La jerarquía implicacional de Berlin y Kay a partir de frecuencias

Identificador:  imarina:9443070

Handle: https://hdl.handle.net/20.500.11797/imarina9443070

Autores:  Antoni Brosa Rodríguez; María Dolores Jiménez-López

Resumen:
This study is concerned with the relationship between implicational hierarchies in semantics, in particular the one proposed by Berlin and Kay about colours, and the new possibilities offered by corpus linguistics and computational linguistics. The starting hypothesis is that a quantitative analysis of frequencies based on actual language use coincides with Berlin and Kay’s proposal, thought of in evolutionary terms. This leads to the following objectives: to allow the verification of whether the order of the original implicational hierarchy is respected on the basis of frequencies; to provide greater precision in the levels of the original hierarchy where there is more than one color; to make it possible to show the preference for the different synonyms in relation to the color in English proposed by Berlin and Kay (if any, in specific languages). The methodology employed combines the use of PanLex for lexical retrieval and Kontext for processing occurrences in all real texts in a universal and multilingual way. This leads to promising results, where it will be shown how all the problems encountered in a research of this caliber are being overcome. First, the evolution of this hierarchy, its subsequent implications, its current use and its advantages and limitations are detailed. Subsequently, a proposal for a colour research methodology is offered from a quantitative perspective and focused on linguistic typology. The protocol offered is designed, first of all, to provide data from different languages that allow for a review of the hierarchy based on the concept of salience. This quantitative complement allows the existing model to be enriched. Secondly, the data extracted from the action guide presented here allow for greater precision when hierarchising colours which, in Berlin and Kay’s canonical proposal, are on the same level. Finally, we also aim to reduce the degree of Anglocentrism of the proposal by expanding the English colours into the various language-specific divisions (or synonyms) with a weighting of these labels in terms of their frequency.