For instance in the word literature /ˈlɪtərəʧər/, does the lip shape change for each syllable with schwa, so they become more as if we were pronouncing /e/, /æ/ and /u/ respectively, or some other vowel sound?

Or, on the contrary, is the lip shape always the same?

In the case of the latter being affirmative, I'd also like to know why some syllables with schwa seem to sound closer to other vowels?

  • I'm no linguist, but I'd have thought the whole point of the "neutral" vowel is that it doesn't require any help from the mouth, lips or tongue to articulate (apart from having the mouth open, so the unmodified drone from the larynx actually makes it out where it can be heard). Actually, even that's not true - I assume non-committal Um through the nose (with the mouth closed) counts as a "schwa" vowel. – FumbleFingers Jul 21 '17 at 16:16
  • @FumbleFingers That's my point too. So, is there something else that causes the schwa vowel to sound different? maybe the consonant that precedes it? or am I being deceived because of how the word is spelled? – rraallvv Jul 21 '17 at 16:25
  • I think you're probably more at risk of being misled by the phonetic transcription /lɪtərəʧər/, which implies a four syllable word. As it happens, I got a degree in literature 40 years ago, and will doubtless have said that word tens of thousands of times. But it's nearly always a three syllable word for me (the first schwa in your transcription simply isn't articulated at all). – FumbleFingers Jul 21 '17 at 16:29
  • 2
    The full OED gives the pronunciations Brit /ˈlɪt(ə)rᵻtʃə/, and U.S. /ˈlɪdər(ə)tʃər/. From which it would appear that (in their opinion?) Americans usually discard the second schwa where Brits discard the first. But really, if you want to master pronunciation in these areas, you should listen to native speakers saying the words, rather than focusing on inevitably imperfect attempts to reflect the sounds of speech using written symbols. – FumbleFingers Jul 21 '17 at 16:43
  • 2
    At the end of the day, I expect your question should more properly be addressed by a linguist. But I still don't see how it would help you improve your ability to speak English - which I can speak perfectly well despite not having a clue what your question really means. – FumbleFingers Jul 21 '17 at 16:52

Short version: It must change, but nowhere near enough to become /e/, /æ/ or /u/. The change is not significant enough that you should take it into consideration. In fact, the less you think about it when speaking, the better. Your brain and tongue will take care of the tiny differences for you. :)

I consider this a valid language learning question, because seeing how sounds alternate can be part of a phonological approach to a new language. For example, if an English learner is having difficulty knowing when to aspirate a voiceless stop (i.e. when to say /p/, /t/ or /k/ with the infamous "puff of air"), it will be helpful and probably not too technical to know that this happens at the start of a syllable when it is not in a cluster. That is not to say that this information is necessary for every learner of English. Most people get by fine without it. But it can be helpful.

However, in the case of English schwa, the answer is that it's not very relevant to your learning.

I'll give a little more detail, though, if you're curious (though this part more or less belongs on Linguistics Stack Exchange).

Certainly all articulation is affected by its context. One of the major forces is assimilation. Assimilation is where one sound shares its own features with another sound. There are various reasons, but an obvious one is that we try to save effort. Your tongue doesn't like jumping from place to place too often; your vocal cord vibration doesn't like being switched "on" and "off" too often; your lips don't like to round and un-round too often. So we unconsciously cut corners.

Often, we tolerate the variation. For example, try saying "key" and then "cool", very slowly. If you pay close attention, you'll find that /k/ before /i/ is farther forward, almost palatal, but /k/ before /u/ is farther back, more velar. Your tongue anticipates where it will need to be for the next vowel.

But we are equipped to ignore such small differences. We have to ignore them. The resulting acoustics differ very slightly. In any given conversation, there might be more interference from, say, the pitch of the speaker's voice, whether they have food in their mouth, background noise... So we have an unconscious funnel that takes in a bunch of versions of /k/ and all our brain tells us is that we heard /k/.

In linguistics notation like the IPA, we represent that reality by a kind of fiction, even when being very careful. We write [k] for this whole range of sounds. However, we draw a line for certain features. For example, we write an alveolar [t] differently from a dental [t̪].

Moreover, when we see that in a given language, one "structural" sound is pronounced as different "surface" sounds, we call it allophony. Then, we use the broad notation for the underlying "phoneme": /t/. And we link that to the fine notation for the surface "phones": [t] and [t̪].

The relevant part is this: Most speakers of a language can only ever perceive the phoneme, because their brains have learned that the difference doesn't change the meaning. Learners are likely to be aware of the different phones, because their brains haven't learned to ignore the difference. (For example, English speakers don't notice the aspirated stops in their own language, but they do notice the alternation between [ɾ] and [l] in Japanese even though most Japanese speakers don't.)

And for the things that are too fine to be notated... only acoustic technicians and X-rayers bother with them. :) As far as that goes, I haven't heard of the schwa in English being allophonic. Not in terms of its openness or backness, nor in "lip shape" (rounding?). So don't worry about it.

I will say that in some dialects, there are different sounds we normally call schwa. For example, a US English speaker may pronounce Rosa's differently from Rose's or roses. They might say the first one /ɹoʊzəz/, but the other two /ɹoʊzɨz/ with a reduced vowel similar to the vowel in sit. But this is tied to a change in meaning, so it can't represent the same phoneme.

  • I was wondering how spelling bees find out how to spell a word, but now that I've read your answer, it could be somehow related to the ability to distinguish those tiny differences you mentioned. – rraallvv Jul 23 '17 at 3:15
  • 2
    @rraallv Indeed. Certainly those different layers of pronunciation awareness were very much involved when words were first written down, and have something to do with the highly variable spellings in early writing. But of course, a huge amount of modern English spelling also has to do with etymology, as well as how letters have been used historically to capture different languages' sounds. (Which is why they often allow contestants to ask for the language of origin. Each one has its own "subsystem" within English spelling!) – Luke Sawczak Jul 23 '17 at 3:27

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.