[Grammar] modals of probability

[yuriya]

Wilber: Someone's knocking.
Gertrude: That must be Sydney. (HIGH CERTAINTY)
That will be Sydney.
That should be Sydney.
That may be Sydney.
That could/might be Sydney. (LOW CERTAINTY)

The above is an excerpt from the The Grammar Book. I don't think I completely agree with the scale of probability presented above and I have some more doubts related to this use of modals. To me, they are quite subtle points so I'm not even sure I'll be able to get them across. Please feel free to enlighten me!

First of all, I'm a learner and I'm more exposed to American English and I believe that to my ear the sentence with will sounds more certain than the one with must. Moreover, That would be Sydney would sound better than That will be Sydney.

Secondly, the choice of modals of probability seems to depend more on situations than on the scale itself. For instance, let's say that I know Sydney very well (Sydney being my daughter's friend) and we've been expecting Sydney because she called to come over, I'd say, "that should be Sydney." On the other hand, I haven't met Sydney but have heard of her and when I answered the door, there's a Sydney-looking girl on the porch asking for my daughter then I would say, "you must be Sydney." If Sydney has been regularly visiting us around this time and I hear someone knocking on the door, then I would say, "that would(will) be Sydney."

Thirdly, while can can be used in the negative and in questions to express probability, I wonder why can can not be used in the affirmative. For instance, "that can't be Sydney" and "can it be Sydney?" sounds OK but "that can be Sydney" sounds not quite right unless it means something potential as in "that can be tricky." Any comments on these? Thanks in advance!

 

[Raymott]

Wilber: Someone's knocking.
Gertrude: That must be Sydney. (HIGH CERTAINTY)
That will be Sydney.
That should be Sydney.
That may be Sydney.
That could/might be Sydney. (LOW CERTAINTY)

The above is an excerpt from the The Grammar Book. I don't think I completely agree with the scale of probability presented above and I have some more doubts related to this use of modals. To me, they are quite subtle points so I'm not even sure I'll be able to get them across. Please feel free to enlighten me!

I certainly agree with you that too much emphasis is placed on the comparitive probability of modals.
I'd say, in this case, that the top three are highly probable, and the bottom two are of low probability, and leave it at that.
I'd also add: "That is Sydney" (100%) and "That's not Sydney" (0%).
Of course, the person is guessing, so it's not an exact situation.

First of all, I'm a learner and I'm more exposed to American English and I believe that to my ear the sentence with will sounds more certain than the one with must. Moreover, That would be Sydney would sound better than That will be Sydney.
There's no difference to me. They both mean "That's probably Sydney".
Whether one means 77% likely and the opther is 87% is beyond the semantic competence of these words to convey.

Secondly, the choice of modals of probability seems to depend more on situations than on the scale itself.
Exactly. The scale has been devised for learners who insist on it; sometimes giving definite rules is the only way to get students off the teacher's back. She therefore gives them a rule, and lets the students deal with the consequences several years later.

Thirdly, while can can be used in the negative and in questions to express probability, I wonder why can does not be used in the affirmative. For instance, "that can't be Sydney" and "can it be Sydney?" sounds OK but "that can be Sydney" sounds not quite right unless it means something potential as in "that can be tricky." Any comments on these? Thanks in advance!
You have to say, "That could be Sydney".

Note that a probability can be assigned to "can, could, might, may, is possibly". The probability is "greater than a zero chance".
By using "will, would, should, must, is probably", the probability is likely to be greater than 50% - it is more probable than not that it is Sydney (if the person speaking is a competent judge of probability).