Girl, Interrupting

a sociolinguistic state of the union

2020: Mike Pence, debating Kamala Harris, tries to speak over Harris. She stares him down. “I’m speaking,” she says, firmly, calmly. The moment went viral. You can understand why. A famously misogynistic white man demonstrating disrespect interrupting a highly qualified woman of color; the woman asserting her right to speak.

August 2024: Kamala Harris, now the official presidential candidate of the Democratic party, is delivering a speech. She is disrupted by activists protesting Israel’s genocidal campaign in Gaza. The activists are unknown, their names and faces and races obscured in the footage. Harris makes a callback to her viral 2020 debate moment and proclaims, “I’m speaking.” To some, this is a perfect parallel to the 2020 moment. Again, a qualified woman refused to let others interrupt, disrespect, and silence her. To others, it was the opposite: one of the most powerful politicians in the country deciding who gets to hold the floor.

September 2024: Donald Trump, debating Kamala Harris, mocks his opponent’s own viral moment. When Harris tries to speak over him, he says, “I'm talking now, sound familiar?” The quip fails to land. It sounds like whining rather than an assertion of power, and, frankly, it sounds a little girly.

---

If you would like a break from thinking about the more urgent and depressing elements of election season, you’re in luck. A presidential election is a battle over many things. One of those things is language. With its sound-bite pageantry and symbolic verbal sparring, election season gives us a distilled look at the current politics of speech in America. 

Interruption is a pop-linguistics and pop-feminist hobbyhorse. Men, the thinking goes, interrupt women. They do this because interruption is a hostile form of domination, and men like to flaunt their power over women. And this is how we read the high-profile examples I listed above. Pence interrupts Harris because he is a misogynistic old white guy attacking a youngish woman of color, and Harris, in refusing to be interrupted, enacts a symbolic triumph not merely for herself but for women in general. Trump, four years later, mimics Harris’s refusal to be interrupted. In doing so, he demeans women and feminism—but also, maybe, makes himself look fussy and feminine. 

I don’t think it’s only liberals and feminists who think this way. Conservatives abide by the same premises, namely: that interruption is a form of domination, and is therefore the domain of men. It’s just that conservatives, as open misogynists, think that this state of affairs is very good. This is why Trump, one of our most openly anti-woman public figures, took such evident enjoyment in mocking Harris’s “I’m speaking” moment. It didn’t work, but he was taking aim at the idea that women should speak publicly at all.

But the middle example—Harris shutting down protests with the phrase “I’m speaking”—should alert us to the fact that our premises are flawed. Your beliefs and affiliations will inform your reading of that situation: who here is interrupting who? Who is unfairly dominating? Is the disruption of the protestors a reinforcement of the status quo that silenced women like Kamala Harris for most of American history, or is it a revolutionary tool to give voice to the people of Gaza, displaced and dying and ignored by the American political establishment that is funding their annihilation? Who, exactly, is speaking out of turn?

Our national ideas about interruption are, while not strictly inaccurate, oversimplified to the point of uselessness. They are ultimately themselves misogynistic, and quite racist too. And they bring us to obviously ridiculous conclusions, like: these anti-genocide protesters are being chauvinists by disrupting a politician’s speech, and she is being a feminist by refusing the disruption. When your premises make those kinds of readings possible, it’s time to start over from scratch.

---

A quiz. Which of the following would you refer to as “interruption”?

1. One speaker cuts another off midsentence, changing the topic entirely.

2. One speaker murmurs “pass the salt, please” while someone else tells a story across the table.

3. Three friends, maybe a little tipsy, laugh and talk over one another for minutes at a time, telling a story as a group.

4. One speaker tries to cut another off midsentence, but the first speaker refuses to be cut off, and the two go on speaking simultaneously for a while until one gives up.

5. A friend exclaims “seriously?” and “wow!” and “Mmm-hmmm” while you tell a story. You continue speaking at the same pace.

I would personally apply the label “interruption” only to examples 1 and 4. Others might apply it more broadly—I’ve come across people who would consider example 5 to be obnoxious rather than supportive. What’s clear is that interruption is a contested and subjective idea. 

For this reason, linguists usually use the term “overlap” rather than “interruption.” “Interrupt” carries a negative charge. It implies hostility, rudeness, and a failure to follow social norms. “Overlap” is scientifically measurable, and simply refers to any situation in which two or more participants in a conversation are speaking at the same time, for any amount of time.

---

What counts as interruption depends on the circumstance, and it depends on the speaker. The linguist Deborah Tannen theorizes that speakers exist on a spectrum from “high-involvement” to “high-considerateness.” These two types are characterized by opposite attitudes toward overlap and silence. We can think of them as either anti-overlap/pro-silence, or pro-overlap/anti-silence.

