Liberalism and Its Contradictions: Democracy and Hierarchy in Mestizaje and Genomics in Latin America

This article explores Latin American genomic studies of mestizaje and the way mestizaje’s inherent contradiction between equality and hierarchy—a contradiction typical of liberalism—is managed in genomics. In Latin America, ideologies and practices of mestizaje may be seen as an antidote to hierarchies of race and class, but also as a terrain for the enactment of these hierarchies. Mestizaje mediates this contradiction between equality and hierarchy first by deploying the idea of sexual intimacy and family kinship across racial difference, and second by representing blackness and indigeneity as spatially peripheral and temporally backward-looking, thus naturalizing them as other. Multiculturalism can be seen as a recent variant on these themes, as well as a departure from them. Recent genomics research in Latin America strongly reiterates these ideas, while also adding some new twists. Despite its apparent connection with progressive politics and policies (antiracism, better health for all, protection of human rights for victims of oppression), genomics here appears as a mainly conservative force.

This article explores Latin American genomic studies of mestizaje and the way mestizaje's inherent contradiction between equality and hierarchy-a contradiction typical of liberalism-is managed in genomics. In Latin America, ideologies and practices of mestizaje may be seen as an antidote to hierarchies of race and class, but also as a terrain for the enactment of these hierarchies. Mestizaje mediates this contradiction between equality and hierarchy first by deploying the idea of sexual intimacy and family kinship across racial difference, and second by representing blackness and indigeneity as spatially peripheral and temporally backward-looking, thus naturalizing them as other. Multiculturalism can be seen as a recent variant on these themes, as well as a departure from them. Recent genomics research in Latin America strongly reiterates these ideas, while also adding some new twists. Despite its apparent connection with progressive politics and policies (antiracism, better health for all, protection of human rights for victims of oppression), genomics here appears as a mainly conservative force.
In political orders governed by liberalism, there is a powerful tension between hierarchy and democratic egalitarianism. In this article, I argue that in Latin America, this tension is mediated in a specific way, via ideas and practices of mestizaje, which contain both the possibility of racial hierarchy and the possibility of its overcoming through racial democratization, a possibility that has been long highlighted by many intellectuals both inside and outside the region. These dual potentials are enacted by an emphasis on the sexual intimacy and familial aspect of mestizaje, which both de-emphasizes racial difference and hierarchy and acts as an arena for their reiteration. The contradiction between the two possibilities is further mediated by a process of spatiotemporal othering of the subordinate ingredients that produce mestizaje: black and indigenous peoples are portrayed both as included in the nation as contributors to mestizaje and as excluded by virtue of their peripherality and backwardness. The recent turn in Latin American toward multiculturalism, while seen by many as a rupture with ideologies of mestizaje, is better seen as a new way to handle the classic liberal tension between equality and hierarchy: difference is included apparently on an equal basis but actually in a hierarchical fashion, thus reconfiguring but not erasing the basic lineaments of mestizaje.
Among other depictions of the mestizo nation and its human diversity-literary, geographical, anthropological, and so forth-genetic science has played a role since the mid-twentieth century, measuring the African, European, and Amerindian genetic ancestry of "original" populations and of mestizos . In this article, I argue that recent genomic science highlights histories of sexual mixture and portrays indigenous and black people as spatially and temporally other, and thus reinforces traditional and conservative views of the nation. hierarchies, including in romantic and sexual relationships, and inside families. Conceptual links between mestizaje and democracy were made primarily in the nineteenth and early to mid-twentieth centuries, when Latin American nation-building elites confronted their majority black, indigenous, and mixed populations in a context in which Euro-American thinkers and scientists described these populations as biologically inferior and uncivilized and saw mixture as a degenerative process. Latin American elites found various possible ways forward (Appelbaum, Macpherson, and Rosemblatt 2003;Larson 2004;Lasso 2007;Peard 1999;Stepan 1991;Wade 2010). One option was to encourage European immigration in an attempt to racially "whiten" and thus supposedly improve the population. Another was to simply map the country's diversity, while also hierarchizing it, thus retaining the elite's status (Poole 2011). A third was to recalibrate mixture as a positive force, able to produce constitutional vigor in the population, and cultural richness and social democracy in the society-albeit aided by social hygiene and education. The third option was deployed in some countries in the context of early to mid-twentieth century eugenics, but usually it was still infused with the superior value attributed to whiteness (Stepan 1991). Thus the possibility of linking mixture to democracy remained in tension with racial hierarchy.
