Science Has Come Up with a Way to Measure Cultural Preferences
By BETH HARPAZ
“How can we understand contemporary popular photography that grows by billions of images every day?” Manovich asks in his introduction. “Or contemporary music as represented by hundreds of millions of songs shared by twenty million creators on SoundCloud? Or the content of four billion boards on Pinterest?”
You can’t understand these massive archives until you can “see” them, Manovich says, and you can’t see them without data science. The term “cultural analytics” refers to the use of computational analysis and pattern visualization to measure, comprehend, and compare cultural preferences, styles, behaviors, and artifacts.
“If we do not learn to see at sufficient resolution what people today create and how they behave culturally, any theories or interpretations we may propose based on our intuitions and received knowledge are likely to be misguided,” he writes. When Manovich’s lab analyzed data on 16 million Instagram images, a million manga pages, and 5,000 impressionist paintings, all “assumptions … based on intuitions and accepted knowledge were overturned.”
For example, when Manovich’s lab digitally analyzed 5,000 impressionist paintings for characteristics like light and color, only about 25% of them exhibited the light tones, color, and modern subjects found in the genre’s 140 most reproduced images. The rest displayed darker, monochromatic palettes more typical of other 19th century art. While the standard canon of 140 prized works may represent impressionism’s avant-garde “best,” the analysis of the larger archive shows “continuities” with other art of the era and “makes visible the larger context.”
“Our common histories of culture often focused too much on the original elements and the new inventions,” Manovich writes. “This focus on the avant-garde in human history comes at the expense of the norm, the typical, the conventional.” Cultural analytics, in contrast, can “look both at the boring and the exciting, the norms and the inventions.”
And while it may be radical to use data science to compare vast collections of cultural artifacts, themes, techniques, and topics, Manovich says it’s also a way of continuing “the humanities’ most basic and oldest way of thinking.”