If you watched any of the debates on CNN, you saw two worms at the bottom of your screen. Well, they looked to me like worms, or maybe caterpillars, scrunching and stretching throughout the 90 minutes. Actually they were real-time graphs — with one color for men, another for women — recording the instant reactions of members of undecided focus groups to what they, and we, were watching. As they listened to the debate, these influential voters turned their hand-held dials up into the plus zone when they liked what they saw, and down when they didn’t. CNN calls this its “exclusive on-air undecided voters meter.” What they should call it is junk journalism.
How many voters? In the second presidential debate, 35 people in Ohio were wired to the worms. Actually, they weren’t completely undecided. As CNN anchor Erin Burnett explained, half of the 35 were for Obama, and half for Romney, but they said they might change their minds. I’m guessing the split was actually 18 to 17, or vice versa, unless they turned up one voter in Ohio who was split in two, half for Obama and half for Romney, or maybe they found someone all for Obama and all for Romney simultaneously — Burnett didn’t say. (The 35 were also split between men and women, suggesting that the odd man out was more precisely the odd man-woman out.)
I don’t know whether the TV screens these focus groups were watching carried the same CNN feed that I or anyone else in America might have been watching. If they were — if their instant reaction to their own instant reaction could, in turn, instantly affect their own reaction — then Jorge Luis Borges and the makers of “The Matrix” have nothing on CNN.
Even for those of us in the audience not controlling the caterpillars, watching these meters’ ups and downs has been a strange experience. If you’re tuned to CNN, which brands itself as the only news network not committed to a candidate, your view of the debate is literally framed by the scrolling political vital signs of a non-nationally representative focus group. I bet it’s been pretty much impossible for anyone to watch the debates without paying attention to, and even being affected by, the impact of the candidates’ and moderators’ words and body language on this sample of a teeny tiny but immensely empowered sliver of the American electorate.
This made me a little bit crazy, especially when I found myself yelling “Yes!” to some things, like the president’s rediscovered willingness to nail his opponent, which the yellow line of undecided women didn’t much like at all. I was torn between feeling genuinely good about my guy getting his mojo back, but also wanting him to win over these voters who still can’t make up their minds despite all they’ve heard; whose belief in can’t-we-all-get-along comity is a suicidal strategy for countering ruthless Republican obstructionism; and who nevertheless are the magical swing voters in the magical swing states with the muscle to decide the election.
Framing a successful debate performance as the successful seduction of 35 undecided Ohioans disses other criteria for success. The meter readout of a group of people that, say, regularly consumes newspapers or watches “The Daily Show,” would likely take a different path. That graph might not predict how swing voters will break on Election Day, but it also might not discount the premium that at least some citizens want other citizens — and journalists — to put on facts, context, reason, history and reality.
Of course no one’s being forced to watch CNN’s swing-o-meter. But it can’t be long before real-time tabulation of the sentiments of various audience segments becomes an expected and common element of all infotainment. As we watch the TV screen, we’re already learning in real time what topics and attitudes are trending on social media, either because we’re simultaneously checking out another screen, like the Twitter feed on our smartphones (guilty), or because that information is embedded in the crawl at the bottom of the TV screen. The most popular news Web sites are already telling us which of their stories are the most popular right now so that we can check them out and make them even more popular. Self-surveillance is entertaining; we enjoy learning about us. But when technology puts a finger on the civic scale, when it skews what we esteem in political discourse, when it privileges popularity over other criteria for worth, an instant-reaction gizmo isn’t just fun, it’s potentially as subversive as the Electoral College, Citizens United or the ascendance of post-truth politics.
This election will likely come down to the last-minute decisions of a few thousand people in a handful of states. Both campaigns conduct nightly tracking polls sensitive enough to detect each passing zephyr in undecided voters’ minds. They’re constantly testing phrases and issues to figure out what will move the meter for single noncollege undecided women in the suburbs of Columbus and Orlando, or whoever the decisive ones turn out to be at the end of the trail. Media organizations are also collecting increasingly subtle data about their audiences, some of them swapping editorial judgment for real-time metrics about what their customers want so that they can give them more of it. Micro-pandering: that’s how you win elections and ratings these days, and yes, winning is what counts. But I can’t help fantasizing about an alternative reality where candidates and coverage don’t routinely blow off the highest common denominators in their publics.
Marty Kaplan is the Norman Lear professor of entertainment, media and society at the USC Annenberg School for Communication and Journalism. Reach him at firstname.lastname@example.org.