In Colombia, in 1861, the politician José María Samper wrote that "this marvelous work of the mixture of races . . . should produce a wholly democratic society, a race of republicans, representatives simultaneously of Europe, Africa and Colombia, and which gives the New World its particular character" (Samper 1861, 299). At the same time, he characterized indigenous people as "semi-savage" and "of primitive race" (1861, 88), and described in racist terms the black boatmen of the Magdalena River whom he encountered on his travels. Their "savage features, fruit of the crossing of two or three different races" indicated that these people "had of humanity almost only the external form and the primitive needs and forces." He added that they would "only be able to regenerate themselves after many years of civilizing work" (Samper 1980, 88-94). In principle, then, mixture could produce democracy; in reality, it could produce the almost subhuman boatmen.
Mestizaje was also linked to democracy by Jorge Bejarano in his contribution to a 1920 book of essays, Los problemas de la raza en Colombia (The problems of race in Colombia), when he stated: "What is the result of this variety of races? Politically [it is] the advent of a democracy, because it is proven that the promiscuity of races, in which the element socially considered inferior predominates, results in the reign of democracies" (Muñoz Rojas 2011, 245). Another contributor to the volume, Colombian politician Luis López de Mesa, took a more ambivalent stance in his later writings. On the one hand, Colombians were "Africa, America, Asia and Europe all at once, without grave spiritual perturbation," in a country which was no longer "the old democracy of equal citizenship only for a conquistador minority, but a complete one, without distinctions of class or lineage" (López de Mesa [1934Mesa [ ] 1970. On the other hand, "the mixture of impoverished bloods and inferior cultures [i.e., African and indigenous people] brings about unadaptable products" (cited by Restrepo 1988, 380). López de Mesa believed mestizaje in general produced democracy, while also seeing indigenous and black people as racially inferior, and mixture between them as disadvantageous for the nation.
In Mexico, from the early twentieth century, intellectuals depicted mestizaje as a way to dismantle hierarchies of race and create national unity. The writer Andrés Molina Enríquez (1868-1940) saw the "mestizo liberals" as the leaders capable of forging a modern nation. Indigenous peoples and the criollos (the Mexican-born descendants of the Spanish conquistadors) were not adequate to this task, but the mestizos, with their "egalitarian" tendencies, could accomplish it (Molina Enríquez [1909] 2004, 42). Molina did not erase racial hierarchy: he also saw mestizos as a "race" that was "inferior" by birth. But he clearly linked mixture to equality and modernity.
This view of mestizaje was held by the politician and intellectual José Vasconcelos (1882Vasconcelos ( -1959, who was nevertheless influenced by contemporary eugenic beliefs and their racialized hierarchies. In his view, a fifth "cosmic" mestizo race would eventually displace the existing four races. The principles underlying Latin American independence movements had affirmed "the equality of all men by natural right [and] the social and civic equality of whites, blacks and indios"; the leaders of independence had formulated the "transcendental mission assigned to that [Iberoamerican] region of the globe: the mission of uniting [all] people ethnically and spiritually." But Vasconcelos also saw black people as a "lower type" of the species who would be gradually absorbed by a "superior type": blacks would disappear "through voluntary extinction, [as] the uglier breeds will gradually give way to the more beautiful" (Vasconcelos 1997, 59, 72). Still, the link between mixture and democracy was a cornerstone for his overall approach. In the 1948 edition of his book, Vasconcelos noted in a prologue that UNESCO had recently "proclaimed the necessity of abolishing all racial discrimination and educating all men in [conditions of] equality"; he saw this as a return by dominant political doctrines to "the recognition of the legitimacy of mixtures . . . and interracial fusion," which Vasconcelos had promoted twenty years earlier (1997,43).
The idea of a Brazilian racial fraternity has been used by some commentators from Brazil and North America alike to make contrasts with the racial segregation of the US South (Seigel 2009). In 1920s Rio de Janeiro, plans to erect a statue of the Mãe Preta (black slave mother) provoked some debate. According to some people, the mother figure was a symbol of black resistance; for others, she belonged to a family dominated by a white patriarch. Nevertheless, all agreed on the ideal of racial fraternity. Washington Luis, elected president in 1926, wrote that "fraternity, the sentiment that unites all men as brothers, with no distinctions whatsoever, will be the work of the South American peoples" (cited by Seigel 2009, 217).
