Yet another batch of q's (American pronunciation)

Would like some American input on the following.

- "He's out TO lunch" or "out FOR lunch"? "We're going out FOR lunch, or going out TO lunch?"

- How do you pronounce "How're you doing". There's always some kind of a problem for me with this phrase, and most other non-native speakers. Is it "How-r ya DOOing" or no "r" sound, just "How you doin"?

- When you pronounce "str-" as in "straw", "strong", does the sound originate so far back in the mouth that it almost makes an "sh-", e.g. "shtraw", "shtrong"?

- I've been told that the vocal difference between "can"/"can't" is that "can" is pronounced with a simple 'e' as in "ken", but "can't" is pronounced with the jaw slightly wider as in "cat". The difference in sounds, phonetically, is e/ae. Is that true?

- Having some problems with the "t" sound after "n", as in: "center," "Internet," "intimate," etc. Do you aspirate the 't' here in fast speech, or is it a flap as in 'butter'?

Okay, I'm a native AmEng speaker from Massachusetts.

1. Your first particular example is actually more of an idiom, so I wouldn't suggest trying to use it in a literal sense: saying that someone is "out to lunch" means that they are clueless or "out of it". As for literal usages, I think I would prefer "for" - e.g. "We're going out for lunch" - but "to" wouldn't sound glaringly unnatural either.

2. This phrase can be subject to some variability. (By the way, are you familiar with IPA? If not, see http://en.wikipedia.org/wiki/IPA .) The most formal pronunciation would be along the lines of [ˈhao ɚ jə ˈdu:ɪŋ], and I think this would be likely to denote a literal question as to how someone is doing (as opposed to a greeting). The less formal [ˈhao ɚ jə ˈdu:ɪn] seems like it could be used as a literal question, or as a greeting; the more informal [ˈhao jə ˈdu:ɪn], "How you doin'?" would almost always be a greeting.

3. In canonical General American, and in the speech of a majority of AmEng speakers such as myself, no, those words just use a regular [s], e.g. [ˈstɹɒ:ŋ]. However, there's a fair number of speakers throughout several regions for whom the influence of the /r/ leads to a retraction of the /s/, yielding something like [ˈʃtɹɒ:ŋ] or [ˈʂʈɻɒ:ŋ]. You can hear this feature in the speech of comedian Stephen Colbert, who otherwise uses a pretty standard General American.

4. No, it's not. The difference is that "can" is usually reduced [kən] or [kn] when it occurs in real speech, whereas "can't" is always pronounced as something like [kʰænt], [kʰænʔ] or [kʰæn], or even [kʰæ~ʔ] or [kʰæ~], without vowel reduction. However, when the word "can" can also be pronounced as [kʰæn] in certain situations (e.g. when it's being emphasized), and it's true that this can occasionally lead to ambiguity. So some speakers do pronounce the unreduced "can" as [kʰɛn], which I think is the result of treating the reduced form as the default and then extrapolating from it a new, reconstructed pronunciation that's more clearly distinct from "can't". This [kʰɛn] pronunciation might not be standard strictly speaking, but I think it's become very common in AmEng, for example in a sentence like "Can we do it?" So to that extent, yes, there is a bit of truth in what you've heard, even though I do think the assertion, as you've put it, is misleading.

5. It isn't a question of "aspirating" the /t/, it's a question of *pronouncing* the /t/. /t/ is generally only aspirated at the beginning of a word or at the beginning of a stressed syllable; it's perfectly possible to pronounce an (unstressed) /t/ as a plosive [t] without aspiration, and although I don't think there are any speakers who would naturally *aspirate* the /t/s in those words, but there are many speakers (like me) who would, nonetheless, pronounce them as plosives. My own treatment of /nt/ words seems to be more conservative than the North American average; I often use a plosive [nt] in "center", I usually use a plosive /nt/ in "Internet", and I always use a plosive /nt/ in "intimate". As a general rule, it's common in North American English for /nt/ to be reduced to [n] or to a nasalized flap [ɾ~]; you might see a transcription [nɾ], but I don't think that's an accurate description. The problem is that there seems to be a rather complex continuum of allowability: the /nt/ is usually reduced to [n] in the word "twenty", very often so in words like "center" or "county", and often in words like "Internet"; but there are some words, as examples of which I would suggest "banter", "hunter", "lantern" and "intimate", where I think an /nt/ reduction would sound quite unnatural. If I were learning AmEng, I think I would err on the side of unreduced [nt]: an unreduced [nt] never sounds wrong (although I suppose it might, on occasion, sound a bit too formal; nonetheless this could be written off as British influence), whereas I think using a reduced [n] in a word like "lantern" might strike a lot of AmEng speakers as quite unnatural.

