Early Understanding of Pragmatic Principles in Children’s Judgments of Negative Sentences

ABSTRACT

Adults find negative sentences difficult to process, but an informative context can facilitate processing substantially, suggesting that much of this difficulty may come from the pragmatics of negation. Are children sensitive to the pragmatics of negation as well? Although children perform poorly on many tests of negation comprehension, we argue that these past findings are due to children’s sensitivity to general pragmatic principles that govern communication, rather than the conceptual difficulty of negation. In Experiment 1, replicating previous work, we found that adults rated negative sentences as more felicitous in more informative contexts. In Experiment 2, we showed that children are also sensitive to the contexts of true negative sentences, with three- and four-year-olds also rating true negative sentences higher in more informative contexts. We discuss children’s understanding of negation and pragmatics in light of these results, arguing that the felicity of negative sentences for both adults and children is determined by the informativeness of these sentences in context.

Acknowledgments

Portions of this work were presented in Nordmeyer and Frank (2015a; 2015b).

Notes

¹ The experiment can be viewed at http://anordmey.github.io/neg-tablet/experiments/exp1/negadult.html.

² The model specification was as follows: rating context referent type negation frame + (referent type negation frame | subject) + (referent type negation frame | item). All categorical variables were deviation coded. Significance was calculated using the standard normal approximation to the distribution (Barr, Levy, Scheepers, & Tily, 2013). Data and analysis code can be found at http://github.com/anordmey/neg-tablet.

³ The experiment can be viewed at http://anordmey.github.io/neg-tablet/experiments/exp2/negkids.html.

⁴ The model specification was as follows: rating context truth + (1 | subject) + (1 | item).

Additional information

Funding

This work was supported by the National Science Foundation [GRFP].