Beyond Words - Language Blog

Recent Studies Examine Categories of Language and Thought

Three centuries ago, in his Essay Concerning Human Understanding, John Locke examined the idea of linguistic and psychological categories by juxtaposing them with a hypothetical language in which every object that exists would have its own name. Instead of using the word “bird” to mean any member of its species, people would be forced to refer to each bird by a separate name. The most daunting and impractical element of such a lexicon is its complexity: While it would no doubt be useful to have two discrete words for a bird in flight and one perched in a tree, speakers of such a language would be burdened with a staggering amount of data to memorize.

This amusing mental exercise finds resonance in a recent study published in the journal Science by Carnegie Mellon University’s Charles Kemp and the University of California at Berkeley’s Terry Regier. The study analyzes data collected by linguists and anthropologists for 566 world languages, examining how those languages organize spheres of activity and understanding into categories. Dr. Kemp uses as an example the term “grandparent” to illustrate linguistic variations in kinship categories. While some languages have distinct terms for each maternal and paternal grandparent, others have only one or two words that refer to those family members. Here, the trade-off is between simplicity and usefulness: The more complex linguistic kinship systems are more specific and thus more useful in distinguishing between individuals, while the more sparse systems are simpler, but less informative.

The researchers have labeled the intersection of simplicity and usefulness the “optimal frontier.” Previous studies have demonstrated that this juncture exists within a number of categories, including those that house our understanding of color and space. In a study of spatial categories across languages, researchers from the University of Chicago, the Max-Planck Institute, and the University of California at Berkeley asked speakers of English and Dutch to organize objects into groups based on the similarity of the spatial relation exhibited. The way in which the objects were organized aligned with the way in which the two languages employ prepositions, demonstrating just how well each individual language suits the needs of its users.