Some notes on the EPP (by Jeffrey Punske)

The EPP can be a vexing principle for early practitioners of Chomskyan syntactic theories. This is, in part, because the theoretic apparatus that underlie the EPP have been shifted away from. Yet, the EPP in some form generally remains. Thus, the goal of this blog post is to sketch a brief history of the EPP and outline its current status. I will also discuss some remaining potential issues with the EPP.

What is the EPP?

In an introductory syntax class, the EPP is generally taught as the requirement that a sentence have a subject. It is supposed to be the reason why English expletive subjects like it exist in examples like (1).

(1)  It seems John is happy.

(2)  John seems to be happy.

(3)  *seems John is happy.

In sentence (1), John is receiving its theta-assignment from the predicate happy and Case from the embedded T; seems only requires a propositional complement which is satisfied by John is happy. An expletive it is inserted into the otherwise empty subject position. One goal of the EPP is to explain why such expletives are required. In sentence (2), John still receives its theta-assignment from the predicate happy but there is no Case available in the embedded clause so John raises into the matrix clause to get Case. Sentence (3) is ungrammatical because there is no expletive.

In many introductory syntax textbooks, the EPP (be it a feature, constraint, or principle) drives movement of the subject to the SPEC of TP. Adger (2003, 215), who utilizes an EPP feature to drive subject movement to SPEC of TP, refers to this nomenclature as a “historical accident”.  Carnie (2013, 238) provides a principle which examines grammatical outputs:

(4)  Extended Projection Principle (EPP)
All clauses must have subjects (i.e. the specifier of TP must be filled by a DP or CP) and lexical information is expressed at all levels.

Using both these discussions from Adger and Carnie—we can begin to unpack and understand the history part of the “historical accident” of the EPP. We will do so by examining the elements of the Carnie’s Principle.

Carnie provides the standard unabbreviated form of the EPP—the Extended Projection Principle. (A group led by David Pesetsky has argued that the EPP should be reacronymized the ‘Extra Peripheral Position). Of course, for something to be extended there should be a non-extended version of it. Carnie’s definition of the EPP cheats this a little and combines the definition of the Extended Projection Principle with the Projection Principle.

The Projection Principle, as stated in Chomsky (1982) states that (informally) “the theta-marking properties of each lexical item must be represented categorically at each syntactic level: at LF, S-Structure and D-Structure” (p. 8).

An illustration of the Projection Principle at work is found in Chomsky (1988: 74-76). In this discussion, Chomsky examines the Spanish verb afeita ‘to shave’ which is obligatorily transitive. He, observes following the Projection Principle that a sentence like (5) below must be underlyingly (6) with an empty argument (a trace) in the object position to preserve the theta assignment of afeita (and obey the Projection Principle).

(5)   Juan se    afeita
 Juan self  shave
 Juan shaves himself

(6)  Juan se [afeita t]

Thus, the Projection Principle is most easily expressed as the idea that lexical information—particularly theta-assignment—is preserved across syntactic representations.

What then is the relationship between the Projection Principle and the Extended Principle? To unpack this, we need to place ourselves in Chomsky’s (1982) analytic environment—which in many ways is distinct from today. In particular, Chomsky critically assumes that verbs directly theta-marked their subjects (1982: 10). This assumption alone puts the idea that clauses must have subjects well within the domain of Projection as previously defined by Chomsky, which was fundamentally about the preservation of theta-assignment.

However, subjects are generally trickier than objects and as Chomsky observes it is possible for pleonastic (expletive) subjects to arise (these are subjects with no theta assignment) and for nominalized heads to lack overt subjects. Following examples from Chomsky 1982:10.

(7)  It is clear that…

(8)  I expect it to be clear that…

(9)  The claim that…

For Chomsky, the presence of these non-theta marked subjects moves this out of being directly under the Projection Principle. Further, he notes that passives remove both the subject theta role and promote the object to subject position again fitting broadly with Projection but not fitting neatly within it. Thus, Chomsky termed the idea that clauses must have subjects as the Extended Projection Principle which captures the then real connection between the Projection Principle and the requirement of sentence-level subjects.

