People signal their gender identity by using gendered and gender-neutral pronouns pronouns. It’s increasingly popular to include one’s preferred pronouns in email signatures, Twitter bios, etc. This usage provides an opportunity to research how people understand language. In this interview, I speak with Jennifer Arnold (pictured below) about her paper on the topic recently published in Psychonomic Bulletin & Review. She discusses the experiments, the impact of training, possible individual differences, and future research.

Transcription

Wolfe: You’re listening to All Things Cognition, a Psychonomic Society podcast. Now, here is your host, Laura Mickes.

Mickes: I’m talking with Jennifer Arnold about her paper published in Psychonomic Bulletin & Review called, “My pronouns are they/them: Talking about pronouns changes how pronouns are understood.”

Hi Jennifer. Thanks for talking about your research.

Arnold: Hi, thanks for having me.

Mickes: So you published the paper with a couple of authors, who were they?

Arnold: My coauthors are Heather Mayo and Lisa Dong [pictured below], and they were both undergraduate, senior students at UNC last year. And we started this research project together in the fall, and we didn’t really know what we’re going to do. We were interested in gender and language broadly. We actually hadn’t even specified the topic of non-binary pronouns. And we had reading groups every week and we eventually settled down on this topic and the project came together very quickly. So it was really good experience working with them.

Mickes: They have a Psychonomic Bulletin & Review paper as undergraduates. That’s really impressive.

You got ahead of me a little bit. It’s increasingly popular to add one’s preferred pronoun to email signatures or Twitter and social media bios. And your research has to do with how people interpret the pronoun “they” as singular or plural, is that right?

Arnold: That’s exactly right. Yeah.

Mickes: Will you give us a bit of background?

Arnold: Yeah, sure. Pronouns are very popular these days because everybody’s talking about using pronouns to signal gender identity, but it turns out that pronouns have been in the language for a very long time and their primary function is not to signal gender identity.

They’re one of the most frequent words in the English language and they’re used to connect ideas, right? We use them to refer. We just use them to refer to people and things. And many times those are human people. And it turns out that many languages, including English, use different pronouns depending on the gender of who you’re referring to. And so we say that they’re marked for gender.

And in English, they’re also marked for number. So “he” means just a single person who is male; couldn’t mean a plural person. That’s a part of the meaning of the word. And “she” means just a single person; it can’t mean a plural person, you know, plural people. And most people think that the pronoun “they” is a plural pronoun. You’ll find a lot of people who say it’s ungrammatical to use “they” to refer to just a single person. But it’s actually not true, …

Mickes: Is that right?

Arnold: Yeah. We actually use “they” all the time to refer as a singular pronoun.

Mickes: Oh. I would have been wrong too.

Arnold: The thing is we use it in very specific contexts. We’ll say things like, “I was walking down the street and I saw somebody, but I couldn’t see their face.”

Mickes: Right.

Arnold: That sounds natural, right?

Mickes: Yeah. That sounds fine.

Arnold: Somebody called, but they hung up. Okay.

Mickes: Yeah.

Arnold: So those are cases where you don’t know the gender of the referent, but you also find it in cases where you do know the gender. Like, “I walked into a store and saw the clerk and they told me where to find the milk.”

Mickes: Okay.

Arnold: Probably if you walk into a store, you’re going to see a clerk who you identify as male or female, but you don’t necessarily talk about it. So you might then tell the story and just kind of ignore … It’s not really important to the story, whether the clerk was male or female. So you just sort of say they, and it’s a genderless pronoun that’s being used in a very natural way. And people do this all the time.

But, more recently, people have started to talk a lot about using pronouns to signal gender identity and individuals who don’t identify as either a man or a woman often choose another pronoun that they ask other people to use. And one of the most common ones is the pronoun they/them, which is a very natural choice since it’s already used as a singular pronoun in the language, even though a lot of people aren’t aware of that. It is actually a very natural pronoun to choose if you want to be referred to with an ungendered pronoun.

Mickes: Right. So that’s recent, fairly recent that people have been using that is that right?

