"The cat that ate my homework for fun will upset my teacher."

Hello! I created this sentence to help me understand the concept of EPP. Assume this is how the major components should be correctly labelled after moving the entire DP from Spec-VP to Spec-TP: [TP [DP The cat that ate my homework for fun] [T'[T will][VP upset my teacher]]] enter image description here

Now I am realizing that I don't know what is happening within the DP.

Here is my attempt at parsing this DP, without including any movements: [DP [D the] [NP [NP [N cat]] [CP [C that] [TP [T ] [VP ate my homework for fun]]]]] enter image description here I am confused as to what is / whether there is a subject in this VP and TP. Is "the cat" a subject selected by "ate" that got moved to where it is now?

Edited with trees and additional context: We are currently learning about the EPP, which states that a tensed TP must have a subject (DP). I understand the movement of the main DP from Spec-VP to Spec-TP. However, I am unsure if this requirement also applies to a TP within a CP adjunct, and if it does, what the subject would be and how it moves in my sentence. (I understand that different analyses may label the phrases differently, but that is beyond my concern. Please assume that the first tree I provided is correct.)

  • 1
    "The cat that ate my homework for fun" is an NP, not a DP.
    – BillJ
    Commented Feb 20 at 12:10
  • 1
    @BillJ The analysis I was taught treats this as a DP in X-Bar Theory, but I don't think this even matters for my question
    – hangrycat
    Commented Feb 21 at 0:08
  • Better find another teacher then!
    – BillJ
    Commented Feb 21 at 17:41
  • @BillJ oh no guess I need to ask for a tuition refund from my uni and return the textbooks I bought for my ling classes!
    – hangrycat
    Commented Feb 22 at 7:41
  • 2
    The DP analysis is completely standard. Only the point of attachment of the CP with respect to D and N is a problem
    – Alazon
    Commented Feb 22 at 20:26

1 Answer 1


It is not so much a question for the EPP but for the analysis of relative clauses. You analysed the "that" as a C-head, and that looks ok to me. However, there is some uncertainty in the literature whether it is a conjunction or a pronoun. If it were a pronoun, it would be the subject in your example. If not (which looks more probable to me), then we basically have the case of a relative clause with a gap. Another such construction with a gap would be: the cat [I own -- ]. An argument expression may therefore be missing in a relative clause, no matter whether there is a visible C° or not.

In the Chomskyan model, "empty operators" have been invented for this purpose, so the structures would be:

  • cat [OP [I own t ]]
  • cat [OP that [t ate my homework ]]

The "t" is the trace you get by moving the operator from an argument position to the SpecC relative clause marking position.

So the main part of the answer is that the EPP can be satisfied also with a phonetically empty subject. Big-PRO in infinitives is another example.

As I said, there may be controversies around the specific structures.

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.