Можно ли пропустить букву t в словосочетании but when?

[krutoichuvak]

Или ее обязательно произносить?

[Andrea]

krutoichuvak,
Можно.

[Chaika]

You can do this. But in AmE there is usually something like a glottal stop in place of the /t/.

If the following word is a vowel the /t/ is replaced by a /d/. This a general rule, e.g., water, slated, wetting, later, latter, butter, sitter, seated. these are all /d/.

[Dragon27]

It's called No audible release.

[Andrea]

This a general rule, e.g., water, slated, wetting, later, latter, butter, sitter, seated. these are all /d/.

I think, yours is a different case as it describes change of "t" in an intervocal position.

No audible release.

I guess the TS's example may be the case of elision, or just "missing sounds", put simply:

In linguistics, an elision or deletion is the omission of one or more sounds (such as a vowel, a consonant, or a whole syllable) in a word or phrase. However, these terms are also used to refer more narrowly to cases where two words are run together by the omission of a final sound.[1] An example is the elision of word-final /t/ in English if it is preceded and followed by a consonant: "first light" is often pronounced "firs' light" (/fɜrs laɪt/).[2] Many other terms are used to refer to specific cases where sounds are omitted.

https://en.wikipedia.org/wiki/Elision

[Dragon27]

I guess the TS's example may be the case of elision

Is it elided, though? The elision of T (and D) is common when it's between consonants, here we have a different phonetical environment.

[Andrea]

Is it elided, though?

I think this is pretty much the same as "black and white", which actually sounds as "blacken white":
https://dictionary.cambridge.org/pronun ... -and-white

The elision of T (and D) is common when it's between consonants, here we have a different phonetical environment.

I'm sorry, I'm afraid I don't get how exactly it is different. Will you be so kind as to tell us how?

[Dragon27]

I'm sorry, I'm afraid I don't get how exactly it is different. Will you be so kind as to tell us how?

it's in the citation you've provided

if it is preceded and followed by a consonant

Other examples of consonant clusters where T and D are easily elided (from A Handbook of Pronunciation by L. Canepari): mostly, handsome, postman, perfectly, next day, mashed potatoes, posts, tests, text selection.
In "but when" T is preceded by a vowel.

"and" being pronounced as /ən/ is just a regular reduced form, it's barely ever pronounced fully.

Relaxed pronunciation of "but" can also be considered a reduction (at least, the vowel is reduced to /ə/), but I don't think T is elided, it's either not audibly released or replaced with a glottal stop.

[Andrea]

it's in the citation you've provided

What are you talking about?

[Dragon27]

What are you talking about?

The difference in phonetic environments. Haven't I clarified it enough in the subsequent passage?
1) T between consonants (first thing)
2) T between a vowel and a consonant (but when)

[Andrea]

Dragon27,
I'm sorry, but doesn't my quotation read

these terms are also used to [i]refer more narrowly to cases....[/i]

Thus, the term "elision" refers to any sound/sounds omissions regardless of the preceding sound (vowel/consonant etc.) Which is the very first sentence of the quotation I've provided.

Next, why neglect "and" as a case of elision while doesn't it only illustrate the phenomenon perfectly well, but also happens as often as with other elision-prone words?

[Dragon27]

Thus, the term "elision" refers to any sound/sounds omissions regardless of the preceding sound (vowel/consonant etc.)

Which doesn't mean that any sound can disappear in any conditions. We know that in English T/D easily disappears between consonants, so it's no wonder that T in "first thing" is subject to this, but I don't see any rule describing the elision of T in "but when". In my original quote I specifically mentioned the rule:

The elision of T (and D) is common when it's between consonants, here we have a different phonetical environment.

to which you answered:

I'm afraid I don't get how exactly it is different

Hence, my explanation of how the phonetic environment of T in "but when" is different.

Next, why neglect "and" as a case of elision while doesn't it only illustrate the phenomenon perfectly well, but also happens as often as with other elision-prone words?

Because it's conceptually different from a general elision in a variety of words with similar phonetic environment in that it applies only to a specific word when it's unstressed. Besides, I didn't neglect it, I've accepted it, but I've questioned the presence of elision in "but when" in this post:

Relaxed pronunciation of "but" can also be considered a reduction (at least, the vowel is reduced to /ə/), but I don't think T is elided, it's either not audibly released or replaced with a glottal stop.

[Andrea]

Which doesn't mean that any sound can disappear in any conditions.

No, they can't. Did anyone claim the opposite? Yet, a sound can definitely be elided when preceded by a vowel. Consider the examples:
'I don't know' /I duno/ , /kamra/ for camera, and 'fish 'n' chips' are all examples of elision.
https://www.teachingenglish.org.uk/prof ... n%20speech.