Arnold: In my experience, yes. I think it’s only come into the mainstream fairly recently.

In 2019, the American Dialect Society decided that “they” would be the word of the decade. That suggests that “they” really started to rise in use in the, you know, 2010-2020 decade. And in my personal experience, that kind of feels right. Although I do believe that in trans communities, it’s probably been used for a lot longer than that.

Mickes: Yeah. This brings us to your research question then. What exactly was the research question?

Arnold: In our experience, people are saying, my pronoun is she-her, or my pronoun is he-him, or my pronoun is they-them. And we wanted to know what kind of impact this has on the ways that people are actually understanding pronouns.

My background is in trying to understand what’s going on in your head as you’re using language, what are those cognitive steps? What are the cognitive mechanisms underlying language processing? Does it matter that people are going around commenting on the intended referent of the pronoun?

We know that a lot of language processing is automatic. And so how is that discussion impacting the automatic processing? Would we see any difference between a context where you had just heard this person uses they-them versus a situation where it hadn’t been explicitly mentioned, but you’d figured it out in some other way.

There has been some research on the use of singular “they,” but most of the other work in this area has not examined situations where you are already familiar with who the referent is, who the person who’s being referred to is.

So a lot of psycholinguistic work uses stories about unknown people. Like, “The waitress went to the store and she…” right? So some unknown kind of person. Whereas we wanted to study what would happen with so-called known people. In our case, they were fictional known people, so we told stories about made-up people. And so we had characters named Liz, Will, and Alex, and we introduced our participants to these characters at the beginning of the experiment. And they would see a little picture [shown below], so this was supposed to make this person seem somewhat real; when we say real, like, in a fictional sense, right?

And so we wanted to know how people would resolve the ambiguity between “they” as a singular pronoun and “they” as a plural pronoun, even though “they” is a well-established singular pronoun, it is much more frequently used as a plural pronoun. So the dominant interpretation of “they” is plural, but potentially if you have somebody in the context who uses they-them pronouns, that in and of itself, will be a contextual constraint that will lead you more to interpret “they” as a singular pronoun. And so we wanted to see how people would resolve those two interpretations.

Mickes: So you introduced them to these fictional characters and then you gave them a scenario and said choose, or how did you test this?

Arnold: Right. So we used actually a very simple technique where they would read little stories. When I say little stories, they’re really two sentence stories. So very, very short. Things like, “Alex went to the store, they bought milk.” Okay?

Mickes: Okay.

Arnold: And that was one of our little stories. And then they got two questions [example below]. And one of the questions was just kind of an attention check, make sure that actually read the story. So something like, “Where did Alex go?” And they had two choices: to the store or skiing, right – so something that’s very obvious and that would just help us make sure that they were paying attention to the stories. And the other question asked, “Who went to the store?” And there’d be a picture of Alex, or there would be a picture of Alex together with someone else. So maybe Will. And people were instructed that they were supposed to pick the picture that matches who went to the store.

So if it was Alex all by themselves, you should just pick Alex. And if it was Alex together with someone else, then you would pick the two of them. This allowed us to measure whether they got the singular or plural interpretation.

This is what we call an offline task. It’s not probing what’s going on during the comprehension process, it’s probing what their final interpretation is. It’s also – because we’re asking pretty explicitly, they’re probably somewhat aware of it – but it still provides an empirical measure of how frequently people get that singular versus plural interpretation, and how that changes across different contexts.

Mickes: Okay. So you did measure different contexts across three experiments.

Arnold: Right. We had three different experiments, but we were basically looking at the same question in all three experiments.

So our primary question was: how does that singular-plural judgment change, depending on whether you have been explicitly instructed that Alex uses they-them pronouns. In one condition, at the very beginning, when they were introduced to the characters and they would see a little picture of each of the three characters. They would read something like “This is Alex. Alex uses they-them pronouns. This is Liz. Liz uses she-her pronouns. This is Will. Will uses he pronouns.” That’s in the explicit condition. The explicit pronoun introduction condition. In the other condition, they just got a picture of the person, say, “this is Alex, this is Will, this is Liz,” no mention of the pronouns, but we then give them an opportunity to observe what pronouns they use.