In the 1930s, Gilberto Freyre, writing on the formation of the Brazilian nation, developed this idea. In his 1933 book Casa-grande e senzala (The Masters and the Slaves), he wrote that "miscegenation and the interpenetration of cultures . . . have tended to mollify the interclass and interracial antagonisms developed under an aristocratic economy" (Freyre 1986, xiv). He said that the "social effects of miscegenation," the contributors to it, and its products all "exerted a powerful influence for social democracy in Brazil" (1986, xxx). Freyre did not use the term "racial democracy" in the book, but others in Brazil began to use it by the 1940s: the image of Brazil as a racial fraternity was promoted as part of the nationalist policies of the Getúlio Vargas administration (1930)(1931)(1932)(1933)(1934)(1935)(1936)(1937)(1938)(1939)(1940)(1941)(1942)(1943)(1944)(1945) and during the military dictatorship . Despite academic studies severely criticizing the idea of racial democracy, Freyre continued to espouse ideas linking mixture to racial democracy. In the 1970s, he talked of a "meta-race" of morenos (brown people): "The concept of meta-race [is] linked to that of brownness, as a Brazilian response-beyond sectarian or archaic racist ideologisms-to whitenesses, blacknesses and yellownesses" (cited by Hofbauer 2006, 252).

Mestizaje as a Mediator of the Contradictions of Liberalism
Ideologies of mestizaje have been linked strongly to democracy while also permitting the operation of racism and the persistence of racial hierarchy. This tension is a variant of the one at the heart of liberal social orders, but while in other such orders the tension is mediated by ideologies of meritocracy, in Latin America the tension is also mediated by two processes: the formation of mestizo families through sexual relations; and the spatial and temporal othering of indigeneity and blackness, which are defined as peripheral and backward in relation to the mixedness to which they contribute.
First, ideas of mestizaje draw much of their power from practices of having sex and making families, which engage with the emotional and psychological aspects of personal life, giving many people a stake in the day-to-day realities of mixedness. Being mixed is the norm in countries such as Mexico, Brazil, and Colombia, and those who are mixed can usually recount their parentage, in terms of racialized categories, up to at least the grandparental generation. If asked, most people readily recognize that they are a mixture ( However, many studies have shown that sexual relations between people of different racial identities are no guarantee of racial democracy-instead they can be an arena for sexual violence and predation (Powers 2005;Smith 1997). Also, conjugal relations and families can be the terrain on which racial hierarchy is enacted, in terms of skin color, beauty and ugliness, and racial stereotypes (Fernandez 2010;Hordge-Freeman 2015;Moreno Figueroa 2012;Sue 2013;Viveros Vigoya 2015;Wade 2009). But the frictions generated by these hierarchies-and those caused by racial hierarchy in the markets and politics-are mediated by the kinship relations established by mestizaje, which also reduce intersubjective agreement on racial identifications. Latin America illustrates that structural intimacy across race lines can coexist with racism and racial hierarchy: they are simultaneous realities. For Brazil, Telles (2004, 229-231) argues that "horizontal" relations of intermarriage and residential proximity coexist with "vertical" relations of exclusion in the markets; "cordial" social interaction coexists with racial hierarchy.
Second, spatiotemporal othering defines the relatively "pure" components, whose mixing creates mestizos, as naturally prior to or beyond the inclusiveness of mestizaje, insofar as they appear still relatively pure. Temporally, mestizaje is construed as progress towards modernity, with blackness and indigeneity being left behind as the nation moves towards mixedness (Stutzman 1981). In the dominant "structures of alterity" (Wade 2010), indigeneity is othered most clearly, while blackness may, as in Brazil, be seen as a less valuable but nevertheless integral part of the nation. Still, blackness is given a naturalized association with backwardness. Whiteness is also othered insofar as it appears "pure"-claims to pure whiteness may be decried as racist in Latin America (e.g., Schwartz-Marín and Wade 2015)-but whiteness is also powerfully linked to modernity and progress, the desired goals of mestizaje itself, so this gives whiteness a different status.
Reinforcing this naturalized temporal otherness, indigenous and black communities, especially those perceived as typical exemplars of their categories, are often seen as located in isolated, rural, or peripheral areas seen as temporally backward. In Brazil, blackness is prototypically located in the quilombos (rural communities) or the favelas (peripheral urban spaces). In Colombia, the "black region" par excellence is the Pacific coast-poor, underdeveloped, and isolated. In Mexico, when blackness is not defined as literally foreign (Sue 2013), its prototypical location is the Costa Chica, perceived as isolated and rural. Where blackness is urban-in Salvador or Rio, in Cartagena, or in Veracruz-it is by definition seen as increasingly mixed and is subject to appropriation as part of the national culture, for example through music (Sue 2013;Vianna 1999;Wade 2000).