Sorry if I seem a bit rambling above; you could ask for an expansion or clarification on anything if you like.

Thank you very much Lazar. Very informative and comprehensive answers.

Everything's clear except the can/can't difference. In essence, are you agreeing that the difference there is between a short-e and wide-a? Sorry, I don't have access to that phonetic alphabet.

On the /nt/ sound. I completely agree. But it's hard to make a plosive 't' when you're talking fast and it's not in the beginning of a word, in words like "intimate" and "hunter." So, is there a 3rd /t/, which is rarely discussed, where you simply press the tongue on the alveolar ridge but DON'T aspirate, as in "star"? Is it exactly the same /t/ in those examples?

Thanks again, much appreciated.

<<In essence, are you agreeing that the difference there is between a short-e and wide-a?>>

No. What I'm saying is that we have this:

"can" - reduced with schwa, OR unreduced with /æ/ ("flat" or "wide" a)
"cant" - always unreduced with /æ/

So no, going strictly by standard usage, "can" can have either a schwa or a flat a, but not a short e.

*But*, if we want to examine popular semi-standard usage, it has become common for some speakers to replace the flat a in unreduced "can" with a short e, at least in some contexts, meaning that a sentence like "Can we do it?" might sound like "Ken we do it?" On the other hand, I don't think that kind of vowel change would sound so natural in the sentence "Yes we can!" If I were learning AmEng, I think I would just stick to the standard schwa or flat A.

<<But it's hard to make a plosive 't' when you're talking fast and it's not in the beginning of a word, in words like "intimate" and "hunter." So, is there a 3rd /t/, which is rarely discussed, where you simply press the tongue on the alveolar ridge but DON'T aspirate, as in "star"?>>

Let me make it clear that "plosive" is not the same thing as "aspirated", because I think people sometimes confuse them. A plosive can be either aspirated or unaspirated; "plosive", here, simply means that the /t/ is pronounced as a [t], and not as a flap.

Let me try to illustrate some examples of /t/ allophony in American English. (Unfortunately, it is somewhat complex.) At the beginning of a word, or at the beginning of a stressed syllable (primary or secondary stress), /t/ is both plosive and aspirated:

"tune" - [ˈtʰu:n]
"attain" - [əˈtʰeɪn]
"military" - [ˈmɪləˌtʰɛɚi]
"tomato" - [tʰəˈmeɪɾoʊ]

And of course, we have the exception for words like "star", which you're aware of:

"star" - [ˈstɑɚ]

When /t/ is in intervocalic position and it's not at the beginning of a word or a stressed syllable, it's pronounced as a flap, which is neither plosive nor aspirated:

"butter" - [ˈbʌɾɚ]
"Carter" - [ˈkʰɑɚɾɚ]

When "to" occurs between vowel sounds in the middle of a sentence, it's generally pronounced with a flap:

"go to the store" - [ˈgoʊ ɾə ðə ˈstɔɚ]

At the end of a word in isolation, /t/ is typically pronounced as an unreleased plosive [t]: there's a buildup of air as if you were going to make a [t], but you don't release it. (However, you could pronounce it merely as an unaspirated plosive [t], which might sound a bit more formal, but not totally unnatural.)

"cat" - [ˈkʰæt]

When /t/ is at the end of a word and it's followed by another word starting with a vowel, it's generally pronounced as a flap:

"The cat is happy." - [ðə ˈkʰæɾ ɪz ˈhæpi]

Now, when we come to /nt/, the traditional pronunciation uses [nt] with a [t] that is plosive but unaspirated, just like the /t/ in "star" - this is the rarely discussed third type that you were talking about.

