seed vs. seat pronunciation, AmE

[whitemo...@gmail.com]

How do we pronunciation seed and seat exactly in AmE? I'm confused even with just the simplest case. Consider the two following sentences...

"I have a seed." vs.
"I have a seat.".

When I search the internet, I got a few different suggestion for the ending t or d sound. Some of those are...
1) a stop t for both t and d sound (seed and seat would sound the same, @@? I supposed...) https://youtu.be/AzZswz8na5c?t=4m54s
2) omit t or d all together, and distinguish them by with/without an abrupt cutoff of the vowel. https://youtu.be/ltDpWioRYjM?t=7m13s

3) By the way, when I tried to do the listening myself (with the video online), I heard people are saying t-sound for both the ending t and d.

To keep things simple, let's just consider ONLY the two sentences "I have a seed" and "I have a seat". Will you pronounce the ending d-sound and t-sound audibly for the sentences in concern? If not, how's it sounds exactly?

[Peter T. Daniels]

Probably the most useful cue for distinguishing syllable-final "voiced" and
"voiceless" segments is that the preceding vowel is longer before the
voiced one than before the voiceless one. This is the case regardless
of the phonetic realization of the stop.

/t/ has the more varied possibilities. Most often in AmE(American English),
it is an "unreleased" closure of the airstream by the tongue tip ("apex")
against the alveolar ridge (behind the upper teeth).
For special emphasis, it may be "released," i.e. aspirated (with a
puff of air). To a Chinese-speaker this will be immediately obvious
because aspiration is phonemic in Chinese, but an English-speaker will
normally be unaware of the existence of aspiration.
A third possibility, found in some AmE varieties and increasingly in
BrE (British English), is to replace the apico-alveolar closure with a
glottal closure.

The special characteristic of /d/, on the other hand, though nominally
it's "voicing," is in fact "laxness" -- relaxation of the oral muscles
more than is the case for /t/. A /t/ will never be voiced, a /d/ may
be voiced. It would be unusual for a voiced stop to be aspirated, because
that represents a conflict between two physical mechanisms involving
the vocal cords, but that is nominally the description of "breathy-voice"
sounds in some languages of South and Southeast Asia.

[Don Phillipson]

<whitemo...@gmail.com> wrote in message
news:1a7ff8d1-da1e-4ab3...@googlegroups.com...

> How do we pronunciation seed and seat exactly in AmE?

>. Consider the two following sentences...
>
> "I have a seed." vs.
> "I have a seat.".

1. No uniform norms determine American pronunciation.
2. The audible difference (a phonologist would explain)
is that terminal D is "voiced" and T "unvoiced." Theory
treats the difference as binary/absolute: our ears may
suggest it is a matter of degree.
--
Don Phillipson
Carlsbad Springs
(Ottawa, Canada)

[Richard Yates]

On Fri, 25 Dec 2015 09:39:26 -0500, "Don Phillipson"
<e9...@SPAMBLOCK.ncf.ca> wrote:

><whitemo...@gmail.com> wrote in message
>news:1a7ff8d1-da1e-4ab3...@googlegroups.com...
>
>> How do we pronunciation seed and seat exactly in AmE?
>>. Consider the two following sentences...
>>
>> "I have a seed." vs.
>> "I have a seat.".
>
>1. No uniform norms determine American pronunciation.
>2. The audible difference (a phonologist would explain)
>is that terminal D is "voiced" and T "unvoiced." Theory
>treats the difference as binary/absolute: our ears may
>suggest it is a matter of degree.

And also beware the McGurk effect it you are watching someone speak.
It can completely override your perception the sound that is produced.

http://www.slate.com/blogs/lexicon_valley/2014/06/27/mcgurk_effect_you_think_you_re_hearing_da_when_you_see_ga_and_hear_ba.html

[Jerry Friedman]

On Friday, December 25, 2015 at 9:51:06 AM UTC-5, Don Phillipson wrote:
> <whitemo...@gmail.com> wrote in message
> news:1a7ff8d1-da1e-4ab3...@googlegroups.com...
>
> > How do we pronunciation seed and seat exactly in AmE?
> >. Consider the two following sentences...
> >
> > "I have a seed." vs.
> > "I have a seat.".
>
> 1. No uniform norms determine American pronunciation.
> 2. The audible difference (a phonologist would explain)
> is that terminal D is "voiced" and T "unvoiced." Theory
> treats the difference as binary/absolute: our ears may
> suggest it is a matter of degree.

I'd have reversed "theory" and "our ears" in that last sentence, but
it might depend on where you get your theory and how well-trained
your ears are.

