More than words

               Both of this week’s readings discussed “scalar implicatures” and children’s and adults’ abilities to reason about them.  Scalar implicature is the concept that the use of a weaker term on a scale as opposed to a stronger one signifies that the stronger one is untrue, even though logically it could be true.  For example, if someone said that “most of the tables have four legs”, one would assume that not all the tables have four legs (even though if all had four legs, so would most).

                The first paper focused on figuring out the explanation for why young children have trouble excluding the stronger form of the scale when they hear a weaker statement, a task adults have no trouble with.  It explored theories such as counterfactuality and linguistic alternatives. One concept presented that appealed to me was the pragmatism of the fact that we tend to describe things by their rarest features.

               The experiment of the Barner paper focused on children’s knowledge of scalar alternatives and the inclusion of the word “only” on their performance on scalar implicature tasks.  One very helpful thing this paper did was outline theoretical steps necessary to “derive a scalar implicature”.  A concept that came up repeatedly in the paper was the difference between numerical scalars like 1-10 and context-independent scalars like “some” or “all”. Considering that children tend to learn scalar implicature for numbers sooner, I am led to believe that “real-world” context is one of the driving factors of children’s difficulty with scalar implicature. Five of something will always be the same amount, but “some” varies with context. 

               The idea of scalar implicature reinforces one of the main ideas that I’ve learned and we’ve discussed in the course: a lot of information that we communicate through language isn’t present directly in the words we say. While this concept does vary from last week’s discussion and readings about how information about speakers is carried through language, they reinforce the same idea.  Before this course, I wondered why a computer that could speak like a human hadn’t been created. It seems now that every week I see more complexities in language.  Do you all agree a computer would have to be able to reason about the world and the behaviors of others in order to speak correctly?

               Another trend I’ve noticed emerging over the past few weeks is that speakers tend to try to save time and energy in many ways in their speech, and sometimes sacrifice perfect clarity of meaning in doing so.  This shows up in how we cut sounds out of words (although sometimes a word with every word enounced how it is spelled sounds wrong) or how we don’t say “some but not all of the tables have four legs.”  It even shows up in dialects; some dialects like AAVE involve the elimination of some words or sounds from speech, which can make it more difficult for others to understand.  Are these shortcuts really worth the time we save?