"center" - [ˈsɛntɚ]
"hunter" - [ˈhʌntɚ]
"intimate" - [ˈɪntəmət]

The thing is that in modern American English, these instances of /nt/ are sometimes (not always) reduced to [n]. Here are some examples based on my own intuition as an AmEng speaker:

Usually reduced to [n]:

"twenty"

Often reduced to [n]:

"plenty", "center", "county", "Internet",

any verbal form ending in "n't", like "don't", "can't", "won't", "isn't", "wouldn't", when followed by a vowel

Rarely reduced to [n]:

"hunter", "banter", "intimate", "lantern"

As I noted in my previous post, if I were learning AmEng, I would err on the side of [nt], because unreduced [nt] will never be wrong (at worst, it might sound a little formal), whereas using a reduced [n] in a word like "banter" or "lantern" might strike people as quite unnatural.

Thanks a lot Lazar. Much appreciated. I understand it now.

Just to make sure: in the "hunter"/"banter" non-reduced words, the /t/ is like the one in "star", but not in "ten", correct?

Yes.

I would say something here, but the above is a good explanation of the matter within the context of North American English, and my own dialect is just too damn weird to be suitable with respect to providing guidance for learners... (even if I have quite the tendency to do just that).

Lazar, I always enjoy reading your posts because you explain things so clearly.

I have a question about your cat example. The way I pronounce the 't' in "cat" is more like a glottal stop than [t], even when it's followed by another word starting with a vowel. Do you know if this pronunciation is found just in the Wisconsin area?

Thanks! ;) Using a glottal stop for final /t/ is definitely within the realm of North American English variation - it's a feature of the speech of New Britain, CT, for example - although I don't know how common it is to use a glottal stop when the following word starts with a vowel. Travis could certainly provide you with more information on Wisconsin dialects.

I have some more questions on the topic of /t/.

In phrases like "Had to", "supposed to", "allowed to", as well as words like "after" and "victim", is it really the case in American English that you're supposed to aspirate the "t" in fast speech? When you're talking fast, you don't have time to aspirate, so it comes out as a simple 'tap' rather than an aspirated /t/.

Today at work, I heard two American-born people say "had to" in their speech where the word "to" was not an aspirated /t/, but more like a blunted quick'n'dirty tap on the ridge, same as in "star".

Also, I said the word "victim" today in conversation with someone where I aspirated the "t", and the person looked at me funny and repeated the word but with a 'fast tap' "t" without the aspiration.

<<Today at work, I heard two American-born people say "had to" in their speech where the word "to" was not an aspirated /t/, but more like a blunted quick'n'dirty tap on the ridge, same as in "star".>>

What you're referring to isn't a tap, though, in the phonetic sense, but rather an unaspirated [t]. A tap or flap, transcribed as [ɾ], is a sound very similar to a Spanish or Greek r, which really only occurs between vowels in American English.

<<In phrases like "Had to", "supposed to", "allowed to", as well as words like "after" and "victim", is it really the case in American English that you're supposed to aspirate the "t" in fast speech? When you're talking fast, you don't have time to aspirate, so it comes out as a simple 'tap' rather than an aspirated /t/.>>

In those cases, you would use a softer unaspirated [t]. Aspiration can only occur at the beginning of a word or stressed syllable, so it wouldn't be natural to use it in "victim". As for "to": it could be pronounced aspirated if you were speaking deliberately or emphatically, but generally, a sequence like "had to" would use a soft, unaspirated /t/.

Thanks Lazar. That's what I meant.

By the way, when I say 'tap', I don't mean 'flap' ;-)

A flap is the 't' in butter, I know. But a tap, to me, is the soft unaspirated /t/ (which I distinguish from an aspirated /t/).

Yes, that's what I thought. The problem is that the words "flap" and "tap" are used basically interchangeably to describe the voiced sound transcribed as [ɾ]. My (sort of arbitrary) practice is to describe the English sound in "butter" as a flap and the Spanish sound in "pero" as a tap, thus not using "tap" at all in discussions of English.