Social-Cognitive and Affective Antecedents of Code Switching and the Consequences of Linguistic Racism for Black People and People of Color

Abstract

Linguistic racism shapes the psychological antecedents of code switching and its consequences for Black people and other people of color. We highlight mentalizing as an antecedent of code switching. We posit that stereotype threat arises in contexts where racism is salient, prompting scrutiny of others’ mental states (i.e., mentalizing) when making choices about linguistic self-presentation. Additionally, we posit that sustained appraisals of stereotype threat add cognitive load and reinforce self-protective code switching. We highlight potential consequences of linguistic racism for Black people and other people of color, including reduced opportunities for authentic self-presentation, increased emotional effort, and stress. Finally, we outline paths forward for research and practice: (1) recognizing the heterogeneity of language and thereby reducing linguistic racism, (2) implementing changes that promote racially affirming environments that reduce demands for self-protective code switching, and (3) adapting and creating scalable psychometric tools to measure linguistic choices and linguistic racism.

Introduction

Consider the following vignette: Nia is excited about her first week in graduate school.

She wants to do her best work, grow as a scientist, and make new friends and colleagues. However, she’s also very aware that she is the only Black student in her program. Nia feels comfortable with herself and the way that she speaks but understands that she might be perceived differently by others in her program, depending on how she communicates (e.g., her words, grammar, tone, mannerisms). One day, she decides to attend the first-year happy hour at a popular student bar nearby in hopes of getting to know her cohort better. Her arrival to the bar goes smoothly—members of her cohort say hello and she sits down with a group. But then, she gets a call from her sister. Normally, Nia would pick up immediately and greet her sister enthusiastically. Instead, she answers in a hushed voice, “Can I call you back later?”, concerned that her usual greeting would be seen as inappropriate in this new context. Her sister asks, “You don’t sound like yourself. Is everything ok?”. Feeling unsure about whether she can be herself and be embraced as a scholar by others in her program is unsettling and distracts Nia from fully participating in her cohort’s discussion of their first-year projects.

Nia’s situation is not unique. Graduate students from all racial backgrounds may face concerns about how they are perceived when first interacting with a new group of people.

However, in this context, Nia faces the additional risk of her speech practices being marked (or assessed by others) as improper or non-standard by listeners because of linguistic racism, a form of discrimination targeting the cultural and linguistic practices of Black people and other people of color (Baker-Bell, 2020; Dovchin, 2020). If a listener holds bias against Black people and/or their linguistic practices, this may color how they perceive Nia’s speech, regardless of the manner in which Nia is actually speaking (Rosa, 2019; Rosa & Flores, 2017). In other words, Nia could speak using linguistic practices that would go completely unmarked were they used by a White speaker and still be interpreted as speaking improperly (Samy Alim, 2007; Flores & Rosa, 2015). Although White graduate students can and do face concerns about how their speech is perceived, it is likely that their speech practices will go unmarked1 (i.e., will be perceived as standard/normative) by listeners, thereby inoculating them against linguistic racism as it manifests itself for Black people and other people of color (Cutler, 2003). In short, though Nia’s situation is not unique, she faces a particular risk above and beyond the challenges that come with interacting with a new group of people. These risks are largely due to the thoughts and beliefs surrounding race and language located in the hegemonic modes of perceptions listeners are socialized into (Flores & Rosa, 2015). The material consequences that follow carry risk for Black and other people of color in terms of career opportunities and feelings of authenticity (McCluney et al., 2019) within a cultural context that centers whiteness.

Objective

In this review, we explicitly link a raciolinguistic perspective that examines the ongoing social and historically constructed relationship between language and race in the minds of perceivers (Flores & Rosa, 2015), to psychological literature on mentalizing, stereotype threat, stress, and emotion. In making these links, we highlight some psychological mechanisms that may be involved in the linguistic choices people make, and the consequences of these linguistic choices for Black people and other people of color, due to the impact of linguistic racism. We acknowledge that there are many different antecedents to code switching, which may include processes common to many individuals’ linguistic decisions, such as mentalizing, as well as added vigilance (McCluney et al., 2019) demanded for Black people and other people of color in contexts where group identity is salient (Giles & Johnson, 1987) and specifically in racist environments (Hall et al., 2012). At the structural level, we argue for the need to implement changes that promote racially affirming environments that allow people to employ their full linguistic repertoire in social interactions and reduce external demands for self-alterations during interactions that do not feel authentic to the speaker’s sense of self. In service of these goals, we also seek to (1) adapt and create more scalable psychometric tools to measure linguistic choices and linguistic racism and (2) measure the antecedents and consequences of these choices.

