Understanding imprecise absolute adjectives: role of context and processing cost
The meaning of absolute adjectives such as 'full' or 'straight' has been claimed to be based around relating objects to maximal degrees of their described property: A full cup of water is only full if it contains the maximal amount of water in it (see Kennedy, 2007). However, these adjectives are often used imprecisely to mean 'almost full' or 'straight-ish'. What are the repercussions of imprecision for sentence processing? One straight-forward hypothesis, put forth by Syrett et al. (2010), is that even though imprecise usages might be tolerated, they come at a processing cost relative to precise usages, resulting in longer processing time for understanding imprecise vs. precise usages of absolute adjectives. An alternative hypothesis, suggested by Aparicio et al. (2016), is that a precise interpretation of absoluteadjectives could be costlier to process than an imprecise interpretation given that, in general, there are more contexts that support an imprecise vs. precise interpretation. According to this view, contextually licensed imprecise interpretations should be generally preferable and less costly than precise ones, a view similarly advocated by Krifka for the interpretation of round numbers (2002). Finally, a furtherpossibility, argued for by Van der Henst et al. (2002) as well as Gibbs and Bryant (2008) for the production of round numbers, is that processing cost of precision will be mediated by considerations of relevance, as defined by Relevance Theory (Sperber & Wilson, 2002). The current work in progress is an investigation on the processing cost of contextualized imprecise adjectives. I will be presenting preliminary results from 3 pilot studies, showing how comprehension rate and processing cost of imprecise absolute adjectives is mediated by context.