Before I knew that school buses were yellow, I experienced them as tangerine. I argued with my betters about this. If a good part of education is training the judgment to accept the consensus view, then I was educated to understand what adults were convinced of: that school buses are yellow in the same way roses are red and skies are blue. They certainly aren’t tangerine, or cheddar, or as the Crayola box of my youth judiciously instructed, "yellow-orange."
Alma Deutscher might sympathize. From the time she began speaking, she was the subject of a linguistic experiment: her father refrained from associating "sky" and "blue" to see if she would spontaneously put them together—a natural equation, like two and two making four. It didn’t happen, so eventually her father started asking her point-blank what color the sky was. She didn’t answer until she was twenty-three months old: "white." "It took another month until she first called the sky ‘blue,’" her father wrote,
and even then it had not yet become canonically blue: one day she said "blue," another day "white."… In short, more than six months had passed from when she was first able to recognize blue objects confidently until she named the blueness of the sky. And it seems that her confusions were not entirely over even by the age of four, because at this age she once pointed at the pitch-black sky late at night and declared that it was blue.
Had Alma seen that the sky is blue the way school buses are yellow and roses are red, she might not have given her father, Guy Deutscher, a linguist at the University of Manchester, further proof of what he suspected: that the cultural nuances of our mother tongue influence our perceptions and possibly whole structures of thinking. In Through the Language Glass: Why the World Looks Different in Other Languages (Metropolitan; $28), the history of color perception provides the thin edge of the wedge into a topic that is surging again to the forefront of linguistics—not whether but how much language tells us what to see, and hence what to think.
This is a hypothesis that seems intuitively correct, but it is one that has had to fight its way back from intellectual ignominy. In 1936 Benjamin Lee Whorf published An American Indian Model of the Universe, which asserted that "the Hopi language is seen to contain no words, grammatical forms, constructions, or expressions that refer directly to what we call ‘time.’ " The idea that the Hopi experienced time in a radically different way from Europeans soon became the cornerstone of "linguistic relativity," or the Sapir-Whorf hypothesis, the new dominant paradigm in linguistics. (Edward Sapir, Whorf’s teacher, was an armchair linguist influenced by Bertrand Russell and Ludvig Wittgenstein’s work on the limits of language. Einstein’s theory of relativity was also an ambient influence.) In 1983 the linguist Ekkehart Malotki published Hopi Time, in which he demolished Whorf’s research "in 677 pages of small print." This complete reversal of cherished assumptions induced a revulsion proportionate to the excitement Sapir-Whorf once generated. For the past several decades we have accepted the Chomskyan version of language—that it is a genetic and therefore universal component of the human brain—and have seen it championed in the pop science press by the untergiversating Steven Pinker.