6

I hope this question fits the group.

As a teacher of EFL I have come across this question several times:

Question: Does "/kləʊðz/" provide the right phonemic representation of the final sound in the word "clothes"?

"Oxford Learner's Dictionaries" gives four variants. In both BrE variant recordings, what I hear is a final unvoiced /s/ sound, and this is also the way I pronounce it. The dictionary seems to transcribe the final sound in both cases of BrE as voiced /z/-sound. (see Oxford Learner's Dictionaries, "Clothes")

Collins Dictionary, explicitly states that their phonemic representions follow Received Pronunciation:

The accent represented by the pronunciations in this dictionary is Received Pronunciation.

gives the same transcription of "clothes".

It seems practically impossible to me (let alone to my students) to actually realise the pronunciation suggested by the phonemic representation provided.

Ashwin Schumann
  • 434
  • 1
    That depends a lot on dialect. – marcellothearcane Jul 13 '17 at 12:11
  • Thanks for your input @marcellothearcane - I've specified my answer: I am discussing the pronunciation of "clothes" in Received Pronunciation. – Ashwin Schumann Jul 13 '17 at 12:20
  • I guess this question of mine can only get a final and objectified answer by measuring the sound and its specific occurrence by means of instruments. – Ashwin Schumann Jul 13 '17 at 13:11
  • 3
    It's not a phonetic transcription but a phonemic representation. Instruments won't tell you anything relevant. – StoneyB on hiatus Jul 13 '17 at 13:24
  • 3
    I (in the US) would voice the s in clothes. Speaking quickly, it may even sound like close (with s voiced). – GEdgar Jul 13 '17 at 15:06
  • 2
    Possibly related: Are “whores” and “horse” homophones? – herisson Jul 13 '17 at 18:02
  • @GEdgar The question referred to standard BrE pronunciation. In the audio samples of the standard AmE pronunciation variants, I can indeed hear the /z/ quite clearly. – Ashwin Schumann Jul 13 '17 at 18:13
  • 3
    @GEdgar it doesn't just sound like close when speaking quickly in some US dialects. The /ð/ is just omitted. In fact, that's what I thought the question was going to be about from the title. – trlkly Jul 14 '17 at 10:59
  • 1
    For the /kləʊz/ pronunciation that Oxford Learners gives, you can contrast it with /kləʊs/ here--the difference should be quite apparent. For their recording of /kləʊðz/, the final "z" is devoiced much more than I (native Br Eng speaker) would consider "normal"--I think most RP speakers would pronounce the final sound as a clearer "z" (though it may be less voiced than usual as others have noted). Also note that there you will find considerable variation between speakers/dialects - eg /kləʊdz/ is also relatively common. – psmears Jul 14 '17 at 11:57
  • 1
    @psmears My gut feeling is that final /ðz/ quite commonly gets partially, if not almost fully, devoiced in BrE. I find it quite difficult, truth be told, to pronounce clothes with two voiced final consonants: I either drop the /ð/ and realise is as [kl̥əʊz], or I devoice both (the /ð/ partially, the /z/ fully), realising it something like [kl̥əʊð̬s]. An actual [kl̥əʊðz] takes effort for me. – Janus Bahs Jacquet Jul 14 '17 at 12:25
  • @Janus Bahs Jacquet You nailed the point I was trying to make in my question. The pronunciation suggested by the transcription provided sets a standard that seems impossible to fulfill, at least for me and probably all of my foreign students. – Ashwin Schumann Jul 14 '17 at 12:29
  • 1
    @JanusBahsJacquet: Yes, it certainly gets devoiced to an extent--my point is that the extent of the devoicing varies, both by speaker and by surrounding sounds. Ashwin: Yes, it's a tricky word to pronounce, even for native speakers. But remember that it's a phonemic transcription - meaning you have to produce a final sound that's close enough for listeners to perceive as a /z/, not one that is necessarily an accurate [z] in practice (and the fact final /-ðs/ never occurs helps with this). Also note that variants such as /kləʊz/, /kləʊdz/, /kləʊvz/ are common and unlikely to be remarked on. – psmears Jul 14 '17 at 12:54
  • @psmears After a little bit of research, I understand the fundamental difference between phonemic and phonetic. I was (obviously) totally ignorant of the difference in meaning. – Ashwin Schumann Jul 14 '17 at 13:04
  • 1
    The sound byte is more modern. [kləʊðz] is probably a pronunciation that went into the OED decades ago and it is much more representative of mid-century RP (Think Margaret Thatcher, for example, who really forced a hyper-proper RP accent for the time.). – J... Jul 14 '17 at 13:08
  • @J...: No, real people without old-fashioned accents really do still pronounce it with a [ðz]. As well as many other variants, as I've noted :-) – psmears Jul 14 '17 at 13:45
  • @psmears Agreed, but it is a syllable that has been de-emphasized to a measurable degree amongst modern RP/SBE/BBC/etc speakers, I suppose I mean to say. ʊ would be a bit higher, more rounded, and perhaps more foreward than today; ð would be held longer and enunciated more strongly. Typical standard pronunciation today would probably win the transliteration a few extra diacritics, I think. I'm not talking about regional accents, also, to be sure - you wouldn't have to go far in west london, even, to find [kləʊvz], I'm sure... – J... Jul 14 '17 at 15:20
  • @J...: Right - I almost added a comment to the effect that the vowel sound was probably the biggest difference between "old RP" and "new RP" (for want of better names)! But whatever the differences in realisation, I think /kləʊðz/ is still how most speakers would think of it phonemically. And yes, I mentioned /kləʊvz/ above :) – psmears Jul 14 '17 at 15:42
  • @psmears hah, didn't notice that, apologies! – J... Jul 14 '17 at 15:58

