Support The Bulwark and subscribe today.
  Join Now

Biden, Trump and the Science—or Instinct—of Persuasion

The thinking behind both parties’ attempts to influence voters’ thinking.
September 1, 2020
Biden, Trump and the Science—or Instinct—of Persuasion
Joe Biden speaks on stage during the 2015 Concordia Summit at Grand Hyatt New York on October 1, 2015 in New York City. (Photo by Leigh Vogel/Getty Images for Concordia Summit)

With the number of voters still up for grabs in the election dwindling—a late-July poll found that only 13 percent of the electorate remain truly undecided—and the battle for support intensifying, the persuasive strategies used by the two parties become increasingly important, and key differences become more striking.

The Democrats, perhaps not surprisingly, are leaning heavily into the science of persuasion.

A recurrent theme at the Democratic convention, emphasized by speaker after speaker—Kamala Harris, Barack Obama, Hillary Clinton, among others—was the urgent need for viewers to “plan” their vote. This reflects a research strategy developed through iterative experimentation over a decade ago by Todd Rogers and colleagues.

As one of his Ph.D. advisers (Max Bazerman) describes it (along with co-author Michael Luca), in 2003 Rogers was a “young Democratic pollster and recent Williams College graduate” who started graduate studies in social psychology at Harvard, then switched to the university’s organizational behavioral program, where he felt he could better pursue his interest in applying insights from behavioral science to the political process.

The rising Harvard star was soon connected with—and apparently brought critically needed pragmatic experience to—the “Consortium of Behavioral Scientists,” a group of scholars hoping to “apply the results of academic research to win elections and turn more of the electoral map blue.” These efforts were felt to contribute to Obama’s 2008 election, and were celebrated in Sasha Issenberg’s 2012 book The Victory Lab: The Secret Science of Winning Campaigns.

Rogers extended his work after his graduation, serving as founding executive director of the Washington, D.C.-based Analyst Institute, described by Luca and Bazerman as a “left-leaning organization” created to use “field experiments to establish a set of empirically proven best practices for interacting with voters.” One of this effort’s key findings was based on the psychological concept of “implementation intentions”—the idea that you’re more likely to do something if you have a concrete plan. Rogers and colleagues found that they could goose participation simply by phoning voters and asking several questions around their specific plans around getting to the polls; the actual answers didn’t matter—the point was simply to motivate them to think in concrete detail about the process.

The emphasis on voter intention on display at the 2020 Democratic convention reflects only the latest application of the science Rogers and other behaviorists have developed and refined. (Today, Rogers is a professor of public policy at Harvard’s Kennedy School of Government.)

The impact of psychological science can also been seen in the approach taken by organizations like Republican Voters Against Trump (RVAT), focused on persuading voters who supported Trump in 2016 to change their minds in 2020. At the core of this strategy is creating what the organization’s leaders refer to as a “permission structure,” a mental construct that makes it easier for someone to do something unfamiliar.

RVAT has solicited and published a remarkable collection of testimonials, often gritty and unpolished, from actual Republicans across the country—including many previous Trump supporters—explaining in their own words, directly to the camera, why they’re no longer supporting the president. Similar testimonials were also seen during the Democratic convention.

It’s difficult to locate the specific term “permission structure” in the psychological literature (a search of “PubMed,” for instance, reveals no items), but the concept is familiar. “That’s a good strategy,” Duke’s Dan Ariely, a professor of psychology and behavioral economics, told me. “There’s no question that showing [voters] that people from the ‘in’ group, and who are like them (connected to them or related to them) have changed their mind. That is incredibly important.”

The term itself has apparently been around “for years” in fields like marketing, according to Reuters, but it achieved particular prominence in 2013 when President Obama publicly used the term a handful of times, largely in the context of how his team was thinking about winning GOP support for a range of proposals he felt many Republicans agreed with in principle but resisted because of “politics.”

As Obama explained, “We’re going to try to do everything we can to create a permission structure for them to be able to do what’s going to be best for the country. But it’s going to take some time.”

The New York Times traces Obama’s embrace of the concept to chief strategist David Axelrod, who “cut his teeth helping to elect Black mayors in cities like Cleveland, Detroit and Philadelphia,” where success depended on “attracting a sizable percentage of the white vote.” Axelrod was famously successful in achieving this, thanks in part to a strategy of providing “third-party authentication”—endorsement from individuals and institutions “that white voters trusted to make safe, conventional decisions about whom to vote for,” offering in effect “mainstream validation.”

