Labeling through percolation

International Journal of Chinese Linguistics, vol. 12, pp. 38–65

Abstract

Abstract In the field of generative syntax, the literature proposes conceptual notions of labeling or labeling algorithms to interpret the categorization / or parts of speech of a given syntactic object. Building on this mechanism, we put forth the argument that a revised feature percolation principle (FPP) is a necessary supplement to effectively categorize linguistic items, and it should be applied at each merge operation. The study demonstrates that s-selection and categorial features (binary specifications of [V] and [N]) serve the crucial role of labeling from lexical representations to phrasal representations. This process illustrates the categorization process for merge (X, Y) → {H {X, Y}}. To test the minimalist treatment of this principle, we conduct an examination of data from both English and Mandarin Chinese languages. I compare several approaches by which labeling can be done in syntax. These approaches include: (a) through merge ( Chomsky 1995 ), (b) through movement / or minimal search ( Chomsky 2013 , Epstein et al. 2014 , and Rizzi 2012 ), and (c) through probing ( Cecchetto & Donati 2010 ). Through empirical evidence, this study demonstrates that a feature percolation analysis effectively addresses the issues arising from a minimal search in both merge (XP, YP) and the merge (X, Y) problem. Furthermore, we derive constructions like VP ellipsis and VP fronting. Theoretically, this approach is well-motivated by the principle of compositionality found in linguistic items, and it offers a more economical labeling method. In empirical testing, we apply this configuration and mechanism to examine additional structures.

Authors 1

1. Liching Livy Chiu corresponding Aachen

   Leiden University

   Affiliation as printed

   Leiden University