A. More evidence for BE+COMMA being parenthetical:
(1a) She_1 said*(,) why shouldn't she_1 leave early?
(1b) Why, she_1 said, shouldn't she_1 leave early?
(1c) Why shouldn't she_1 leave early, she_1 said.
(2a) She_1 was wondering*(,) why shouldn't she_1 leave early?
(2b) Why, she_1 was wondering, shouldn't she_1 leave early?
(2c) Why shouldn't she_1 leave early, she_1 was wondering.
(3a) The dispute has always been, what gives her the right to
leave early?
(3b) What, the dispute has always been, gives her the right to
leave early?
(3c)(?)What gives her the right to leave early, the dispute has
always been.
(4a) The dispute has always been, she abuses the right to
leave early.
(4b) She, the dispute has always been, abuses the right to
leave early.
(4c)(?)She abuses the right to leave early, the dispute has
always been.
B. Diachronically, 2BE looks like an obvious amalgam of BE+COMMA
plus ordinary Subj+BE+complement. Synchronically, it's hard to
see what's going on. If I understand correctly, (5-6) are
possible.
(5) The issues have to be, are that she has the right to leave
early, and abuses it.
(6) The issues have to be, are why shouldn't she leave early and
have the right to do so?
A possible analysis is that _the issues have to be_ is still a
parenthetical, with _are_ the complement of _be_ and extractee of
_have_, and _the issues_ is subject of _are_, even though it is
subordinated within _the issues have to be_, much as in:
(7) Why have God only knows how many people been invited?
where _have_ agrees with _people_, which is subordinated within
_God only knows how many people_.
But the snag here is with (6), since the non-2BE structure (8) is
not okay, the reason being that wh-inverted clauses can't be
subordinate.
(8) *The issues are why shouldn't she leave early and have the
right to do so?
That should mean that the wh-inverted clause in (6) isn't subordinate,
and hence leaves unresolved its syntactic relationship with _are_.
--And.
|