Roger wrote:
> Imagine being confronted by a group of people, known to be couples, but not
> so identified, and you want to sort them out: who(m) does who love? with
> normal intonation could be a perfectly valid question.

Yeah, for me, that would be bad;  in such a situation, I'd say "who
loves whom?".  Now, I would not be surprised if there are dialects
of English in which there are no superiority effects, and yours
may be one of them.

I've gotten to the point where I can no longer rely on my own
grammaticality judgements.  When you hear/read things that at first
sound ungrammatical enough times, they begin to sound grammatical,
and you just have to go find someone else to test it on.

> > So, the interesting thing is that in German, even not in echo
> > questions, it is according to some native speakers possible to
> > say "Wen liebt wer?" ("Whom loves who?").
> What is the context in German? My limited experience suggests that most such
> questions would be echo/emphatic/contrastive, except possibly in the
> situation I devised, or similar. We can certainly do it with "what"--
> e.g. at a potluck supper, we might well ask "who brought what?"

These things are supposed to happen with "what", as well. *"What did
who say?" is definitely bad for me.  As for the German facts, Grewendorf,
the author of one of the articles I've been reading, is a native speaker
and he explicitly denies that these are echo-questions.

Charlie wrote:
> Tom wrote:
> > A: "You'll never guess:  of all people, John saw Mary at
> >    the library today."
> > B: "John saw WHO at the library?!?"
> You need to explain to me this in-situ language.  It seems to me
> that, regardless of emphasis, "who" is still the object of the
> verb "saw" and should be "whom" (for those of us who still use whom).
> Let me illustrate by continuing the conversation, but substituting
> Sam for Mary.

The case marking is quite irrelevant for our purposes.  It's called
"in situ" because the wh-word is located in the normal grammatical
position for objects, right after the verb.  "John saw WHOM at the
library?" is still in situ, and an echo-question. The point was that
unlike Japanese, in English we normally front wh-words away from
their normal position (however one theoretically analyzes it).  Thus,
in "Who(m) did John see at the library?", "who(m)" is still the
direct object, but in a noncanonical position for direct objects.

Andy wrote:
> There's nothing wrong with that sentence in the English I speak.
> Well, nothing except I would say "Who did who see?"

You see, that begs the question, doesn't it? If it's not natural
for you to say it, then that's something the grammar should be
Thomas Wier	       "I find it useful to meet my subjects personally,
Dept. of Linguistics    because our secret police don't get it right
University of Chicago   half the time." -- octogenarian Sheikh Zayed of
1010 E. 59th Street     Abu Dhabi, to a French reporter.
Chicago, IL 60637