Living on "The Edge" in Derivational Syntax
In some respects this is stating the obvious, but the concept of “edges” has played a prominent role in modern syntactic theory. To see this, one need only note how extensively early generative treatments of syntax looked into the effects that structural edges had on extraction—Ross (1967), Bresnan (1972), and Chomsky (1973, 1982), for example, studied in depth the properties of structural boundaries (edges) that limited syntactic movement, such as the presence of specified subjects; it is fair to say, we believe, that explaining these edge effects became the central preoccupation of derivational syntax from the Extended Standard Theory (EST) model to the Government-and-Binding (GB) framework. To account for edge effects, derivational syntax throughout the 1970s and 1980s proposed that phrases had positionally defined edges—the COMP position in EST and the X-bar SPECifier position in GB—and that edge-sensitive filters of various sorts (the Specified Subject Constraint, the Tensed S Condition, Bounding Theory, etc.) circumscribed permissible syntactic movements.
As syntactic theory shifted from the representational GB-framework to the derivational Minimalist Program (MP) model, top-down phrase structure rules have been replaced by a structure building operation known as Merge, which compiles structures from the bottom-up rather than top-down. One of the not-well-recognized consequences of having Merge operations, however, is that the ability to define “edges” has been compromised. Although Chomsky (2001) does offer a positional definition of an “edge”—defining it as all the lexical material in a phrase that is to the left of the phrase-head—this definition is problematic: it is stipulative (designed only to serve Chomsky’s Phase Impenetrability Condition); it is indiscriminate (in the way it lumps together all material to the left of the head, specifiers and adjuncts alike); and it is representational/backwards looking (the “edge” cannot be determined until the phrase is completely built). Further, as troubling as the above notion of “edges” is in the Minimalist Program, the uninterpretable “Edge Features” (EFs) that Chomsky uses to activate the Merge operation and to bring lexical material to the edges of a derivation raise even more serious questions—about the nature of features and about the necessity of having lexical features be interpretable by the Conceptual-Intentional and/or Sensorimotor performance systems.
Despite the fact that there are problems with defining what “edges” and “edge features” are and what functions they serve, it is still widely assumed, however, that syntax largely happens on the peripheries of derivations, and that if we can unravel how edges and their features work, we will understand syntax at a much deeper level than we do now. The papers in this volume take on this important challenge, trying to identify and explain the properties and functions of “edges” and “edge features.”
Scope and Content of this Volume
The contributions in this volume address critical aspects of the notion of “edge,” focusing on refining the definition of this concept and on exploring ways in which “edges” shape syntactic structures and their interpretations. Timothy Osborne’s article Edge features, Catenae, and Dependency-based Minimalism investigates edges from the perspective of a Dependency-based version of the Minimalist Program. Osborne argues that “edge features” mark the syntactic functions of lexical items and that these features play an instrumental role in building syntactic structure left-to-right in terms of dominance relations (rather than bottom-up in terms of constituency relations). For Osborne, syntactic operations use Edge Features to form dominance units known as catenae.
Halldór Sigurðsson’s article On UG and Materialization proposes that the Universal Lexicon (the “lexical” component of UG) contains two elements: an initial Root—Root Zero—and an initial functional feature, Feature Zero, identified as the Edge Feature (duly named “zero” since they are void of any content). According to this analysis, human language emerges from the following sequence of operations Copy, Merge of Root Zero and Merge of Feature Zero, ad infinitum. Sigurðsson argues that his analysis of UG provides a natural explanation for language externalization and language variation.
Michael Putnam and Thomas Stroik develop an argument against the uninterpretable Edge-Features advanced by Chomsky (2004 et seq.) and Sigurðsson (this volume). In their article Syntax at Ground Zero, Putnam and Stroik expose a conceptual flaw in using vacuous Edge-Features as the chief means of structure building: namely, the inability of such features to explain how a derivation begins. Putnam and Stroik contend that structuring building, from beginning to end, involves the “Edge Features” of lexical items, where an Edge Feature is the highest unchecked feature contained in a hierarchically-arranged set of interpretable features of lexical items.
The second set of papers in this volume investigates the properties of edges of clauses. In Moving Towards the Edge, Michelle Sheehan and Wolfram Hinzen argue that reference in human language is shaped grammatically; in particular, it is an edge phenomenon in which the more “edge-heavy” (their term) a DP, v*P, or CP phase is, the more referential force the phase will carry. By showing that there is a strong correlation between grammatical structure (especially phase structure) and semantic interpretation, Sheehan and Hinzen offer strong support for Chomsky’s 2001 phase-theoretic syntax.
The final contribution in this volume, Tejre Lohndal and Paul Pietroski’s essay Interrogatives, Instructions, and I-languages: An I-Semantics for Questions, argues that movements to edge positions, such as wh-movement to the edge of CP, provide semantic instructions for how to assemble mental representations from grammatical structures. Lohndal and Pietroski cash in on a long-held axiom that the leftmost edge of a sentence can license a kind of abstract that makes it possible to use sub-sentential (mood-neutral) expressions to form questions. As a result, Lohndal and Pietroski maintain that the edge of a phrase is a locus for a “secondary” semantic instruction that provides information about how to use the mental representation assembled by the “primary” instruction encoded in the rest of the clause, cp. Chomsky (2005: 14).
The papers in this volume take on the challenge raised by Ross (1967) and Chomsky (1972) of investigating the effects that “edges” have on structuring building and on interpreting syntactic structure. They look at the structure building properties of “edges” from the perspective of the Minimalist Program (in Sigurðsson, and in Putnam and Stroik), and of Dependency Grammar (in Osborne); and they probe the consequences that “edges” have on semantic interpretation in Sheehan and Hinzen and in Lohndahl and Pietroski. These papers provide compelling arguments for, and evidence for, placing “edges” at the center of syntactic analysis.
- Bresnan, J. 1972. A Theory of Complementation in English Syntax. MIT doctoral dissertation.
- Chomsky, N. 1973. Conditions on Transformation. In S. Anderson and P. Kiparsky (eds.), A Festschrift for Morris Halle. New York: Holt, Rinehart, and Winston, 232-86.
- Chomsky, N. 1982. Lectures on Government and Binding. Dordrecht: Foris Publications.
- ––. 2001. Derivation by Phase. In M. Kenstowicz (ed.), Ken Hale: A Life in Language. Cambridge, MA: MIT Press, 1-52.
- ––. 2004. Beyond Explanatory Adequacy. In A. Belletti (ed.), Structures and Beyond. Oxford: Oxford University Press, 104-131.
- ––. 2005. Three Factors in Language Design. Linguistic Inquiry 36: 1-22.
- Ross, J. 1967. Constraints on Variables in Syntax. MIT doctoral dissertation.