"can" and "can't". How and why?

Question

The main question is how to tell them apart?

I've heard two main explanations from native speakers:

1. Because can't has that t sound in it.
2. Using the context

To what I think:

1. Yeah, no way in earth you can tell that t most of the times. If the pronunciation is right and there's no background noise, it's mostly audible. But that's a big if.
2. Well, sometimes it's obvious, but too many times it's just not. Precisely because of what the verb is used for it could either be one or the other in the same context or situation.

Now, I know you can say cannot, which makes perfect sense, but I don't see people using it much.

And on top of that I need to ask why. Why are them too phonetically close? Who would do such a thing?

For someone who has been brought up with english, imagine a language that had "yes" for yes and "yess" for no. Kind of silly isn't it?

Answer 1

You are correct in saying that the /t/ is rarely heard in can't. /n/ and /t/ are articulated at the same point in the mouth, so there is no distinct movement of the tongue to produce the /t/.

However, there is a great deal more to phonology than the sounds selected to be represented by the writing system. There are four associated 'environmental' phenomena which clearly distinguish can from can't:

1. A /t/ (or other voiceless stop) at the end of a syllable is spoken with a 'coarticulated' glottal closure, so there is always a distinct interruption of airflow. Can, by contrast, ends with a voiced continuant which does not interrupt the airflow: air continues to emerge through the nose.
2. The vowel which precedes a voiceless consonant is 'shorter', as voicing is relaxed in anticipation of the consonant. Although the vowel before /-nt/ is nominally followed by a voiced consonant, it is timed as if it were followed by the voiceless stop, so the vowel in can't is shorter than that in can.
3. If can't is followed by a word beginning with a vowel, the /t/ re-emerges through ordinary English syllabification, which binds a terminal consonant to an onset vowel in the following syllable. With can, of course, there is no /t/, and the /n/ is bound to the following syllable.
4. Most importantly, can't is always stressed; contraction is only permitted on words following stressed words. Can, however, is almost always unstressed. In consequence, the vowel in can is 'reduced' to something closer to a centralized /ɛ/ or even /ɪ/. When can is stressed, this will usually be because it contrasts with a preceding negative; in this case, the discourse contrast will supplement the other environmental factors.

So the contrast isn’t between /kæn/ and /kæn_(t)/ but between /kɛ·n/ and something like /’kæ̃ʔ/. Native speakers 'hear' the /t/ not because it's actually there but because everything else in the environment points to its being there!

Answer 2

This probably isn't what you want to hear, but the answer is that native speakers can hear the t.

'Cannot' is the original form, but over time the 'no' has been dropped specifically because it's not really needed in casual conversation between fluent speakers.