What is a descriptive representative? The study of descriptive representation by ethnorace—that is, the presence of elected officials who reflect descriptive characteristics of their constituents (Dovi Reference Dovi2002; Mansbridge Reference Mansbridge1999; Pitkin Reference Pitkin1967)—is a hallmark of American politics research (Brown Reference Brown2014a; Butler and Broockman Reference Butler and Broockman2011; Canon and Posner Reference Canon and Posner1999; Casellas Reference Casellas2010; Gay Reference Gay2002; Grose Reference Grose2011; Hardy-Fanta et al. Reference Hardy-Fanta, Lien, Pinderhughes and Sierra2016; Lublin Reference Lublin1999; Minta Reference Minta2011; Minta and Sinclair-Chapman Reference Minta and Sinclair-Chapman2013; Rouse Reference Rouse2013; Swain Reference Swain1993; Tate Reference Tate2003).Footnote ¹ Although the literature tends to treat ethnorace as a binary construct and focuses on intergroup diversity (e.g., Sen and Wasow Reference Sen and Wasow2016, but see Bejarano Reference Bejarano2013; Brown Reference Brown2014b; Hardy-Fanta et al. Reference Hardy-Fanta, Lien, Pinderhughes and Sierra2016), there are abundant opportunities to pursue agendas that focus on intragroup diversity and consider the wide variation within ethnoracial categories. Examining such diversity serves two purposes: (1) it un-essentializes members of group categories (Haywood Reference Haywood2017; hooks Reference hooks1991); and (2) it more accurately reflects the malleability of “ethnorace” (Davenport Reference Davenport2020; Masuoka Reference Masuoka2017; Sen and Wasow Reference Sen and Wasow2016).

I argue that amid conversations about electorate diversity, increasing inter-ethnoracial marriage, legislative organizational diversity, and attacks on the relevance of ethnorace to policy making and scholarship, future research on representation must engage this question. Without considering this question in studies of ethnorace and legislative politics, our work neglects the hierarchies that exist within ethnoracial groups (Bonilla-Silva Reference Bonilla-Silva2004; Hunter Reference Hunter2007; Nadal Reference Nadal2019) and sidesteps questions about why those hierarchies exist at all (Haywood Reference Haywood2017; Masuoka and Junn Reference Masuoka and Junn2013; Omi and Winant Reference Omi and Winant1994; TallBear Reference TallBear2013, 31–61). Ethnoracial categories and processes are a part of everyday “common sense”—you know someone’s ethnorace when you see it (Omi and Winant Reference Omi and Winant1994)—yet ethnoracial logics, such as notions of ethnoracial purity and ethnorace mixing, actually make little sense (Spencer Reference Spencer1999; Spickard Reference Spickard and Maria1992). I contend that scholars often flatten the non-sense of ethnorace in the study of descriptive representation—from theorizing the meaning of descriptive representation (Pitkin Reference Pitkin1967), to conceptions of descriptive representatives as unidimensional (Griffin Reference Griffin2014), to the coding of who counts as a descriptive representative (Shah and Davis Reference Shah and Davis2017).

