The Intervocalic /t/
We will now think about a more recent development which has only affected some accents of English: this is the intervocalic /t/. By this term, we mean the /t/ that occurs between vowel sounds (or perhaps more accurately between voiced sounds).
A voiced sound is produced when the vocal cords vibrate. Therefore all vowel sounds like [a], [uː] or [aʊə] are voiced, as well as some consonants like [b], [d], [z] and [m], are voiced. (Feel your throat when you produce these sounds.) When there is no vibration, the sound is said to be voiceless or unvoiced, eg [p], [t] and [s].
If you have a voiceless sound surrounded by vowels, what you need to do is to stop the voicing for the voiceless sound before starting it again, so that perhaps some effort is required. This was exactly what happened with the OE phonemes of /θ/, /f/ and /s/ mentioned earlier. When they occur between vowel sounds, the voiced allophone was used: [ð], [v] and [z].
What happens then when we have /t/ between vowels, such as in little
or satay? In General American, the /t/
phoneme is voiced in such a context. We will simplify matters and represent the
sound as [d];
the words would then be pronounced [ˈlɪdl] and [ˈsadeɪ].
This voiced /t/
can also be heard in more informal versions of Australian,
However, in some other British accents such as Cockney
(working class
(We need to add that for the /t/ to be voiced, it needs to occur at the end of a stressed syllable. So if your accent allows /t/ voicing, it would occur in ˈletter but not in atˈtend. Stress has been indicated by the " symbol before the stressed syllable.)
We could use the argument about simplification and economy of effort to explain the use of the voiced version of /t/. However, it doesn’t explain why /p/, which is also voiceless, is not voiced in similar contexts (eg capping).
When you’re ready to take the quiz based on this topic, go to the Canvas and click on ‘Quiz’ on the left, and then on ‘Phonology’.