See this word:

doctor /ˈdɑːktər/, the /t/ in this case seems to be like a mild aspirated T (that is there may have a bit air coming out of your mouth) Source.

But expected /ɪkˈspektɪd/, the /t/ in this case seems to be like an unaspirated T (that is there is no air coming out of your mouth) Source

& attractive /əˈtræktɪv/, the /t/ in this case seems to be like an unaspirated T (that is there is no air coming out of your mouth) Source

Is there any source that mentions about this issue?


I recorded my own voice of unaspirated T & aspirated T in attractive, please check it

unaspirated T: https://drive.google.com/file/d/0B9o9iLQ2SktLZjVzVG4taHBsUVk/view

aspirated T: https://drive.google.com/file/d/0B9o9iLQ2SktLWnR5SGM1Tlh4NVk/view

1 Answer 1


I think you may be imagining differences that don't exist in the minds of native speakers. The actual amount of air released there will vary with every utterance and occasion. No one hears these as one bit different from one another.

Or perhaps you have not heard enough examples.

If your first language considers the two sounds to be completely different phonemes instead of merely two allophones of the same phoneme, then you may be being overly sensitive to something you should not "notice", since native speakers do not do so.

  • I updated my question & I recorded my own voice of pronouncing unaspirated T & aspirated T of "attractive". I think they are very different. Nonnative sees things differently from native. I am sure they are very different.
    – Tom
    Apr 12, 2016 at 5:11
  • We think that /l/ and /r/ are very different. Lots of people whose native language is Japanese can't tell them apart. For /kt/ before an unstressed syllable, we don't notice whether or not the /t/ is mildly aspirated. Apr 12, 2016 at 11:17
  • 1
    @Tom Do you understand what allophones are? They’re any of several different ways that a single "mental" phoneme can be physically pronounced. You can swap around allophones in a word's sounds but until you instead change to a different phoneme. Until then, the listener will not "hear" a different word being said. That's why this does not matter.
    – tchrist
    Apr 12, 2016 at 11:48
  • 2
    There are different kinds of allophones, though. If they are in free variation, then it really does not matter at all which one a non-native speaker uses. If they are conditioned allophones, it's important to get the right one. Aspirated and unaspirated /t/ are often considered allophones, but a non-native speaker who uses unaspirated [t] in a word like "time" is at risk of being misunderstood as saying "dime," because this is not a position where unaspirated [t] normally occurs for native speakers. I think Tom is asking if there are any conditions that affect the level of aspiration here.
    – herisson
    Apr 12, 2016 at 14:13
  • 2
    I haven't actually seen this written down anywhere, but this is my perceptions: Aspirated and unaspirated /t/ are in free variation at the beginnings of unstressed sylables after /k/, as well as in some other positions. They are not in free variation after /s/ or at the beginnings of stressed syllables. Try listening to some other recordings of those words, and see if they have the same aspiration. Apr 13, 2016 at 17:30

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.