about rule 3 is that it applies even when you're adding a genitive to a plural (and you're pronouncing the genitive "-s", which not everyone does):
[hɔɹsɪz] -> [hɔɹsɪzɪz]
[dɔgz] -> [dɔgzɪz]
[kæts] -> [kætsɪz]

And like linguoboy said, when the genitive/plural "-s" is preceded by and followed by an unvoiced consonant, in regular speech it's probably devoiced.
Follow my TAC 2017 here.

(N)  (en-CA) | (B2)  (fr) | (B1)  (pt-BR) | (A2)  (es-CO) | (A1)  (ja) (ko) (sv) (ta-LK)
(A0)  (de) (fy) (haw) (hi) (hu) (id) (it) (oc) (oj) (pl) (ro) (sq) (tl) (tr) (zh)

User avatar
Woods
Posts: 275
Joined: 2007-11-14, 12:43
Gender: male
Location: София
Country: BG Bulgaria (България)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby Woods » 2017-02-17, 12:32

This reply has been waiting as a draft on my computer, but I haven’t had the time to finish it.

Should we conclude that the genitive ‘s is always pronounced [z], except in words ending in unvoiced consonants like cat - cat’s – as I can see dEhiN has explained below?

Interestingly, I think I pronounce the genitive ‘s when added to words ending in -s as /ɛs/ in all circumstances… which may be another thing I should think about.


linguoboy wrote:So, prince’s is [‘printsəz]?
I don't think many people have a true [ə] there. Despite what dEhiN says, I think in General American [ɪ̈] is much more common. (This sound is sometimes referred to as "s(c)hwi" by analogy with s(c)hwa.)


I guess it’s that you’re just pronouncing the s and the z, and the schwi/schwa is just the very minimum amount of filling necessary to separate them.


linguoboy wrote:And [r] sounds pretty unusual here unless you're Scottish.

You know the IPA much better than me. When I write [r] I mean normal English r, I really don’t know which way I should turn it around to match the exact IPA symbol… Indeed, both dictionaries I have on my computer (Cambridge Advanced Learner’s, which is pretty basic, but also Oxford English Dictionary, which is very detailed), use the simple [r] in their transcriptions.

And when somebody posts something here, asking if a certain thing r or ɹ for example (without specifying r like in this language or dialect and ɹ like in that one) – I’m just wondering what this person means and not able to reply or understand. Indeed, dictionaries use the same letters for different sounds – my Danish dictionary uses [a] for [ɛ] and [α] for [a] – I just know that when I see an a it’s ɛ and don’t ask myself the question if it’s IPA-correct or not (I know it isn’t). And both my Danish and English dictionaries use r for the two languages’ fricatives, even though they have nothing to do with each other.

Indeed, is there such a thing as a standard IPA where every sound in any language and dialect has its own symbol and everybody uses it and knows it?


linguoboy wrote:There's a subphonemic distinction in English whereby short vowels are slightly longer before voiced consonants than before unvoiced ones.

Interesting. I can’t really imagine it without hearing it – can you give some good examples so that I can try to check them by pressing the hearing icon on the dictionary?


linguoboy wrote:Very view native speakers are aware they do this and it's not taught explicitly in English courses as far as I know, but it can affect comprehension.

You have no idea how little of pronunciation is taught in language courses. I studied French in high school for five years and no one ever told anything about its phonology. It was kind of similar in the very few extra English courses I’ve taken. It’s a disaster. Language schools and the whole language education system seems to put very little value on pronunciation – and people don’t mind speaking and sounding strikingly un-native.


linguoboy wrote:How about the prince’s castle? Does the ‘s remain [z]?
In rapid speech, it might be devoiced to [z̥], but I don't feel this is mandatory.

I have no idea what the difference between [z] and [z̥] is.

I think I’ll have to search for a good book or website that explains English phonology in detail. Does anyone know such one? Even though I think it’ll be pretty different from place to place – so I guess if your English is good overall and do something like pronouncing [s] instead of [z] for the genitive, natives may just think it’s another sort of dialect? I found two such things on my favourite torrent site: English Phonetics and Phonology by Peter Roach, edited by Cambridge (so this one should deal with British English) and an Atlas of North American English by Labov, Ash and Boberg (edited by Mouton de Gruyter, whatever that is – this one is without audio). I’ll download them and check them when I have time for this.


dEhiN wrote:One thing about rule 3 is that it applies even when you're adding a genitive to a plural (and you're pronouncing the genitive "-s", which not everyone does):
[hɔɹsɪz] -> [hɔɹsɪzɪz]

Is it even allowed? I say [ho:rsɪs] in both nominative and genitive and if somebody makes it [ho:rsɪsɪs/-ɪzɪz] it’ll sound strikingly wrong to me.