3 Answers3

22

You need to keep a couple of things in mind:

  • The glyphs employed in the pronunciations you find in dictionaries are not "phonetic transcriptions" but phonemic representations (note that they are enclosed in //, not []). That is, they do not represent actual, infinitely variable acoustic phenomena but elements in the finite set of structurally categorized entities onto which hearers map what they hear. /z/ is the 'meaning' of the phone uttered, not its physical realization.

  • Pronunciation—physical realization—is environmentally conditioned: actual acoustic output of any phoneme is determined by the context in which it appears. With clothes, for instance, the voicing of the terminal /z/ will be sustained if the /z/ liaises with a following voiced phoneme but will slide off into /s/ if it liaises with a following unvoiced phoneme:

    I have clothes in my closet → /kləʊð zɪn/
    I took clothes from my closet → /kləʊð sfrɒm/ (sorta—it's actually more like .../zsf/...

    But what is heard in both cases, by a Real Hearer attending to the discursive meaning rather than the acoustic actuality, is the phoneme /z/.

The recorded pronunciations you find in dictionaries are artificially abstracted from context, like the wretched example sentences in grammarbooks and exams. But in practical terms there is no such thing as a 'null context' in Real Speech; the actual context of these pronunciations is a by-definition-voiceless silence following the phoneme /z/. So the Real Speaker slides off into the patent voicelessness which concludes these utterances.

StoneyB on hiatus
  • 68,905
  • Thanks for your answer! I see and agree that the environment of the word needs to be taken into account. And what about "clothes" then at the end of a sentence, where no sound is following the word immediately? (to get back to my question) What you at one point say is that when I don't attend to the sound as much as to the discursive meaning, I will hear the phoneme /z/. Need to try that, put on my Platonic meditation gear and think of the meaning of "clothes" while listening to the audio! Or is that meaning not discursive anymore? – Ashwin Schumann Jul 13 '17 at 13:33
  • 1
    I'm not sure what the question is. Are you asking if the real pronunciation is /kləʊðz/, /kləʊðs/, /kləʊz/, or /kləʊs/? – Mitch Jul 13 '17 at 13:42
  • 1
    @AshwinSchumann It's unlikely to be discursive, since there's no discourse context. --It's a matter of what you're attending to rather than what you're hearing. As Polanyi says "[W]e attend from the subsidiary particulars to their joint focus". – StoneyB on hiatus Jul 13 '17 at 13:44
  • Thanks for your well-informed answers @StoneyB is that Michael Polanyi? Which book? – Ashwin Schumann Jul 13 '17 at 15:35
  • @AshwinSchumann Yes, it's Michael Polanyi. That particular quote is from 'The Structure of Consciousness', but it's pretty a much a central theme in all his work that I'm aware of. The Tacit Dimension is his most accessible treatment. – StoneyB on hiatus Jul 13 '17 at 16:01
  • 1
    Hi. I can't work out how I took clothes from my closet → /kləʊð sɪn/ works. "Clothes from" (for me) is bound to have something labiodental going on in there even if said rapidly. The pronunciation you give doesn't seem to work with "from". – Francis Davey Jul 13 '17 at 16:03
  • 1
    @FrancisDavey Oops, I didn't finish changing in to from in the phonemic rep -- I'll fix. – StoneyB on hiatus Jul 13 '17 at 16:12
  • @StoneyB Thanks for the book recommendation. I wrote a thesis on Descartes' early philosophy and worked on the topic of language awareness. I'll read that article! – Ashwin Schumann Jul 13 '17 at 17:45
  • @StoneyB I want to add this: I don't think that it is the silence of a missing context that weakens the alleged /z/ sound, because to me the final sound is at no point a /z/, hence never sliding off from a /z/. I rather think that the presence of a /ð/ obstructs the appearance of a /z/ sound. – Ashwin Schumann Jul 13 '17 at 18:32
  • @AshwinSchumann I'd say just the opposite. The realization of the /S/ plural affix as /z/ is structurally entailed by the voicing on /ð/; consequently actual physical voicing becomes unnecessary, and voicing can glide off as the tongue moves from the dental to the post-dental position. (And in my own speech, which pronounces clothes as /kloʊz/, with no /ð/, the same glideoff to /s/ is discernible when I utter the word into silence.) – StoneyB on hiatus Jul 13 '17 at 19:12
11

In words like grieves, clothes, many speakers stop the voicing of the final /z/ earlier than you might expect from the phonemic transcription. Most English speakers still hear the phonemes /vz/ anyway, because at the end of a word, /vs/ is a combination of phonemes that does not appear in English. We only need to distinguish between griefs and grieves: /griːfs/ versus /griːvz/. We do this on the basis of whether the penultimate consonant is voiced.

