Conventional wisdom would indicate that the verb speak cannot be used in deictic there constructions, which supposedly only allow for existence and motion predicates. However, this semantic constraint fails to account for the acceptability of the verb-specific there speak construction. Therefore, this study provides a detailed investigation of contextual situations to identify appropriate conditions in which this unique construction can be felicitously used along with a consideration of its pragmatic and intersubjective functions. Thus, the study emphasizes both the role of the particular verb and the critical significance of context in accounting for the there speak construction.

This paper deals with the head transparency that is exemplified by the following compounds: the healing-time of all ills (Boase-Beier (1987)) and a guidebook to modern linguistics (Namiki (1985)). Their right-hand constituents are “transparent” in that the left-hand constituents ignore them and license the PPs, contradicting the Right-Hand Head Rule. Naya (2016, 2017) syntactically analyzes healing-time-type compounds by arguing that only semi-lexical nouns can be transparent. However, this analysis cannot be applied to guidebook-type compounds. Clarifying the semantic relation between the constituents of the guidebook-type, this paper proposes that the head can also be transparent when it represents a part of the lexical meaning of the non-head. Accordingly, head transparency in compounds can stem from these two different factors.

This article starts with a review of Blümel (2017), which addresses the question of successive cyclicity, attributing it to a labeling failure in which the set of two maximal projections is unlabelable. Blümel also shows that an item stops moving when it shares the same type of feature with another item. His analysis is attractive but faces some challenges when we observe infinitival constructions and there constructions. I then propose a mechanism of partial labeling under which each feature consisting of φ-features is entitled to be a label. I also show that partial labeling is observed in past participle agreement in Romance languages and in double object constructions.

Based on Chomsky’s (2001) Uniformity Principle, Miyagawa (2017) advances the thesis called Strong Uniformity (SU), which states that all languages are equipped with the same inventory of grammatical features, and that those features are overtly manifested in one way or another in each language. The present article places under careful scrutiny some of the major findings and proposals that Miyagawa puts forth, especially those concerning the syntax of ‘why’ and Nominative/Genitive Conversion, with particular attention paid to the syntactic operation known as Feature Inheritance. It is also demonstrated that working with Miyagawa’s core ideas and implementing them in a slightly different way may yield fruitful results.