This has been bothering me for a long time. I know there is “stop t”, as we find in the word:

  • wait

But I’m still not sure about the /t/ before the word the in these phrases:

  • lift the cat
  • trust the process

Is the t a stop t? It sounds strange or different to me.

  • If I understand you, yes. What else might it be? – WS2 Aug 19 '17 at 7:22
  • 3
    @Araucaria In most cases, it's not even there at all because of where it falls between the two fricatives: too complex a consonant cluster for common speech. – tchrist Aug 19 '17 at 14:51
  • 4
    Consonant clusters like /ftð/ in lift the duck are usually simplified in ordinary speech. The easiest way to simplify /ftð/ is to eliminate the stop and let the fricatives flow into one another, resulting in /fð/. If it's not there, it can't be a stop at all. The technical name for this phenomenon is pronounced /fæspitʃrulz/; is this an example of what's being asked? – John Lawler Aug 19 '17 at 15:20
  • 1
    @WS2 It could be a dental stop instead of an alveolar one caused by anticipatory assimilation. It can't be a glottal stop here because there's no preceding sonorant. It can't be voiced because it's not intervocalic. It's unlikely to be an alveolar plosive because it's difficult to transition from an alveolar sound to a dental one. – Araucaria Aug 19 '17 at 15:32
  • @tchrist, John Yes, quite so. In the examples the OP has given the /t/ can be elided completely (if the preceding consonant's voiced though, then it'll have to stay.) So out of the five or six possibilities commonly available for /t/, there's three available here 1) alveolar elision, 2) a dental stop, 3) a normal alveolar stop (but this third one's quite unlikely). – Araucaria Aug 19 '17 at 16:03

The English phoneme /t/ is most often realised as an unvoiced alveolar plosive, [t]. The term plosive is just another way of saying that the consonant is a stop. To make this sound the tip and blade of the tongue make a seal around the alveolar ridge. That is the little shelf-like part of your mouth behind your upper teeth. The vocal folds (also known as vocal cords) are open and not vibrating. Air is forced up from the lungs and compresses behind this seal. Then the tongue suddenly leaves the ridge, breaking the seal causing shock heard as audible plosion and the compressed air is released.

However, /t/ has some other common allophones (different realisations which also count as a /t/). For example it may be realised as a voiced alveolar tap, [ɾ], when occurring intervocalically (i.e. between two vowels). For this realisation the vocal folds continue to vibrate for the whole vowel-/t/-vowel sequence and the tip of the tongue strikes the alveolar ridge in a percussive movement without actually blocking of the airstream for long enough for there to be any compression of air. The /t/ in the Original Poster's examples cannot be this sort of /t/ because it does not occur between two vowels.

Another realisation of /t/ in English can be a glottal stop, [ʔ]. This is when the vocal folds come together to block the air coming up from the lungs. In other words it is caused by a closure of the glottis. However, this can only be used as a realisation of /t/ in English after voiced sounds. In the Original Poster's example there is no voiced sound preceding the /t/. So this is not an option here. (It would be in a string like paint the cat, though: [peɪnʔ ðə]).

Before dental fricatives—the sounds we make using our tongue on the back of our upper teeth which we represent in English writing using the th digraph—it is difficult to move the tongue quickly enough from the alveolar ridge to the back of the teeth. In other words it is difficult to get a smooth transition from an alveolar articulation to a dental one. For this reason we often see a special allophone of /t/ before dental sounds. This involves blocking off the airstream by making a seal using the tip and blade of the tongue on the back of the upper teeth instead of on the alveolar ridge. We call this a dental t. The IPA symbol for this uses a little tooth diacritic under the 't', [ t̪]. When we do this we usually cannot hear the release of the [ t̪] as it is masked by the following fricative.

However, there is another likely candidate outcome for the realisation of the Original Poster's examples. The /t/ here occurs at the end of a word (i.e. at the end of a morpheme) and is surrounded by consonants on both sides. In addition, like /t/, the preceding consonant is voiceless. This means that we have the ideal conditions for eliding the /t/ altogether. In other words we don't need to pronounce it at all. This is equally as likely as using a dental [t] here. So we might see the Original Poster's examples pronounced as:

  • /lɪf ðə kæʔ/ 'lif the cat'

  • /trʌs ðə prəʊses/ 'trus the process'

  • 1
    I like that intervolcanically bit. :) There's also a denti-alveolar possibility by the by. – tchrist Aug 19 '17 at 16:03
  • A lot of us just have [ˌpʰẽʔ.ð̪̺ˈkʰæ̞ʔ] for paint the cat. – tchrist Aug 19 '17 at 16:15
  • @tchrist Sure thing. I reckon there's quite a few possibilities there ... :) (What's the bit after the thorn symbol there, btw?) – Araucaria Aug 19 '17 at 16:37
  • The eth has two combining diacritics: ‭◌ ̺ U+033A COMBINING INVERTED BRIDGE BELOW to mean an apical tongue position and ◌ ̪ U+032A COMBINING BRIDGE BELOW to indicate a dental placement. Unfortunately when those mated up they are visually the same as ◌ ̻ U+033B COMBINING SQUARE BELOW which would mean a laminal tongue position so the opposite of what I meant by saying it’s apical. I wanted people to realize it was not denti-alveolar nor with the flat of the tongue; perhaps there are better ways of saying that though. Main points: elision, glottalization, nasalization, monophthongization. – tchrist Aug 19 '17 at 16:52
  • You're using /phonemic/ solidi where I’m sure you’re only talking about non-phonemic phonetic allophones in the last bullet-pair, and so those should be [brackets]. I also wish you wouldn't “truss” (sic) the cat. :) – tchrist Aug 20 '17 at 16:38

Connected Speech Processes

The OP's question is dealt with by what is called connected speech processes. Below is list of the most common features for final /t/ in English. 4) is the one being asked about here.

For English:

These are from a list for the /t/ at the end of a word, they are said to be assimilated [basically, become completely attenuated]

1) *Word-final /t, d, n/ become bilabial before bilabial consonants.

[my example: dirtbag, street boy]

2) *Word-final /t, d, n/ become velar before velar plosives.

[cat gut or bright cat] [velar plosive here: the /g/ as in gut, and the /k/ as in cat

3) *Word-final /t, d, s, z/ become post-alveolar before the post-alveolar phonemic approximant /j/, with possible disappearance of /j/.

[[also written as ⟨ɹ̠⟩ in IPA, and ⟨ɹ⟩, ease of typing. [this is the r in red by the way]] [[the transcription of this post-alveolar approximant is discussed at length on Wikipedia, but not worth going into here.] [ light room or street route]

Also for the /t/: there is elision [omitted]:

4) Word-final /t, d/ preceded by a consonant and followed by a consonant in word-initial position are omitted.

[lift the cat or trust the process]

[full list ] from: Table 2 summarizes all the processes affecting pronunciation in rapidly articulated connected speech that we have discussed in this section.1

I thought it might help the OP to have a simplified list. Also, here is a longer article on the subject along with what is called co-articulation, which is explained in some detail in the article: connected speech and co-articulation

If my examples contain errors, I'd be glad to fix them. Though I am a linguist, I am not a phonetician and, therefore, am probably prone to slips between my ear and my writing.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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