User avatar
dEhiN
Posts: 3512
Joined: 2013-08-18, 2:51
Real Name: David
Gender: male
Location: Toronto
Country: CA Canada (Canada)
Contact:

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby dEhiN » 2017-02-17, 13:21

I'm not going to quote because I'm being lazy, but I'm going to try and respond to your different points/questions.

1) In the linguistics course I took they basically broke down the pronunciation rule for the plural s, which would be the same for the genitive s, which perhaps one extra rule:
a. unvoiced stop + s = /s/
b. voiced stop + s = /z/
c. sibilant + s = /z/ (actually I think it was /ə/)

The same rules above would apply with 's. The extra rule would be that if you have a plural word and you're adding 's, you could pronounce it or not. If you don't, then I think it's usually just said with /s/ the same as the plural word. I've heard some people try to geminate the /s/ or maybe even add an unvoiced schwa to denote somehow the genitive factor. And then I've heard others pronounce the genitive giving rise to the pronunciations I indicated with horses's (see point 4).

2) There's a difference between broad and narrow transcription in IPA. Broad, which uses /, is meant to give a rough idea. For example, in English it's known that stops are pronounced slightly different when at the beginning of a word, in the middle, and (for some people) at the end. In broad transcription, those variations wouldn't be shown: /pɪg/, /hæpi/, /slæp/ all use /p/. In narrow transcription, which uses [..], variations would be shown and the exact pronunciation is given: [pʰɪg], [hæpi], [sl̪æp̚] (I use a dental /l/).

That's could be one reason why your Danish dictionary uses the IPA symbols differently than in English. Also, with vowels there's a slight range and variation as to when one vowel becomes another that's close to it. And that range could be different depending on the language.

These are some reasons why I used schwa and linguoboy talked about the schwi. (Another reason is that he's a professional linguist while I'm not, so he's had more training that I in distinguishing different phones and phonemes.) Typical English /r/ (meaning what most varieties have, and what is taught as standard) is actually an alveolar approximant [ɹ]. Apparently Danish /r/ is a uvular fricative or approximant [ʁ].

But otherwise, IPA is supposed to be a standard that everyone can use in the same way.

3) An example of the voiced/unvoiced factor on lengthening of short vowels is this pair: bat and bad. If I transcribe them, I would probably write [bæt] (or [bæt̚]) and [bæˑd] (where ˑ means half-long).

Also, [z̥] means a devoiced or voiceless [z], which in my eyes is effectively an [s].

4) The whole [hɔɹsɪz] -> [hɔɹsɪzɪz] thing isn't common and I think nowadays it's rare, at least in my dialect/sociolect/idiolect. But I have heard it in the past. I believe there's an English grammar rule about when you're supposed to orthographically add 's to a plural and when you just add the apostrophe. I imagined that whenever the apostrophe + s was added to a plural, some people who then pronounce both the plural and the genitive s.
Follow my TAC 2017 here.

(N)  (en-CA) | (B2)  (fr) | (B1)  (pt-BR) | (A2)  (es-CO) | (A1)  (ja) (ko) (sv) (ta-LK)
(A0)  (de) (fy) (haw) (hi) (hu) (id) (it) (oc) (oj) (pl) (ro) (sq) (tl) (tr) (zh)

User avatar
linguoboy
Posts: 19327
Joined: 2009-08-25, 15:11
Real Name: Da
Location: Chicago
Country: US United States (United States)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby linguoboy » 2017-02-17, 15:49

dEhiN wrote:1) In the linguistics course I took they basically broke down the pronunciation rule for the plural s, which would be the same for the genitive s, which perhaps one extra rule:
a. unvoiced stop + s = /s/
b. voiced stop + s = /z/
c. sibilant + s = /z/ (actually I think it was /ə/)

What about after unvoiced obstruents which are neither stops nor sibiliants? I doubt you say *[tʰɪfz], for instance.

dEhiN wrote:2) There's a difference between broad and narrow transcription in IPA. Broad, which uses /, is meant to give a rough idea.

There's a difference between phonetic transcription and phonemic. The convention is to put phonemic transcriptions between /forward slashes/ and phonetic transcriptions between [square brackets].