What is Code Switching?

Code switching is described in a variety of different ways across disciplines. Within sociolinguistics, code switching describes the act of alternating between languages, or dialects of the same language (Gumperz, 1977). Its practice at the individual level, and at the level of a community, aids in establishing social boundaries during interactions and between group members (Heller, 1988; Myers-Scotton, 1995). In the field of education, code switching has been conceptualized as a tool to teach students to distinguish between dialects in appropriate scenarios (Wheeler et al., 2004). Within psychology, the definition of the practice expands.

Psychologists describe code switching not only as a linguistic practice, but also as an impression management strategy that incorporates adjusting aspects of one’s appearance, behavior, or even expressions in different contexts (McCluney et al., 2019).

Related constructs to code switching within psychology include cross-cultural code switching (Molinsky, 2007), cultural frame switching (Benet-Martínez et al., 2002), shifting (Jones & Shorter-Gooden, 2009), and passing (DeJordy, 2008; Newheiser & Barreto, 2014). All of these constructs encompass the idea of altering aspects of oneself (e.g., speech, behavior, mannerisms) in service of conforming to context-specific standards of appropriateness (Flores & Rosa, 2015).

This review calls attention to two foundational ideas related to the definition of code switching. The first is a raciolinguistic perspective (expanded below), which argues that linguistic heterogeneity and switching between modes of communication is a neutral practice in itself and one that nearly everyone engages in during social interactions (Flores & Rosa, 2015; Rosa & Flores, 2017). This perspective highlights that heterogeneity in speech practices becomes marked as divergent when employed by groups that already face marginalization or othering.

The second idea is that although adapting communication styles in different contexts is normal and universal, for some groups, these adaptations do not have the same psychological antecedents and consequences. For instance, Black Americans and other marginalized groups who experience structural inequality (viz., linguistic racism) may face conditions that precipitate demands for them to shift (Jones & Shorter-Gooden, 2009), particularly away from their authentic selves (which varies across people and contexts) towards dominating2 cultural norms.

A Raciolinguistic Perspective

In this review, we put code switching research into conversation with the recently developed raciolinguistic perspective that has emerged in sociolinguistics (Rosa & Flores, 2017). At the core of a raciolinguistic perspective is a translingual orientation, which points to the inherent heterogeneity of all language practices (Guerra, 2016). Race is a key dimension that is used to determine what counts as “switching.” In this way, the linguistic heterogeneity of communities of color come to be marked as “switching” in ways that the linguistic heterogeneity of White populations often are not—even when engaged in seemingly similar linguistic practices (Rosa & Trivedi, 2020).3 This bundling of race and language practices often serves to frame the language practices traditionally associated with communities of color, such as African American English (AAE), as inappropriate for school and professional settings (Baker-Bell, 2020) even while White speakers often use similar linguistic practices (e.g., Dupree & Fiske, 2019) to frame themselves as cool, hip, or down to earth (Cutler, 2003). In short, while code switching has historically been conceptualized as an objective linguistic phenomenon, a raciolinguistic perspective brings attention to the ways race is integral to the ideological production of what is and is not perceived as code switching within any particular context. In this review, we do not presuppose that code switching is an objective linguistic phenomenon but rather seek to focus our attention on the role of race and, in particular, anti-Blackness, in marking the heterogeneity in the linguistic practices of Black people and other people of color as fundamentally different from the unmarked heterogeneity of the White norm and the consequences of this on the psychological well-being of Black people and other people of color. We argue that understanding how people make choices about their language practices and how those choices are perceived is key for creating more racially affirming environments.

As we espouse the raciolinguistic perspective, which is in line with the notion that everyone makes strategic choices about self-presentation and language use, we also highlight the ways that linguistic racism marks certain language choices as divergent. For this reason, many Black people and other people of color make strategic decisions about how to linguistically present themselves across different contexts (e.g., work, school, home) that weigh risks and benefits that are less salient or not present for other groups (McCluney et al., 2019). For some Black Americans, as with other groups, making strategic language decisions can be a way to index certain aspects of their cultural identity through the way that they speak (Myers, 2020). For others, however, these decisions, which may or may not feel consistent with a person’s authentic sense of self, constitute a form of impression management to navigate racist environments where the threat of linguistic racism is salient (Hall et al., 2012; Jones & Shorter-Gooden, 2009). Given the ongoing impact of structural racism in the USA (Bailey et al., 2017; McGee, 2020), and linguistic racism in particular, it is crucial to study how decisions about how to linguistically present oneself to others map onto individual and structural level experiences of racial inequity (Alim et al., 2016). It is also key to understand how this relates to a person’s experience of well-being, authenticity, cognitive load, and stress.