Multiculturalism and Mestizaje
Has this scenario changed with recent multicultural reforms in Latin America? Multiculturalism is a way to address the dilemmas of liberalism, which set equality and democracy against hierarchy and inequality. On the one hand, multiculturalism is based on the concept of a society consisting of distinct and bounded cultures, which have a right to recognition and respect. It envisages these separate cultures interacting on equal and inclusive terms, thus enhancing democracy. On the other hand, critics of multiculturalism accuse it of masking inequality and ignoring evident hierarchies of race and class by simply claiming different cultures are of equal standing. Furthermore, it is said, the language of cultural difference reinforces these hierarchies, exposing multiculturalism to critiques that it essentializes, reifies, divides, and excludes-critiques which are sometimes one-sided (Barry 2000;Lentin and Titley 2011;Modood 2007). In Latin America, the post-1990 turn to multiculturalism supposedly dismantles and even reverses previous valuations-black and indigenous groups are recast as at least equal in value to mestizos and whites. But in practice multiculturalism cannot escape the tension between democracy and hierarchy that affects liberal political orders: like mestizaje, multiculturalism is a variant of the dynamic interplay of sameness and difference and it does not escape the operations of power that always shape this dynamic (Hale 2002;Speed 2005).
Multiculturalist reform represents a significant turn in Latin America: it facilitates political organization among indigenous and Afro-descendant peoples; it has resulted in important land reforms in countries such as Colombia and Brazil; it has raised the national and transnational public profile and political leverage of Afro-descendant and indigenous peoples in the region. But multiculturalist reform does not necessarily dispense with mestizaje as a set of ideas and practices that enact the mestizo nation and the mestizo individual. This is evident in two ways.
First, it is still unquestioned that Latin American nations, such as Colombia, Mexico, and Brazil, are fundamentally mestizo. For example, the Colombian 2005 census, product of multiculturalist negotiations, effectively categorizes people as either Afro-descendant or indigenous, without a category for the mestizo majority, left unmarked and taken as "normal." Second, multiculturalism reinforces the spatial and temporal othering of black and indigenous minorities, by focusing a good deal of legislation on communities located in rural and isolated areas, understood as "traditional": the Pacific coast in Colombia (Wade 2012); rural indigenous communities in Mexico, ignoring the large number of urban indigenous people (De la Peña 2006; Del Popolo et al. 2007); and quilombos in Brazil (French 2009). The major exception to this are the race-based quotas for university admission in Brazil, which are resolutely urban and linked to modernity. This is perhaps part of the reason this policy has generated such furious debate, as it strikes at the heart of dominant conceptions of the nation. Although Colombia has some elements in its legal reforms that move in the same direction as Brazil-and address its large urban black population-these are still overshadowed by the focus on the Pacific region (Wade 2012).

Genomics, Multiculturalism, and Mestizaje
From early in the development of genetics in Latin America, techniques were developed to ascertain the genetic ancestry of mestizo populations in terms of continental origins-African, European, and Amerindian (Ottensooser 1944;Saldanha 1962). This work was conceived as a contribution to the study of biological "races" and the development of techniques for analyzing the genetic ancestry of mixed people as a biological problem in itself. A further goal was to contribute to understanding the history of colonization in the region and elucidating the characteristics of indigenous populations and their place within the modernizing nation (Duque Gómez 1944;Sans 2000;Suárez-Díaz and Barahona 2013). More recently this ancestry testing has been driven primarily by genomic medicine, for which it is helpful to know the ancestral composition of sample populations (Burchard et al. 2005;Fujimura, Duster, and Rajagopalan 2008), and in which there are suggestions of links between particular ancestries and specific disorders (Kahn 2013;Montoya 2011).
The post-1990s science differs from earlier studies in that many more genetic markers are now used to measure ancestry, and this can be done for individuals as well as populations. There is similarity between the two periods, however, in the use of "parental populations" as reference benchmarks for African, European, and Amerindian ancestry (typically samples taken in Africa, Europe, and among indigenous peoples); and in the presentation of the results as percentages of each ancestry (with a stated margin of error). Critics of genetic ancestry testing fear that it reifies and biologizes outdated notions of race (Bolnick 2008;Duster 2011;Fullwiley 2011;Weiss and Lambert 2014). I am less concerned with race per se and more with how the discourse of genomics in Latin America reiterates key aspects of mestizaje (especially in the multiculturalist context).