--
Jerry Friedman

[Peter T. Daniels]

On Friday, December 25, 2015 at 3:22:33 PM UTC-5, Richard Yates wrote:
> On Fri, 25 Dec 2015 09:39:26 -0500, "Don Phillipson"
> <e9...@SPAMBLOCK.ncf.ca> wrote:
> ><whitemo...@gmail.com> wrote in message
> >news:1a7ff8d1-da1e-4ab3...@googlegroups.com...

> >> How do we pronunciation seed and seat exactly in AmE?
> >>. Consider the two following sentences...
> >> "I have a seed." vs.
> >> "I have a seat.".
> >1. No uniform norms determine American pronunciation.
> >2. The audible difference (a phonologist would explain)
> >is that terminal D is "voiced" and T "unvoiced." Theory
> >treats the difference as binary/absolute: our ears may
> >suggest it is a matter of degree.
>
> And also beware the McGurk effect it you are watching someone speak.
> It can completely override your perception the sound that is produced.

There's unlikely to be a visible difference between [t] and [d].

http://www.slate.com/blogs/lexicon_valley/2014/06/27/mcgurk_effect_you_think_you_re_hearing_da_when_you_see_ga_and_hear_ba.html

[whitemo...@gmail.com]

Peter T. Daniels於 2015年12月25日星期五 UTC+8下午10時29分45秒寫道：

I see...

I merely thought that /d/ was a voiced sound so we had to voice it at all cost.

Is it correct to say... "the voicing of the ending /d/ could be barely noticeable, and the ending /d/ could differ from an aspirated ending /t/ just by aspiration" ?

[Peter Moylan]

On 2015-Dec-25 22:42, whitemo...@gmail.com wrote:
> Oops... just in case, please consider three sentences...
>
> "I have a seed."
> "I have a seat."
> "I have to sit."

The third is a red herring, I believe, because in most Englishes it has
a different vowel from the other two.

In AusE vowels are shorter before unvoiced stops: [si:d] vs [sit]. Is
that not also true in AmE?

--
Peter Moylan http://www.pmoylan.org
Newcastle, NSW, Australia

[Micky]

On Fri, 25 Dec 2015 03:38:16 -0800 (PST), whitemo...@gmail.com
wrote:

>
>How do we pronunciation seed and seat exactly in AmE? I'm confused even with just the simplest case. Consider the two following sentences...
>
>"I have a seed." vs.
>"I have a seat.".

These two are easy to distinguish. They're pronounced as they are
spelled.

Where you'd have trouble is with seeded and seated, or more commonly,
seeding and seating.

Who's in charge of the seating/seeding?

Then I can't distinguish the two even when I myself say them.

>When I search the internet, I got a few different suggestion for the ending t or d sound. Some of those are...
>1) a stop t for both t and d sound (seed and seat would sound the same, @@? I supposed...) https://youtu.be/AzZswz8na5c?t=4m54s
>2) omit t or d all together, and distinguish them by with/without an abrupt cutoff of the vowel. https://youtu.be/ltDpWioRYjM?t=7m13s
>
>3) By the way, when I tried to do the listening myself (with the video online), I heard people are saying t-sound for both the ending t and d.
>
>To keep things simple, let's just consider ONLY the two sentences "I have a seed" and "I have a seat". Will you pronounce the ending d-sound and t-sound audibly for the sentences in concern?

Yes, just like they're spelled.

[Peter T. Daniels]

But that wouldn't help distinguish the words, because the aspiration would
be quite rare. The best cue, as I said, is the lengthening of the preceding
vowel.

[whitemo...@gmail.com]

Mmm... in this case, I guess I really need to work on my rhythm all over again.

In case you might be curious, Chinese characters are all pronounced with a single syllable, and each of them are equal in length. That's why I had a really hard time to distinguish the words differ only by duration. Indeed, even for the obvious cases like /I/ vs. /i:/, it already took me quite a bit of attention.

Anyway, could you give me some advices about the rhythm concerning the ending consonants? Put it in terms of your subjective feeling would be excellent. ( trust me, I had done good enough theory )

[Dingbat]

On Friday, December 25, 2015 at 5:08:18 PM UTC+5:30, whitemo...@gmail.com wrote:
> How do we pronunciation seed and seat exactly in AmE? I'm confused even with just the simplest case. Consider the two following sentences...
>
> "I have a seed." vs.
> "I have a seat.".

[sɰi:d] is what I hear when a Malayalam speaker says <sweet>. It's a challenge to distinguish terrminal [t] and [d] when the context provides no clues. In English, one clue is that /i/ has a long realization of [i:] when followed by a voiced plosive but a shorter one when followed by an unvoiced plosive. Another clue in this particular context can be that /t/ has a glottally reinforced realization.