There's also a different between broad and narrow phonetic transcriptions. So, for instance, in your examples you did not notate the subphonemic length distinction between [pʰɪg] and [sl̪æp̚]. (Or perhaps your idiolect lacks this?)

Many sources blur the distinction. Wiktionary, for instance, tends to use broad phonetic transcriptions for English but put them in between forward slashes. But I think it's useful to know what it should be, with the understanding that not everyone adheres to linguistic convention.

dEhiN wrote:Also, [z̥] means a devoiced or voiceless [z], which in my eyes is effectively an [s].

I wouldn't expect most English-speakers to hear the difference. But I've studied two languages (Osage and Alemannic) with [z̥] (and voiceless lenis consonants in general), so I do notice it.
"Richmond is a real scholar; Owen just learns languages because he can't bear not to know what other people are saying."--Margaret Lattimore on her two sons

User avatar
linguoboy
Posts: 19327
Joined: 2009-08-25, 15:11
Real Name: Da
Location: Chicago
Country: US United States (United States)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby linguoboy » 2017-02-17, 16:18

Woods wrote:I guess it’s that you’re just pronouncing the s and the z, and the schwi/schwa is just the very minimum amount of filling necessary to separate them.

Essentially. Alveolar sounds (like [s] and [z]) are produced with the tongue tip near the very front of the mouth, which is the same place it is for high front vowels like [i], so it takes less articulatory effort to produce a shwi here than a shwa (which is a mid-central vowel).

Woods wrote:
linguoboy wrote:And [r] sounds pretty unusual here unless you're Scottish.

You know the IPA much better than me. When I write [r] I mean normal English r, I really don’t know which way I should turn it around to match the exact IPA symbol… Indeed, both dictionaries I have on my computer (Cambridge Advanced Learner’s, which is pretty basic, but also Oxford English Dictionary, which is very detailed), use the simple [r] in their transcriptions.

I don't know about the CALD, but the OED gives transcriptions which are phonemic but not phonetic. (See my reply to dEhiN for more on the distinction between the two.) That's partly because they are supradialectal. English /r/ has a wide range of realisations, so using a single symbol to cover them all makes more sense than a narrow phonetic transcription. And if you're going to do that, it makes sense to use the easiest one to type.

Woods wrote:Indeed, is there such a thing as a standard IPA where every sound in any language and dialect has its own symbol and everybody uses it and knows it?

As dEhiN explained, that's what the IPA is. When used for phonetic transcriptions, each symbol has a single accepted value. However, most people aren't phoneticians, and even the best phoneticians make mistakes, so the usage of IPA will never be 100% consistent and accurate.

When it comes to phonemic transcriptions, IPA is just a jumping-off point. There are a lot of considerations which come into play, but one of the principles is to choose something which is easy to type. (This was even more important in the days before word processors.) To interpret them properly, you have to understand the phonological analysis that's being used, which may require referring to an entire independent work.

Woods wrote:
linguoboy wrote:There's a subphonemic distinction in English whereby short vowels are slightly longer before voiced consonants than before unvoiced ones.

Interesting. I can’t really imagine it without hearing it – can you give some good examples so that I can try to check them by pressing the hearing icon on the dictionary?

bad vs bat is a good one. Try also bid vs bit and bed vs bet.

Woods wrote:
linguoboy wrote:Very view native speakers are aware they do this and it's not taught explicitly in English courses as far as I know, but it can affect comprehension.

You have no idea how little of pronunciation is taught in language courses.

Oh, I do. I'm just appalled by how little effort most USAmerican students make to sound remotely like a native, but I can't really blame them because of how little time their teachers generally spent drilling them. It's just something you're expected to "pick up" on your own.

Woods wrote:
linguoboy wrote:How about the prince’s castle? Does the ‘s remain [z]?
In rapid speech, it might be devoiced to [z̥], but I don't feel this is mandatory.

I have no idea what the difference between [z] and [z̥] is.

A ring below indicates lack of voicing.

Woods wrote:I think I’ll have to search for a good book or website that explains English phonology in detail. Does anyone know such a one?

Start here: https://en.wikipedia.org/wiki/English_phonology. If you need more detail, there's an extensive bibliography and I can give you some feedback on the works I've consulted.

Woods wrote:Even though I think it’ll be pretty different from place to place – so I guess if your English is good overall and do something like pronouncing [s] instead of [z] for the genitive, natives may just think it’s another sort of dialect?