The Narrative of Sexual Asymmetry in Genomics
A common finding in Latin American genomics is that most people have some-often a lot-of Amerindian ancestry in their mitochondrial DNA (mtDNA) and high levels of European ancestry in their Y-chromosome DNA (Y-DNA). Mitochondrial DNA is inherited unchanged via the maternal line, thus potentially connecting living individuals unilineally to an unknown Amerindian woman-or "mother" in popular science-who lived in early colonial times. Y-DNA is inherited by men through the paternal line, connecting them genetically to an unknown European man, or "father," who lived long ago. Analysis of these tiny portions of the genome-which reveal little about the overall genetic ancestry of the individual or the population-can be used to make inferences about relations between racialized categories of men and women during early colonial times.
The presence of indigenous markers in the mtDNA and European markers in the Y-DNA is evidence of the well-known fact that early colonial encounters were mainly between European men and indigenous and African women. In the technical language of genetics this is "sexual asymmetry" (Salzano and Bortolini 2002, 309), "strong sex-biased genetic blending" (Gonçalves et al. 2007, 257), "asymmetric mating" leading to the "introgression of [European] genes through native women" (Bortolini et al. 2004, 1-2), and "a biased pattern of mating" (Bedoya et al. 2006, 7234). Geneticists note that "European men preferentially mated with African and Amerindian women" (Palha et al. 2011, 477) and that there was "directional mating between European males and African females" (Ribeiro et al. 2009, 355). The data show "directional mating" (Campos-Sánchez et al. 2006, 560) and "strong gender-biased admixture history between European males and Native American females" (Martinez-Cortes et al. 2012).

Popular Science: Mothers and Fathers
Popular publications used less technical language and more kinship terms. Geneticist Emilio Yunis in a book written for a general audience talks about a Colombian population as having "Amerindian mothers [and] European fathers" (Yunis Turbay 2009, 117). An online journal based in the United States, in an article on Yunis, described the "indigenous founding mothers of the Americas" (Kearns 2007). The writer continued: "The indigenous roots of Colombia are coming into focus, as it is yet another Latin American nation learning about its true history: the founding mothers of Colombia were indigenous." The Colombian newspaper El Tiempo said that Yunis was "concerned exclusively with the genetic load transmitted by Colombian women" (Bejarano 2006). The article stated that the data indicated that "85.5 percent of Colombian mothers are of indigenous origin" (in fact the data showed that 85 percent of the haplogroups 1 found in the mtDNA of the people in the sample had Amerindian origins).
In a publication of the São Paulo Research Foundation (FAPESP) devoted to disseminating the results of scientific research to the public, a report on the work of Brazilian geneticists Sérgio D. J. Pena (e.g., Pena et al. 2000) and Maria Cátira Bortolini (e.g., Hünemeier et al. 2007) said the geneticists' research showed that "the first groups of European colonists . . . were almost exclusively men [who] had children with the Indian women. Later, with the arrival of slaves . . . they went on to impregnate African women too" (Zorzetto and Guimarães 2007, 41). Perhaps inevitably, the reporters made reference to Freyre, saying that in Casa-grande e senzala African women were depicted as having "exercised a special enchantment, of a sexual kind, over the masters of the [sugar] mills of European origin" (Zorzetto and Guimarães 2007, 38). The report cited Freyre's observation that "all Brazilians, even white ones with blond hair, carry in their soul, if not in the soul and the body . . . the shadow or at least a mark [pinta] of the indigenous or the black [person]" (Zorzetto and Guimarães 2007, 41).

Narrative Effects
What implications do these scientific and popular narratives have, bearing in mind my argument that mestizaje works to mediate the tensions between democracy and hierarchy (1) by counterpointing hierarchy with images of kinship and family, and (2) by spatially and temporally othering indigenous and black people and thus naturalizing hierarchy?

Hierarchy and narratives of kinship
Both the technical language of "sex-biased genetic blending" and the popular language of "mothers and fathers" actually erase hierarchies of class, race, and gender. Although the time frame of the early colonial period could directly elicit the context of conquest, there is virtually no mention of coercion and rape, nor even of sexual exploitation. The genetic narratives often make some mention of the sex ratios of the early colonial European population, which had more men than women, implying that sexual relations between European men and indigenous and African women were a normal, consensual result of demographic ratios. There is almost no reference to the surrounding context of very unequal power relations, or to any violence and coercion. Nor is it recognized that structural processes would have constrained "consent" for indigenous and African women. 2 The multiple dynamics of male-female sexual exchanges that filtered across differences of race and class are "grayed out." Like inaccessible options on a drop-down menu in a computer application-still visible but deactivated-the attendant phenomena of sexual predation, social climbing, calculations about the instrumental value of intimacy, reckonings of honor and reputation, and the erotic charges and emotions attached to sex are all placed in the background (for details and relevant literature, see Wade 2009). These phenomena could perhaps be inferred by Latin American readers, aware that the innocuous references hide a history of sexual and racial exploitation, but the meanings are merely an absent presence behind the power-evasive language, which evokes mestizaje as an inclusive process and the starting point for racial democracy.