[Peter T. Daniels]

No, that's not why. There's nothing about pronunciation _inherent_ in a
Chinese character -- if there were, Japanese couldn't give a single
character both an _on_ and a _kun_ reading.

You need to be able to divorce your analysis of the Chinese language
from your knowledge of characters! We would deal with 詞 _cí_ 'word' rather
than with 字 _zì_ 'character'.

> I had a really hard time to distinguish the words differ only by duration. Indeed, even for the obvious cases like /I/ vs. /i:/, it already took me quite a bit of attention.

English-speakers also are not aware of vowel length differences in
their native language. They are a cue to the identity of the words that is
below the level of consciousness.

> Anyway, could you give me some advices about the rhythm concerning the ending consonants? Put it in terms of your subjective feeling would be excellent. ( trust me, I had done good enough theory )

Speakers just "know" what word is being said without knowing why. It
probably only takes practice for the learner. Are you able to detect
and produce word stress patterns, without making them into pitch differences?

[Jerry Friedman]

On Friday, December 25, 2015 at 7:47:57 PM UTC-5, Peter Moylan wrote:
> On 2015-Dec-25 22:42, whitemo...@gmail.com wrote:
> > Oops... just in case, please consider three sentences...
> >
> > "I have a seed."
> > "I have a seat."
> > "I have to sit."
>
> The third is a red herring, I believe, because in most Englishes it has
> a different vowel from the other two.
>
> In AusE vowels are shorter before unvoiced stops: [si:d] vs [sit].

You have that and phonemic length differences? I'd never be able to
learn that.

> Is that not also true in AmE?

From what the linguists say, it's true and it's the main way we
Americans tell voiced from voiceless consonants at the ends of words.

--
Jerry Friedman

[Jerry Friedman]

On Friday, December 25, 2015 at 6:30:46 PM UTC-5, whitemo...@gmail.com wrote:

If you speak that way, just about all English speakers will understand
you.

> Is it correct to say... "the voicing of the ending /d/ could be barely noticeable, and the ending /d/ could differ from an aspirated ending /t/ just by aspiration" ?

If you want to be mistaken for a native speaker, you have a very hard
row to hoe.

I don't think I normally aspirate a final /t/. I agree with what Peter
Daniels said.

--
Jerry Friedman

[Peter Moylan]

On 2015-Dec-28 14:47, Jerry Friedman wrote:
> On Friday, December 25, 2015 at 7:47:57 PM UTC-5, Peter Moylan wrote:
>> On 2015-Dec-25 22:42, whitemo...@gmail.com wrote:
>>> Oops... just in case, please consider three sentences...
>>>
>>> "I have a seed."
>>> "I have a seat."
>>> "I have to sit."
>>
>> The third is a red herring, I believe, because in most Englishes it has
>> a different vowel from the other two.
>>
>> In AusE vowels are shorter before unvoiced stops: [si:d] vs [sit].
>
> You have that and phonemic length differences? I'd never be able to
> learn that.

It's not something that you learn consciously. It's likely that the
majority of speakers are completely unaware that they're producing
vowels of differing lengths.

Anyway, from what you say below you've already learnt how to do this
kind of length difference. It's not phonemic for us in this case, of
course. That it is phonemic in some other pairs happens almost as a
side-effect.

>> Is that not also true in AmE?
>
> From what the linguists say, it's true and it's the main way we
> Americans tell voiced from voiceless consonants at the ends of words.

On reflection, it occurs to me that this has an interesting implication.
If the consonant difference is insignificant, then seed/seat is a
minimal pair that demonstrates a phonemic vowel length difference in AmE.

(But not in AusE. For us there's a clear difference in the consonants.)

[Peter Moylan]

In my experience this is a major problem for Chinese speakers who come
to Australia and have to work on improving their fluency in English. (I
include both Mandarin and Cantonese speakers. I haven't met enough
speakers of other Chinese languages to have an opinion on them. It seems
to be less of a problem for speakers of other South-East Asian
languages, e.g. Vietnamese, but I don't know why.)

In principle we could use both voicing and aspiration to produce four
different consonants in the t/d family. In practice most languages I'm
aware of make only a two-way distinction, but they make it in different
ways in different languages. (This duplicates some of what Peter Daniels
said.) In AusE the 'd' is always voiced and unaspirated, and the 't' is
almost always unvoiced and aspirated, although the aspiration can be
weak depending on what follows the 't'.

(Of course, all of this also applies to pairs like p/b and th/th, and
possibly k/g.)

What I have noticed, repeatedly, is that the native AusE speakers notice
the voicing and ignore the aspiration, while the Chinese speakers notice
the aspiration and ignore the voicing. This means that they mishear
words, because they fail to notice features that are obvious to us and
insignificant to them. There is no solution for this other than a lot
more practice in speaking and listening.

