Scalar Implicature

Abstract

We review experimental evidence regarding the development of scalar implicature in children. Scalar implicatures are inferences that arise when utterances like “Mary ate some of the cakes” are interpreted as “Mary ate some but not all of the cakes.” The evidence suggests that, even though the mechanism for generating scalar implicatures in children is in many respects adult-like, children nevertheless face limitations in computing such conversational inferences from what the speaker said. We highlight the importance of the findings for the development of pragmatic inference, language acquisition, and communication in general. We also identify open questions and promising areas for future research.

Children's attitude problems: Bootstrapping verb meaning from syntax and pragmatics

“…But if children have access to speaker meanings, their difficulty with scalar implicatures cannot be due to an inability to assume speaker meanings that go beyond literal meaning. Instead, it might stem from difficulty figuring out what alternatives are relevant in a given context (for overviews, see Lewis, ; Pouscoulous, ; Papafragou & Skordos, ). Our results thus add to the growing body of literature that argues that young children's pragmatic abilities have been underestimated: children's pragmatic sophistication is present early on—and is, in fact, necessary to get attitude verb learning off the ground.…”

Some Pieces Are Missing: Implicature Production in Children

“…In all these experiments, the great majority of children accepted the weaker term as compatible with a stronger one, whereas adults would either consider them to be incompatible or at the very least be equivocal. Taken together, these findings might suggest that young children are unable to derive pragmatic inferences prompted by scalar expressions (for reviews on developmental findings on scalars, see Siegal and Surian, 2004 ; Pouscoulous and Noveck, 2009 ; Katsos, 2014 ; Papafragou and Skordos, 2016 ).…”

“…It is worth noting that these accounts are not necessarily mutually exclusive. The first three, in particular, are sometimes presented by their supporters as potentially complementary (Katsos, 2014 ; Papafragou and Skordos, 2016 ). The debate to establish the best account of children's early difficulties with scalar implicatures is still very much raging.…”

The Role of Speaker Knowledge in Children's Pragmatic Inferences

“…also Foppolo, Guasti, & Chierchia, ; Guasti et al., ; Skordos & Papafragou, ). Moreover, children are more likely to infer that weak statements exclude a stronger alternative in tasks that do not involve explicit, binary acceptability judgments or in tasks that make scalar alternatives more easily accessible (Barner, Brooks, & Bale, ; Katsos & Bishop, ; Papafragou, ; Papafragou & Skordos, ; Papafragou & Tantalou, ; Pouscoulous, Noveck, Politzer, & Bastide, ; Stiller, Goodman, & Frank, ). Furthermore, 5‐year‐olds seem to distinguish between natural language expressions in terms of their informational strength: When given a choice between a weak (e.g., or ) and a strong (e.g., and ) logical term to describe the outcome of a story, children prefer the strong, more informative one when they know it to be true (Chierchia et al., ; cf.…”