At the same time that direct references to the hierarchies of race, class, and gender are avoided, an indirect reference is made to the obstinate presence of these hierarchies by the persistent foregrounding of the sexual agency of European men. Y-DNA data affirm the dominant role of the Spanish and Portuguese conquerors, who physically colonized the Y-chromosomes of the men who would go on to forge national populations in the region. 3 European men were apparently the ones busy having sex, making mestizos and ensuring their place as paternal ancestors of the nation. On the one hand, then, the multifaceted relationships between lighter-skinned men and darker-skinned women-mediated by sexual predation, whitening, materialistic motives, social climbing, desire, and resistance-are grayed out and deactivated by technical and kinship terminology. On the other hand, race-gender hierarchy remains obstinately-but indirectly-present by foregrounding the sexual agency of European men. Meanwhile, African and indigenous women are cast as accepting, passive "mothers," graying out histories of resistance (Behar 1989;Powers 2005); female agency is limited to consensual participation. African and indigenous men are barely mentioned, even though the genetic data for some samples show that their contribution to Y-DNA is substantial or even in the majority (Martinez-Cortes et al. 2012, table 2;Palha et al. 2011;Rojas et al. 2010, table 3;Ruiz Linares 2014, 93). This contribution is swamped by the generic assertion that European men were having sex with African and indigenous women. In sum, the hierarchies and skewed power relations are denied by technical and kinship language that make mestizaje seem consensual, obvious, unproblematic, and inclusive, while the inevitable tension between democracy and hierarchy is reflected in subterranean fashion in the persistent privileging of European men as having sexual agency.

Spatial and temporal othering
In relation to the spatial and temporal othering of black and indigenous people, genomic research has this effect in a number of ways. First, there is a tendency to highlight the fact that the vast majority of the national population is mixed; being a mestizo is the norm, while indigenous and black people (and white people) are outside that norm . In Mexico, the Mexican Genome Diversity Project sampled people in several states, going to major cities and using mainstream university and governance channels to advertise for volunteers; these were assumed to be mestizos, even though people were not asked to identify themselves as such. In contrast, when indigenous people were sampled, different governance networks were mobilized, involving local-level doctors and anthropologists, and indigenous people were identified as such, because they lived in communities defined by the residents and others (including the state) as indigenous, and also because sampled individuals had to have grandparents who spoke the local indigenous language. When some of these sampled individuals turned out to have some degree of European ancestry, this did not make them into mestizos, for the purposes of presenting data. For example, graphs showed Zapotec samples as clearly separate, in genetic terms, from mestizos (Silva-Zolezzi et al. 2009). In short, the spatial boundary between indigenous and mestizo was forcefully reiterated (García Deister 2011;López-Beltrán and García Deister 2013;Wade et al. 2014). In Mexico, African ancestry received little attention in genomic studies because of its low contribution, but studies tended to reinforce the links of Costa Chica and Veracruz to African ancestry (Martinez-Cortes et al. 2012).
In Colombia, this spatial othering occurred in relation to indigenous people, who were sampled in communities already identified as indigenous, which were rural and isolated and thus relatively unmixed. Afro-Colombians were also spatially othered. One study selected fifteen samples, of which eight were classified as Native American, one was Afro-Colombian, and the rest were mestizo. The Afro-Colombian one came from the Pacific coast region, an isolated area traditionally known as Colombia's black region (Rojas et al. 2010). Other studies showed that the common image of Colombia as a "country of regions" was borne out by DNA data on the differing genetic ancestry profiles of regional populations-and the most obvious correspondence between racialized ancestry and region was for the Pacific coastal region (Sandoval, De la Hoz, and Yunis 1993;Yunis Turbay 2009). Some studies genetically reified a categorical distinction between "Caucasian mestizos" and people of "African descent," who were sampled in the Pacific coast region (Yunis et al. 2013).