The syntactic framework surrounding discussion issues related to the Projection Principle and the Extended Projection Principle have shifted greatly. With the introduction of minimalist reasoning (beginning with Chomsky 1993, but chiefly Chomsky 1995 and beyond) there is a marked shift away from multiple representational levels. Thus, the underlying architecture of the Projection Principle no longer exists within the operative framework of most working syntacticians. We cannot reference D-Structure and S-Structure if they do not exist.

Similar issues arise with the Extended Projection Principle. Beyond being the extended version of a principle that is no longer in force. In particular, one of Chomsky’s chief underlying assumption which connects the Projection Principle and the Extended Projection Principle is the idea that the verb assigns the subject theta-role; however, since at least Kratzer (1996) it is been widely assumed that the external argument theta role is not assigned by the lexical verb.

So, in a way, we arrived at the “historical accident” that Adger referred to. But to get fully there, the EPP must remain as a theoretic tool—removed from context.

Chomsky (1995, chiefly Chapter 4) begins to strip away the connection between the EPP and the (Extended) Projection Principle. Under this view, EPP is a strong feature on T which attracts the subject (or an expletive in the right construction). In Chomsky’s 1995 view the EPP may be parameterized with some languages having the EPP and others not (as made explicit McCloskey’s (1996) discussion of Irish). See Lasnik 2001 for a more through discussion of Chomsky’s 1995 view of the EPP and a great example of an EPP-based syntactic argument.

Once the EPP was converted to a feature, the connection between its grammatical/analytic role and its name became less and less clear. EPP features became the driver of otherwise unmotivated movements into specifiers. The underlying principles and the explanatory purpose of the EPP weren’t always clear.

For Chomsky, the connection between the EPP and the original formulation was never totally lost. In two papers addressing Projection (2013: ‘Problems of Projection’, 2015: ‘Problems of Projection: Extensions’), he revisits issues related to the EPP. For Chomsky, the need to have the specifier of TP filled is not driven by a specialized feature or special principle—but issues of labeling. He suggests that the category T is “too ‘weak’” to label by itself and thus must have its specifier filled (Chomsky 2015:9). See Gallego (2017) for an alternative labeling view of the EPP that doesn’t utilize the weak/strong effect.

Adger, David. 2003. Core syntax: A minimalist approach. Oxford: Oxford University Press.

Carnie, Andrew. 2013. Syntax: A generative introduction. 3^rd edition. Oxford: Wiley-Blackwell.

Chomsky, Noam. 1982. Some concepts and consequences of the Theory of Government and Binding. Cambridge, MA: The MIT Press.

Chomsky, Noam. 1988. Language and problems of knowledge: The Managua Lectures. Cambridge, MA: The MIT Press.

Chomsky, Noam. 1993. A minimalist program for linguistic theory. In Ken Hale and Samuel Jay Keyser (eds.) The view from Building 20: Essays in Linguistics in honor of Sylvain Bromberger. 1-52. Cambridge, MA: The MIT Press.

Chomsky. Noam. 1995. The Minimalist Program. Cambridge, MA: The MIT Press.

Chomsky, Noam. 2013. Problems of Projection. Lingua 130: 33-49

Chomsky, Noam. 2015. Problems of Projection: Extensions. In Elisa Di Domenico, Cornelia Hamann and Simona Matteini (eds.) Structures, strategies and beyond: Studies in honour of Adriana Belletti. 1-16. Philadelphia: John Benjamins.

Gallego, Ángel. 2017. Remark on the EPP in Labeling Theory: Evidence from Romance. Syntax 20: 384-399.

Kratzer, Angelika. 1996. Severing the external argument from its verb. In Johan Rooryck and Laurie Zaring (eds.) Phrase structure and the lexicon. 109-137. Dordrecht: Springer.

Lasnik, Howard. 2001. A note on the EPP. Linguistic Inquiry 32: 356-362.