Thus, the two pronunciations [griːvz] and [griːvs] represent the same underlying phonemes, which dictionaries represent as /griːvz/.

I'm not a trained phoneticist, so I don't entirely trust my ears. But I think you're right in that the Oxford dictionary's British speaker for the pronunciation /kləʊðz/ is actually saying [kləʊðs]. In the Collins dictionary, for this pronunciation, I think the voicing stops part way through the final /z/.

I can't say whether the "correct" pronunciation for RP is [griːvz] or [griːvs] or somewhere halfway in between, or all three possibilities. To decide that, you would have to collect recordings from a lot of RP speakers and examine the statistics for when they stop voicing the final consonant. Quite possibly, somebody has already done this.

The dictionaries give the phonemic transcription, which is /kləʊðz/ (and which is consistent with both phonetic transcriptions [kləʊðz] and [kləʊðs]).

Peter Shor
  • 88,407
  • 3
    Actually, we mostly distinguish between griefs and grieves on the basis of whether the preceding vowel is lengthened or not. The /v/ can (and does sometimes) get pretty much fully unvoiced as well and the word is still recognisable. – Janus Bahs Jacquet Jul 14 '17 at 12:29
0

Marcellothearcane gets it right - the IPA transcription of the pronunciation of a word is highly dependent on the dialect of the language. Certainly, my reading of the transcription you provide isn't wrong - but in the dialect(s) in my home city, it's just as likely that the /ð/ would be omitted, and I wouldn't guarantee /əʊ/ for the vowel; /ou/ is quite common, perhaps more so than /əʊ/.

Jeff Zeitlin
  • 6,776
  • 2
  • 21
  • 38
  • My question is not whether the transcriptions are right or wrong as such, but whether they correctly transcribe the given audio samples that are provided by the dictionary and are supposed to represent the standard of Received Pronunciation. – Ashwin Schumann Jul 13 '17 at 12:25
  • Unfortunately, the quality of the speakers on my computer are such that I can't say conclusively whether that last sound is /s/ or /z/; it really could be either. You might want to try to compare with "ozone" - but (on my computer) I still can't tell whether the /z/ is really /z/ or /s/. – Jeff Zeitlin Jul 13 '17 at 12:30
  • Do you mean the pronunciation of the word "ozone"? But that's clearly a /z/-sound! – Ashwin Schumann Jul 13 '17 at 12:31
  • There seems to be no controversy sounding the transcription of "clothes". At least, I haven't found anything on the internet yet. But maybe, sometimes, we believe so strongly in what we read - and maybe this is not only true of reading letters but also of reading phonetic transcriptions -, that we might not have noticed that it actually belies what we perceive?! – Ashwin Schumann Jul 13 '17 at 12:36
  • And I suspect that on better speakers than mine, I'd also hear a /z/ at the end of 'clothes' from that source. I chose 'ozone' more-or-less with malice aforethought; it has the same vowel before the /z/ that 'clothes' does - and I wanted to hear it without the intervening /ð/, since that consonant is most often omitted in my dialect, and the /z/ is quite distinctly a /z/ in common conversation. – Jeff Zeitlin Jul 13 '17 at 12:36
  • To me, that shift from /ð/ to /z/, as the transcription suggests and almost proscribes, seems physically challenging, almost as if it was against the economy proper to English phonology. But maybe I am totally mistaken about this! – Ashwin Schumann Jul 13 '17 at 12:45
  • If the question is whether the final sound on the recording is the same z as 'apples,' then yes that is the sound. – Yosef Baskin Jul 13 '17 at 12:50
  • I don't see it as any different than /ð/ to /z/; to my ear and voice, /s/ and /z/ differ only in that /z/ is voiced - that is, /z/ is voiced /s/ (or /s/ is voiceless /z/). Since /ð/ is also voiced, transitioning from /ð/ to /z/ should be no more of a challenge than /ð/ to /s/ - and possibly less, since the latter includes a transition from voiced to voiceless. – Jeff Zeitlin Jul 13 '17 at 12:53
  • (and the IPA chart agrees with me re: the relationship between /s/ and /z/) – Jeff Zeitlin Jul 13 '17 at 12:55
  • I think that the difficulty of reproducing both sounds in a sequence is rather a difficulty of moving between the tongue positions of /ð/ and /z/. – Ashwin Schumann Jul 13 '17 at 13:18
  • Entirely possible, and it may be the reason that the /ð/ is so often omitted entirely in my area. – Jeff Zeitlin Jul 13 '17 at 13:23