This article reviews my interventions in this area by focusing on mixed legislators, Black women candidates, and Black members of Congress. I show that critically examining intragroup diversity raises questions about exactly what is a descriptive representative, and I invite more research in this area. Like many scholars of legislative politics, Fenno (Reference Fenno1978) influenced my work. I take a multimethod approach to this question among different ethnoracial groups. A few years ago, I conducted one-on-one interviews with state legislators and legislative staff (Lemi Reference Lemi2018). Using a most-similar-cases design, I sought to compare legislators who were similar in gender, partisan affiliation, and district composition but who differed in ethnoracial background—categorized in either one ethnorace (non-mixed) or at least two ethnoraces through parentage (mixed). My goals were to allow the potential choice of ethnoracial identity among legislators to vary and to discern whether mixed legislators were distinct from their non-mixed counterparts. I found that mixed legislators may leverage their ethnoracial background by joining multiple ethnoracial caucuses—potentially amassing individual power within the legislature by drawing on their heritage. However, questions surrounding belongingness and loyalty were apparent for mixed Black legislators. For example, one lighter-skinned legislator was asked to clarify his ethnorace. At the same time, people in the Capitol wondered about a darker-skinned legislator’s status in the Latino caucus. These questions did not emerge acutely for non-Black mixed legislators. Issues of securing policy support did not occur, and this finding suggests that mixed legislators may leverage multiple racial backgrounds to create coalitions. Nevertheless, their relationships within those identity-based caucuses may be tenuous. This research illustrated that despite the widespread expectation that the United States will be post-ethnoracial as the mixed population grows, the legacy of the American system of ethnoracial stratification via classification remains rigid for Black legislators (Lee and Bean Reference Lee and Bean2010). Incorporating mixed legislators into sampling procedures and studies of representation thus reveals consequential intragroup dynamics that otherwise would be missed. What is a descriptive representative if others question one’s group membership based on one’s descriptive characteristics? What is a descriptive representative if some mixed legislators, particularly those belonging to two nonwhite categories, can strategically leverage multiple ethnoracial caucuses?

Although the traditional one-on-one interview yields deep insights into legislators’ thoughts and feelings about representation, they do not allow the researcher to observe the interpersonal conflicts that occur in legislative settings (Brown Reference Brown2014b; Tyson Reference Tyson2016). Focus groups permit this observation (Krueger and Casey Reference Krueger and Casey2014; Onwuegbuzie et al. Reference Onwuegbuzie, Dickinson, Leech and Zoran2009). Recently, Nadia E. Brown and I worked with the Texas-based Black Women’s Political Action Committee to conduct a focus group with current, former, or potential candidates for political office—or current and future lawmakers. We intended to gather information on the styling choices that Black women make on the campaign trail; however, we observed organic conversations unfold between different age cohorts of participants on how newcomers should best package their candidacies to gain access to political networks (Brown and Lemi Reference Brown and Lemi2020). Using the focus group and exploring heterogeneity among Black women, we observed another layer of intragroup nuance in the study of ethnorace and legislative politics that is less considered: intergenerational differences (but see Fenno Reference Fenno2003). What is a descriptive representative if the representatives themselves do not agree on a single standard for feminized self-presentation standards of descriptive characteristics that communicate identity and ethnoracial group membership (e.g., hairstyles) (Lemi and Brown Reference Lemi and Brown2019; Sims, Pirtle, and Johnson-Arnold Reference Sims, Pirtle and Johnson-Arnold2020)?

Despite emphasizing descriptive characteristics in theories of representation (Mansbridge Reference Mansbridge1999), few studies consider how descriptive representatives substantively represent their group differently based on actual descriptive characteristics related to their social and political experiences (e.g., skin tone and colorism) (but see Burge, Wamble, and Cuomo Reference Burge, Wamble and Cuomo2020; Lemi and Brown Reference Lemi and Brown2019; Orey and Zhang Reference Orey and Zhang2019). In a working paper, Jennifer R. Garcia, Christopher T. Stout, and I examine the relationship between the skin tone of Black members of the 114th Congress and their substantive representation of Black interests through bill introductions and press releases (Garcia, Lemi, and Stout Reference Garcia, Lemi and Stout2020). What is a descriptive representative if representatives within descriptive groups look differently and may behave differently?

My research shows the complexity of descriptive representation that previously was masked by a monolithic understanding of ethnoracial-group membership. Considering this diversity encourages us to shift from asking whether members of marginalized groups should represent their own along a single dimension (Mansbridge Reference Mansbridge1999) and who is most preferred to represent their own among multiple dimensions (Dovi Reference Dovi2002) to asking: What is a descriptive representative when we seriously interrogate the non-sense of ethnoracial logics? Future scholarship may move beyond phenotype, mixed status, and gender to examine how other markers of intragroup ethnoracial membership may matter for the effects of descriptive representation on substantive representation.