Code Switching and Mentalizing

In line with the idea that all people make choices about their linguistic practices during communication, we introduce mentalizing as an antecedent to decisions about how to linguistically present oneself. We propose that most people may engage in mentalizing when making choices about how to use language.

Mentalizing is a cognitive process by which a person infers and reasons about the thoughts, feelings, and/or intentions of another person (Apperly, 2010; Frith & Frith, 2003; Malle, 2020; Saxe et al., 2004); mentalizing has been associated with the development of successful communication in children, (Fan et al., 2015; Hale & Tager-Flusberg, 2005) as well as successful persuasion and negotiation in adults (Baek et al., 2018; Galinsky et al., 2008).

Although the relationship between code switching and mentalizing has so far received little attention in affective science, recent theoretical and empirical work has shed some light on the relevance of mentalizing to decisions about how to linguistically present oneself.

We draw from the Interactive Mentalizing Theory (Wu et al., 2020) in proposing that mentalizing is recruited on an ongoing basis in order to simultaneously manage one’s own impressions in the eyes of others while also figuring out the thoughts, feelings, and intentions of one’s interaction partner. This theory describes how real-world social interactions involve a dynamic interplay between first-order mentalizing (inferences about the thoughts and feelings of others) and second-order mentalizing (inferences about how others are inferring one’s own thoughts and feelings; Wu et al., 2020). In other words, thinking about others’ thoughts is intimately tied to how we perceive the way others are thinking about us.

Following this logic, we suggest that mentalizing processes are likely recruited at different stages of social interactions. Prior to an interaction, a speaker may mentalize in an attempt to form an impression of whether or not the listener is likely to perceive their speech practices favorably or unfavorably. If the speaker presumes that the listener will perceive their speech unfavorably, they may make an a priori decision to alter their ways of speaking in order to align with the communication goals and needs of the listener. So, to revisit the case of Nia, uncertainty about the motives of her new peers would likely trigger additional mentalizing in an effort to get an accurate read on their past and present thoughts, feelings, and intentions, in order to determine how she should engage linguistically.

As discussed in the next section, this process may be recruited more frequently before and during conversations by people whose speech practices are likely to be marked, due to concerns about linguistic racism and confirming stereotypes associated with one’s own group (Richeson & Shelton, 2012). These concerns about confirming a stereotype, coupled with uncertainty about how to present oneself linguistically to avoid discrimination, constitute additional cognitive effort that is expended by Black people and people of color.

Consequences of Linguistic Racism

For Black people and other people of color, beyond universal human motivations to be accepted and connected to others (Leary, 2010; Panksepp, 2004; Ryan & Deci, 2000), attempting to adhere to dominating cultural norms to avoid racial discrimination carries added pressure linked to perseverance as well as academic and professional success (Baugh, 2010; Jones & Shorter-Gooden, 2009; Schmaling et al., 2017). As in our introductory vignette about Nia, we contend that above and beyond general risks associated with navigating everyday scenarios, Black people and other people of color may have an awareness that their speech can be perceived as inappropriate in certain contexts (e.g., Baker-Bell, 2020). If the context prompts concerns about how their speech is perceived (e.g., by colleagues, clients, supervisors, family, or peers), they may begin to mentalize in an active attempt to understand others’ thoughts or feelings. Considering other people’s thoughts and feelings may also raise concerns about confirming negative stereotypes about one’s group (i.e., stereotype threat). The person’s appraisal of the likely costs and benefits of adapting their linguistic expressions then motivates their eventual decision to engage in code switching or discontinue code switching.

Chronic decisions about how to linguistically present oneself to others in order to avoid discrimination (DeJordy, 2008; Newheiser & Barreto, 2014) may then (1) diminish the congruence between a person’s perceived authentic self and outward presentation, (2) require additional cognitive effort, and (3) result in added stress and negative health effects. Here, we focus on these psychological consequences for Black people and other people of color who face discrimination, due to linguistic racism, particularly in contexts where there are demands for presentations of self that deviate away from a person’s authentic self.

