By now, you'll have heard one of the most distinctive statistics from the first presidential debate between Hillary Clinton and Donald Trump Monday night: Trump interrupted Clinton 51 times, while Clinton interrupted him only 17 times. Part of this is doubtless down to Clinton's strategy: frequently, instead of pushing back at Trump, she allowed him to ramble on and create his own difficulties. It also reflects the fact that Trump was reactive rather than proactive in many areas of the debate, working off Clinton's responses. But it's important to make it clear that this is not just an isolated example of candidate behavior, or indeed merely an example of sexism (which it is). It reflects a larger truth borne out by behavioral science about how men and women interact in professional spaces, and one that Clinton, I suspect, knows exceptionally well through experience.
Vox, commentating on the interruption pattern, calls it "exhaustingly familiar;" but it's also an unprecedented new aspect of the presidential campaign because of Clinton's gender. As the first female candidate for the presidency debating a male, Clinton's experience shows the interruption problem, and other issues surrounding our treatment and assessment of female leaders, on a new stage. It's linked to a broader issue: Sheryl Sandberg, in an op-ed for the New York Times, pointed out that in a study of senators, male senators with higher rankings spoke more on the floor, while female highly-ranked senators didn't, probably because of societal expectations of being "pushy" or "aggressive." How much women speak, even in spaces where they have power and are expected to put their views across, is still significantly curtailed by expectation and their own self-censoring — and, as we saw Monday night, by men.
Here's the science behind Trump's interruptions.
How Trump Represents A Larger Phenomenon
Scientific findings are pretty clear on this one: when it comes to interactions between men and women, men are far more likely to interrupt women. One of the latest manifestations of this was a study done in 2014 in which 40 participants, 20 male and 20 female, had conversations with trained participants; both the male and female participants interrupted more, and used more dependent clauses, when they were talking to women. Dependent clauses are usually a sign of intricate sentences; the researchers, talking to New Republic about the results, think that people of both genders might switch to a more elaborate way of speaking when they talk to women, because it's seen as a more feminine way of expressing yourself.
But the real study that established the prominence of interruption as a societal gender problem was published back in 1975. Called "Sex Roles, Interruptions And Silences In Conversation," it was done through observation of 31 different conversations recorded across a campus, which were then analyzed for interruptions and speech patterns. The conversations were mostly casual, between undergraduates, and covered a wide variety of relationships. The researchers found that when men talked to men and women to women, the interruption rate was pretty even. But when men talked to women, men took up 96 percent of the interruptions. And these weren't arguments in front of 100 million people between two individuals under a lot of pressure; they were just undergrads chatting over coffee.
More recently, Women In The World, which investigated interruption as a phenomenon after the famous incident at SXSW where Google CTO Megan Smith was repeatedly interrupted by executive Eric Schmidt on a panel about gender equality, dived into the issue in 2015. Professor Stanley Deetz, who specializes in interpersonal communication, told them that “a tremendous amount of academic research (as well as personal experiences) support the claim that men interrupt women more in meetings (and that women interrupt women more)." But there's more to the story than simple statistics.
There Are Certain Situations Where Men Are More Likely To Interrupt
The 1975 study wasn't uncontroversial. A follow-up study by other researchers claimed that interruption was actually pretty gender-neutral, and a debate raged about whether the claim was correct at all. In 1998, a meta-study published in Sex Roles looked at 43 different interruption studies to see if the 1975 study was actually correct, and found that it actually gets pretty complicated.
Overall, the review found, men do actually interrupt women more than they interrupt men; but how and with what frequency depends on a lot of different factors. One was how people defined "interruption;" if they meant "usurping the speaker's turn at talk with the intent of demonstrating dominance," rather than just overlapping the ends of their sentence, then men showed a definite increase in interruptions. Another factor was the size of the group: men are far more likely to interrupt women in big groups of three or more, or in front of large audiences (hello, Trump), as a way of demonstrating "group dominance."
Activities also mattered: men interrupted a lot more in "open-ended and unstructured situations," rather than ones where everybody was focused on doing a certain task. The researchers said that a lot of interrupting behavior seems dependent on context and what we view as interruption. Some of us, for instance, may jump in on the end of a friend's sentence because we enthusiastically agree with what they said and want to emphasize it, which isn't really a dominance gesture.
One would think that in professional contexts, when a person is consulting a female expert (an accountant, perhaps), the interruption rate would be lower. But this context seems to actually do the opposite. Several different studies, from 1992 to 2001, show some striking trends: both female and male doctors interrupted their female patients more than their male ones, male doctors interrupted everybody more than female ones (including other doctors), and female doctors were far more likely to be interrupted by male patients than male doctors were. Being female in a professional scenario, whether as a client or as an expert, will lead you to being interrupted. This casts a light on the Trump-Clinton scenario; if one looks at them as competing professionals, this data indicates Trump's significantly more likely to interrupt Clinton.
The Bottom Line
Interruption is, a lot of the time, about power: controlling the conversation, the room, the direction of peoples' thinking, and unsettling, silencing or derailing the other person. The fact that it may occur unconsciously (one cannot assume that Schmidt was consciously interrupting Megan Smith as a show of dominance on a gender equality panel) makes it even more intriguing as a social measure. This fits with what we've observed to be Trump's psychological combat plan: use strategies of dominance to try and grab hold of the conversation.
But Clinton's performance against that strategy looks to have been successful; she didn't look rattled by the interruptions, didn't let them throw her off track, and despite a heavy bias towards Trump when it came to actual speaking time, is seen as the winner in many analyses of the debate's outcome. Her lack of response made him look petty and childish; it's a demonstration that interruption may not be the only way to gain the upper hand in a conversation.