From:    John Cowan <[log in to unmask]>
> Thomas R. Wier scripsit:
> > Just look at English, for example.  Mark Baker has argued (IMO
> > plausibly) that English has a kind of "abstract incorporation"
> > with verbs which take prepositional complements become phonologically
> > fused to them, and then this process feeds passivization, which
> > cannot normally occur with prepositional complements:
> >
> > (1) a.  John has slept in the bed.
> >     b.  The bed has been slept in.
> > (2) a.  David was writing on Tuesday, but not Thursday.
> >     b.  **Tuesday was being written on.
> But 2b is perfectly perspicuous if Tuesday is the topic rather than the
> date,

Whether something is "perspicuous" is rather beside the point; the
question is whether it is sensed to be grammatical.  And I'm
pretty sure I can't get your topic reading. :)

> and if it were (somewhat improbably) the name of a blackboard,
> then 2b would be excellent.

Right, which rather proves the point (given my gram. judgements).

> Yet even if you want to say that "writing
> on" is fused in the latter case, it seems exceedingly unlikely to me
> that it is fused in the former case.
> > (1b) is fine for most, if not all, speakers, but (2b) is
> > ludicrously ungrammatical.
> Ah yes, like "Spiro conjectures Ex-Lax."  :-)  See below.

Ah, but you'll remember that the problem with this famous sentence
(perhaps tied for second with "Tabs are being kept on Jane Fonda")
is not that there are differing contexts, but that "conjecture"
requires a phrasal complement, and "Ex-lax" is allowed only in the
marked case of elision, which has a different structure. Thus
(2b) is bad because the structure semantic mapping being imposed
onto it is bad.

> > My understanding is (to be brief) that some
> > languages are like English (the goal can passivize, the patient
> > can't), some are like some Bantu languages (the patient can
> > passivize, the goal can't), and others are like IIRC Norwegian
> > and some English dialects (where both arguments can passivize).
> Can you be a little less brief here, and illustrate what works
> and what doesn't, at least for English?

To simplify, let's use English and some pseudo-Englishes.  Take
a basic ditransitive like "give":

  (1) a. Liberal activists gave the NEA money.
      b. The NEA was given money.
      c. *The money was given the NEA (where "the NEA" is still the

Now, in Standard English, only "the NEA" could passivize to become
subject, as in (1b). The alternative in (1c) is right out in standard
English and most dialects.  In some Bantu languages, precisely the
opposite is the case:  the equivalent of (1c) would be fine, but
(1b) would be ungrammatical.  In some other Bantu languages and IIRC
in Norwegian, *both* arguments can become the subject of the
corresponding passive (though not both at once, obviously).  These
facts about English have been used to argue that English does not
contrast direct with indirect objects, but rather primary with secondary
objects.  Only primary objects can passivize, secondary ones can't. In
the case of languages where the theme argument is the only one that
can passivize, such a language would contrast direct with indirect

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