High-Considerateness speakers are anti-overlap/pro-silence. They prefer longer “inter-turn pauses,” or breaks between speakers in a conversation. The High-Considerateness speaker will take a nice long time before they start talking, just to make sure that their interlocutor is all done. When the High-Considerateness speaker is finished saying what they have to say, they will go quiet, because they assume that others will take their silence as a cue to jump in. In a hypothetical group of totally typical High-Considerateness speakers, there would be either one or zero people speaking in a given moment. There would be a pause between each speaker’s turn. Any overlap would be construed as impolite.

Then there are the High-Involvement speakers. They are pro-overlap/anti-silence. To the High-Involvement speaker, an inter-turn pause is evidence of awkwardness and disinterest. In terms of turn-taking, the High-Involvement speaker will continue talking until someone else overlaps with them. Once another speaker’s voice is heard, the High-Involvement speaker feels they feel they can safely hand things off. They are also more likely to interject throughout, adding exclamations or affirmations to signal that they are paying attention. In a hypothetical group of totally typical High-Involvement speakers, there would always be at least one and often more people speaking in a given moment. There would be overlap between every single turn. Any pause or silence would be construed as impolite.

The problem arises when people with dramatically opposed conversational styles meet. According to Tannen (and in my experience this is absolutely the case), each speaker unwittingly pushes the other to the most extreme end of the spectrum, precisely because they unconsciously assume that their interlocutor’s style will resemble their own. The High-Considerateness speaker will pause in order to “give the floor” to their High-Involvement interlocutor. The High-Involvement speaker, however, will hear that pause and panic, taking it as a sign that things are going badly. They will jump in as quickly as possible, making the High-Considerateness speaker feel as if they are being cut off and talked over. The overwhelmed High-Considerateness speaker will wait for their interlocutor to pause, since they see a pause as the signal that it’s their turn to talk. “Jesus,” they think, “this person’s still talking. I guess I won’t say anything yet.” But the High-Involvement speaker is thinking, “Jesus. This person really isn’t going to say a word. I guess I have to keep this going all by myself.” Seeing that their interlocutor isn’t jumping in at all, they will continue talking—in my experience, with increasing animation and intensity—to avert the possibility of a terrible silence. And so you end up with one person speaking when they’d really like a break, and the other silent when they’d really like to speak.

---

For the past few years, I have thought about Tannen’s theory of conversational styles almost every single day.

That is because I moved to the United Kingdom three years ago. I did not expect much of a language barrier, because I speak English, and many people in London also speak English. It’s the same language, I thought. I was wrong. As an American, my conversational style was and sort of still is wildly out of sync with the normative conversational style of the UK, or at least of Southern England.

In the UK, I am on the extreme end of the highly-involved side of the spectrum, and it takes a great and sometimes conscious effort for me to avoid overlapping the way I’m used to. If I use my normal turn-taking habits here, I will be perceived as incredibly rude. Most Brits by American standards, slot into the far end of the highly-considerate side of the spectrum, and would run the risk of being judged awkward wallflowers in a conversation with a group of Americans. I have learned to slow down my turn-taking only after several years of practice.

Previously, I lived briefly in Bulgaria, a place where I encountered what any person would recognize as a language barrier: most of my neighbors did not speak English and I, frankly, did not speak very good Bulgarian. But, though I speak native English and clumsy Bulgarian, I often felt more easily able to express myself in Bulgaria because Balkan norms around turn-taking more closely resemble the American ones I grew up with. Bulgarians are extremely high-involvement speakers—much more so than Americans, actually.

Of course, I’m acting as if these habits follow strictly national lines, but they don’t. It’s only in a very broad and relative sense that Brits are High-Considerateness and Americans are High-Involvement and Bulgarians are Extra-High-Involvement: the splits are regional and cultural and racial, too. Tannen’s original theory emerged from a Thanksgiving dinner, in which she observed Jewish speakers from New York in conversation with WASPs from California. If you have ever encountered these two groups of people, you can probably guess which speakers were more High-Involvement. She notes that Midwestern Americans would be considered very high-considerateness compared to East Coasters— but in a study comparing a group of Midwesterners with a group of Native American speakers in Alaska, the Midwesterners were the ones who displayed a lower tolerance for inter-turn pauses. And on it goes. It’s all relative, an enormous spectrum that never quite ends.

---

This is why the election-time pop-linguistics debates feel so brittle. These conversations assume that overlap has a fixed meaning: overlap is always a sign of boldness in conversation. It always carries the same tonal implications. Maybe it’s justified and maybe it isn’t, but it’s always a sign of dominance, or of the desire to wrest dominance away from someone else.

In this formulation, men interrupt because men are dominant, either rightly or wrongly. Women tolerate and endure interruption because they are sweet and demure, or because they are downtrodden and oppressed. And this is sort of true. The studies aren’t completely consistent, but there’s still plenty of research indicating that men overlap with women much, much more than women overlap with men. There is also plenty of research indicating that men overlap with women more than men overlap with other men. 

I don’t hesitate to read this as a sign of misogyny, disrespect and disinterest. I do think that men like to try and dominate women, and language is one arena where this happens.  What I object to, and what sociolinguists (especially in the subfield of Critical Discourse Analysis) tend to object to, is the one-to-one equivalence between gender-based oppression and the very act of overlap. Yes, men are dismissive of women. But that isn’t because they overlap with them in conversation.

