Allowed surface locations of [+wh] phrases apparently depend on semantics—if so, how and why?

Question

Consider

Harvey's machine can resemble a human completely or not at all.

1a) ... The extent to which it resembles a human is determined by its software.

1b) ... To which extent it resembles a human is determined by its software.

The placement of "to which" in the surface form is optional, and both sentences (appear to) carry the same meaning. But we can't do this in general:

2a) ... The machine to which I'm referring is in the room.

2b) ... # To which machine I'm referring is in the room.

At first, it seems like it might be to do with "extent" vs. "machine". But consider

Harvey has made several machines.

3a) ... The machine to which I'm referring is unclear.

3b) ... To which machine I'm referring is unclear.

which makes me suspect this is at least in part a more general semantics issue.

(If these judgments don't sound correct, try un-pied-piping the "to", and then they definitely do sound correct.)

In 2, we're talking about a particular machine that's in the room. But in 3, there's a sense in which we haven't "evaluated" 'which machine' yet—we're talking about my denotation of the machine, not a particular machine itself. It seems almost like a de re/de dicto distinction. Or maybe there's just a [+Q] somewhere. But if so, where?

How is this formally modeled? Is there a syntactic distinction between the two situations? If not, how do we model the acceptability of having a [+wh] phrase in a different position under certain semantics but not others?

Or just generally—what's going on with the placement of the wh-phrases in these sentences?

Answer 1

I think you have to be more careful with the examples you're using here. Your parallelism seems odd (at least to me). There are issues of transformation (passivization), surface PP order, and argument structure (transitivity) that should be considered. You didn't take this into consideration. I believe this is the source of your confusion.

Let me explain. Your examples (1a)/(1b) contain passivized sentences with the verb determine. They also contain an embedded ditransitive resemble (X resembles Y to Z extent). However, this is not the case in (2a)/(2b). There’s no passivization nor ditransitivization in (2).

(2) (also (3)) pairs are examples of a CP specifier (clausal subject) where wh-movement is not allowed. It’s not only restricted to occur, but it doesn’t exist at all. It doesn’t exist because the [+Q] would never have scope over the whole clause given its embedded position as a specifier. This is why your comparison sounded a bit odd to me.

These points are critical to determine the difference between (1) and (2) pairs.

The active version of (1a) is (1a') and the active version of (1b) is (1b') below. The difference between (1a) and (1b), as you mentioned, relates to the optional movement of wh-phrase (PP). I think you’re wrong here. In both active sentences (1a’)/(1b’) below, there is an indirect wh-movement. The difference is that it surfaces as DP+PP (the extent to which) in the first, but as PP+DP (to which extent) in the second. This is why when these examples are passivized, as you did in (1), not only the wh-phrase but the whole CP moves.

(1a') Its software determines [CP the extent to which it resembles a human t]
(1b') Its software determines [CP to which extent it resembles a human t]

I hope you can see the picture now. So your question: “is there a [+Q] in (2) (or (3))?”, the answer is NO, because although the movement would take place, there would be no wh-scope over the clause.

If you want to compare two sentences, always remember that they fall within the same syntactic environment. (1) is passive with an embedded ditransitive verb. (2) is active with no ditransitive verb. Comparison between them would not result in anything.

Answer 2

(1a) [the extent to which it resembles a human] is a noun phrase. Here you have a noun extent which is modified by a relative clause [to which it resembles a human]

(1b) [to which extent it resembles a human] is an embedded question. it's a CP in which [to which extent] moves to Spec-CP.

Here are the two structures schematically:

Let me give some evidence for this before moving forward. The verb wonder can have an embedded question object, but not a full NP object. You can see this here:

(i)
a. I wonder [_CP which person caused the problem].
b. *I wonder [_NP the problem].

Notice that (1b) [to which extent it resembles a human] works with wonder, while (1a) [the extent to which it resembles a human] does not. Therefore, (1b) is an embedded question, while (1a) is a noun phrase.

Both work as the subject in (1), because (1) is a passive derived from:

(ii) (someone) determined ____ by its software.

The verb determine can have either an NP or an embedded question as its object. We can see this here:

(iii)
a. we've determined [_NP the problem].
b. we've determined [_CP [which person] caused the problem].

The reason (2a) works, but (2b) doesn't, is that an NP can be the subject here, but an embedded question can't. We can see this here:

(iv)
a. [_NP the problem] is in the room. (semantically odd but clearly grammatical)
b. *[_CP [which person] caused the problem] is in the room. (semantically sensible but clearly ungrammatical)

The reason both (3a) and (3b) work is that either an NP or an embedded question can be the subject of this sentence. We can see that here:

(v)
a. [_NP the problem] is unclear.
b. [_CP [which person] caused the problem] is unclear.