The form of the experiment was like a survey where they would go through each of these stories and then answer questions about them. And they were not aware that the survey was broken up into different parts, but the way we structured it as experimenters, the first part of the experiment had single person stories that were our kind of example stories. And so they would read these stories like “Alex went to the store and they bought milk” where there’s only one person in the story. Like, it should be pretty obvious that the person buying milk is Alex. Right? And so you should be able to figure out, Oh, this must refer to Alex. Alex must use they-them pronouns. Okay?

Mickes: Okay.

Arnold: And so there were four of those examples for Alex and we then also took a pretty severe analytical strategy where we said, we want to make sure that we’re only looking at people who, first of all, are willing to take that singular interpretation, because there are people out there who say, nope, for ideological reasons, they are opposed to the idea of using “they” as a singular pronoun.

And there were some people who would just systematically say plural across the board, even if it really didn’t make sense for the context. And so we tongue-in-cheek, we labeled these people “resistors.”

Mickes: [laughs]

Arnold: We said, these are the resistors. And we excluded them from the analysis. One of the reviewers or the editor pointed out that it could have been that, you know, maybe they’re somebody who speaks a dialect of English that doesn’t use “they” as a singular pronoun. I don’t know of any. It was quite possible. It’s possible that such a thing exists. The other thing is that if they were in the condition where it wasn’t pointed out explicitly, they could have just been a little clueless. I mean, that’s also possible. And they could have just not quite figured it out for whatever the reason, we just excluded those people.

So what that meant was that we had a sample of people in our analysis, where we knew for sure that they knew that Alex use they-them pronouns because they had performed perfectly on those early training questions, whether they were in the explicit or the implicit condition. So then the real question is, now we’re going to give them some test questions, which in some cases were harder. And we want to know, does it matter if they had seen that explicit introduction or not at the beginning of the experiment?

In the first experiment, we had two conditions in the test questions, the critical test questions. So half of the critical test questions were just like those training questions where there’s just one person. And not surprisingly people just answered the singular all the time. So these were people who had already answered singular perfectly to all of the training questions. So now they also answered singular perfectly to all the test questions. But the harder condition was when we put them in a context where there were two people. So now it is actually plausible that they is being used in a plural sense. So we give someone a sentence like, “Alex went skiing with Liz, they fell down.” It could be Alex, or it could be the two of them, both are perfectly reasonable interpretations.

This is the kind of ambiguity that people encounter all the time in language. I mean, this is just very normal linguistic ambiguity. And, you know, one of the things that psycholinguists are excited about studying is the fact that people resolve these ambiguities quite easily. And so part of what we’re trying to figure out as a field is how do we do that? What are the conditions that push you more towards a singular or the plural interpretation? That’s what we want to know.

And because it’s so popular to talk about pronouns, we wanted to know, is this actually changing the way that we’re interpreting the pronouns? It turns out that in fact, yes, it does. So when you’re given these harder, two person contexts, if you had previously been told that Alex uses they-them pronouns, you’re more likely to take that singular interpretation.

Mickes: Oh, you will say it was Alex who fell.

Arnold: You’ll be more likely to say Alex. Yeah. Even though I should say in this, in the two person context, the rate of getting that singular interpretation plummets, it’s much lower because now that plural interpretation is very attractive because it’s plausible, it’s it very well could be what the speaker meant.

Mickes: Right. And that people do use “they” and “them,” like you said to mean plural more often; that that’s the more common usage, right?

Arnold: Yeah. I mean, in our lifetimes, we’ve encountered “they” as a plural pronoun, probably more often. And in particular, in the context where you’re referring to somebody who is a specific and familiar and known person. Like I said, at the beginning, we do use “they” as a singular pronoun. Absolutely. It’s a real thing, but we use it in a very restricted way. We tend to use it only when you’re referring to somebody whose identity is backgrounded not specific or not salient.