From what I'm learning in this thread, it appears that the AmE speakers
ignore both the voicing and aspiration, and instead hear vowel length as
the principal difference.

[Peter T. Daniels]

No, because "voicing" (really tense/lax) operates throughout the consonantal
system but length doesn't operate significantly throughout the vocalic system.

> (But not in AusE. For us there's a clear difference in the consonants.)

Do you _also_ non-phonemically lengthen vowels before voiceds? Or is that
precluded by phonemic, and greater, length difference elsewhere?

[Peter T. Daniels]

But we don't _consciously_ hear it. We hear the difference as voicing --
quite possibly because we know how to spell.

[Peter Moylan]

Yes, we do.

> Or is that precluded by phonemic, and greater, length difference elsewhere?

As far as I know the length difference does not depend on whether it's
phonemic, but to be sure one would have to analyse recordings.

I think, too, that we have to distinguish between "phonemic" and
"habitual". There are relatively few pairs (in AusE) where vowel length
makes a difference to the meaning. There are a great many more words
where length is not strictly phonemic, but where nevertheless getting
the length wrong could cause a listener to wonder what the word was. For
example, if you shortened the vowel in "see" [si:], the result would
sound like a foreign word.

This does not seem to be the case for all vowels. For a word like "big"
some speakers seem to have a short vowel and others have a long vowel,
and that difference does not cause any comprehension problems, so we
probably perceive [I] and [I:] to be the same vowel. On the other hand
we seem to perceive [A] and [A:] to be different vowels, and likewise
for [&] and [&:].

It's entirely possible that "lengthen vowel before voiced consonant"
accounts for the majority of our long vowels, but I'd have to go through
a couple of random pages in a dictionary to be sure of that. There are
certainly exceptions. For example, "seafood" has a long [i:] but an
unvoiced [f].

We definitely notice that AmE has fewer short vowels than AusE does.

[Peter T. Daniels]

On Tuesday, December 29, 2015 at 4:03:04 AM UTC-5, Peter Moylan wrote:
> On 2015-Dec-29 14:52, Peter T. Daniels wrote:
> > On Monday, December 28, 2015 at 7:14:57 PM UTC-5, Peter Moylan wrote:
>
> >> On reflection, it occurs to me that this has an interesting implication.
> >> If the consonant difference is insignificant, then seed/seat is a
> >> minimal pair that demonstrates a phonemic vowel length difference in AmE.
> >
> > No, because "voicing" (really tense/lax) operates throughout the consonantal
> > system but length doesn't operate significantly throughout the vocalic system.
> >
> >> (But not in AusE. For us there's a clear difference in the consonants.)
> >
> > Do you _also_ non-phonemically lengthen vowels before voiceds?
>
> Yes, we do.
>
> > Or is that precluded by phonemic, and greater, length difference elsewhere?
>
> As far as I know the length difference does not depend on whether it's
> phonemic, but to be sure one would have to analyse recordings.

If any two words are distinguished by nothing but vowel length in your
language, then vowel length is phonemic in your language.

> I think, too, that we have to distinguish between "phonemic" and
> "habitual". There are relatively few pairs (in AusE) where vowel length
> makes a difference to the meaning. There are a great many more words
> where length is not strictly phonemic, but where nevertheless getting
> the length wrong could cause a listener to wonder what the word was. For
> example, if you shortened the vowel in "see" [si:], the result would
> sound like a foreign word.

There's no such thing as "not strictly phonemic." If /si:/sounds like "see"
and /si/ doesn't (even if /si/ happens not to be a word), then length is
phonemic.

If it just sounds like someone saying "see" with some unidentifiable or
foreign accent, then it hasn't been shown to be phonemic.

> This does not seem to be the case for all vowels. For a word like "big"
> some speakers seem to have a short vowel and others have a long vowel,
> and that difference does not cause any comprehension problems, so we
> probably perceive [I] and [I:] to be the same vowel. On the other hand
> we seem to perceive [A] and [A:] to be different vowels, and likewise
> for [&] and [&:].

Four phonemes, then.

> It's entirely possible that "lengthen vowel before voiced consonant"
> accounts for the majority of our long vowels, but I'd have to go through
> a couple of random pages in a dictionary to be sure of that. There are
> certainly exceptions. For example, "seafood" has a long [i:] but an
> unvoiced [f].

You shouldn't even be able to perceive that length difference if you haven't
been trained in phonetics. (If you have, your observations aren't too useful.)

> We definitely notice that AmE has fewer short vowels than AusE does.

Hey!

Maybe you mostly meet Westerners, with their CiC and MiMiM.