This week, the assigned papers addressed children's ability (or lack therof) to interpret various types of scalar implicatures and compared these abilities to those of adults. The Barner paper showed that children failed to correctly interpret scalar implicatures unless they were either able to generate relevant scalar alternatives (e.g. numbers) or were presented with these alternatives. This explains why children do not succeed in interpreting some as not all; they have not realized that some and all are scalar alternatives. The Stiller paper further specifies this finding by showing that children succeed at above chance levels in ad-hoc scalar implicatures due to contextual factors. Furthermore, adults and children succeed at scalar implicatures due to statistical information about the rarity and informativeness of descriptive features.
These papers made me think of a story I read some time ago about a mother and her young son. The son reported to his mother that he had made friends with a new boy at school, and he asked if he could invite his new friend over for a playdate. She said yes, and asked for a description of the boy. Her son said that his new friend had curly hair, dark eyes, and that he liked to play. When the new friend came to their house, the mother was surprised to encounter a black child since her son had not told her that his friend was black and so she had expected to meet a white child. They lived in a primarily white area, so according to the Stiller paper, that the the new boy was black should have seemed a more informative (statistically rare) feature to the mother and son than his curly hair, dark eyes, or playfulness. Interestingly, it seems that the boy's race did not seem particularly salient to the son, while race was indeed a salient feature to the mother. Perhaps this is due to the fact that the son had not learned how to categorize people by race but he did know how to categorize people by other features. This relates to implicature because "curly hair" and "dark eyes" are weaker descriptors of the friend's appearance relative to "black." Because the son did not use the stronger descriptor "black," the mother assumed that the friend had the mentioned features without being black. It is likely that the son did not have access to "black" as an alternate descriptor of appearance, and thus he did not intentionally apply whiteness when neglecting to mention blackness.
It is interesting to note how our expectations and interpretations are influenced both by what we are explicitly told and the negations that we assume in cases ranging from numerosity to race.
Interesting example story! This ties in perfectly to the statement in the Barner paper that people are more likely to explicitly say the unusual things. Since children haven’t had the life experience to know what is unusual (their probabilities of encountering certain things does not match the actual probability of encountering that thing in the world), their language usage doesn’t align with people who have seen more of the world.
ReplyDeleteI thought your story was a fascinating example of scalar implicatures! It's also interesting to note the child's deductive reasoning instead of inductive reasoning when describing his friend. There are not a lot of negative examples (African American) in this white community so the boy used deductive reasoning instead.
ReplyDeleteI really liked the story as well because it shows that this phenomenon of using children not using inferences is not contained to just linguistics. I definitely do think that this inference is developed with age as one grows accustomed to what is deemed "normal" and rare.
ReplyDelete