And why there's a superscript ə? just found this on the dictionary.cambridge.org ...enter image description here

  • Did you try clicking on the loudspeaker symbols at antonym to hear how it sounds?
    – Andrew Leach
    Dec 22, 2015 at 11:10
  • Yeah, I tried.. but couldn't get it...
    – user152435
    Dec 22, 2015 at 11:12
  • 1
    Pronunciation and related themes are on-topic. We have a number of technical questions about physiological necessities for getting sounds right..
    – Andrew Leach
    Dec 22, 2015 at 11:45
  • @A It's vaguely interesting that the "British" and "American" interpretations of the identical US pronunciation are different. One uses /-t̬ən.ɪm/ and the other has a superscript [which I can't do in a comment].
    – Andrew Leach
    Dec 22, 2015 at 12:13
  • @AndrewLeach, in the American transcription, the t with a little under cup seems to be for a flap, and the last syllable starts with the vowel instead of n. These are both very strange ideas about the pronunciation. Maybe that transcription is simply mistaken.
    – Greg Lee
    Dec 22, 2015 at 16:31

4 Answers 4


This is a very helpful question. Many members here are unsure about what the superscript schwa indicates in dictionaries, and there is a lot of inaccurate information given about it.

The superscript schwa indicates that there are two different pronunciations in common use. One uses a schwa here the other doesn't. So the American version of antonym can be pronounced in the following ways:

  • 'æn.tn.ɪm
  • 'æn.tə.nɪm

The superscript schwa is not a "small" schwa sound. Neither does it indicate that people disagree about whether there is actually a schwa sound present. It is also not a non-syllabic vowel. As a rule of thumb, whenever we see a syllabic /n, m, l, r/ there will be an alternative pronunciation available with a preceding schwa occurring in the nucleus of the syllable.

Some tips for the production of schwa:

(1) Stand in front of a mirror. Don't close your teeth or open your mouth, just relax your face. Make a sound as if it's coming from your throat or chest (in reality it will be coming from your vocal folds). This should be a schwa sound. In the mirror you should not be able to see your face move at all. If you recorded a video of you practising schwa, but with no sound, we would not know when you were making a sound and when you were silent, because your tongue, jaw and lips - and your face in general - should all be relaxed and not moving at all.

(2) Try to make the sound /b/ as in the word big, but just /b/ on its own. Now try the sound /d/ as in dog and then the sound /g/ as in girl. Do this two or three times. When we say these sounds on their own, we automatically put a little vowel on the end - we have to because they are voiced. If you said the sounds correctly, then you probably said /bə/, /də/ and /gə/. The little vowel that you made by accident after the consonant is a schwa. This is because you were not trying to make any special vowel there.

Schwa only occurs in unstressed syllables. The reason we make schwa like this is because we need to make unstressed syllables shorter than other ones in English. We need the other stressed syllables to be longer and to stand out. The schwa is very quick to make because we do not need to move any of the articulators (the parts of our mouth that we use to make consonants or change the sounds of vowels). If we make a big articulation, a big movement of our mouths, like we do for /æ/ in cat, we need to move our articulators a long way. For /æ/, for example, we have to spread our lips very wide, and drop our jaw very low and move the 'front' (that means the middle) of our tongue so it raises slightly up towards the roof of our mouth. This all takes a lot of time. Because of this, /æ/ is actually quite a long sound, even though it belongs to the so-called 'short vowels'. For a schwa you do not need to move anything! In conclusion then, what you need to do to make a good schwa sound is: nothing!

  • 1
    Your paragraph "(2)" is more confusing than helpful. Your shouldn't use slashes when you're talking about sounds rather than phonemes, for one thing. For another, the position of the tongue for the [b,d,g] in big, dog, girl will be determined by the following vowel in those words -- it will not be a schwa. If it were actually a schwa, you would be able to hear a central on-glide in the following vowels. When people attempt to pronounce a [b,d,g] in isolation, they may use a schwa tongue position, but there is no guarantee of that.
    – Greg Lee
    Dec 22, 2015 at 14:32
  • @GregLee In isolation they may not. But having had several hundred (at a conservative estimate) students over the years, nearly everybody does. But your right, not everybody. Of course it's the saying the sound in isolation that makes the following sound likely to be a schwa. It's usually a shcwa because there's no following vowel intentionally articulated. Dec 22, 2015 at 14:45
  • @GregLee Us phoneticians don't use // and [] the same way as you. We use // for broad transcriptions that give little or no other detail about the sound other than which language-specific phoneme is occurring. We use [] for narrow transcriptions where we give more phonetic detail. When we do this we don't use language specific symbols. So we would have to use ɐ instead of ʌ for example for the vowel in cup. Dec 22, 2015 at 15:03
  • 2
    How can you have non-phonemic transcriptions that give "no other detail about the sound other than which language-specific phoneme is occurring"?
    – Greg Lee
    Dec 22, 2015 at 15:27

What it means and what it is are different matters. Making the schwa a superscript here apparently means the sound is non-syllabic. While (except for some diphthongs) you ordinarily have one syllabic sound per syllable, which is the "nucleus" of the syllable, there are some difficult or marginal cases. Syllabic sounds are generally vowels, but liquids like l and r and nasals like n and m can also be syllabic. Syllabic sounds are longer in duration than non-syllabics and acoustically have a "steady state" -- there is a period of time, a few milliseconds, during which the sound doesn't change very much.

