Page 2 of 2

Re: /s/ vs. /z/

Posted: 2019-01-14, 16:29
by Dormouse559
Woods wrote:According to this website:

When the letter ‘s’ is after an unvoiced or quiet consonant, it is
pronounced as a /s/. e.g., hats, tops, works, laughs, what’s, moths.

When the letter ‘s’ is after a vowel, another ‘s’, or a voiced consonant,
it is pronounced as a /z/ sound.
e.g., logs, tubes, beds, moves, clothes, was, becomes, he’s, passes.

Pronounce /s/ as ‘əz’ after the consonants /s /z/ ‘ch’ ‘sh’ ‘j’ ‘zh’ ( as in
vision)- ( in uz the ‘u’ is said very short and quickly –the schwa sound).
e.g., passes, beaches, washes, packages, noses.


How far is that true?

It's mostly correct for pluralizing nouns, contracting "is", and forming the possessive of singular and occasionally plural nouns.

But if the final vowel is a silent E, you should really ignore it when choosing between /s/, /z/ and /əz/. The important thing is the final pronounced sound, normally a consonant in the case of silent E. For example, "brake" /brejk/ pluralizes as "brakes" /brejks/, not /brejkz/.

Those rules don't account for a final S that doesn't represent plural, "is" or possession.

Woods wrote:At the moment, I'm listening to Walking In My Shoes by Depeche Mode and the singer is clearly saying "/traj wo:kiŋ in maj ʃu:s/" (even though I myself would rather pronounce it as /ʃu:z/). Is it just a British thing, or will it be almost impossible for a foreigner to get when to say /s/ and when /z/? I guess this is the single biggest issue in my pronunciation.

I don't know, but I hesitate to draw many conclusions from someone's singing pronunciation. The singer, Dave Gahan, adopts an accent that overall sounds a lot more American than his native one (Compare this interview I found).

Woods wrote:One more thing about the quote above: I'd say /'klowθs/, not /'klowθz/ (clothes) - okay, there's a vowel between th and s, but it's just in writing. Am I right?

Not in my dialect. I say /'klowðz/. I think I also say /'klowz/ for the noun (as opposed to the third-person singular of "to clothe") in fast speech. I wouldn't expect a /θ/.

Re: /s/ vs. /z/

Posted: 2019-01-15, 21:52
by linguoboy
I listened to the DM song again and I can see what Woods is saying. I don't think the /z/ is completely voiceless since I still perceive it as a /z/ even though phonetically it more closely resembles [s]. I have a couple of hypotheses:

1. The sound is actually [z̥], i.e. a sound that is unvoiced but lenis. I'm familiar with this sound from languages like Osage and Alemannic and I tend to map it to /z/ because it contrasts with a (unvoiced fortis) /s/.
2. The sound is [s] but I'm identifying it with /z/ on the basis of other cues. This seems less likely since the preceding vowel is /uː/ and English long vowels show less variance in length before voiced and unvoiced consonants than short vowels. (I know from experience with non-native speakers of English that I'm quite likely to identified a voiced sound as voiceless if the preceding phonemically short vowel is also phonetically short, e.g. when my German friends say [bɛd] rather than [bɛˑd], I hear bet and not bed. I'm not sure it works the same way in reverse, however.)

tl;dr: Gahan sings the word shoes with /z/ even though phonetically it sounds more like [s]. I don't know if this specifically British or just a vocal affectation or what.

Re: /s/ vs. /z/

Posted: 2019-01-16, 9:01
by Woods
linguoboy wrote:Do you just want the rules for pronunciation of the the default plural ending in English or for the pronunciation of s more generally?

Just the rules about the plural s :)

(I guess they're complicated enough ;)

Re: /s/ vs. /z/

Posted: 2019-02-09, 11:24
by Woods
Could you guys say how s should be pronounced after l, m, n, ng and r?

Re: /s/ vs. /z/

Posted: 2019-02-09, 15:10
by linguoboy
Woods wrote:Could you guys say how s should be pronounced after l, m, n, ng and r?

