In this “Conversation with Authors,” we spoke with APSR authors Marques G. Zárate, Enrique Quezada-Llanes and Angel D. Armenta about their open access article “Se Habla Español: Spanish-Language Appeals and Candidate Evaluations in the United States.”

APSR: The first question is two-fold. Where did you get the inspiration for this paper? How did this co-authorship come about?

Enrique Quezada-Llanes: The inspiration came from what we saw at the beginning of the Presidential election of 2020, specifically with the Democratic primary race. During the first primary debate there were several candidates, but 3 of them spoke Spanish. I was watching the debate, seeing this use of Spanish, and thought “Oh, this is almost the exact thing that I’ve been thinking about in my work!” I was working on a different project at the time at the time related to the use of Spanish language in campaigns, but the project didn’t have this particular component: variation in language proficiency.  Marcus and I had talked about this research idea, but it wasn’t yet a component of the paper I was working on. Since we’d been discussing it, I asked if he wanted to work on it together. Once we started working together, the paper took a different shape, thinking more about accents and fluency and language proficiency, which led us to where the paper is now.

Marques Zárate: For me, many of the research questions I become interested in are based off of what I see around me and in politics today. I’m from El Paso, Texas, where Beto O’Rourke is from. After one of those first debates in 2020 he got accused of pandering. However, being from El Paso I know he’s always spoken Spanish. This motivated the question of “Why do people react in certain ways to this language use?” So I went into the literature and I found an interesting piece by Alex Flores and Alex Coppock that looks at how people react to Spanish language appeals, finding mixed effects. This led me to wonder what might explain those mixed effects and the proficiency with which the candid spoke Spanish seemed like a clear direction to explore. Enrique and I both went to Rice for a graduate school and started this paper in a class together. Enrique was working on a project very similar to this, and was also working with the Baker Institute, did a podcast, and had audio recording experience. With our prior research and experience, we figured we’d manipulate the accent of the candidate speaking. The project certainly shifted a bit throughout the process but that was our main motivation.

APSR: What did you find most difficult in researching or writing about how Hispanic Americans perceive Spanish language appeals?

Enrique Quezada-Llanes: I would say that the most difficult, but also the most rewarding, was the research design itself. Once we decided that we wanted to explore the impact of proficiency and how this moderated Latinos response to these candidate appeals, our question was “How do we actually do this?” We thought that audio was the best way test proficiency’s impact. We needed a design featuring speech that is heavily accented Spanish, and then one that sounds like a native speaker. The issue we had was finding someone who could do both, and finding someone who would not immediately be perceived as either white or Hispanic.  Sometimes there are indicators, perhaps in the accent used when one speaks English, that gives a sense of one’s race or ethnicity. That’s one of the challenges about language, you know, it can communicate so many different things. We needed a voice actor whose voice and accent was such that if we tell you this person is white, you’ll believe us, and if we tell you this person is Hispanic, you’ll also believe us. Finding someone who fits that description makes this type of research so difficult. But, once you find them and you get it, you realize how cool the design is.

Marques Zárate: When we were first starting this project, it was right around the same time that Diana O’Brien, Matt Hayes, and Bryce Dietrich came out with their APSR article “Pitch Perfect.” Based on this article and our need to keep a true experimental setting, we wanted to keep things as controlled as possible. This meant we wanted one voice actor to do all 3 of the treatments. To better understand what we needed in a voice actor, we recorded some of our classmates in our graduate program. This allowed us to get an idea of what we needed to be thinking about, particularly what others heard in the speakers’ voices. For example, because accent is correlated with race and ethnicity, people assume the racial or ethnic identity of a speaker based on how they sound. That process took a while, but going through it was also very fun. This is like one of our first projects we co-authored so figuring out the design process of the research was difficult, but it was also fun and rewarding at the same time.

Enrique Quezada-Llanes I’ll just add that it was challenging and interesting to look at what people were thinking about when they heard these different voices. This led us to record many different voices to get exactly what we were looking for.

APSR: What was the major thing that surprised, confounded, or excited you as you were conducting your research or writing the article?

Enrique Quezada-Llanes: The design was certainly a highlight, but the results were exciting, too. Experiments excite me: seeing the treatment in practice, getting the data back, and seeing how a small difference in treatment, or in this case, difference in audio, makes a difference in how people evaluated the candidates. This result was not necessarily surprising. We expected, for example, that the Hispanic candidate who spoke with a heavy accent would be punished compared to the Hispanic candidate speaking English. This person was not really meeting the expectations that you would have of someone who’s Hispanic. While we did expect this, I also wondered if respondents would perhaps be a little more forgiving of accents. But we did see the punishment for speaking with an accent. Respondents were less likely to identify that speaker as being Hispanic when they spoke with an accent, even though we had explicitly told them they were listening to a Hispanic candidate. In fact, the voice actor we ended up using is Hispanic. Even if not surprising, it was interesting something so small as change in accent can make quite a difference for Hispanic candidates. This leads to interesting questions about what these results mean for Hispanic candidates.