Sometimes when a sound that is ordinarily syllabic (a vowel that is) is non-syllabic, that is indicated in a phonetic transcription by making it a superscript. That is perhaps what has been done, here. Schwa is a vowel and is ordinarily syllabic, but in the American pronunciation, the dictionary makers think it is non-syllabic, so they have made it a superscript. Then, comparing the British and American pronunciations given, the British pronunciation has three syllables, "an-to-nym", while the American pronunciation has two syllables, "anton-im". (I am not using phonetic transcriptions here, but rather modified conventional spellings.)

Another possibility here is just some confusion on the part of the dictionary makers. If you click through to this page, http://dictionary.cambridge.org/pronunciation/english/antonym, you'll see another transcription for the American pronunciation that does not have the schwa as a superscript.

As for what the American pronunciation actually is, personally, all I hear in the sound recording is the aspiration after the t, with no vowel sound at all between the t and the n. That is consistent with a syllabification "an-tnym" with just two syllables and the medial vowel lost. Syllable-initial t is aspirated in English, and so aspiration of t would be expected here.

  • 1
    I agree about the voiceless schwa in the american example. But a superscript schwa just indicates two alternative pronunciations. You can either have a syllabic consonant or a schwa and a consonant. This convention is widespread in dictionaries. LPD and Oxford ALD both use it for example. You can find it in the guide to phonetic symbols Dec 22, 2015 at 14:40
  • While the superscript schwa is used for non-syllabic schwa, I don't think any English speakers pronounce antonym this way. I suspect that @Araucaria is correct: this indicates two possible pronunciations. Dec 22, 2015 at 14:42
  • @PeterShor I can't even interpret what "two pronunciations" means here. There's just one sound recording given. Which is it? Is this a phonetic transcription or a phonemic one? In "antonym", the t and the n are both alveolar. If one "reading" of the superscript schwa is that it's not there, but the following n is syllabic, does that mean that alveolar closure can be continuous going from the t to the n? (It clearly is not continuous in the sound recording.)
    – Greg Lee
    Dec 22, 2015 at 14:59
  • @GregLee The transcriptions given in a dictionary are not a description of the audio. They are a description of how the word is pronounced by speakers of a particular variety of the language. People pronounce the word in two different ways. They only give one audio, which could be a version of either. The audio clearly is using a schwa (albeit a devoiced one). As you say, there is no syllabic n there. Dec 22, 2015 at 15:08
  • I didn't say there was a voiceless schwa in the American example. I said there was an aspirated t in the recording. (I don't say it this way -- I say [æ̃nʔn̩ɪ̃m].)
    – Greg Lee
    Dec 22, 2015 at 15:11

The Cambridge Dictionaries Online has a key where they discuss how they use IPA phonetic symbols. They say:

əl, əm, ən can be pronounced either: əl or ḷ etc.:

Here /ḷ/ stands for a syllabic /l/.

So the IPA stands for two different pronunciations. The t with a vee underneath it is a flapped t. The two pronunciations are:

/ˈæn·təˌnɪm/ or /ˈæn·tṇˌɪm/

where the /t/ is flapped in both pronunciations, and the n with a stroke under it is a syllabic /n/.

The /n/ followed by a flapped /t/ would actually probably be a nasalized flap in real life pronunciations (despite the fact that they're shown as being in different syllables), but Cambridge Dictionaries Online doesn't have a special symbol for that, so they represent it as /nt̬ /.

  • +1 Quite right. Clear too. There's several possibilities there. You could get an n and a flap, you might get a single nasalised tap, or you could get a regular /t/ in this position too (as in the audio given by CDO). Dec 23, 2015 at 9:16
  • @Araucaria: Although I might use either a nasalized flap or an /nt/ in antonym, I can't follow an /n/ with a flap. My tongue really doesn't want to pronounce that combination. Are you sure people use it? (Going by all your great answers here, I expect you're right—I just find it somewhat hard to believe when I have so much trouble pronouncing it.) Dec 23, 2015 at 20:52
  • Well, I've done some research on the web, and I have to say that so far after a hundred or so examples, I haven't been able to find an example of antonym with any kind of flap at all! Every example apart from maybe the "Southern American" example from this dictionary has used a regular [t]. It's certainly possible to get a voiced flap following a /n/. For example, it's often heard in the word Toronto. Of course, the approach to the tap isn't tap-like in the same way that the release of a /t/ at the end of word may not be plosive. Dec 28, 2015 at 10:33
  • What you might get in those situations is a kind of tap which starts of nasalised and then has a velic closure during the contact phase, so that the beginning is nasalised and the end isn't. Dec 28, 2015 at 10:36

You really have to ask the editors of the dictionary what they were trying to convey, because choices are inevitably made when applying the IPA. There is even, in fact, a school of thought (that of the Atlas of North American English) that North American English and British English should be transcribed with different systems of vowel notation to account for the fact that North American English does not differentiate vowels in terms of length.

  • You don't actually have to ask them. You can Google and find their IPA key. Dec 22, 2015 at 17:46

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.