Lexical comparisons of signed languages and the effects of iconicity

Abstract

Lexical comparisons of signed languages present new methodological challenges not found in comparisons of spoken languages. Two standards for comparing wordlists are examined using a sample of four European sign languages that are not known to be related to each other and a second sample of different dialects of the signed languages of Spain. The use of different standards is shown to affect the numerical results; comparing signs on the basis of probable historical relatedness typically yields percentages that are 5-10% greater than comparisons on the basis of similarity. The amount of iconicity inherent in signed languages affects the wordlist scores even more. Comparing lexical items that were chosen for their low potential for iconicity resulted in significantly lower scores among unrelated languages than did word lists of basic vocabulary or highly iconic signs. Conversely, the non-iconic word list comparison showed greater similarity between closely related language varieties. Therefore, wordlists that are low in iconicity give more insightful results than wordlists that include significant numbers of iconic items. 1. Similarity and cognate studies There are two different approaches to lexical comparisons that have been used in the study of sign language variation, each with distinct objectives. Studies of LEXICAL SIMILARITY investigate to what extent the words of two languages are similar, often with the hopes of making a further correlation to the intelligibility between languages. For example, family and the Spanish equivalent familia are very similar to each other. If an English speaker heard the word familia, he might be able to guess the correct meaning. In most cases, the greater the lexical similarity between two variations, the more likely it is that they will be able to understand each other. Lexical similarity is only one of many factors that determine intelligibility; nevertheless it is a relatively easy place to start. The second main approach looks for HISTORICAL RELATEDNESS. Two words that are historically related are called COGNATES. While lexical similarity is most concerned with how languages appear at the present time, cognate studies are most concerned that the two varieties had the same historical root. It is possible that at one time two words may have been historically very similar, but with the natural changes that occur over time, the two words have evolved into forms that are so distinct as not to be easily recognizable. For example, the words eight and the Spanish equivalent ocho do not look or sound at all similar, yet they can both be traced to the Latin word octo (Campbell 1998). In making judgements about similarity, the assumption is that a monolingual Spanish speaker would not understand the English word.