Does full devoicing occur?

Status
Not open for further replies.

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
Hello.

I wonder if, in normal speech, full devoicing of usually voiced consonants can occur? For example, I often hear native speakers correctly say /ɪz/ (is), but sometimes, especially if the word is not emphasized, it sounds to my ears like /ɪs/. This seems to happen especially at word endings.

I've noticed similar devoicing with other consonants, and it seems sometimes one is only left with other clues like vowel lengthening or context.

Is this bad hearing on my part or can full devoicing indeed occur in English depending on how much attention the native speaker pays to the correct pronunciation?

Thanks!
 

Raymott

VIP Member
Joined
Jun 29, 2008
Member Type
Academic
Native Language
English
Home Country
Australia
Current Location
Australia
Hello.

I wonder if, in normal speech, full devoicing of usually voiced consonants can occur? For example, I often hear native speakers correctly say /ɪz/ (is), but sometimes, especially if the word is not emphasized, it sounds to my ears like /ɪs/. This seems to happen especially at word endings.

Thanks!
It's not clear what you mean. Are you talking about the word 'is'? If not, then some words do end in /ɪs/, eg. 'this'.
Can you give some examples?
 

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
Yes, I'm taking about the word "is". Or "eyes", "games", "season", etc., where the bold s should be pronounced /z/. Sometimes I hear that very clearly, sometimes I don't.

So I wonder if that's bad hearing on my part, or if an usually voiced consonant can indeed by fully devoiced.
 

Raymott

VIP Member
Joined
Jun 29, 2008
Member Type
Academic
Native Language
English
Home Country
Australia
Current Location
Australia
Yes, I'm taking about the word "is". Or "eyes", "games", "season", etc., where the bold s should be pronounced /z/. Sometimes I hear that very clearly, sometimes I don't.

So I wonder if that's bad hearing on my part, or if an usually voiced consonant can indeed by fully devoiced.
None of these words should be devoiced.
The only time this should happen, is when the /z/ if followed by /s/ for example in "His sister", but even then most people manage /hɪz sɪstə/
You might hear it happening when the following phoneme is another unvoiced sound, ("his toy") but it's not what we are "supposed" to say.
 

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
None of these words should be devoiced.
The only time this should happen, is when the /z/ if followed by /s/ for example in "His sister", but even then most people manage /hɪz sɪstə/
Even here, I doubt if the /z/ is fully devoiced. The preceding vowel will ensure that at least the onset of the /z/ is voiced.
 

mykwyner

Key Member
Joined
May 13, 2005
Member Type
English Teacher
Native Language
English
Home Country
United States
Current Location
United States
If an "s" sound is produced immediately after an unvoiced (the vocal chords are not used to produce it) consonant, it is usually pronounced . Example: cat + s = cats.

If an "s" sound is produced immediately after a voiced consonant or vowel, it is usually pronounced [z]. Example: dog + s = dogz.
 

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
None of these words should be devoiced.

Sorry to be pedantic, but does "should" here mean, that native speakers would never ever fully devoice?

The only time this should happen, is when the /z/ if followed by /s/ for example in "His sister", but even then most people manage /hɪz sɪstə/
Here I'm again confused by your use of "should". Does this mean, you ought to devoice the s in his, if followed by an unvoiced consonant?

You might hear it happening when the following phoneme is another unvoiced sound, ("his toy") but it's not what we are "supposed" to say.
So native speakers do fully devoice if the next sound is unvoiced, even though it's not correct?

I apologize for asking for clarification, but my interpretation of the three quotes of your post seem to contradict each other.

Is it allowed to upload some audio files here? I'd love to collect some examples where I hear no voicing although there should be. Then you could tell me if you do hear voicing or not.
 

BobK

Moderator
Staff member
Joined
Jul 29, 2006
Location
Spencers Wood, near Reading, UK
Member Type
Retired English Teacher
Native Language
English
Home Country
UK
Current Location
UK
Sorry to be pedantic, but does "should" here mean, that native speakers would never ever fully devoice?
....

I doubt it. In one case I can think of, the de-voicing can be fossilized and used - by mistake - when the phonetic context doesn't justify it. When 'have' is used in the sense of necessity rather than possession, it is usually (always...?) followed by 'to', causing devoicing of the /v/. Geoffrey Sampson, in Liberty and Language (or maybe Making Sense, it's over 30 years since I worked on them), reports hearing a politician saying 'It's a matter of simply /'hæfɪŋ/ to'. (This is not correct or usual, and is certainly not to be copied; but it shows how strong the tendency to devoice can be.)

b
 

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
Geoffrey Sampson, in Liberty and Language (or maybe Making Sense, it's over 30 years since I worked on them), reports hearing a politician saying 'It's a matter of simply /'hæfɪŋ/ to'. (This is not correct or usual, and is certainly not to be copied; but it shows how strong the tendency to devoice can be.)
Some fifty years ago, I was corrected at school for using /f/ in 'have to'; this was supposedly the sign of an uneducated speaker. I shudder when I think of some of the artificial 'rules' I was taught in those days.