Because it's conceptually different from a general elision in a variety of words with similar phonetic environment in that it applies only to a specific word when it's unstressed.

Does that mean "and" is not elided? Isn't it still the same phenomenon, though regular?

Relaxed pronunciation of "but" can also be considered a reduction (at least, the vowel is reduced to /ə/),

Following your logic, "but" is also not an example of reduction due to being commonly pronounced that way. Or is it?

glottal stop

I'm sorry to say this again, but at least according to Chaika's post in the thread glottal stop takes place in an intervocal position (which is not the case with "but when")

[Dragon27]

No, they can't. Did anyone claim the opposite? Yet, a sound can definitely be elided when preceded by a vowel.

Some sounds can in some words. Does T get commonly elided (i.e. completely dropped) when preceded by a vowel? It does between consonants.

Consider the examples:
'I don't know' /I duno/ , /kamra/ for camera, and 'fish 'n' chips' are all examples of elision.

"don't" is a frequent auxuliary verb, which has its peculiar set of reduced variants. In "camera" we have a vowel (schwa) elided in an unstressed syllable, not sure why this example is here (is it preceded by a vowel? it seems to be surrounded by consonants). In 'fish 'n' chips' we have the reduced "and" again.
Didn't really get the point of these examples. Are you telling me that I deny the appellation of "elision" in these cases? I just separate different patterns of elision (of different nature) to avoid drawing false conclusions. Otherwise, we can fall prey to the same trap, that the inventor of the "ghoti" respelling of the word "fish" did.

Does that mean "and" is not elided? Isn't it still the same phenomenon, though regular?

"and" is not elided, some sounds in "and" are elided (dropped) or reduced (the vowel turned into schwa). The phenomena are markedly different, in one case we have a consonant cluster simplification across a multitude of similar sound patterns, in another we have a commonly used word, which still gets recognized in its simplified form, become reduced due to its constant usage.

Following your logic, "but" is also not an example of reduction due to being commonly pronounced that way. Or is it?

Can't see what kind of logic you've applied here (care to elaborate?). "But" in an unstressed position is obviously reduced from its full, stressed form /bʌt/.

I'm sorry to say this again, but at least according to Chaika's post in the thread glottal stop takes place in an intervocal position (which is not the case with "but when")

According to Chaika's post, in intervocalic position T is replaced with D:

If the following word is a vowel the /t/ is replaced by a /d/. This a general rule, e.g., water, slated, wetting, later, latter, butter, sitter, seated. these are all /d/.

It's not entirely correct (the sound is actually an alveolar flap, it happens not only between vowels, there are some important constraints based on stress position), but it'll do for a first glance at this typically American phenomenon.
In British English, though, the glottal stop replacement can take place in an intervocalic position:

Among speakers of Britain, especially younger ones, glottal replacement of /t/ is frequently heard in intervocalic position before an unstressed vowel. It is most common between a stressed vowel and a reduced vowel (/ə/, /ɪ/):

getting better [ɡɛʔɪŋ bɛʔə(ɹ)] (in GA, this is [ɡɛɾɪŋ bɛɾɚ]);
societies [səˈsaɪəʔiz], detail [ˈdiːʔeɪl] (these are slightly less likely to be glottalized).

https://en.wikipedia.org/wiki/T-glottalization
The same article also talks about glottalization before a consonant:

In RP, and in many accents such as Cockney, it is common for /t/ to be completely replaced by a glottal stop before another consonant, as in not now [nɒʔnaʊ] and department [dɪpɑː(ɹ)ʔmənʔ].

Pretty sure, it would also happen with "but when".
Americans don't usually do intervocalic glottalization, but it can still happen in some special cases (from the same article):

(Less commonly) across word boundaries.
"Right ankle" [raɪʔ‿æŋkəl]
"That apple" [ðæʔ‿æpəl]

Which is marked as "less common". Also:

Word finally or before a syllabic /n/
Latin [læʔn̩],
Important [ˌɪmˈpɔɹʔn̩t]

According to Canepari (from the above-mentioned source) it can happen before consonants, with examples like "Chutney", "lately", "Atlantic" (the first T).
Anyway, this is a general phonetic change, but since we're considering a singular word (a commonly used preposition), it may have its own separate sound changes that we're looking at here.

[Andrea]

It does between consonants.

Not only that, and you keep swinging between two opposing opinions. Please consider deciding upon the issue.

In "camera" we have a vowel (schwa) elided in an unstressed syllable, not sure why this example is here (is it preceded by a vowel?

"Camera" just demonstrates that "sound environment" is irrelevant concerning elision.