When we examine “men” and “women” as two cultural groups, the way we might contrast British and American English speakers, we see that these two groups follow different norms when it comes to overlap and conversational style. Overlap is as much a “feminine” speech norm as it is a “masculine” one, if not more. For one thing, in single-gender groups, it’s women who tend to be the overlappers. The linguist Jennifer Coates writes that “all-female talk is characterised by,” (among other things), “collaborative turn-taking strategies (jam sessions)” while all-male talk is instead characterized by (among other things) “one-at-a-time turn-taking patterns” and “monologues.” 

Women are also more likely to make their own voices heard when they hear multiple other voices speaking at once. In “singly-developed floors”—situations set up so that one individual speaks at a time—women speak less than men. But in “collaboratively-developed floors,” where speakers overlap and jump in as they please, women and men speak the exact same amount. In other words, women seem to feel more comfortable in settings where multiple people are already speaking at a time.¹

This does not simply mean, however, that women are ruder, bolder, or more dominant than commonly believed. The better takeaway is that overlap has no fixed meaning. No single linguistic act has a fixed social meaning. When women overlap with one another, it isn’t usually an attempt to “wrest the floor,” change the subject, or override. Rather, groups of women tend to engage in what linguists call “cooperative overlap.” They interject to show interest, to affirm, exclaim, or agree. 

Coates theorizes that one reason women tend to overlap more in single-gender groups—whereas all-male conversations involve monologues—is because women avoid “playing the expert.” If we are to believe Coates’s interpretation, this means that women overlap, not because they are more intense and egotistical, but rather because they are more cautious about appearing to dominate. Women overlap with one another to help avert the uncomfortably possibility that one of them might come across as too authoritative.

“Power and solidarity are bought with the same currency: The same linguistic means can be used to create either or both,” writes Tannen. By the same token, linguists have observed instances in which husbands refuse to speak while their wives cajole and beg. We often speak about women being silenced by men. But in these instances, Tannen notes, it is silence rather than speech that serves as a weapon of patriarchal manipulation.

---

When we equate overlap with a set of personality traits, we end up negatively stereotyping many groups of speakers as aggressive and uncouth. It goes the other way too, of course. When we treat our own culture’s conversational norms as the default, we also end up stereotyping certain  speakers as shy, mysterious, or inexpressive. When we assume that overlap is always masculine and indeed misogynistic, despite the fact that women overlap more than men in many situations, we wind up characterizing a good number of women as undesirably masculine. And, when we treat language in facile identitarian terms, we end up in insane situations, asserting that a small group of protesters are somehow behaving misogynistically or unfairly towards the standard-bearer of a major political party.

---

Cooperative overlap can look aggressive to people who aren’t used to it, of course. But even when speakers really are trying to wrest the floor or disagree or talk over one another, this isn’t necessarily a show of hostility. In cultures where overlap is the norm, this behavior is a sign of intimacy. Sometimes, when I read linguists’ fieldwork transcripts of overlapping speech, I feel a prickle of familiarity and comfort. The formal notation of overlapping voices looks like how my own Jewish, Southern mostly-female family talked to one another growing up, especially in moments of warmth or affection.

I am not, obviously, trying to say that politicians on a debate stage are engaging in a pleasant moment of linguistic play. But national politics are a culture too. These individuals are taking part in a highly context-specific mode of discourse that both is and is not part of our broader American vernacular.

Increasingly, linguists have focused on male and female, not as fixed categories, but as performances in constant construction through speech acts. Men and women alike, to put it simply, play characters. They act up various masculine or feminine types. There is some validity to the idea that men interrupt women. At the same time, the more we make interruption a pop-feminist issue, the more people will perform the roles as created by this discourse: the demure interrupted feminine, the interrupting and dominant masculine. Lately, I’ve been watching the way that both men and women consciously steer clear of these roles, as if to avoid getting typecast. Male friends will earnestly apologize for interrupting me, which I find sweet but off-putting: I took the overlap as a sign of solidarity and interest! And of course women will sometimes lean, in a self-aware way, into the role of the indignant interrupted: this is precisely what Harris managed to do, to skillful effect, in 2020. 

National politics supercharges the performativity already driving gendered language use. Every campaign conversation is a performance in the literal sense, existing for the benefit of a quiet audience. Technology exaggerates this paradigm further, turning each of these political conversations into simulacra of real ones. The politicians on a debate stage aren’t talking to one another at all—they’re talking to a sea of silent interlocutors. Most of the people involved in the conversation can’t overlap, they can only observe and then participate in a meta-conversation, stitching chosen moments into virality. The conflict is one of interpretation. It is not a question of who interrupts, but a question of what that interruption means, and of who gets to decide.

1. Incidentally, I wonder what video-call technology will mean for women. A Zoom call is always a singly-developed floor because the technology doesn’t allow multiple people to speak at once. Does this have an effect on who feels comfortable taking the floor?