Asymmetric inference towards the antonym: Experiments into the polarity and morphology of negated adjectives

Abstract

In this paper, we investigate the interpretation of negated antonyms. A sentence such as Peter is not tall can be understood as meaning either that Peter is not tall tout court or that Peter is rather short (inference towards the antonym; ITA). We present the results of two experiments, in which we test two theoretical predictions. First, according to Krifka (2007), it is reasonable to expect a stronger ITA effect for positive versus negative adjectives. Second, elaborating on Krifka (2007), we expect ITA strength asymmetry to be greater for morphological antonymic pairs than for non-morphological pairs. In the first experiment, ITA strength was assessed implicitly by having participants judge the pragmatic acceptability of sentences involving a negated adjective and its antonym. In the second experiment, we collected explicit inferential judgments. The results of both experiments support the two predictions. We also discuss theoretical and methodological issues concerning the different notions of polarity.

Not unreasonable: Why two negatives don't make a positive

“…We've extended earlier work showing that negated lexical antonyms (e.g., not dull) are distinct from positive adjectives (e.g., sharp) when presented simultaneously (Giora, Balaban, Fein, & Alkabets, 2005) to both the case of morphological antonyms (e.g., not unhappy) and when presented in isolation (single utterance conditions). We also provide convergent evidence that supports the phenomenon of negative strengthening or inference towards the antonym, where negating an adjective (e.g., not intelligent) can lead to the stronger interpretation than literally implied (e.g., rather stupid; Ruytenbeek, Verheyen, & Spector, 2017;Gotzner, Solt, & Benz, 2018).…”

Scalar Diversity, Negative Strengthening, and Adjectival Semantics

“…The third line of Figure 1 shows the effect of negative strengthening on the stronger scalar item: If the statement John is not stunning is negatively strengthened, the resulting meaning is inconsistent with John being attractive (i.e., it is consistent with him being average looking or ugly). In a similar vein, an asymmetry between positive and negative expressions has been pointed out (Brown and Levinson, 1987 ; Horn, 1989 ; Ruytenbeek et al, 2017 ). For example, while an utterance like John is not tall is often interpreted as John being rather short, the statement with the antonym ( John is not short ) is unlikely to be strengthened in order to convey that John is rather tall.…”

“…The assumption is that the positive adjective denotes a desirable property in contrast to its antonym, relating the negative asymmetry to euphemism and understatement (Brown and Levinson, 1987 ; Horn, 1989 ; Krifka, 2007 ). However, it is easy to find counterexamples to this asymmetry and it is unclear which notion of polarity is the relevant one (e.g., emotional valency vs. negative morphology) or which adjective constitutes the positive form (see especially Ruytenbeek et al, 2017 ). We will return this issue below.…”

Typicality and Graded Membership in Dimensional Adjectives

“…The observation that in antonym pairs typicality varies more for the positive member than the negative member is as far as we can tell a new one. Being unmarked, positive adjectives might have broader semantic meanings than negative adjectives do in that the latter conjure up a subsection of the underlying scale, whereas the positive adjectives can be associated with the entire scale (at least in some linguistic environments, such as questions; “How short is Mary?” asks one to consider low heights, whereas “How tall is Mary?” does not, see Ruytenbeek et al., ) . Furthermore, all the adjectives we consider are associated with ratio scales in which the zero point is a true zero, with the exception of the pair warm / cold where participants were given only an interval scale, namely the Celsius scale (which admits of negative values).…”