I am not the first person to wonder, “What is the function of asking Right? in the middle of a sustained flow of verbal behavior?” As the internet is abuzz with remarks about the recent surge of its use, I may be the last. People have been griping about it since as early as 2018, so it presumably became widespread in English the last 10 years, or so. As my own speech is distinctive for its dysfluencies, its pauses, and its sprinkling of ums and ahs, as I play for time to figure out what I am trying to say, it ill behooves me to find fault with the diction of others, but my purpose is not to criticize, or to campaign for the suppression of human Right?s, but to understand it. I ask, not as a sociologist, linguist, or expert on diction or public speaking. but as a behavior analyst. Right? Verbal phenomena of this sort are not important enough to justify an experimental analysis, and it is hard to imagine how one might go about doing such an analysis. A plausible guess is good enough for my purposes.
According to Wikipedia, Right? is a “filler word,” like ah, uh, um, you know, and I mean, utterances that serve to hold the floor so that your interlocutor will not butt in while you are organizing your next remark. Perhaps sometimes. But clearly Right? is not interchangeable with other filler words. We might say, His name is…uh…um…don’t tell me…Harrison! but we would never say, His name is…Right?…Harrison. Let’s see how far casual observation will take us toward an alternative:
First the topography of the response: It is reliably a discrete response, inflected as an interrogative, with a rising intonation, characteristically preceded by a pause or change in inflection that sets it apart as a unitary operant. As for function, surely it is simply a mand for agreement, presumably functionally equivalent to, and an abbreviated form of, Am I not right? and under control of audience variables suggesting the uncertainty of reinforcement.
Unfortunately, this straightforward analysis, however obvious, appears to be inadequate. Am I not right? is indeed characteristically a mand for agreement, as shown by the fact that it is followed by a distinctive pause signaling the passing of the microphone to the audience. That is, it is a turn-taking question that demands an answer, after which the listener usually agrees or disagrees. In contrast, Right? is characteristically followed by further verbal behavior on the part of the speaker. It often appears to be functionally equivalent to saying, “I’m right. Now sit there and keep quiet while I continue my rant.” If listener behavior is required, it is nonvocal.
Might the expression of the listener—or change of expression—serve as nonvocal assent (or dissent) and thereby maintain the behavior in strength? No doubt such variables can be powerful verbal responses. Continuous eye contact, and open expression, an incipient nod, suggest encouragement; furrowed brows, tight lips, incipient attempts to interrupt suggest disagreement. But they would do so with or without a mand for agreement, and two kinds of observations suggest that such responses are not necessary to maintain the verbal response Right? in strength.
1) Right? can be observed in speakers with large and heterogeneous audiences. At the 2024 ABAI convention I attended a symposium featuring three speakers, and I recorded the number of times they each said Right? in 5-minute samples of their presentations.
Speaker 1: 0.4 times per minute
Speaker 2: 0.0 times per minute
Speaker 3: 1.6 times per minute
The speaker in a symposium, facing a sea of unfamiliar faces, is likely to feel ill at ease. Such conditions are likely to evoke a tendency to seek reassurance, so I don’t fault Speaker 3 for seeking it 8 times in a 5-minute interval. But although a large audience can exert a powerful effect on a speaker, it seldom does so differentially. Town Hall audiences may boo the lies of a corrupt politician, but the audience at a professional meeting is not likely to respond systematically to a verbal tic, or even an off-hand request for reassurance in a way that the speaker can detect.
2) I find it difficult to monitor the use of Right? in the wild, so to speak, because verbal contingencies change so rapidly in conversation, and because casual conversation seldom takes the didactic turn that seems to foster it usage. But podcasts offer a good medium for the purpose because they are taped and can be repeatedly rewound. The sort of political podcasts I listen to usually feature speakers who are in separate settings from one another, speaking over Zoom, or comparable media. Nonvocal feedback is considerably weakened under such conditions. Perhaps because the subject matter favors argumentation and persuasion, the response Right? occurs frequently even in the presence of a relatively generalized audience variable.
The following passage is a continuous utterance by one speaker in a recent dialog:
“We need better food for our kids. Who doesn’t want that? Right? Oh yeah, there’s too many processed foods. Right? As a parent, I lean into that. Right? Shouldn’t we wonder about our environment and its impact on us? Yes, we should! Right? But then it comes with a whole passel of crazy! Right?”1
In this context, the speaker was discussing the health-care stance of a controversial cabinet minister, a stance with which she generally vigorously disagreed, but here she was emphasizing that even the worst people have some good ideas. The Right?s seemed to serve, not as fillers (um…), not as mands (Please say you agree with me!), but as intensifiers (Very much so!). At the very least, they were emitted without any response from her interlocutor and without any expectation of a response.
Here are two other examples:
“I don’t want to make this out to be just a Democratic thing. Right?But it’s clear that out of the desire—part of which is very well grounded—to make sure bad things don’t happen, you end up with a lot of measures put in place that stop anything from happening—including good things. Right? But I also think…”2
So let’s take a 21-year-old barista at Starbucks. Right? What are her political views?”3
Here (as above) the Right?s seem to serve as a kind of punctuation, as exclamation points, rather than as requests for agreement. (Who could disagree with taking a 21-year-old barista as a case in point?) Specifically, they hold the floor while inserting a pause to give the listener a moment to digest and firmly fix the previous statement: “What I say is true; let that sink in before I go on.” “Picture your barista before you predict her political views.”
To the extent that this analysis is correct, Right? is an autoclitic; it modifies the listener’s response to other verbal behavior, specifically, the preceding item. Despite its intonation, it is an instruction rather than a question, and that is why the speaker does not wait for a response from the listener before going on.
So is the analysis correct? Who knows? The purpose of a behavioral interpretation is not to determine truth but to show how puzzling everyday phenomena are within the compass of our principles and do not require us to flee to other paradigms.
- Kara Swisher, interviewed by Sarah Longwell: The Focus Group Podcast:2/21/25. https://www.youtube.com/watch?v=SXRBSy5xgl0&t=1116s ↩︎
- Pete Buttigieg, interviewed by Ezra Klein: 2024 and the ‘Crank Realignment; https://www.youtube.com/watch?v=8JDMKyTpXWc ↩︎
- Preet Bharara, interviewed by Harry Litman: Talking Feds Podcast: 9/28/24. ↩︎