The role of extralinguistic knowledge in the derivation of scalar implicatures

The paper reports on an experimental study that reveals a difference between two types of under-informative utterances with the quantifier "nekotoryj" ('some') in Russian.The experiment was conducted as an inference task based on [Geurts, Pouscoulous 2009; Val Tiel et al. 2016] with some modifications.The experiment showed that both types of utterances more often have inference-based interpretations rather than logically based interpretations.This finding is in line with the mainstream experimental research on scalar inferences in various languages.At the same time, the experiment demonstrated a twofold difference between the two types of utterances.On the one hand, under-informative utterances like "Some elephants have trunks" yield scalar inferences (scalar implicatures) like "Not all elephants have trunks" to a lesser extent than under-informative utterances like "Some books are on the shelf" (the scalar inference is "Not all books are on the shelf").On the other hand, the former utterances are processed longer than the latter ones.The reason is that the scalar inferences from the former conflict with the extralinguistic knowledge shared by the speaker and the hearer.The hearer might infer that the speaker does not have this belief and, consequently, the hearer might not draw a scalar inference.However, this reasoning conflicts with extralinguistic knowledge and is time-consuming.Our results contribute to the ongoing debate on utterances with quantifiers within the field of theoretical Natalia A. Zevakhina, Polina A. Ponomareva, Sofia V. Krasnoschekova 271 and experimental pragmatics and support the contextual theory of scalar inferences.The present study is the first experiment testing scalar inferences in Russian.