Opportunities for Self-Verification and Social Belonging

The ability to effectively affiliate oneself with a group has a variety of positive outcomes (Leary, 2010; Panksepp, 2004; Ryan & Deci, 2000). Membership in social groups can improve self-esteem (Jetten et al., 2015), increase levels of social support that may help buffer the negative effects of stress (Cohen, 2004), and strengthen social ties that may provide valuable advantages for emotional, physical, and mental health (Sbarra & Coan, 2018; Seeman, 1996; Umberson & Karas Montez, 2010). For example, in a study with an intervention to increase perceptions of social belonging for African American students, students reported higher levels of overall health and academic performance (Walton & Cohen, 2011). Perceptions of language are one way in which people determine ingroup and outgroup status (Baugh, 1996; Doss & Gross, 1994) and therefore facilitate discrimination or acceptance.

For these reasons, in describing the consequences of linguistic racism in affect, cognition, and behavior, we draw from DeJordy’s (2008) description of “passing” in noting that the potential dissonance created by the gap between the “me” that is experienced by others and the “I” that encapsulates a person’s subjective understanding of themselves might be modulated by the extent to which a person is code switching away from a self that they experience as true and authentic (Mead & Morris, 1935). In other words, code switching may increase or decrease the distance between a person’s actual self and their presented self. For example, a job applicant may code switch as a means of downplaying negative associations with her racial identity or may code switch as a means of fitting in with a specific peer group. Downplaying one’s racial identity in resumes has been linked to a greater likelihood of getting hired, even in companies that explicitly value diversity (Kang et al., 2016).

So, if a person frequently code switches away from their actual self (e.g., towards a dominating norm), this could decrease their opportunity for others to see them as they see themselves (i.e., self-verification; DeJordy, 2008). Increased demand on individuals to self-alter during interactions may also require additional cognitive and emotional labor (McCluney et al., 2019) and constitute negative consequences arising from pressures to switch away from one’s genuine self (Boulton, 2016).

Stereotype Threat and Cognitive Load

An individual experiences stereotype threat if they believe they are at risk of confirming an unfavorable societal stereotype about a group to which they belong (Spencer et al., 2016; Steele & Aronson, 1995). We posit that stereotype threat is salient in a number of contexts including broadly within academia (Spencer et al., 2016) and is one of several appraisals that underscores motivations to adapt one's speech in an attempt to subvert unfavorable stereotypes about one’s group.

Black students and other students of color may strategically adapt their speech practices in an attempt to avoid negative perceptions (Baker-Bell, 2020). This self-conscious regulation of speech, which is triggered by appraisals of one’s perceptions, may consume mental resources. Further strain on these resources may result from online mentalizing about whether or not one’s impressions are successfully being managed.

Additional empirical work linking stereotype threat and code switching comes from qualitative work (Boulton, 2016). In a qualitative study on stereotype threat and code switching in advertising interns, Boulton (2016) highlights tensions experienced by Black advertising interns. On the one hand, Black interns often (but not always) felt pressure to adapt their speech practices among White peers in order to be embraced as professionals and individuals. On the other hand, when speaking with Black peers, there were sometimes judgments that they were “too white” or pretentious (Durkee et al., 2019). Accusations of “acting White” can have negative consequences for the mental health and well-being of people of color (Durkee & Williams, 2015). Future research that creates links between stereotype threat and marked forms of speech would be fruitful in helping to lessen tensions often experienced by speakers.

Although frequently studied in the context of test performance (Pennington et al., 2016; Steele, 1998; Steele & Aronson, 1995), stereotype threat is also known to sabotage interracial interactions (see Richeson & Shelton, 2012). Black participants who were primed to be concerned about White racial prejudice (vs. not primed to have this concern) showed more ingratiating behavior towards White interaction partners (Richeson & Shelton, 2012). Although Black participants’ behavior had the effect of making their White partners feel more positively (see also Shelton, 2003), this came at the expense of Black participants feeling more negatively about the interaction (Shelton et al., 2005). Although this work indicates that stereotype threat can motivate people of color to accommodate White interaction partners to avoid prejudice, more quantitative affective science on how such prejudicial attitudes manifest and are involved in interracial interactions in the context of language is needed.

