Abstract
Chomsky's extension of his Problems of Projection proposal retains the labeling algorithm and deduces EPP and ECP effects from labeling. However, the notion of weak heads, the crucial assumption in this extended proposal, is problematic both conceptually and empirically. Conceptually, the framework threatens to complicate Minimal Search, which ought to be simple, as an instance of the “third factor” in language design. Empirically, the framework cannot explain the derivation of English infinitival clauses, nor can it explain labeling in languages like Japanese. I resolve these problems by eliminating the notion of weak heads. In exploring syntax without weak heads, I reconsider Feature Inheritance. Given that a central argument for obligatory Feature Inheritance (involving simultaneity of feature valuation and Transfer) is invalid on Chomsky's approach, I propose that Feature Inheritance is an optional operation. Then, I revise a proposal that External Pair Merge of heads is involved in the derivation of verbs with a CP complement and in the passive construction. I pose two problems for this proposal, one involving theta roles and one involving the derivation of the passive of verbs with a CP complement. My framework derives not only instances deemed problematic under the External Pair Merge proposal but also the unergative and ergative constructions, whose derivations are otherwise unclear.
Original language | English |
---|---|
Pages (from-to) | 275-294 |
Number of pages | 20 |
Journal | Syntax |
Volume | 23 |
Issue number | 3 |
DOIs | |
Publication status | Published - Sept 1 2020 |
Externally published | Yes |
All Science Journal Classification (ASJC) codes
- Language and Linguistics
- Linguistics and Language