Questions & Answers

This is the page where I will put questions people ask me and the answers to those questions. It might be useful to check this page if a question arises; it is quite likely that somebody else has the same question as you. They will be listed with the most recent first.

Sat Oct 19. On the homework #5 key, there's a discrepancy between sentence (a) and sentence (e) as to where the adverb always is placed in the tree. What gives? True, this was an oversight on my part, but neither one is really wrong. Adverbs like always are attached at the V' level, and in both of these sentences there are two VP's (one for have and the other for the main verb). What's different between the two cases is that I attached always in the auxiliary's VP in (a) and in the main verb's VP in (e). Both are actually possible in English (as seen in I will always have finished my homework by Monday and I will have always finished my homework by Monday), although there is a subtle difference in meaning between them (since in one case always modifies just finish and in the other it modifies have finished).

Sat Oct 19. On the practice midterm, in The President believes the economy will recover, the V believe is shown assigning two theta roles, one to the DP the president and one to the CP the economy will recover. I thought that only one theta role assignment was allowed...? Basically the theta roles are just an indication of "required arguments" -- so, if it were a function in a program, it's what you need to provide it in order to get a value back. For believe, there are two things you need to know (in order to evaluate whether the statement it makes is true or false): the person experiencing the belief and the belief they're experiencing. So, verbs can have a number of theta roles (not just one-- Rain has none, give has three, kick has two, walk has one) to give out, but the uniqueness part comes in where you consider what gets those theta roles -- in a grammatical sentence, no argument can get more than one theta role (nor can it get no theta roles), and no verb can either fail to assign one of its theta roles or assign any of its theta roles twice. There's a certain class of verbs that take whole propositions (clauses or the equivalent) as one of their arguments, like think, say, believe, consider, and so forth. Most other verbs take just DPs as arguments.

Fri Oct 11. Does movement of a head over the trace of a head violate the Head Movement Constraint? Yes; actually, you probably want to think of movement for the most part as starting from the bottom of the tree, so the situation wouldn't arise-- at the point where you would be moving over the head (which would later become a trace as you moved it to somewhere even further up the tree), you would have violated the Head Movement Constraint.

Tue Sep 24. Do we treat place names ( London, Paris, etc.) like proper names or like bare nouns? Treat it like a name. Same thing actually with New York, The Hague, etc., despite the availability of another analysis. Or La Jolla. That is to say, La Jolla is La Jolla is La Jolla, whether it's a Spanish or English sentence you're analyzing (to the extent that you're referring to the city, at any rate). Names are taken to be basically unanalyzed strings of sounds that refer to something.

Sat Sep 14: Concerning homework 1: Should we call auxiliaries "verbs"? Well, it would actually be better to call them "T" (for Tense). We'll go into auxiliaries in more detail within a few weeks, but they do seem to be in a different place in the sentence than verbs are, and you can't really just replace an auxiliary with a verb:

1. Mary is leaving
2. Mary is not leaving
3. Mary will not leave
4. Mary did not leave

(It seems like auxiliaries generally live to the left of not, while verbs generally live to the right of not). There's a sentence or two about this in the chapter -- it's probably around p. 40, though I don't have the actual book onhand right now, but it should be right around a discussion of Cedric might crash the long-boat. Auxiliaries like be and have seem to sit in the same place in the sentence as modals like might, so we're going to put them all in the same category and call them "T".

As a preview, I'm actually going to back off that slightly in later classes. Have and be have properties both like verbs and like T, and we'll see why later on, but I'll explain it when we get there.

Tue Sep 10: Concerning homework problem #1.1: for sentences that are descriptively ungrammatical, how can we decide if the error is semantic? After all, if there is an error in syntax, the sentence won't make sense semantically either. So is something like "His purple was bike" a syntactic or a semantic error?

Fair enough question. Let's say we limit the scope of "semantic errors" to just those sentences which are otherwise syntactically well-formed, but have interpretation problems. That is, let's suppose that we can only evaluate whether a sentence is semantically well-formed if the sentence is first syntactically well-formed. In short, syntax wins, at least in this course. If a sentence is syntactically ill-formed it's a problem with syntax. If a sentence is syntactically well-formed, then we can evaluate whether it is semantically well-formed or not.

If you ever teach a yodeling class, probably the hardest thing is to keep the students from just trying to yodel right off. You see, we build to that. —Jack Handey, Deep Thoughts.