Labels and roots ed. by Leah Bauke, Andreas Blümel (review)

Compositionality, order, projection/labeling, and displacement are core properties of concern when inquiring into universal grammar (UG), and the relevant questions about labeling need further exploration (Chomsky 2013). Basically, labeling provides the necessary information for syntactic objects (SOs) to be interpreted at interfaces. Chomsky (2013, 2015) ignited the discussion when formulating two proposals for labeling: symmetry-breaking movement and shared feature + Agree. Labels and roots, edited by Leah Bauke and Andreas Blümel, contributes to the intensive discussion about the properties of labeling and its interaction with roots.

The volume is based on the Labels and Roots Workshop in Germany in 2014. It not only elucidates the syntactic and semantic contributions that lexical categories (e.g. roots) could make to syntactic derivation and generation, but also delves into issues closely related to roots and labeling. Thus the roles labeling plays within syntax (or morphology) and semantic and phonological interfaces are revealed, addressing issues regarding the properties of roots and labeling/endocentricity that Chomsky (2013, 2015) demonstrated. In the following, we outline the key points of the four parts of the book, examining how this volume contributes to the exploration of ontological and procedural issues in grammar.

The chapters by Samuel David Epstein, Hisatsugu Kitahara, and T. Daniel Seely (EK&S), Chris Collins, and Aleksandra Vercauteren constitute the first part, framing a theoretical background for the volume. In ‘Merge, labeling and their interactions’, EK&S summarize their previous papers related to Chomsky’s (2013, 2015) analysis of labeling by minimal search, and then explicate the labeling algorithm (LA) as creating the relation ‘member of’ for X and Y (SOs), forming a set {X, Y}. Labeling does not exist in narrow syntax (NS), but is created during Transfer via seeking for relevant object-identification information within unlabeled syntactic outputs generated by simplest Merge. Labeling in this way can explain ‘obligatory exit’ and ‘obligatory halting’ in English and Japanese, without resource to Move-over-Merge, lexical array and subarray, or even phase (CP/v*P), realizing the simplest means to reach the maximum explanatory force.

Collins (‘Merge(X,Y) = {X,Y}’) mainly elaborates Transfer. First, Collins shows how phrase structure rules and transformations are unified under Merge and summarizes thirteen properties of Merge. Among them, the labellessness of Merge projects, which drives unbundling of the syntactic operation, that is, endowing Transfer with the duty to create structures with labels. Transfer is specified as Transfer_SM (SOs into the sensory-motor system) and Transfer_CI (SOs into the conceptual-intentional system), with the latter further unbundled into Transfer_PF and Transfer_SM. Crucially, Transfer_SM externalizes SOs one after another, with reference to phase and the asymmetrical relation among SOs. Consequently, a simple derivation fits with labelless Merge, without appealing to lexical array but by listing lexical items as their own line of derivation. There is no Agree in UG, and no parameterization in Transfer but in different instances of internal Merge.

In her contribution (‘Features and labeling: Label-driven movement’), Vercauteren demonstrates, with English and Italian data, the mutually beneficial relation between LA and cartographic syntax, which assumes a universal hierarchy of ordered functional heads (features). To wit, the hierarchy of functional heads provides landing places for XP movement, because functional features could label SOs via minimal search and establishment of agreement between SOs, and moved constituents would find a label in NS. Movement comes for free, but is not triggered by features. Meanwhile, the fact that LA seeks only features permits cartography to maintain its assumption.

EK&S and Collins clearly provide readers with a historical examination of labeling and relevant issues, and most importantly clarify how to transfer SOs at/into interfaces. This enables readers to understand how the proposals of labeling and the key factors conjectured successively for syntax, like interface conditions, free Merge...