Does Language Change How We Experience Color?Print
By Jessica Love
May 3, 2012
According to one accepted view, living a life without language would be a lot like building a house without a tape measure: we can absolutely do it—there’s nothing magical about centimeters or feet, after all—but having one is surely more efficient than eyeballing everything, or conjuring up one’s own unit of measurement (bales of hay, perhaps, or footsteps). This is especially true when it comes time to order materials.
But, as I hinted in last week’s column aboutWhorfianism, some very smart people believe that language does more than assist us in sorting our own thoughts and communicating them to others. My favorite study supporting this view was published in 2007 by Jonathan Winawer (then at MIT, now at Stanford) and his colleagues. The researchers recruited a group of native English speakers and a second group of native Russian speakers for a simple computer-based color discrimination task. In this task, participants were repeatedly presented with three blue squares arranged in a triangle. The squares ranged in shade from very light blue to very dark blue, and one of them—the top one—was always identical in color to one of the others. Participants were instructed to decide which of the bottom squares matched it, and to press one key for “the square on the right” or another for “the square on the left” as quickly as possible.
Now, in Russian, the colors light blue and dark blue have different names, and are as different as pink and red are in English. Were a Russian speaker to use a single term for both, she would be seen as weird and annoying. But in English, with the possible exception of towel manufacturers (who must master the terms breeze, Pacific, quartz, steel, hydrangea, Hawaii, lagoon, marina, peacock, crystal, denim, and of course, bluestone), all blues are just blue.
The researchers reasoned that the matching task would be easier for participants if the two different colors had different names. That is, they predicted that the Russian speakers would be faster to discriminate between a light blue and a dark blue than between two shades that were equidistant from one another on the color spectrum but would nonetheless be classified using the same lexical term (e.g., a very pale light blue and a darker light blue). The English speakers, on the other hand, should find the two decisions equally difficult. And this is precisely that pattern that the researchers found.
This finding is so intriguing because it suggests that language doesn’t just serve as a handy mnemonic for recalling and sharing complex events. No, the way the Russians categorized the colors via language seemed to affect how they perceived those colors. This, in a simple task that didn’t explicitly involve language at all—in a task, indeed, where participants couldn’t have consciously accessed the color terms even if they wanted to: they didn’t have the time.
But the operative word here is “consciously” because linguistic knowledge was definitely being accessed. When the same task was repeated, but participants were given an additional instruction to silently recite a string of numbers, the effect disappeared. Why? Repeating “1, 2, 6, 8, 7, 3, 4 … 1, 2, 6, 8, 7, 3, 4 …” is linguistically demanding: without using language, we just can’t remember a string like this. But when our linguistic knowledge is put to use reciting numbers, it’s not available for other things, like categorizing colors.
Such studies suggest that the metaphor of language as tool might not be sufficient. Beyond helping us build that house, language—if ever so slightly, and if we’re not too busy silently rehearsing something else—may change how we see the house in the first place.
Jessica Love is a contributing editor of the SCHOLAR. She holds a doctorate in cognitive psychology and edits Kellogg Insight at Northwestern University.