Marques Zárate: I’ve always kind of been interested in how people would choose preferable descriptive representatives. I think our results give you an insight into the different types of cues that people might use in order to try to identify preferable descriptive representatives. But I think another thing that really surprised me is the fact that among the white Democrats, we see a preference for the Hispanic candidates and candidates who spoke Spanish. There’s certainly more to dive into there. Additionally, among the white Conservatives we find that there was really no effect, regardless of the candidate’s race or of the language with which they spoke. White conservatives didn’t necessarily punish or reward the candidate in any particular way. With the way we’re talking about Latino voters now, and how the national media are discussing how Latinos don’t always vote Democrat, I think  there are some interesting avenues to explore. What non-Hispanic Republicans think about candidates appealing to Latino voters? Do they view Latino voters as a coalition? How are they strategically targeting this group in order to try and win an election? These are some things that stick out to me.

APSR: How do you see your work speaking to additional scholarship on candidate accents and/or non-English campaign appeals? Do you think your results here speak to a larger trend in the US that non-English speakers are beginning to experience less backlash for using their native language in society?

Enrique Quezada-Llanes: More broadly than language is cultural competency, something we discuss in the paper. There are other ways to signal inclusivity. We’re testing how cultural competency works for both Hispanic candidates and white candidates, so we’re talking about people who are part of that group and people who are not part of that group. For one of the candidates, this is a signaling game of saying “I care, I want to represent you, and I’m trying to understand the group.” I think these dynamics speak to a broader view of cultural competency that we demonstrate in the paper, but also that we see in contemporary politics, at least in the Democratic Party.

Marques Zárate: My initial reaction to this question is that this is heavily dependent on how much exposure local populations have to foreign languages. If you’re like walking around in New York City and you hear a foreign language, I don’t imagine you’re going to have much of a reaction. However, if you’re in an area without many demographic changes that hasn’t been exposed to hearing other languages, then perhaps you would still find more backlash. As far as political elites using non-English language appeals, I think it’s interesting to see how candidates use these as the United States becomes more diverse. Candidates are trying to find a way to signal that they want these different groups of votes, so it’ll be interesting to see how language continues to factor into political campaigns.

APSR: Finally, where do you see this research going in the future? There are many options for future research in the paper. Do you have plans to expand this paper and theory further?

Enrique Quezada-Llanes: Right now we’re working together on a project that is in a bit of a difference direction than this paper. We recently learned of some scholars taking on one the ideas we discussed in the paper. There are different varieties of accents in Spanish within Latin America. One of the things that’s sometimes taken for granted is this “unity” within the Latino community, even though Latinos have many different national origins and, with those different nationalities come different accents. It’s yet to be seen how representation functions across these different origins. For instance, let’s say you have a Mexican American candidate speaking Spanish with a Mexican accent. Are other Latinos who may have different backgrounds going to feel the same level of representation, or have that experience of being included, compared to a potential candidate who spoke with their same accent? We’re talking with some colleagues who are working on this, and I’m excited to see what they find. For us, we’re going in a different direction, currently focusing on group unity but not as much on the language component.

Marques Zárate: I agree, I’m excited to see that group taking up the mantle and exploring that angle. Enrique’s and my current work is looking at the way that Latino politics does and does not make sense, trying to figure out when the group can be seen as more homogeneous or more heterogeneous. Another thing I’m working on is very much in the thinking phase, but I have talked to Enrique about this idea of a perceived coalition. Why did Republicans not have a negative backlash to the Spanish language appeals in our study? Regardless of whether this group is high in racial resentment or not, I would expect some level of backlash simply because this group knows that Spanish language appeals are targeted towards a group that is not their own. Regardless of whether non-Hispanic Republicans have a negative sentiment towards that group, I should expect them to still have a lower evaluation of the candidate making those targeted appeals.  As I mentioned earlier, we’ve seen more recently about Latinos voting for Republicans and things like that. This, combined with our results, makes me wonder if the knowledge that candidates need to pull Latinos in to win elections condition the way people respond to identity-based appeals. I haven’t started the project yet, but it’s on my radar.

Enrique Quezada-Llanes: Another thing that Marques reminded me of as we were talking about this paper was that we only tested how Latinos and whites respond to these appeals. But how do other marginalized groups respond? Is this a zero-sum game where other groups view Latinos as taking away from attention that they might get? Or are groups thinking of a broader coalition? I think this is very important to think about, and there is already some really interesting work on this.

– Enrique Quezada-Llanes, Rice University

– Marques G. Zárate, Rice University

– Angel D. Armenta, University of Kentucky