Consider the sentence "Who will John meet in the classroom?". What would the (s-structured) x-bar tree of this sentence look like? Does "will" still simply go in the T node, or is it a CP-specifier, or what? I'm having trouble charting it since "will" and "meet" are separated by "John".

I'm new to the site, so I'm not sure how plausible this is, but I'd appreciate a depicted tree if possible.

On an unrelated note, in a sentence such as "Which picture of himself will John hate?", what is considered the reflexive noun phrase (as it pertains to Condition A of traditional Binding Theory)? I know it'd normally be "himself", but is it "John" in this case, since "himself" precedes it?


  • "will" sits in C head position in this example May 29, 2014 at 18:26

1 Answer 1


Kiyoshigaang points to the direct answer to the question. On a GB-style analysis (Government and Binding), who is in spec of CP and will is in C of CP; both move to those positions from lower in the tree.

But the question references T of TP. Thus the question seems to be located in the generative tradition after about 1995, since the generative tradition did not start acknowledging TP until about that time. Producing an analysis that can be agreed upon becomes more difficult in this regard, since as the generative tradition has progressed, more and more functional categories have been added, e.g. AgrP, FocP, TopP, etc. A "correct" analysis and direct answer to the question is therefore increasingly debatable.

This site is inundated by this type of parse tree question. In my view, the questions bear witness to a major problem with the state of mainstream Chomskyan syntax and modern syntax courses in linguistics in general. Students are being required to produce parses of sentences using an increasingly complex system; they are therefore understandably frustrated and thus seeking help where they might find it. But due to the complexity of the system, it is often difficult to produce coherent answers to the questions. The experts often disagree.

This is of course not the answer that the question is seeking. Producing a clear answer would, though, be difficult, and drawing a modern X-bar-style tree that acknowledges TP is laborious. Perhaps this answer will, however, help generate skepticism about the value of producing such parses to begin with, in a system of syntax that is (in my view) overly complex.

  • 1
    The second paragraph is factually wrong: TP was introduced in Syntactic Structures in 1957 and it plays a crucial theoretical and empirical role in the works of Klima, Emonds and Jackendoff from the 1960s and early 1970s.
    – Olivier
    May 31, 2014 at 5:45
  • @Olivier, if you would, please cite a page number where TP appears in a syntax tree from one of the works you mention. If I can, I will check the source. If I'm wrong, I will acknowledge my mistake and correct my answer. May 31, 2014 at 5:50
  • @Olivier, you've raised a serious accusation. I've asked for help verifying that the accussation is correct and my answer "factually wrong", and what I receive is silence. I'm going to poke around the library this afternoon. I'll take a look at Chomsky 1957. I'm pretty sure I will find no syntax trees showing TP in that book. I'll be back! May 31, 2014 at 14:52
  • You should do your own homework, but for the benefit of anyone interested section (5.3) of Syntactic Structures (which is not really an obscure reference) explicitly points out the need of a rewriting rule involving Aux and though no tree appear there, this is only because Chomsky points out that the complete set of rewriting rules carries more information than a tree and points out explicitly how to construct the latter from the former. As for the others, you can have a look for instance at Klima and Bellugi (1966) page 190.
    – Olivier
    May 31, 2014 at 17:55
  • Then Jackendoff (1972), Emonds (1976), Emonds (1978), Roberts (1985), Baker (1988) and Pollock (1989), which, taking for granted that there is a T node above V popularized the hypothesis that there are two nodes above V: v and T. That hypothesis (the three-layered V) was only generally accepted in generative syntax in 1995, and I suspect you might be conflating it with the existence of T, which has been discussed since the very inception of the generative syntax.
    – Olivier
    May 31, 2014 at 18:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.