Assignment 7 -- Gapping
Due Thursday, November 14

1.  Read Ross: Gapping and the Order of Constituents.

2.  Outline, very carefully, the argument that leads to the conclusion
that Turkish is a deep structure SVO language.

What are the key assumptions in this argument?

For each key assumption, what is the empirical support for it?

3.  In a language that has "scrambling", what are the possible orders
between forward gapping and scrambling, according to Ross's assumptions?
(For the purposes of this question, regard "scrambling" as a rule that
changes the order of V and O.)

What are the possible orders between backward gapping and scrambling?

4.  Recall that, in English, and for most speakers, there is a
"two-remnant" constraint on Gapping:

	 Harvey painted the fence red yesterday, and Tom the porch.
	*Harvey painted the fence red yesterday, and Tom the porch white.
	*Harvey painted the fence red yesterday, and Tom the porch white two days ago.

This is true for Turkish (for most speakers) too:

	 Hasan dUn evi yeSil boyadI, ve Mehmet arabayI.
	 Harvey yesterday house[ACC] green painted, and Tom car[ACC]

	*Hasan dUn evi yeSil boyadI, ve Mehmet iki gUn evvel arabayI mavi.
	 Harvey yesterday house[ACC] yellow painted, and Tom two days ago car[ACC] blue

But when the gapping is "backward", the two-remnant constraint seems to disappear:

	 Hasan dUn evi yeSil, ve Mehmet iki gUn evvel arabayI mavi boyadI.

Is this predicted by Ross's analysis?

5.  Read Eisenberg, "A Note on Identity of Constituents".

Here are some similar facts from Turkish:

	 ben yumurtayI yedim, ve sen patlIcanI yedin
	 I the-egg[ACC] eat-PST-1s & you the-eggplant[ACC] eat-PST-2s
	 "I ate the egg, and you ate the eggplant."
	 ben yumurtayI yedim, ve sen patlIcanI
	*ben yumurtayI, ve sen patlIcanI yedin
	*sen patlIcanI, ve ben yumurtayI yedim
	*ben yumurtayI, ve sen patlIcanI yedim

What do you conclude about the relation of forward gapping to subject-verb agreement?
What about backward gapping and subject-verb agreement?