Basic semantic meanings of modal auxiliaries.

[Stephen Jones]

Why would you want to do that when the research at Longman shows that "may" is hardly used for permission in contemporary use? "Can" is used much more.

Does Longman's research include American usage. I get the feeling that it is much more common in the States than in the UK.

[Stephen Jones]

Google (and BNC) hits:

• "Might I come in?" 174 (0)
  "May I come in?" 38.000 (17)
  "Can I come in?" 38,000 (24)

Looks like that Longman 'research' isn't saying what you claim it is.

[metal56]

Why would you want to do that when the research at Longman shows that "may" is hardly used for permission in contemporary use? "Can" is used much more.

Does Longman's research include American usage. I get the feeling that it is much more common in the States than in the UK.

From the Longman site:

This corpus contains a very wide variety of language, from spontaneous conversations in both the UK and the US, through novels and newspapers up to the academic language of scientific textbooks.


Is this also not true the States?

97% of all may occurrences express possibility and only 3% permission ...

[metal56]

Google (and BNC) hits:

• "Might I come in?" 174 (0)
  "May I come in?" 38.000 (17)
  "Can I come in?" 38,000 (24)

Looks like that Longman 'research' isn't saying what you claim it is.

Are all those examples of contemporary usage? Are they all native speaker examples?

Try a concordancer:

CAN I GO 0= 130 (in one hundred million words)

MAY I GO = 17 (in one hundred million words)

CAN I COME IN = 24
MAY I COME IN = 17

The "cans" win.

[Stephen Jones]

And "can" as in "John can ride a bike" is not stating a fact, but is stating potential. "John rides bikes" is a to be read as a fact. Modality steps back from the factuality of non-modal statements. In all cases

You're only giving half the argument, as usual. Let's paraphrase Huddelston on ability:

Two subcases can be distinguished; potential and currrently actualized.

Examples:
(i) (Potential) She can run the marathon in under three hours.
(ii) (Currently actualized) I can hear something rattling.
The latter is found with sense verbs and various verbs of cognition and (ii) differs little from I hear something rattling. By contrast (i) differs sharply from She runs the marathon in under three hours. ...... in this potential ability case, the degree of difference from the unmodalised version will depend very much on the pragmatics of the situation concerned. There is, for example, little effective difference between She can speak fluent French and She speaks fluent French because it is not easy to assert the fromer without repeated acutalisations fo the ability.

Two basic points here

• 1. With 'can' there is sometimes no difference between the modalized and unmodalized versions. I can't understand is the same as I don't understand and I can see it is a factual statement.
  
  2. More importantly, there is no doubt over the factuality of the potential. Look at these pairs to see the difference.
  (i)Factual
  She can speak French.
  She's in Paris now.
  (ii)Possible
  She might be able to speak French
  She could be in Paris now.

Again, can can roughly be paraphrased as:

Can = I assert that it is possible that ...

That is the basic semantic meaning of that modal.

You are the one making the claims that there is a core meanng and that it is what you say it is. I have the following three authorities in front of me:

• The Cambridge Grammar of the English Language by Pullum & Huddleston (2002)
  A University Grammar of English by Quirk & Greenbaum (1975)
  An Introduction to Functional Grammar by Halliday (1985)

None of them speak of core semantic meanings, and Huddleston clearly states what I said at the beginning of the thread

In general it is plausible to regard the deontic uses as more basic, with the epistemic ones arising by extension to the domain of reasoning of concepts primarily applicable in the domain of human interaction, such as compelling and permitting.

That is to say that even if there were a basic meaning, it would not be the epistemic one you, Lewis and Larry are peddling.

[Stephen Jones]

Try a concordancer:

I was; the same one you were and you will see the figures in brackets. your reading skills still not up to scratch today I note.

And the BNC is the British National Corpus. Neither of us is disagreeing that 'can I' is more common for permission in British English.

'Can I go' is not only permission - it is often used for possibility as in "Where can I go to get a good job?" This will skew the figures.

The fact that 'may' is used much more for possibility than permission is irrelevant. The question is whether 'may I' is still fairly common for permission. If Longman has published figures on that please give us them.

[metal56]

And "can" as in "John can ride a bike" is not stating a fact, but is stating potential. "John rides bikes" is a to be read as a fact. Modality steps back from the factuality of non-modal statements. In all cases

You're only giving half the argument, as usual. Let's paraphrase Huddelston on ability:

Two subcases can be distinguished; potential and currrently actualized.

Examples:
(i) (Potential) She can run the marathon in under three hours.
(ii) (Currently actualized) I can hear something rattling.
The latter is found with sense verbs and various verbs of cognition and (ii) differs little from I hear something rattling. By contrast (i) differs sharply from She runs the marathon in under three hours. ...... in this potential ability case, the degree of difference from the unmodalised version will depend very much on the pragmatics of the situation concerned. There is, for example, little effective difference between She can speak fluent French and She speaks fluent French because it is not easy to assert the fromer without repeated acutalisations fo the ability.

Two basic points here

• 1. With 'can' there is sometimes no difference between the modalized and unmodalized versions. I can't understand is the same as I don't understand and I can see it is a factual statement.
  
  2. More importantly, there is no doubt over the factuality of the potential. Look at these pairs to see the difference.
  (i)Factual
  She can speak French.
  She's in Paris now.
  (ii)Possible
  She might be able to speak French
  She could be in Paris now.

Again, can can roughly be paraphrased as:

Can = I assert that it is possible that ...

That is the basic semantic meaning of that modal.