I think that the total devoicing of /v/ in 'have to is a result of our (probably) subconscious feeling that the two words have and to are very closely associated as one idea (similar to must). In an utterance such as 'I've told you a thousand times', even though the /t/ follows immediately after the /v/, I don't think that it is fully devoiced.

The close association of the two parts of 'have to' have led to the written forms of the word(s) hafter/hafta in dialogue that is spoken by supposedly less educated speakers.

Although it's not connected with devoicing, we see a similar result of this close association of to with a preceding verb in such forms as wanna (want to), gotta (got to) and oughta (ought to)
 

Raymott

VIP Member
Joined
Jun 29, 2008
Member Type
Academic
Native Language
English
Home Country
Australia
Current Location
Australia
Sorry to be pedantic, but does "should" here mean, that native speakers would never ever fully devoice?


Here I'm again confused by your use of "should". Does this mean, you ought to devoice the s in his, if followed by an unvoiced consonant?

There are no rules saying that you ought to devoice the /z/ in 'his'. I was saying that it does happen sometimes.

So native speakers do fully devoice if the next sound is unvoiced, even though it's not correct?

No, I'm saying that it sometimes occurs.

I apologize for asking for clarification, but my interpretation of the three quotes of your post seem to contradict each other.

Is it allowed to upload some audio files here? I'd love to collect some examples where I hear no voicing although there should be. Then you could tell me if you do hear voicing or not.
Sorry about the ambiguous 'shoulds'.
If you listen to enough English you will know when phonemes are devoiced, etc. Yes, it's a good idea to upload videos or audio of sounds. It's easier to identify and talk about sounds after you've heard them.
I think you'd have to post them on an external site and give us the URL.
 

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
Some fifty years ago, I was corrected at school for using /f/ in 'have to'; this was supposedly the sign of an uneducated speaker. I shudder when I think of some of the artificial 'rules' I was taught in those days.

I think that the total devoicing of /v/ in 'have to is a result of our (probably) subconscious feeling that the two words have and to are very closely associated as one idea (similar to must). In an utterance such as 'I've told you a thousand times', even though the /t/ follows immediately after the /v/, I don't think that it is fully devoiced.

The close association of the two parts of 'have to' have led to the written forms of the word(s) hafter/hafta in dialogue that is spoken by supposedly less educated speakers.

Interesting. If you go to

en.wiktionary.org/wiki/have_to

you see that /ˈhæf.tu/ is indeed the correct pronunciation in the US. The same devoicing seems to occur (and is also documented) with constructs like "it's".

Also

en.wiktionary.org/wiki/hard

lists "heart" as a homophone of "hard" in some dialects, which seems to support the idea that full devoicing can indeed occur.
 

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
The same devoicing seems to occur (and is also documented) with constructs like "it's".
It's interesting that the contracted form of has and is pronounced with the same voiced/unvoiced forms as the third person singular verb ending and the plural -s (and indeed the possessive 's and 's'). With these other /s/ and /z/ sounds, we cannot really say that the sound is basically /s/. but it is voiced after certain voiceless consonants (or that it is basically /z/, but it is devoiced after vowels and certain voiced consonants). With the contracted form of has and is, however, the basic sound is /z/.

So, in answer to ancor90's original question, there is at least one consonant that can be, and is, fully devoiced.
 

BobK

Moderator
Staff member
Joined
Jul 29, 2006
Location
Spencers Wood, near Reading, UK
Member Type
Retired English Teacher
Native Language
English
Home Country
UK
Current Location
UK
:up: The song lyric springs to mind:

'Swonderful, 'smarvellous, [ʧ]?*you should care for me.
'Sawful nice etc.​

When I first heard this, I found it pretty hard to decipher. If it had been 'Zwonderful 'zmarvellous...' I'd probably still be wondering. ;-)

b

PS *This isn't a glottal stop; it's a question mark - I'm not sure if the [ʧ] is there.
 
Last edited:

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
Okay. As requested I've spent quite some time looking for, extracting, and uploading a couple of sound samples where I find it difficult to hear the voicing of a supposedly voiced consonant.


1) In the first sample, "father-is-a-fairly-religious-man.mp3," I cannot hear any vocal chords vibrating when she says 'is'.

