My question is about the sound /t/ being pronounced more like [ts] in British accent. For example, The words like Tomato, Peter, water, task, Tom, talented, take the /t/ sound is definitely not pronounced as /t/ (it can, but this variation is 100% in place) but with [ts] sound after the /t/ sound. However, there are also several words that don’t actually follow the rule of it, like attach, toast, teach that these t sounds do not go for the rule of adding [ts] sound. The third category would be in the case of the words like task or tooth where either pronunciation is fine. I wanted to ask Is there is rule for it? Thx!

  • 3
    There's no ts in any of those words in any mainstream British accent. It's possible you are hearing strong aspiration as an s sound, but that doesn't seem to fit your examples. I think you'd have to give us some audio samples (via soundcloud maybe).
    – rchivers
    Nov 30, 2020 at 18:58
  • 1
    It would be helpful if you could tell us what your native language is.
    – fdb
    Nov 30, 2020 at 19:20
  • 3
    @rchivers In most forms of English, /t/ is alveolar and does have slight affrication coincident with its aspiration. This is different from, say, German, where /t/ is alveolar and aspirated, but has little to no affrication. So while it’s true that there is no /ts/ (like in bats) in those words, their /t/ can be described as [tˢʰ]. This, of course, is equally true in words like attach and toast, and I have no idea what is supposed to distinguish the two ‘groups’ mentioned in the question. But perceiving English /t/ as [ts] is not unwarranted. Nov 30, 2020 at 20:27
  • 2
    @TKR Full assibilation is not a general feature of BrE, but it is a feature particularly associated with particular dialects, like certain Welsh dialects, Scouse (Liverpool), and particularly some Irish dialects (stereotypically Limerick, I believe), where non-initial /t/ can sometimes become something almost like [ʂ] (as famously chronicled by Frank McCourt in Angela’s Ashes in the phrase ‘gesh oush of ish’, with the severely assibilated t’s Limerick kids were taught by their parents to suppress if they wanted to sound posh). Dec 1, 2020 at 7:41
  • 2
    @TKR I’d agree there’s a higher rate of assibilation among women than men, but full assibilation – i.e., where tell and sell or hit and hiss become homophonous – is not common in my experience. Even in dialects characterised by heavy sibilant frication, /t/ and /s/ normally remain acoustically distinguishable. The only dialect I’m familiar with where complete loss of the alveolar closure is a prominent feature, is Irish English, where the result is the ‘split fricative’ I called ‘something like [ʂ]’ above ([s̠] would probably be more accurate). Dec 1, 2020 at 19:12

1 Answer 1


Affricated realization of /t/ is characteristic of (certain varieties of) London speech (Cockney). Wells (1982: 31) writes:

A common allophone of /t/ in a London accent is a heavily affricated [ts], thus [tsɑɪʔ ~ tsɑɪts] tight, [ˈpʰɑːtsi] party. To an American ear, as mentioned above, this evokes the stereotype of effeminacy, if the speaker is a man; but in London it has absolutely no such connotations, being quite ordinary.

The affrication is indeed noted to be a characteristic of stereotypical speech of gay men (Cameron & Kulick 2003: 90).

Wells continues (322–3):

[I]n broad Cockney at any rate – the degree of aspiration is typically greater than in RP, and may often also involve some degree of affrication.

Affrication may be encountered in initial, intervocalic, and final position. In the latter it is usually preglottalized, as [aʔpɸ] up, [ɑːʔtˢ] art, [ˈna˗θɪŋʔkˣ] nothing. Non-finally affrication of /p/ is rare, but examples for /t/ and, to some extent, /k/ abound, e.g. [tˢəi] tea, [kˣoʊ] call, [ˈbetˢəi] Betty (the latter differing from [ˈbetsəi] Betsy in having a shorter fricative element). Sometimes the voiced plosives, too, are affricated – particularly /d/, as [dᶻɪʔkˣ] Dick, [bædᶻ] bad.

Watson (2007) also reports /t/ affrication "is common word-initially" in Liverpool English.

As for the "rule" you mention, however, I've never heard of anything like it.

  • Thank you very much for your detailed reply! What I mean by 'the rule' i mentioned is basically about the selected manifestation of assibilation in /t/. For example, the words with sound /t/ in intervocalic position, 'attach' and 'water', /t/ sound in attach is never assibilated but /t/ sound in water is always be assibilated instead. Is there a rule for it? I have been trying to find out why this is the case. Thank you very much again and really look forward to your reply.
    – Peteryu
    Dec 2, 2020 at 16:48
  • 1
    I understand that, and I've never heard of anything like it. The examples you name are so all over the place in terms of phonetic environment, stress, and types of vowels, that I don't think one can derive a rule from them, suggesting the observation is defective in some way in the first place.
    – Nardog
    Dec 3, 2020 at 8:26
  • Thx for your reply!
    – Peteryu
    Dec 3, 2020 at 10:40
  • I agree that it’s the observation that’s defective. I have never heard anyone make any audible distinction between the /t/’s in attach and task. I don’t think the categories mentioned in the question exist. Dec 4, 2020 at 17:23
  • 1
    I would contest the observation given by @Peteryu: as a native Londoner from the East End, in causal speech I hear the affricate in attach (intervocalic with the stress on the syllable taking the t-) as well as task, Tom, toma-to (word-initial); a glottal stop would be used in water, Peter.
    – Michaelyus
    Dec 7, 2020 at 16:48

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.