In Brazil, this effect was clear in relation to indigenous communities (Santos 2002; Santos, Lindee, and Souza 2014) but much less so in relation to Afro-Brazilians. Although some early genomic studies targeted isolated "black" settlements-particularly old quilombos (Bortolini et al. 1995)-influential geneticists in Brazil were particularly concerned to show that social categories of race divisions had no genetic basis (Pena 2005;Pena and Birchal 2006). Geneticists showed that people who self-identified as branco (a well-known administrative category) had appreciable amounts of indigenous and African ancestry in their mitochondrial DNA (Pena et al. 2000); they also showed that people who identified as negro or preto had substantial amounts of European ancestry in their genomes (Parra et al. 2003); even in old quilombos people had a lot of non-African ancestry. Everyone-apart from isolated indigenous communities-was more or less mixed. In Brazil, blackness had for many decades been defined as a constitutive part of the nation, and the genetic othering of black people was not as evident as it was in Colombia. Indeed, statements about Afro-Brazilians' genetic mixture were used to undermine the black social movement by implying the nonexistence of a black category (Kent, Santos, and Wade 2014;Kent and Wade 2015). In Brazilian genomics, blackness was not othered; instead it was dissolved into genetic mixedness.
This tendency in genomics to spatially other black and indigenous populations, marking them as separate from the mestizo norm, resonated with the way mestizaje manages the tension between democracy and hierarchy by implicitly restricting the operation of democracy to the mestizo majority and masking the hierarchical subordination of black and indigenous people by placing them outside the national order, but allowing their inclusion when they become mestizos (or are defined genetically as mestizos). This hierarchical othering of blackness and indigeneity, alongside the normalization of the mestizo, was naturalized by being expressed in a genetic idiom, underwritten by the authority of the latest scientific research.
The second way genomic research reinforced the othering of indigeneity and blackness was that it depended on the use of "parental populations"-living populations used as reference benchmarks to assess the amount of each ancestry in the samples of admixed populations. This is characteristic of all genetic ancestry testing, in which "it is assumed that are, or were, such 'pure' human populations" (Weiss and Lambert 2014, 17), even if it is also known that Europe, Africa, and the native Americas are genetically heterogeneous. Parental populations serve as proxies to bridge the temporal gap separating present-day samples from the populations that came into contact five hundred years ago in the Americas. In all cases, the Amerindian parental populations were inferred by sampling local indigenous communities, selected to be as "pure" as possible (although this terminology was not used, the idea was always to select an isolated population that had been as free as possible from the effects of mestizaje). In most cases, the African component was based on internationally used reference populations, such as the HapMap's sample of Yoruba from Ibadan, Nigeria (International HapMap Consortium 2005). Unusually, in one Colombian study, a local black population from the Pacific coast region was used (Rojas et al. 2010). The European component was assessed using international datasets, such as the HapMap's sample of Utah residents of northern and western European descent.
The use of parental populations implies a temporal distancing, taking us back to the "original" encounter between three continental populations. This effect is also strongly produced by the narratives about "asymmetrical mating," which refer to original sexual encounters between European men and African and indigenous women, telescoping the subsequent several hundred years of sexual encounters between mestizos. The temporal aspect of the distancing effect is clearest in the case of Amerindian ancestry, as this is calculated using communities still living today in the national territory: the clear implication is that (some of) today's communities can stand in for those of five hundred years ago. Indigenous people are linked to the past as genetic exemplars.
For African and European ancestry, the effect is slightly different. Most obvious is another spatial distancing effect as both African and European ancestries are linked to distant continents, making them seem foreign. However, they are linked to the past, too, by a temporal telescoping in which five hundred years of mixture, migration, and evolutionary change among populations in West Africa and Europe (or in Utah) are classed as irrelevant for these purposes. This effect operates slightly differently for African and European ancestries due to existing preconceptions of Africa and Europe. Because Africa as a place is generally considered in Latin America to be underdeveloped and backward, the temporal and spatial distances combined simply reinforce the idea that African ancestry belongs to the past and is foreign.
Things are different for European ancestry. Europe still appears as foreign and therefore can be a threat to mestizo national identity-in Brazil, people who are "too white" sometimes feel a need to assert their Brazilianness (Turner 2014, 86). In Mexico, the pejorative term malinchista is directed at those who appear to prefer the foreign over the national. 4 However, whiteness or lightness of skin and other phenotypical features linked to European origins are often highly valued. In the history of Latin America, Europe has often been allied to images of progress and civilization and still stands for all things modern. Thus the temporal link to the past is grayed out and loses significance, being replaced by a link to the future. In sum, things that look ontologically equivalent in genomic terms-African, Amerindian, and European genetic ancestry-are given distinct values by the prevailing social hierarchies, which genomics itself reinforces (e.g., by attributing sexual agency to European men).