2) In the second sample, "it's gonna be a felony charge on the three juveniles.mp3," I cannot hear any vocal chords vibrating in the last part of 'charge' - it sounds to me like /ˈʧɑɹʧ/ and not like /ˈʧɑɹʤ/. However, I can make out a slightly voiced /ʤ/ at the beginning of juvenile.

3) In the third example, an excerpt from the beginning of the American television series "Fringe," it sounds to me like /fɹɪnʧ/ and not like /fɹɪndʒ/.

However, if you go to dictionary.reference.com/browse/fringe and click on the speaker button to hear the word pronounced, I can make out a very clear /ʤ/ at the end.


I'd really love to know if you can hear the voiced consonant in each of the three samples, or if the speakers do indeed (fully) devoice in the examples.


Thanks. :)
 

Attachments

  • father-is-a-fairly-religious-man.mp3
    128.4 KB · Views: 7
  • it's gonna be a felony charge on the three juveniles.mp3
    123.8 KB · Views: 4
  • previously-on-fringe.mp3
    69.4 KB · Views: 5

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
1) In the first sample, "father-is-a-fairly-religious-man.mp3," I cannot hear any vocal chords vibrating when she says 'is'. I hear /z/ there.

2) In the second sample, "it's gonna be a felony charge on the three juveniles.mp3," I cannot hear any vocal chords vibrating in the last part of 'charge' - it sounds to me like /ˈʧɑɹʧ/ and not like /ˈʧɑɹʤ/. However, I can make out a slightly voiced /ʤ/ at the beginning of juvenile. I hear /ʤ/ in all three places.

3) In the third example, an excerpt from the beginning of the American television series "Fringe," it sounds to me like /fɹɪnʧ/ and not like /fɹɪndʒ/. Once again, I hear /ʤ/.
There is a possibility that I have convinced myself that I hear these sounds as voiced because I expect to hear this, but I am fairly sure that there is voicing there.
 

BobK

Moderator
Staff member
Joined
Jul 29, 2006
Location
Spencers Wood, near Reading, UK
Member Type
Retired English Teacher
Native Language
English
Home Country
UK
Current Location
UK
There is a possibility that I have convinced myself that I hear these sounds as voiced because I expect to hear this, but I am fairly sure that there is voicing there.

Same here - although given the way phonemes affect expectations and so perception... :oops: that's just what 5jj said, but using big words, so I won't bother with a finite verb! ;-)

b

PS Generally, if you think about the mechanics of the thing, it takes quite an effort to devoice consonants between vowels (which are necessarily 'voiced' - that's what vowel means). Obviously you can though; but, in that context (VCV) the lazier change would be in the opposite direction (unvoiced -> voiced).
 
Last edited:

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
There is a possibility that I have convinced myself that I hear these sounds as voiced because I expect to hear this

I know what you mean. I know that feeling from my own native tongue. I was pretty certain to hear a specific sound in a German word, until a non-native speaker told me that it isn't there, and after I listened closely he was right. I just expected it to be there (because it actually is there if you really emphasize the word), but it wasn't there when you didn't stress the word.

However, in this case I'm almost convinced that you are indeed right, and my hearing is just flawed.

Maybe I'm looking for the wrong cues?

I'm really keen on getting this right.

What has confused me, however, was that you said "I hear /ʤ/ in all three places.", when according to my dictionary it should only be in two places, at the end of "charge" and at the beginning of "juvenile." The beginning of charge should be voiceless. Did you also hear a /ʤ/ where there should be none?

Could some one maybe give me some tips on how to better hear the voicing? The part that is so frustrating is that I usually do hear the difference between /s/ vs /z/ and /ʧ/ vs /ʤ/ quite clearly, but in some cases, like the ones I attached, I just can't make it out.
 

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
What has confused me, however, was that you said "I hear /ʤ/ in all three places.", when according to my dictionary it should only be in two places, at the end of "charge" and at the beginning of "juvenile." The beginning of charge should be voiceless. Did you also hear a /ʤ/ where there should be none?
No. Careless writing. Sorry.
 

ancor90

Member
Joined
Feb 18, 2011
Member Type
Student or Learner
Native Language
German
Home Country
Germany
Current Location
Germany
Could you maybe explain what to look for? How you distinguish voiced vs unvoiced in these examples? :-(
 

5jj

Moderator
Staff member
Joined
Oct 14, 2010
Member Type
English Teacher
Native Language
British English
Home Country
Czech Republic
Current Location
Czech Republic
Could you maybe explain what to look for? How you distinguish voiced vs unvoiced in these examples? :-(
I am trying to decide how I know. I'll come back when I have a helpful answer - if I find one.
 
Status
Not open for further replies.
Top