Maybe. It depends a lot on the particular mix of features. If you have anything which just screams "foreign accent" (like, for instance, [s] for /θ/), then they'll assume you're non-native no matter what.

Woods wrote:I found two such things on my favourite torrent site: English Phonetics and Phonology by Peter Roach, edited by Cambridge (so this one should deal with British English) and an Atlas of North American English by Labov, Ash and Boberg (edited by Mouton de Gruyter, whatever that is – this one is without audio). I’ll download them and check them when I have time for this.

I haven't used the Roach, but Cambridge does good work. Labov is terrific. There's a website containing some of the same information as the Atlas here: http://www.ling.upenn.edu/phono_atlas/home.html.
"Richmond is a real scholar; Owen just learns languages because he can't bear not to know what other people are saying."--Margaret Lattimore on her two sons

User avatar
Woods
Posts: 275
Joined: 2007-11-14, 12:43
Gender: male
Location: София
Country: BG Bulgaria (България)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby Woods » 2017-02-18, 0:03

Man (or girl, you’re a girl indeed, aren’t you?) – I love how much you know about your language!


linguoboy wrote:As dEhiN explained, that's what the IPA is. When used for phonetic transcriptions, each symbol has a single accepted value. However, most people aren't phoneticians, and even the best phoneticians make mistakes, so the usage of IPA will never be 100% consistent and accurate.

Can you recommend a place to study the different phonetic symbols and the sounds they represent (a website, a book or something?) I’m not a phonetician, but I gradually learn about the different symbols essentially by asking questions about phonology on this forum :)


linguoboy wrote:To interpret them properly, you have to understand the phonological analysis that's being used, which may require referring to an entire independent work.

What?


linguoboy wrote:
Woods wrote:
linguoboy wrote:There's a subphonemic distinction in English whereby short vowels are slightly longer before voiced consonants than before unvoiced ones.

Interesting. I can’t really imagine it without hearing it – can you give some good examples so that I can try to check them by pressing the hearing icon on the dictionary?

bad vs bat is a good one. Try also bid vs bit and bed vs bet.

I checked them all in CALD:

bad is transcribed as /bæd/, but when I press the pronunciation icon, it really is pronounced as if it was a long vowel, for both UK and American English.

bat – (transcribed as /bæt/) – the UK pronunciation is clearly very short, whereas the US one sounds like a long vowel again

I think they’re trying to emphasise the quality of the vowel and not match the actual length.

But when I think about it (I don’t know if it’s a placebo effect or an objective observation), I think I do pronounce the a in bad a lot longer that the one in bat. It’s almost like a long vowel sometimes – I think I would make it sound like a long vowel if it was the last word in a sentence, but keep it short if it was followed by a noun:

[its ə bæd ɪgzæmpəl] vs. [ðæts ri:li bæ:d]

But still, even as an attributive, I think I make it longer than in bat, so I think I’m getting it right :)

Back to CALD:

bid (UK) – sounds half short, half long
bid (US) – sounds clearly like a long vowel

bit (UK) – sounds really short
bit (US) – sounds half short, half long

bed – both UK and US recorded pronunciations sound long, the US one is a little bit longer

I myself wasn’t aware the ɛ is this word is considered a short one.

bet – the UK pronunciation is clearly short, the US one is long – maybe the US speaker is trying to emphasise the quality at the expense of the vowel length again

Another conclusion (if I can make one based on these few examples) – American short vowels seem to be much longer than British ones?


linguoboy wrote: There's a difference between phonetic transcription and phonemic. The convention is to put phonemic transcriptions between /forward slashes/ and phonetic transcriptions between [square brackets].

That’s unbelievably interesting. Many thanks for these observations!


linguoboy wrote: There's also a different between broad and narrow phonetic transcriptions. So, for instance, in your examples you did not notate the subphonemic length distinction between [pʰɪg] and [sl̪æp̚]. (Or perhaps your idiolect lacks this?)

I got the difference between phonetic and phonemic, but what is subphonemic?


linguoboy wrote:
dEhiN wrote:Also, [z̥] means a devoiced or voiceless [z], which in my eyes is effectively an [s].

I wouldn't expect most English-speakers to hear the difference. But I've studied two languages (Osage and Alemannic) with [z̥] (and voiceless lenis consonants in general), so I do notice it.