Are we talking just about |s| (i.e. the pluralising morpheme)?

Re: /s/ vs. /z/

Posted: 2019-02-10, 20:33
by Woods
linguoboy wrote:
Woods wrote:Could you guys say how s should be pronounced after l, m, n, ng and r?

Are we talking just about |s| (i.e. the pluralising morpheme)?

Yes, mainly - even though info about other instances will be appreciated too.

Re: /s/ vs. /z/

Posted: 2019-02-11, 4:24
by linguoboy
These are all voiced sounds so it retains its default voiced quality.

Re: /s/ vs. /z/

Posted: 2019-02-12, 21:16
by TeneReef
visa can have /s/ or /z/?

Re: /s/ vs. /z/

Posted: 2019-02-12, 22:06
by linguoboy
TeneReef wrote:visa can have /s/ or /z/?

It can.

Interestingly, common lexicographical resources don't seem to reflect this. I quickly checked Wiktionary, the OED, and the AHD, and all of them listed only the /z/ pronunciation. But I grew up with /s/ in visa and only learned the /z/ pronunciation later.

Re: /s/ vs. /z/

Posted: 2019-02-17, 11:40
by Woods
linguoboy wrote:These are all voiced sounds so it retains its default voiced quality.

Are they considered voiced? At least in my native language (which has all except for ŋ), they are considered special and apart from all voiced-unvoiced pairs. And I can say both /ŋs/ / /ŋz/. /Ms/ sounds more more natural than /mz/ to me when I'm trying it now, as if I need to put extra energy into the /z/ and it takes it apart from the combination; /rs/ or /rz/ also sound equally easy, as if /r/ is neither voiced nor unvoiced; /l/ sounds more unvoiced than voiced to me, and so does /n/.

Re: /s/ vs. /z/

Posted: 2019-02-17, 14:44
by linguoboy
Yes, they’re voiced. Several languages (Welsh and Burmese among them) contrast voiced and unvoiced versions of them.

English does have clusters like /ns/ and /rs/, just not when |s| is added. For instance, tense /tɛns/ vs tens /tɛnz/ or purse /pʌrs/ vs purrs /pʌrs/.

ETA: I actually spent some time trying to come up with minimal pairs and I couldn't find any for /m/ or /ŋ/. I think the reason is, quite simply, /s/ only seems to occur after sonorants in the same syllable in words originally borrowed from Latin (as tense and purse were). And Latin dislikes these clusters and tends to eliminate them either via regressive assimilation or epenthesis. And example of the former is com- + sonans > consonans "consonant"; an example of the later is hiemps, a variant form of hiems "winter".

English likes epenthesis after nasals as well and, in fact, it's not unusual to hear epenthetic stops in these words in colloquial speech. That is, tense and tents are often homophones. In fact, the tendency is so pronounced that if you were to use an invented word like [tʰɛms] in an English sentence, I would most likely hear this as temps.

That's one reason why it's so important to get the /s/ vs /z/ distinction right. If you substitute /s/ for /z/ in a word like Thames or rings, I'm likely to hear not a mispronunciation of the word you intended but a different word altogether.

Re: /s/ vs. /z/

Posted: 2019-02-17, 14:50
by Naava
linguoboy wrote:English does have clusters like /ns/ and /rs/, just not when |s| is added. For instance, tense /tɛns/ vs tens/tɛnz/ or purse /pʌrs/ vs purrs /pʌrs/.

I thought it's /pɜː(r)s/ :hmm:

Re: /s/ vs. /z/

Posted: 2019-02-17, 15:41
by linguoboy
Naava wrote:
linguoboy wrote:English does have clusters like /ns/ and /rs/, just not when |s| is added. For instance, tense /tɛns/ vs tens/tɛnz/ or purse /pʌrs/ vs purrs /pʌrs/.

I thought it's /pɜː(r)s/ :hmm:

The actual quality of the vowel varies considerably, which is why I’m using a broad phonemic transcription.