There is a study by Camilliere et al. This was done in Dan Grodner’s lab at Swarthmore. And they bring in this label, the label’s kind of funny, they bring in the label, social distance, which I should say they started using before the pandemic.

Mickes: [laughs]

Arnold: So it was before other meaning. And if the referent is socially distant, meaning that it’s not like somebody, you know, personally, you’re more likely to say “they” or “them.”

Mickes: oh.

Arnold: And I’ve observed in interactions with people that this is how it gets used. People will tell me stories. I hear this all the time. People tell stories and they say, you know, “a student I know” – they. “A person I know” – they. And then what happens is actually as the story evolves, they start telling me more about this person. And then suddenly the person becomes he or she. So there’s seeing this about somebody they know who is a he or she, but when they first start the story and the person is kind of backgrounded and hasn’t been identified yet, and hasn’t been really described, they haven’t been specific about the gender and

Mickes: They’re more general about everything then, and then kind of,

Arnold: Well, I wouldn’t say more general about everything, it’s that the gender isn’t relevant at that point.

Mickes: Right.

Arnold: It’s not necessary to comment on it.

Mickes: It’s only when the gender somehow needs to known?

Arnold: Yeah. So it does appear that the language is changing. It used to be that people would use “he” is a so-called generic pronoun, right?

Mickes: Oh, yeah.

Arnold: And you still see it in some cases. In fact, I notice these things. Somebody said this to me yesterday, the context was a programmer – he. Actually, it wasn’t “a,” it was “the,” we were talking about a specific person who was the programmer. And I happened to know that the programmer was actually a woman, but this person didn’t know. And they said,

Mickes: They said he?

Arnold: Yeah. Um, so this happens, right. But I mean, this is also one part of how people refer, but increasingly people are using “they” in, uh, in this kind of non-specific sense. But the kind of condition that we’re looking at in our experiment is where you’re referring to a specific familiar person. And this is the sort of scenario you have when you’re using they or them as the preferred pronoun for individuals who identify as non binary.

Mickes: Right.

Arnold: Right. Because you know, this would be about like, if you know, Alex, Alex is your friend, you say, “Oh, I went out with Alex and they did this and they did this.” And it’s, it’s always going to be about this very specific person. And it feels to me like it’s different. Right?

Mickes: Right.

Arnold: The use of “they” as a singular pronoun is evolving. And there’s a lot of discussion of this in the literature. There are several linguistic accounts of they-them that have talked about the evolution of the use of “they,” but there’s a distinction between what English teachers say is grammatical, which is what linguists call “prescriptively grammatical” or prescriptive grammar versus the way people actually talk, which is what reveals our actual human, cognitive grammar. And within the way people actually talk, “they” has been used for a very, very long time. And it’s not surprising at all, but it’s been very restricted to these cases where either the gender is completely unknown or is not relevant to the context.

And in my observation, the use of they-them is growing. And so in fact, my children who are teenagers use they-them in a much wider set of circumstances I think, than I do. So they’re just much more productive in their use of they-them.

Mickes: Wow.

Arnold: Yeah.

Mickes: You tested in other contexts too. So in Experiment 1, you explicitly said, Alex goes by they-them. What did you do for Experiments 2 and 3?

Arnold: Yeah. So all three experiments were very, very similar. So in all three cases, we were asking the question of how introducing Alex’s pronouns explicitly might change the way people interpreted pronouns. And so what was different was the kinds of critical sentences. So in all three experiments, we had these early kind of training sentences where we allowed them to observe that Alex used they-them pronouns. This allowed us to select a sample of people that were capable of interpreting they as a singular pronoun and willing to respond in that way in our experiment.

Mickes: Ah, the non-resistors.

Arnold: Yeah [laughs]. And then the critical test sentences, they’re intermixed with some fillers, but they came in two conditions in each of our three experiments. So in Experiment 1, half the test sentences were one person sentences. Like, “Alex went to the store, they bought milk.” Half the sentences were two person sentences. Like, “Alex went running with Liz, they fell down.”

In Experiments 2 and 3, the two conditions were both two person conditions, but we manipulated whether Alex came first or second.