You are the one making the claims that there is a core meanng and that it is what you say it is. I have the following three authorities in front of me:

• The Cambridge Grammar of the English Language by Pullum & Huddleston (2002)
  A University Grammar of English by Quirk & Greenbaum (1975)
  An Introduction to Functional Grammar by Halliday (1985)

None of them speak of core semantic meanings, and Huddleston clearly states what I said at the beginning of the thread

In general it is plausible to regard the deontic uses as more basic, with the epistemic ones arising by extension to the domain of reasoning of concepts primarily applicable in the domain of human interaction, such as compelling and permitting.

That is to say that even if there were a basic meaning, it would not be the epistemic one you, Lewis and Larry are peddling.

2. More importantly, there is no doubt over the factuality of the potential. Look at these pairs to see the difference.

And what is epistemic about this?

You must get the eight o'clock train.

Please tell us how the paraphrase does not fit with that deontic example?

Must = I assert that it is necessary that ...

The basic semantic meaning of "must" is someone asserting their view of a necessity.


(i)Factual
She can speak French.
She's in Paris now.
(ii)Possible
She might be able to speak French
She could be in Paris now.

You need to understand the main purpose behind modal statements. The "can" (dynamic) examples have the speaker mainly focusing on potentiality and not on the factuality of the proposition.

That is to say that even if there were a basic meaning, it would not be the epistemic one you, Lewis and Larry are peddling.

Read back on how I showed that Lewis does not focus wholly on epistemic uses.

What is epistemic here?

You must catch the six o'clock train.

Tell us how this paraphrase does not fit well with that sentence:

Must = I assert that it is necessary that ...

[metal56]

Try a concordancer:

I was; the same one you were and you will see the figures in brackets. your reading skills still not up to scratch today I note.

And the BNC is the British National Corpus. Neither of us is disagreeing that 'can I' is more common for permission in British English.

'Can I go' is not only permission - it is often used for possibility as in "Where can I go to get a good job?" This will skew the figures.

The fact that 'may' is used much more for possibility than permission is irrelevant. The question is whether 'may I' is still fairly common for permission. If Longman has published figures on that please give us them.

If you stick your hand in your pocket you might like to buy this:

http://www.longman.com/dictionaries/whi ... lgswe.html

<The fact that 'may' is used much more for possibility than permission is irrelevant.>

Not at all. The prototypical use is important and "permission" is no more than a type of possibility. I give you the possibility to do something because I lift the permissive restraints.

[metal56]

Try a concordancer:

I was; the same one you were and you will see the figures in brackets. your reading skills still not up to scratch today I note.

And the BNC is the British National Corpus. Neither of us is disagreeing that 'can I' is more common for permission in British English.

'Can I go' is not only permission - it is often used for possibility as in "Where can I go to get a good job?" This will skew the figures.

The fact that 'may' is used much more for possibility than permission is irrelevant. The question is whether 'may I' is still fairly common for permission. If Longman has published figures on that please give us them.

Google:

202,000 for "may I help you".
356,000 for "can I help you".

988 for "may I go out".
13,900 for "can I go out".

[Andrew Patterson]

I haven't got anything to say about the recent topic that hasn't already been said, but since we are on the theme of American usage it got me thinking about the use of "come" and "go" as modals.

Earlier I said that all the uses that metal listed were epistemic. I now think I was wrong. "Can" as a proximal positive modal is almost never epistemic IMO.

The test is to see if it can be followed by "have" since such cases are always epistemic (correct me if I'm wrong.) "Can have" doesn't exist, except perhaps as a scarcatic reply to "can't have".

If "can" have can just about slip through, "come" and "go" can only have root modality as they can never be followed by "have."

Go figure!

[lolwhites]

I am finding this debate harder and harder to follow because of the way some individuals feel the need to quote whole posts in their entirety before commenting on them. Could we stick to only quoting the bits we want to talk about? Some people's posts are getting so long and unreadable that it's unclear see who's said what, which rather defeats the object of quoting.

[lolwhites]

Interesting that most of the examples seen here revolve around can, which is probably one of the easier modals to narrow down, though how you can get it down to one basic meaning that encompasses I can see... I can hear to mean I see... and I hear when, let's face it, the former is far more widely used without reference to ability or possibililty (at least in British English) is beyond me.

Noone has attempted to narrow should down because even Lewis admits it can't be done.

I should go. It's getting late
You should be so lucky!
Should we narrow this modal down I'll eat my hat.

[lolwhites]

One more thing:

Surely I'm not the only one here to think there's a halfway house between laundry lists and core meanings? Some modals are easier to narrow down then others, but when the use of a word evolves over time, and other uses fossilise, it's surely almost inevitable that uses will exist that don't fit in the box, however stretchy.

Thinking of more examples to show our points is all very well, but unless someone is going to go to a corpus, look at every single use of can, may, might... etc and find common threads that run through all of them, we're not going to get to the bottom of this. This, obviously, would be a job for a full time researcher.

Those posters who ask us for ideas for things to research for their MAs or PhDs would do well to note that the longest threads on this forum are about modals.

[Stephen Jones]

Do we need to do any narrowing down?

As I said Q & G back in 1975 gave a list of uses for each modal that varied between two and five.

They do not distinguish between epistemic and root necessity, but that would simply require a re-ordering of the list.

The main complication with the modals is not a plethora of meanings; rather it is the number of constraints regarding the use of the modal in each of its uses, and the search for core meanings doesn't help in the least there.

[Stephen Jones]

"Can" as a proximal positive modal is almost never epistemic IMO.

Huddleston in CGEL states that declarative 'can' is never epistemic.