Multiculturalism and Genomics
How do the effects analyzed above operate in a multiculturalist context? Genomic data and multiculturalist images of the mestizo nation interact in two ways. First, genomics reiterates in a biological idiom the multiculturalist idea that within the nation there are distinct minority populations, namely indigenous and, to a lesser extent, black. As stated above, the genomic reification of blackness is most obvious in Colombia and least obvious in Brazil, where genomics reasserts the idea that all Brazilians (except indigenous ones) are mestizos: genomic data have been deployed to challenge the very existence of a black category. In Mexico, black communities also have a low genomic profile, overshadowed by the key division between mestizos and indigenous people. Second, genomics undermines multiculturalism by highlighting the fact that mixture is the basis of the national populations: the basic truth about Brazil, Colombia, and Mexico, authorized by DNA data, is that they are all quintessentially mestizo nations. After thirty years of multiculturalist reform, which focuses on racial minorities and leaves the mestizo majority unsaid, genomic data give scientific ammunition to a powerful reassertion and re-centering of mestizo identity. This assertion is as double-edged as it always has been. In Mexico, for example, being mestizo is potentially associated with major public health issues, such as soaring rates of obesity and diabetes (García Deister and López-Beltrán 2015;López Beltrán, García Deister, and Rios Sandoval 2014).

Conclusion
Genomics is often considered to be a progressive force for social justice: it is antiracist, it promises better health and more effective law enforcement, it allows victims of state violence to lodge claims for restitution (Bliss 2012;Smith 2013). It is also said to have a darker side: the genetic surveillance of citizens, the biometric tightening of national borders, the retooling of the race concept (Duster 2015). Reardon (2011, 25) argues that agendas promoting the antiracist democratic potential of human genomics lead geneticists, driven by "genomic liberalism," to give their research subjects new powers of autonomy and participation. Sampled communities are no longer ants under a microscope; they can define their own identities and terms of participation. Yet these new freedoms can lead to problematic consequences, unanticipated by a liberal view that still separates scientific from social knowledge and does not acknowledge their coproduction. For example, HapMap scientists democratically devolved the messy politics of naming to the subject communities, while they created precise scientific definitions of their samples, aiming to avoid dangerous overgeneralization (and racialization). But their separation of social and scientific knowledge unwittingly left the door open to the generalizing and racializing use of their data by other scientists and observers, because they assumed that the problem of the politics of representation had been solved by their liberal tactics of devolution.
In Latin America, liberal genomics also led to antiracist stances, but there were few attempts to include communities in research in a participatory way, and categories were generally defined by the scientists. 5 This was part and parcel of the conservative bent of a genomics that reinscribed a familiar image of the mixed nation, with black and indigenous peripheries. Hinterberger (2012) argues that, in Canada, the classifications of a multiculturalist politics that normalized whiteness, while recognizing "visible" minorities, were reflected in the classifications used in medical genetics in a process of "categorical alignment" (Hinterberger 2012, 208, citing Epstein 2007. The same can be said of genomic science in Brazil, Colombia, and Mexico, which emerges as a conservative force. The mediations of mestizaje-which hinge on narratives of sex/kinship and spatiotemporal othering-are reiterated in the language of genomics, giving them scientific authority and reaffirming the biological dimensions inherent in mestizaje. Genomics highlights the centrality of mixture to the nation, emphasizing the inclusive potential of mestizaje and downplaying hierarchy. Where difference is noted-by genetically reifying indigenous and (in Colombia) black communities, and by talking in terms of parental populations-it is portrayed as existing on a level playing field, as multiculturalism often presents it. But wider preconceptions of Africa, indigenous America, and Europe stratify this playing field, while genomic narratives about original sexual encounters reinforce hierarchy by attributing sexual agency primarily to European men. Genomics does provide new ways of conceptualizing connections-for example, by revealing indigenous ancestry in the mtDNA of "white" people, and by allowing individuals to make genealogical connections further back than the typical two or three generations. However, these new modes of self-knowing ultimately reinforce narratives of mestizaje. The tensions between democracy and hierarchy that are characteristic of all liberal political orders are still evident in genomic versions of the nation in Latin America.
University Press, 2015), and Degrees of Mixture, Degrees of Freedom: Genomics, Multiculturalism, and Race in Latin America (Duke University Press, 2017).