So it’s very well established in the literature that when we are interpreting pronouns, we tend to link them with the person who was mentioned in the subject position of the previous sentence. So that would be first. If I say, “Anna went running with Liz, she fell down.” Technically “she’s” ambiguous. It really could be either one. We don’t know. It doesn’t have to be one of the other, but people have this tendency to think it’s probably Anna, it’s a pretty stable bias. So the same bias occurs with “they.” If Alex comes first, you think it sounds more likely that they is Alex than if you put Alex second.

So let me say those sentences for you.

“Alex went running with Liz, they fell down”

versus

“Liz went running with Alex. They fell down.”

Mickes: Well, now that I’m all trained up, I would think I would say singular for Alex and then plural when Liz is first.

Arnold: Right. Yeah, because there’s a better match between the pronoun and the first person, so that interpretation competes more strongly in the first case. Now “they” is also still a plausible interpretation. Right. It’s possible. In real life, you’d probably use some other information from the context to try to figure it out. Right?

Mickes: What did your participants do?

Arnold: Not surprisingly, they showed an effect of order. So they were more likely to say that “they” referred to Alex when Alex came first. And in fact, when Alex came second, it was very, very low [figure below]. So the rate of taking that singular interpretation was, that was very hard to get.

Mickes: Oh.

Arnold: Yeah. Whereas if you look at the he/she pronouns, if you look at a sentence, like “Anna went running with Liz, she fell down” to mean it’s Liz, it’s not completely impossible. It’s not that low. When Alex came second, the rate of assigning a singular interpretation was very low.

Mickes: Right. And that’s for Experiments 2 and 3?

Arnold: Yeah. For Experiments 2 and 3.

Mickes: That’s really incredible. Did you predict all of this?

Arnold: Yeah, I would say we did expect an effect of introduction, although like all experiments we didn’t know for sure. And we didn’t know for sure it would be able to measure it using this method, but it turned out to be actually a pretty big effect. We also thought we would get the effect of the linguistic context. But again, we didn’t know whether that would also extend to the non-binary case and we did get that effect.

Now there’s another thing that we looked at that was maybe, maybe a little bit of a surprise. There’s two other things that we looked at. So one of them is whether people from different backgrounds would interpret the pronoun “they” differently. And so it has been reported elsewhere in the literature that people are more likely to get that singular interpretation, for example, younger people, more than older people are more likely to get the singular interpretation, right.

And people who frequently use they-them pronouns. If they know a lot of people who are non binary, they’re, they’re gonna have more practice with it. And so to me, this is a very, very interesting phenomenon because it’s an example of how the way we use language is influenced by our experience, because we have this experience with “they” as both a plural pronoun and a singular pronoun in these non-specific cases. And at least in my own personal experience, the use of it as a non-binary pronoun is relatively recent in my life. So I’d say over the last decade, and it’s pretty low frequency. And so the question is what’s going on in my mind, how’s my mind grappling with these new statistics about the language and how’s it integrating this information? So we kind of hoped to see some individual difference effects. And we didn’t really.

In Experiment 1, we did see an effect where the people who they said that they were more familiar with … they rated on a scale of one to 10, how familiar they were with non-binary individuals. And this did correlate with the rate of their singular interpretations for Experiment 1. And recall that in Experiment 1 is where half of our critical stimuli were the single person context and half were the two-person contexts. Experiments 2 and 3 had harder questions because all of the critical stimuli were two-person contexts, and in those experiments, we did not get any correlations with individual differences. This does not mean that those individual differences aren’t there. They’ve been established in the literature. We know that they’re out there, we just didn’t measure them in our study.

We thought we might get variation among individuals, but in Experiments 2 and 3, we didn’t see any of that. And so we just maybe didn’t have enough participants in our study. In order to find individual differences often you need a really big sample. And this may have been just too small of a sample to be able to detect that. It may just have been a, a sample that didn’t have enough variability amongst our participants. It’s another possibility.