The value of this approach might also be seen in academic research on social norms—the idea that, as Luca and Bazerman write, “people act the way they see other people acting.” Empirical research from Rogers and colleagues demonstrated that voters were more likely to vote if you told them turnout seemed like it would be high than if you said turnout was looking to be low, even though you might think the opportunity for impact would seem higher in the second instance. Such is the attractive power of social norms.

The decision to feature several prominent Republicans—former governors John Kasich and Christine Todd Whitman, former representative Susan Molinari, tech CEO Meg Whitman, and former secretary of state Colin Powell—at the DNC earned Democratic leaders considerable critique from the progressive wing of their party. Yet DNC Chair Tom Perez passionately defended the decision, arguing it created a—wait for it—“permission structure” for disaffected Republicans to vote for Biden.


Republican strategists still working for the party (and not for RVAT or the Lincoln Project) are, of course also keenly aware of the science of political persuasion, even if their leaders are not necessarily steeped in the terminology. (When Obama first surfaced the phrase “permission structure” in 2013, for example, a spokesman for Senate Republican leader Mitch McConnell reportedly googled it because he had never heard it before.)

The emphasis at last week’s Republican convention on highlighting people of color, in particular African Americans (South Carolina Senator Tim Scott, Utah congressional candidate Burgess Owens, Kentucky Attorney General Daniel Cameron, and North Carolina desegregation activist Clarence Henderson, among others) may certainly reflect an effort to win over voters who’ve tended to vote for Democrats (only 6 percent of African Americans voters supported Trump in 2016). But it’s also likely intended to reassure white suburban voters already inclined to support Trump that doing so doesn’t make them racists. It’s arguably the same strategy used by Axelrod, but here used for a markedly different end: providing a permission structure for white voters to elect a Republican candidate.

While the design of the 2020 Republican convention may be informed by strategists steeped in the science of persuasion, the party itself is led by someone who seems to understand persuasion instinctively.

As Stanford Business School professor Jeffrey Pfeffer, author of Power: Why Some People Have It—And Others Don’t, reminds me, a lot of detailed analyses seem to miss something that’s “fundamentally important”—to wit:

non-rational, emotional elements like how people look, how they project power through their language and body language, whether they attract or repel others through their dress, symbolism, etc. Trump uses, and still does, strong language. And a lie, repeated often enough, becomes the truth. His intuition is to focus on a few messages consistently. The Democrats are all over the map.

Pfeffer’s description is borne out by events as recent as Trump’s dishonest (but widely reshared) assertion that the Democrats omitted “God” from the Pledge of Allegiance at their convention, and Trump’s constant evocation of “strength.” For example, in the August 23 press conference announcing that the use of convalescent serum for the treatment of COVID-19 has received an Emergency Use Authorization (EUA) by the FDA (more on that here), the president characterized the EUA as “a powerful term,” said the treatment represented “a powerful therapy that transfuses very, very strong antibodies” into patients.

More generally, the 2020 RNC convention as a whole seemed intent on offering a coherent, if counterfactual, emotionally appealing narrative, one that is easy to explain, easy to understand. It represents an attractive alternative to the often-grim reality outside, and to the incessant censoriousness of the progressive left. Through repetition—along with the consistent discrediting of alternative sources of information—the fable becomes the preferred truth.

The Democrats, predictably, may have landed on a strategy driven by data and logic and guided by reason, but how effectively can they compete with Trump’s marketing instincts, and his one authentic superpower: the intuitive ability to fabricate, hone, convey, and repeat—tirelessly—an easily digestible message appealing viscerally to many voters?

Trump doesn’t need science to recognize what science has in fact demonstrated, according to Baba Shiv, a professor of marketing at Stanford’s business school: virtually all human decisions are shaped by emotion. Shiv explains, “the rational brain is not good at being rational, but instead is good at simply rationalizing what the emotional brain has already decided to do, and this happens non-consciously.”

The narratives Trump constructs, and the pageantry the GOP employs, seems aimed at voters’ hearts, while the case prosecuted by the Democrats tends to focus on voters’ heads. Which will prove more persuasive?

We may have to wait for November to find the answer.

David Shaywitz

David Shaywitz is a physician-scientist at a biopharmaceutical company, an adjunct scholar at the American Enterprise Institute, and a lecturer in the Department of Biomedical Informatics at Harvard Medical School.