This book is a collection of ten papers divided evenly into two parts: “Edges” and “Heads and projections”. The main aim of the book is to bring together the research relating to the role of edges as well as of heads and projections in current theoretical approaches to the study of syntax and the interfaces.
In the first paper in Part I, “Why edges are needed”, Boeckx discusses the three meanings of edge present in current syntactic theorizing and tries to find a common denominator for them. The notion edge in the first sense corresponds to the left periphery of the clause as proposed by Rizzi (1997), which underlies the discourse-oriented interpretational effects. In the second sense, edge is understood as the edge of a phase (Chomsky 2000), and an edge feature of the phase head is held responsible for triggering the so-called indirect feature-driven movement. The third sense of edge figures in current understanding of the operation Merge (Chomsky 2008), where the edge feature is taken to be a trigger for Merge. Boeckx argues that all three senses can actually be unified. He notes that they correspond to the three positions where a moving element can surface, i.e., in the first merge position (the third sense), in the intermediate landing site (the second sense), and in the ultimate landing site (the first sense). He observes that these three positions correspond to three projection levels (minimal, intermediate, and maximal) and concludes that edges are necessary for the same reason that projections are necessary.
In the paper entitled “Bare nouns with different edges”, Ihsane analyses the semantic and structural differences between non-coordinated and coordinated plural and mass bare nouns, as well as indefinites with the determiner du/des ‘of the’ in French. She first examines the readings associated with these three classes of NPs and observes that du/des-NPs can have three different readings: namely, referential, quantificational, and property-denoting; non-coordinated plural bare and mass nouns can have only the property-denoting reading; and finally, coordinated plural or bare nouns can have the property-denoting and quantificational readings. Ihsane links the meaning differences among these classes of NPs with their syntactic structure. She proposes three types of projections at the left edge of the nominal phrase: Property Phrase, Quantifier Phrase, and Subject-reference Phrase. She argues that those NPs [End Page 153] which have all three readings require all three projections at their left periphery, whereas the NPs with two readings project only two of these projections, and finally, the NPs with just one reading have only one projection at their left edge.
In “Implicit agentivity without agents in the syntax”, Cabredo Hofherr and Dobrovie-Sorin analyse passive and middle structures with se in languages such as French and European Portuguese. They argue against Raposo and Uriagereka (1996), who posit the existence of two different types of se in the two relevant structures, one [+arg] and the other [−arg]. The authors propose instead that both constructions host the same type of se. The difference between them, they argue, follows from the position occupied by the theme DP, which is in Spec, IP in middles, but remains in a VP-internal position in passives. An important contribution of the paper lies in distinguishing two types of implicit agents, i.e., those that are syntactically active and those that are syntactically inactive. The authors argue that neither of these agent types is projected in the syntax, but the former is projected at LF if Spec, IP is semantically empty—that is, either filled with the expletive or not projected at all. The agent is then present at LF in eventive se-passives in both French and European Portuguese, but is absent in middles, which lack agentivity and express just a property.
Munaro’s paper called “On the edge of particles, interjections and short answers” confronts Chomsky’s (2008) contention that interjections lack an edge feature. Working with data from Italian, she shows that interjections can precede or follow the clause...