There were two questions in our background questionnaire that pertained to their familiarity with people who use they-them. One of them was how many people do you know? And one was people who identify as neither male nor female, because, you know, not everybody uses they-them. So some people use either neo pronouns or a mix of pronouns. You know, there’s a lot of variability in what people choose to use.

The other thing we looked at was linguistic experience via reading. There’s a little task that’s called the Author Recognition Task. And it’s like a little quiz to see how many author names, you know, this is kind of widely used in the field because it’s a pretty easy quiz to give people. And it’s not literally a test of how much you read, but it happens to correlate very highly. So people who read a lot, tend to know a lot author names. And so you can use the score on this task as a metric of reading experience. And it tends to correlate with a lot of variation in language processing, including variation in pronoun interpretation. A lot of my other work has been exploring that phenomenon with respect to he, she pronouns. So in Experiment 2, we included this task, but we didn’t find any effect of it.

Mickes: You didn’t?

Arnold: No. So, in other experiments, we found that there is a tendency for people who have higher scores on this author task, which is known to be a measure of print exposure – so how much are you exposed to print, to reading – and people who score more highly on a print exposure task tend to show a stronger tendency to assign pronouns to the first character.

Mickes: Oh.

Arnold: Okay. So that bias is stronger in people who read. And in Experiment 2, when we were manipulating whether Alex was first or second, we thought, oh, maybe we’ll see a stronger effect for people who read a lot, but it didn’t show up.

Mickes: What does this all mean? What were your conclusions?

Arnold: What it all means is that the way we talk about language has an effect on our automatic language processing. So the reason I say that this is automatic is that we’re putting people in a situation of conflict where there’s two potentially realistic answers, especially in the two person contexts. It’s like both, both the plural and the singular are kind of reasonable interpretations, and you have to figure out which one it is. This is the kind of conflict situation that we’re, we’re dealing with all the time in language processing. Having recently heard, oh, hey, this is the pronoun that Alex uses is doing something to change that process.

So that’s exciting. And it shows that political movements that people are encouraging new labels, there’s all kinds of political movements that have changed the way we speak. You know, in my lifetime, I’d say the use of words like police officer and firefighter have become more common than policemen, firemen, right?

Mickes: Right!

Arnold: Yeah. And that’s the result of political movements. This is an example of the intersection between political movements about language and the way people actually use language.

Mickes: Wow. So what’s next? Are you following up?

Arnold: Yeah, actually, I’m really excited to follow up on this. So one of the other questions we had in here was about whether the rate of getting a singular interpretation would change if you had additional training questions. In, I think it was Experiment 3, we doubled the number of training questions about Alex. So you’d get more background exposure to the use of “they” as a singular pronoun. So we had eight instead of four and we thought maybe this would change the way people respond to the critical questions and it didn’t. And so what does that mean? Well, it means that the difference between eight and four examples isn’t enough. It doesn’t mean the training overall doesn’t matter. We have a very strong reason for thinking that language experience does guide the way we understand language. And so, you know, at some basic level, you’re not going to be able to interpret “they” as a singular nonbinary pronoun, unless you’ve been exposed to this before. Like you have to have learned about this at some point. The evidence from the literature that people who use these forms frequently then get better at them makes a lot of sense. And so you would expect that training should matter. So, I’m interested in looking at that process of learning and what this tells us about how language changes as a function of exposure in other ways as well. So this is a research theme that, that touches a lot of the areas of research that I’m looking at.

Mickes: Thank you so much, Jennifer, for talking to me about your research.

Arnold: Thank you so much for having me. This was a lot of fun.

Concluding statement

Wolfe: Thank you for listening to All Things Cognition, a Psychonomic Society podcast.

If you liked this episode, please consider subscribing to the podcast and leaving us a review. Reviews help us grow our audience and reach more people with the latest scientific research.

See you next time on All Things Cognition.

Featured Psychonomic Society article:

Arnold, J.E., Mayo, H.C. & Dong, L. (2021). My pronouns are they/them: Talking about pronouns changes how pronouns are understood. Psychonomic Bulletin & Review. https://doi.org/10.3758/s13423-021-01905-0