And:
At 11:25 PM 11/29/1999 -0000, you wrote:
>Carl:
>> At 06:55 PM 11/27/1999 -0000, And Rosta wrote:
>> >Over the last couple of years I have noticed myself
>> >consistently or at least regularly saying things
>> >like
>> >
>> >  So you've got a new car, do you?
>> >
>> >  You've got a new car, don't you?
>> >
>> >  Have you got a light?
>> >  Yes I do.
>> ....
>> >And I also think that I wouldn't use a "do" tag when "have
>> >got" it a bona fide perfect:
>> >
>> >  You've gone and got a new car, haven't/*didn't you?
>> >  You've just got married, haven't/*didn't you?
>> 
>> I'm not sure why these are unacceptable when your first
>> three sentences are fine, but they are definitely bad.
>> (Well, 'gone and got' is bad because you're coordinating
>> different tenses; 'went and got' or 'gone and gotten'
>> work for my dialect.  Still don't know why 'have just got
>> married' is awful, though.)
>
>Doesn't American English require "gotten" when it is a 
>genuine perfect? If so, then where Americans have "gotten",
>I can't have DO in the tag, and where Americans have "got"
>I can have DO in the tag.

That sounds right.  So the first set of 'have got's are more
of a present imperfective?  (Can't think of any other verb
that makes sense for in an English, but it makes a certain
kind of sense for 'get'.)

    You've got it, don't/*haven't you?
    You've gotten it, *don't/haven't you?
which matches up with observed American usage.

On the other hand, I'm not quite sure how to account for:
    You have, *don't/haven't you?
versus
    You have it, don't/*haven't you?

>> Maybe it has to do with restrictions on reaccessing the
>> space to ask the question?  The present space is fine, you
>> can use either have or do, but the perfect makes it inaccessible
>> to 'do'.  Not sure why 'just' has the same effect, though.
>> 
>> >There's no obvious analysis of this leaping out at me. At
>> >first I wondered whether "have got" is underlyingly "do
>> >have got" but that raises as many problems as it solves.
>> >Yet another possibility is that the tag question merely
>> >requires semantic equivalence, but this seems not to
>> >be the case:
>> >
>> >   I should like to, shouldn't/*wouldn't I?
>> >   I would like to, wouldn't/*shouldn't I?
>> >
>> >   I have (got) to, haven't/don't/*mustn't I?
>> >   I must, mustn't/*haven't/*don't I?
>> 
>> The only thing that breaks this is 'have/do'; all of the
>> other cases require the same kind of modal tag.
>
>Though now Dick has raised the example of ought/should
>-- "We ought to go, shouldn't we?" whose occurrence is 
>doubtless helped by the rarity of "oughtn't".

Hard to negate 'ought to' to cover the same ground as 'shouldn't';
'oughtn't' doesn't quite get there.

--
Carl



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%