Oooh, long reply, Jose! Thanks!
Hmm.... I liked your suggestions, but I put the something on puirpose, what's wrong with it anyway?
Your 'Indeed I like W's talk...' sentence was fine with 'may/might'...I just wanted to point out that using your bracketed 'could' would give the resulting sentence a different meaning (different, that is, to what the two of us both apparently think: that Widdowson didn't appear to really be saying anything that new or amazing in his seminar).*
What about Swan's might, is it only a matter of personal choice (using could over might)?
Yes, Swan seems to prefer 'might', whereas I (and it seems, you) would rather go for 'could' (regarding the
possibility that 'you
could've killed/injured/hurt yourself (by doing that)!' - compare 'You might've injured yourself...here, let me have a look, I'm a doctor you know'. I guess I'm always trying to think of a "central" meaning-function=use for any given form
).
Let me see if I got your example: She may have left me, but she's the one I will love forever. => That means she is no longer here but I still love here, regardless what she did or what happened.
You wouldn't use may in She may have told me this before, instead of leaving things come this far., am I right?
Yes, that's what I reckon, right on both counts, Jose!
For the second, again, I like the "There was the chance, the possibility, that she could(/might) have told me this before, back when we were..." kind of meaning (thus 'She
should have told me this before, instead of letting things come this far' would also be OK ~ modal of obligation).
Now, doesn't our little discussion boil down to simply language usage? I mean, you might find the may have odd but some other natives will (would) use it naturally?
What do you say?
Well, the fact that you and me appear to be agreeing is significant, Jose...but yes, it would still be interesting to hear what other competent speakers (non-native as well as native) reckon they say (that being said, however, I do sometimes get a bit tired of people Googling away all the time to "prove" things rather than telling us what
they themselves really think and more importantly say...perhaps we are simply meant to assume that by their presenting the data they are "saying" that they do indeed talk exactly like the people in the data itself do? But I appreciate that the habits in our own speech are not that open to or always reliably revealed by just simple introspection, hence the frequent Googling we often seem to prefer to do
).
I'll post the Nabokov stuff from Widdowson ASAP (later today, after I get home from work).
*If we take the negative 'not' out of the sentence, and perhaps add 'perhaps'
between e.g. 'could' and 'added', the use of 'could' would then be OK and equivalent to the 'may/might'.