Just a note about Binding Theory (which was covered on Sep 29, the first of the handouts labeled “5a”): The textbook does not really cover this, or give you the standard names (Principle A, Principle B, Principle C), it’s just on the handout. But I do expect you to know this a) for the midterm, b) because this is one of the things people who have taken syntax should wind up knowing.
Review: Up through section 5.2 of the textbook
Also, just to reiterate what I said in class: the part of the material that is being covered in the midterm is up to (and including) section 5.2 of the textbook, and the things on the handouts. (This includes the modification that I made, to make M its own category, which is not the way it is done in the textbook.)
Review: Pat must not have been sending flowers to Chris (version 1)
For use in reviewing what’s going on as of now in the syntactic system we’ve put together, I’ve created a fairly elaborate derivation of the sentence Pat must not have been sending flowers to Chris. It includes pretty much everything relevant that happens along the way in creating this sentence.
This includes the information about T that I provided in the review session: Specifically, T always has a [uN] feature, which causes the closest NP to move (that is, copied and then Merged) at the top of the tree.
One note about this: The derivation I provide here is preliminary. There are a few things that we’re about to cover just after the midterm that will formalize a few things. In particular, we will shortly be dividing uninterpretable features into “strong” and “weak” subtypes, and we will spend some time going over exactly how the inflectional endings (like -ing, -en, and past tense endings) get attached to the verbal and auxiliary forms. For the moment, since we haven’t covered this in class, it is left informal. The things that are informal now but which will be formalized after the midterm are listed below. About a week after the midterm, I’ll give you an updated version of this derivation that has the additional formalizations in it.
T always has a [uN] feature, which motivates the movement (copying, then Merging) of the highest NP in the structure to that point (the one closest to T) into the specifier of TP. This is the subject.
If there are any auxiliaries in the sentence (that is, if there is a modal, or Perf have, or Prog be), then the highest auxiliary in the tree (the one closest to T) always moves to T.
V always moves to v.
The inflectional ending on M, Prog, Perf, and v (i.e. V+v) is determined by whichever of T, M, Prog, or Perf is immediately above in the tree. One of these just below T is realized with a tense ending (past or present). One of these just below M is realized as a bare form. One of these just below Prog is realized with an -ing ending. One of these just below Perf is realized with an -en ending. Take a look at the example derivation to see how this works, and maybe think about variations on it to convince yourself that this will yield the right results: Pat was singing, Pat has been singing, Pat must sing. We don’t yet have a way to handle cases where do is inserted, we’ll deal with that after the midterm.
HW4: Key posted and a note about the steps
Ok, I’ve posted the Homework #4 key and linked it to the schedule page.
I’ve gotten a couple of questions about showing the steps for the derivations in the first part of the homework. I’ve put some notes about that in the homework key. I’m pretty sure that when I said in class that you wouldn’t need to show the steps on future homeworks, I said it was either homework #3 or homework #4 that would be the last one. However, regardless, it’s clear that not everybody caught the hedge, and so some of you did not show the individual steps, sincerely thinking that was what was being asked for. The instructions could perhaps have been clearer (though I think they were pretty clear, given that they pointed you to the model in the textbook, and the model showed the steps).
In any event, the instructions I have given the grader are to not penalize anyone specifically for not showing the steps individually. It was still important to indicate what operations happened where (Merge, Adjoin, Move, like in the model in the textbook), and I suspect that this might have been overlooked by people who didn’t draw the steps individually, but in principle it is possible to get all of the available points on the homework even if the steps weren’t given one by one.
Sorry for the confusion. Once the homework is graded and returned to you, I’m happy to discuss the outcome with anyone who wants to make the case for getting lost points back on this, if you think there’s a case to be made.
Practice midterms (and keys) are posted
I have now posted three practice midterms and their keys, linked from the schedule page. They are essentially the actual midterms from Fall 2005, Fall 2006, and Fall 2008. I have tweaked them slightly because there were a couple of things I asked about that we haven’t quite reached yet. I didn’t come up with replacement questions for those, I just took the questions off. These are the keys: Fall 2005 key, Fall 2006 key, Fall 2008 key.
I’d highly recommend going through these to make sure you’d be able to produce the keys yourself. If you are pretty well in control of these three past midterms, you should have no trouble at all on the real one.
HW4: Assume that “to Oliver” is a Goal.
For the first sentence in part 1 Claudia gave mustard to Oliver, we do want to assume here that to Oliver is a Goal. That is to say, it is connected to the tree using Merge, in the same way that we saw in class basically. The other possibility for a PP like to Oliver is that it might have been an adjunct, a modifier, setting the scene. But adjuncts don’t get theta-roles, so if the PP is a Goal, it must have been attached with Merge and not with Adjoin.
Here are a couple of sentences that I think can constitute evidence of this. First, consider a couple of uncontroversial adjuncts like the PPs in the office and by lamplight. There is no problem switching the order of these, so long as they come after to Oliver. But if to Oliver isn’t first, then any order we choose sounds weird.
Claudia gave mustard to Oliver in the office by lamplight
Claudia gave mustard to Oliver by lamplight in the office
*Claudia gave mustard by lamplight to Oliver in the office
*Claudia gave mustard in the office to Oliver by lamplight
*Claudia gave mustard in the office by lamplight to Oliver
*Claudia gave mustard by lamplight in the office to Oliver
I’ll leave it up to you to see why this test shows that to Oliver must be low, inside the VP, and not a PP attached to the VP (or vP) by adjunction. But these facts do suggest that pretty strongly.
HW4: Show the steps for the three sentences in part 1
By the way, in case it wasn’t clear, you should be showing each step for the three sentences in the first part of homework 4. This is what you see if you look at the model on p. 144, as suggested in the problem statement. There aren’t a huge number of steps, I had 6, 5, and 8 for the three sentences.
HW4 v1.1, now with promised first step
It was just pointed out to me that in the instructions to the first section of the homework, I promised that I would provide the first step. But then it wasn’t there. I’ve reposted Homework 4 (version 1.1), including that first step. It’s pretty straightforward:
HW3: key posted
This is coming a bit later than I’d anticipated, but I’ve now posted the key for homework #3 on the schedule page.
HW4: Reports about polls
Having discussed a bit the status of reports about polls in homework #4 with a couple of people, I have a couple of things to observe about that.
I can say right at the outset that the way I intended this to work was basically like letters to Peter on Adger’s p. 107. That is report in this case “needs a P” [uP]. Whether that is the right way to think about reports though (or letters, for that matter) is debatable. If it needs a P, then it should presumably be bad without a P. Except it’s find to say reports by itself (I read reports all night). We could handle that one of two ways. One way is to say that there are two nouns report, one of which needs a P and the other of which doesn’t. That’s kind of how I suggested looking at eat (Pat ate vs. Pat ate a sandwich).
However, it’s nicer not to have to say that, it would be better if there was just the one noun report. And it seems like about polls is really kind of a modifier (like heavy in heavy reports). We know basically how to do that when a PP modifies the verb phrase. You could extend that same idea to noun phrases (adjoin the modifier to the completed NP). And it seems like this is right. On the table is uncontroversially this kind of modifier, and you can say both reports about polls on the table and reports on the table about polls. I’ll leave it up to you to decide why that’s an interesting fact. But it does suggest that the right way to look at about polls here is as a modifier like on the table.
Nevertheless, I’ll take either of these approaches as correct.