American Scientist

To the Editors:

As a psychologist with an interest in human cognition, I read with great attention two articles in the July–August issue of American Scientist : Brian Hayes’s Computing Science column “The Manifest Destiny of Artificial Intelligence” and Dom Massaro’s article "Acquiring Literacy Naturally." They left me asking, what about learning and teaching the social–emotional aspects of words’ meanings? How do we teach human and machine language processors about the emotional valences and appropriate pragmatic uses of offensive words?

Granted, language theory is much simpler if we leave out the negative emotional content of words, but if we do so, we end up with a polite but inaccurate picture of human communication. And what of Massaro’s anecdote about children using obscenities at the breakfast table? Before Johnny and Jane entered school, they knew that “a-s” and “sh-t” were bad words, not to be used at the breakfast table, and that Johnny did not expect to be punished for saying “Cheerios.” Humans live and learn in groups that experience intense emotions, which are conveyed in words that mirror emotional intensity. I’m still wondering when language researchers will factor emotion into theories of language and word meaning.

Timothy Jay
Department of Psychology
Massachusetts College of Liberal Arts
North Adams, MA

Dr. Massaro responds:

Dr. Jay highlights an important aspect of language use that is often slighted in theories and research. We might fault the “language is unique and special” tradition, which is rejected by my view that speech and language understanding is a form of pattern recognition. Emotion in language use is currently a large part of cognitive and brain science research. A Google search of “language and emotion” yielded 2.4 million results. Concerning the “Cheerios” anecdote, it must have originated in earlier generations when children were more innocent and adults less forgiving.