Are you telling me that I deny the appellation of "elision" in these cases? I just separate different patterns of elision (of different nature) to avoid drawing false conclusions

What I'm trying to tell you is that elision is a broad term, or an umbrella term, if you like. What you are doing is narrowing it to a specific sub-category claiming the latter to be the only possible case of elision, or, vice versa, claiming another sub-category not to be elision at all.

Can't see what kind of logic you've applied here (care to elaborate?)

The one described just above.

"But" in an unstressed position is obviously reduced from its full, stressed form /bʌt/.

So why omission of"d" in "and" is not an example of elision, if you please?

Pretty sure, it would also happen with "but when".

I guess you should at least specify the accent you're applying the term "glottal stop" to here.

[Dragon27]

Not only that, and you keep swinging between two opposing opinions.

Between which exactly?

"Camera" just demonstrates that "sound environment" is irrelevant concerning elision.

How is it irrelevant? It's irrelevant to the question "what's an elision?", but not to "where does elision normally happen?".

What I'm trying to tell you is that elision is a broad term, or an umbrella term, if you like. What you are doing is narrowing it to a specific sub-category claiming the latter to be the only possible case of elision, or, vice versa, claiming another sub-category not to be elision at all.

I claim no such thing.

So why omission of"d" in "and" is not an example of elision, if you please?

I said that it's not an example of a specific kind of elision (i.e. the dropping of T/D between consonants).
One more time:

The elision of T (and D) is common when it's between consonants, here we have a different phonetical environment.

i.e. it is common for an elision of T or D to happen, when T is between consonants. In "but when" T is not between consonants. If it were, I would've easily accepted the claim that it's elided. The elision of T after a vowel and before a consonant is pretty unusual, not something that normally happens in the English language. Maybe it can happen for some specific words from the category I mentioned, i.e. commonly reduced "auxiliary" words (in which it probably happens regardless of the phonetic context). Does it happen specifically for the word "but"? I can't seem to find any information about that specific reduction, I can only find stuff about glottalization and no audible release to account for the lack of phonetic T (and what I can personally hear in the actual speech examples seems to corroborate the theory). That's why I'm questioning the possibility of elision in this case.

I guess you should at least specify the accent you're applying the term "glottal stop" to here.

"Glottal stop" is the same in any language, it's not a language specific term. The quote in which I used the term relies on the statement (just before the quote) about the preconsonantal glottalization in RP (and other British accents).

[Andrea]

How is it irrelevant? It's irrelevant to the question "what's an elision?", but not to "where does elision normally happen?".

Ok. But I still have to make it clear (for both myself and the audience.)

So, in the end, the discussion boils down to the following:

I said that it's not an example of a specific kind of elision (i.e. the dropping of T/D between consonants)

A "not specific type" of a phenomenon is still a phenomenon, isn't it? Therefore, omission of "d" in "and" is actually an example of elision, isn't it?

In "but when" T is not between consonants. If it were, I would've easily accepted the claim that it's elided. The elision of T after a vowel and before a consonant is pretty unusual, not something that normally happens in the English language

Yes, I agree that dropped "d" in "but when" is most likely glottal stop. At least we have answered the TS's question) (where are they, btw?) I have to read more on glottal stop, though.

"Glottal stop" is the same in any language, it's not a language specific term.

Yes, there's one in Arabic. I struggled to find out it's an independent consonant (not an allophone)🎉

The quote in which I used the term relies on the statement (just before the quote) about the preconsonantal glottalization in RP (and other British accents).

Thanks for telling. I'll bear that in mind.

. Does it happen specifically for the word "but"? I can't seem to find any information about that specific reduction

Doubt not. "But" is definitely reduced in connected speech, classic weak form.

[Dragon27]

A "not specific type" of a phenomenon is still a phenomenon, isn't it? Therefore, omission of "d" in "and" is actually an example of elision, isn't it?

Sure, you can call it that way. The sound /d/ in "and" is dropped (elided, deleted) in normal speech (even before a vowel - "you and I"), except when the word is emphasized.

Yes, there's one in Arabic. I struggled to find out it's an independent consonant (not an allophone)

Well, in some languages it exists as an independent phoneme, in others it's just a phone arising in some contexts or special interjections, but no matter what role it plays in a particular phonological system, it's still a consonant sound.

Doubt not. "But" is definitely reduced in connected speech, classic weak form.

Yeah, what I meant to say is that its reduced form is commonly represented (in dictionaries, for example) as /bət/ without any specification of what happens to the T sound. In Canepari's work in the examples of reduction we also have the glottalized option ([bəʔ]), as well as some assimilated ones (like [bəp] before B, etc.), but these all just follow from general considerations (glottalization of T before consonant, assimilation of T to a following consonant). And since I can't find anything about true elision, resulting in [bə'wɛn] (' is stress), I would only consider these options as possible for the time being.