This paper analyzes the discourse conditions governing the choice between extraposition and nonextraposition of that-clause and infinitival-VP subjects. On the basis of a large corpus of naturally occurring data, it is shown that nonextraposition requires that the content of the subject be discourseold or directly inferrable. If the content is discourse-new, then extraposition is necessary. The choice between extraposition and nonextraposition for discourse-old and inferrable subjects is examined and is shown to depend on the discourse status of the predicate and on whether it is the predicate or the subject that links to the following discourse. The paper ends with a discussion of the syntactic position of nonextraposed sentential subjects and concludes that it cannot be the same as that of fronted sentential complements. This means that the common discourse properties of fronting and nonextraposition must be linked to their common linear ordering properties, rather than to a common syntactic position.

The Principle of Phonology-Free Syntax (PPFS) is a proposed universal principle of grammar that prohibits reference to phonological information in syntactic rules or constraints. Although many linguists have noted phenomena that appear to them to be in conflict with it, the appearances are misleading in all cases we have examined. This paper scrutinizes four instructive cases in French that appear to falsify the PPFS. Section  deals with the alleged relevance of syllable count to the description of attributive adjective placement ; section  addresses the validity of a rule mentioning consonantality in stating the agreement rule for adverbial tout ; section  turns to the issue of preposition choice (e.g. en vs. au) with geographical proper names ; and section  takes a look at a purported case of phonological reference in stating the rule for ellipsis of a clitic pronoun and an auxiliary in a coordinate structure. In each case we bring independent evidence to bear on the problem in order to show that the analyses employing phonology-sensitive syntactic statements are in error and the prediction of the PPFS is confirmed.

No abstract

Right-Node Raising (RNR, Ross 1967; Chaves 2014) has been claimed to require phonological identity between the missing material and the shared element. Our corpus investigations provide examples of RNR with verb form mismatch with and without syncretism in English and French. Two acceptability experiments show that lack of phonological identity does not affect the acceptability of RNR. We argue further that RNR without phonological identity cannot be taken to be a case of cataphoric VP-ellipsis in French and that it should not be analyzed as such in English. As regards the status of RNR with verb form mismatch, two positions are available: either it is considered to be grammatical, in which case the phonological resolution principle of Pullum & Zwicky (1986) does not hold, or it is considered to be ungrammatical but repaired (or " recycled", Arregui et al. 2006; Frazier 2013). The high acceptability of cases with mismatch compared with ungrammatical controls casts doubt on the applicability of the recycling hypothesis in such cases. In order to account for the broader range of data established by our corpus and experimental results, we propose a new analysis of RNR based on lexeme identity rather than form identity.