So, [z̥] is a half-voiced, half-unvoiced sound between [z] and [s], like in Dutch v is supposed to be something between w (pronounced [v]) and f (pronounced [f])?


linguoboy wrote:
Woods wrote:
linguoboy wrote:How about the prince’s castle? Does the ‘s remain [z]?
In rapid speech, it might be devoiced to [z̥], but I don't feel this is mandatory.

I have no idea what the difference between [z] and [z̥] is.

A ring below indicates lack of voicing.

But lack of voicing is not the same as no voicing (like in a voiceless consonant) in this case? :)


linguoboy wrote:
Woods wrote:You have no idea how little of pronunciation is taught in language courses.

Oh, I do. I'm just appalled by how little effort most USAmerican students make to sound remotely like a native

Same here – European students are the same.


linguoboy wrote:I haven't used the Roach, but Cambridge does good work. Labov is terrific. There's a website containing some of the same information as the Atlas here: http://www.ling.upenn.edu/phono_atlas/home.html.

Thanks for the recommendations!

User avatar
linguoboy
Posts: 19327
Joined: 2009-08-25, 15:11
Real Name: Da
Location: Chicago
Country: US United States (United States)

Re: [ðə ‘printsɛs] – prince’s or princess?

Postby linguoboy » 2017-02-18, 4:13

Woods wrote:Man (or girl, you’re a girl indeed, aren’t you?)

You sussed me out. I just put "boy" in my username to throw everyone off.

Woods wrote:Can you recommend a place to study the different phonetic symbols and the sounds they represent (a website, a book or something?)

Websites are best because they incorporate audio files. I've heard good things about this one: http://www.ipachart.com/

Woods wrote:I’m not a phonetician, but I'm gradually learning about the different symbols essentially by asking questions about phonology on this forum :)


Woods wrote:
linguoboy wrote:To interpret them properly, you have to understand the phonological analysis that's being used, which may require referring to an entire independent work.

What?

Different phonological analysis can produce quite different results. For instance, Standard Chinese can be formally analysed as having five phonemic vowels, two phonemic vowels, or even no phonemic vowels at all! Depending on which theoretical model you favour, it will determine your choice of symbols to represent the underlying phonemes. So, for instance, the vowel in English bait could be phonemicised as /ey/, /ɛɪ̯/, /e/, or /eː/ depending on your analysis. Each choice is defensible provided it's consistent with the choices made for other vowels (and semivowels). But understanding a particular phonologist's choice might require reading a complete phonological analysis of the language, and these can run to hundreds of pages!

Woods wrote:Another conclusion (if I can make one based on these few examples) – American short vowels seem to be much longer than British ones?

It's possible. I don't notice much difference in length between American and southern English accents in this respect. I really notice it in Scottish English, which follows markedly different allophonic rules.

Woods wrote:
linguoboy wrote: There's also a different between broad and narrow phonetic transcriptions. So, for instance, in your examples you did not notate the subphonemic length distinction between [pʰɪg] and [sl̪æp̚]. (Or perhaps your idiolect lacks this?)

I got the difference between phonetic and phonemic, but what is subphonemic?

A phonetic difference which isn't salient for distinguishing phonemes is subphonemic. So, for instance, dialects of English vary considerably in the distribution of [ɫ]. But there are no morphemes distinguished solely by whether they contain [ɫ] or [l] so the difference is subphonemic--even though speakers of languages where they are distinguished (such as Scottish Gaelic) can clearly hear the phonetic contrast between them.

Woods wrote:
linguoboy wrote:
dEhiN wrote:Also, [z̥] means a devoiced or voiceless [z], which in my eyes is effectively an [s].

I wouldn't expect most English-speakers to hear the difference. But I've studied two languages (Osage and Alemannic) with [z̥] (and voiceless lenis consonants in general), so I do notice it.

So, [z̥] is a half-voiced, half-unvoiced sound between [z] and [s], like in Dutch v is supposed to be something between w (pronounced [v]) and f (pronounced [f])?

No. Whether a consonant is fortis or lenis is independent of whether it's voiced or not. [z̥] is not voiced at all, but it is lenis whereas [s] is both unvoiced and fortis.

Woods wrote:But lack of voicing is not the same as no voicing (like in a voiceless consonant) in this case?

Those are different ways of saying the same thing. If a sound is unvoiced, that means it lacks voicing. It is not voiced, i.e. voiceless.
"Richmond is a real scholar; Owen just learns languages because he can't bear not to know what other people are saying."--Margaret Lattimore on her two sons


Return to “English”

Who is online

Users browsing this forum: No registered users and 1 guest