The pressures that arise from stereotype threat also consume working memory capacity as individuals experiencing threats monitor for signs that they are confirming a stereotype (Forbes & Leitner, 2014; Schmader et al., 2008). Combined with a fear of failure, the motivation to disconfirm negative stereotypes (e.g., Kray et al., 2001; Nussbaum & Steele, 2011) constitutes an additional pressure to succeed. Such pressure can sabotage performance by bringing conscious attention to well-learned skills that are best performed without conscious effort (Beilock et al., 2006). Stereotype threat can also trigger a physiological stress response (Mendes & Jamieson, 2011; Schmader et al., 2008) that may be adaptive in the short term, particularly for relatively easy tasks; however, this stress response results in poorer performance on more difficult tasks (Ben-Zeev et al., 2005; O’Brien & Crandall, 2003). In the long term, chronic stereotype threat can also lead to disengagement in jobs (Hewlin, 2009; Slepian & Jacoby-Senghor, 2020; Walton et al., 2015) and career paths (Boulton, 2016; Cheryan et al., 2017; Freeman, 2020) through a reduced sense of belonging and can have health consequences as described below.

Health Consequences and Stress

While people may adapt the way they communicate in different contexts to achieve broad social goals as outlined above and to traverse hostile environments (Hall et al., 2012; McGee, 2020), recurrent incidences of discrimination (e.g., microaggressions) linked to negative perceptions of Black people and people of color (viz., linguistic racism) within these environments may negatively impact psychological well-being (Anderson & Stevenson, 2019). For example, individuals who experience frequent microaggressions can incur increased levels of stress (Smith et al., 2011), in part due to the need for more active vigilance to social cues (McCluney et al., 2019). In addition, past or present experiences of racial discrimination that lead to adverse mental health outcomes (Williams & Mohammed, 2013; Williams et al., 1997)4 including stress may be internalized at a physiological level (Hill et al., 2017).

Indeed, although there may be occupational benefits to code switching as outlined above (e.g., social belonging in professional contexts, career benefits), these benefits come at a cost to Black people in the context of work (McCluney et al., 2019). Black people and people of color may experience greater emotional exhaustion and career burnout through the added burden of decisions related to code switching to avoid discrimination (Hewlin, 2009). These are just a few examples of the potential material consequences of linguistic racism for Black people and people of color as they navigate structural racism throughout their careers.

Future Directions

At the structural level, we argue for the need to implement changes within the field of affective science, and the academy more broadly, that promote racially affirming environments that reduce demands for self-alterations (in written and verbal) interactions. One way affective scientists can do this is by considering a raciolinguistic perspective within their work and outside of it. By recognizing the utility, diversity, and heterogeneity inherent in the communicative practices that people employ in general, and thereby embracing the range of language practices employed by themselves and others, affective scientists can enrich the ways they perceive one another and communicate both within and outside of the context of academic work and science. Yet, it is also key to recognize how and why speech practices of some people of color, including those that are native to non-Western Educated Industrialized Rich Democratic countries (non-WEIRD; Henrich et al., 2010), are marked as divergent and to document and counteract the negative consequences of linguistic racism. Specifically, we encourage affective scientists to consider revisiting policies and practices that structurally disadvantage the speech practices of these populations, likely to be marked as divergent (e.g., curbing comments about “formal academic language” within the peer review and job interview process). Understanding the way linguistic heterogeneity of people of color often is marked can perhaps broaden and expand the ways affective scientists determine which ways of speaking and writing are perceived as “academic” or standard and therefore appropriate.

In service of these goals, future work is also needed to adapt and create more scalable psychometric tools to measure linguistic choices and linguistic racism with a goal to interrogate biases associated with perceptions of speech and understand more about how linguistic racism impacts the psychological well-being of people of color in the USA and beyond, for example, asking “In what situations do the speech practices of people of color and non-WEIRD populations get marked as divergent?” and “how can we better understand impacts associated with having one’s speech practices marked as divergent for individual speakers?” Future research testing these links could support the development of a formal theoretical model of linguistic racism as one antecedent of code switching and longer-term negative health consequences tied to belonging, performance, and well-being. With this foundation, we look forward to future work that will test interventions to buffer negative consequences and raise awareness of these consequences for Black people, people of color, and non-WEIRD populations more generally.

Additional Information

Funding

The authors acknowledge the support from the NIH/National Cancer Institute Grant 1R01CA180015-01 (PI: Falk) and from the Army Research Office under MURI contract W911NF-18–1-0244 (PI: Falk).

Data Availability

Data sharing not applicable to this article as no datasets were generated or analyzed during the current study.

Ethical Approval

N/A

Conflict of Interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Informed Consent

N/A

Disclaimer

The content is solely the responsibility of the authors and does not represent the official views of the funding agencies.