Genetic ancestry testing among white nationalists: From identity repair to citizen science

Abstract

White nationalists have a genetic essentialist understanding of racial identity, so what happens when using genetic ancestry tests (GATs) to explore personal identities, they receive upsetting results they consider evidence of non-white or non-European ancestry? Our answer draws on qualitative analysis of posts on the white nationalist website Stormfront, interpreted by synthesizing the literatures on white nationalism and GATs and identity. We show that Stormfront posters exert much more energy repairing individuals’ bad news than using it to exclude or attack them. Their repair strategies combine anti-scientific, counter-knowledge attacks on the legitimacy of GATs and quasi-scientific reinterpretations of GATs in terms of white nationalist histories. However, beyond individual identity repair they also reinterpret the racial boundaries and hierarchies of white nationalism in terms of the relationships GATs make visible. White nationalism is not simply an identity community or political movement but should be understood as bricoleurs with genetic knowledge displaying aspects of citizen science.

Genetic ancestry tests (GATs) are marketed as a tool for discovery about identity. But what happens when what’s discovered is unwelcome, upsetting or disorienting? Take the case of Craig Cobb, a white nationalist who had gained public notoriety and cult status among admirers for his (ultimately failed) efforts to buy up property in Leith, North Dakota, take over the local government, and establish a white supremacist enclave. In 2013, Cobb was invited on The Trisha Show, a daytime talk show, to debate these efforts. After a guest on the show yelled at Cobb, ‘You have the blood of negroes in your body right now! You are not 100% white!’ he accepted a challenge to take a GAT. Several months later Cobb returned to the show only to be laughed off the stage as the GAT results showed 86% European and 14% African ancestry - results consistent with having one great-grandparent of fully African ancestry. Cobb (2015) later reanalyzed his DNA and reinterpreted his results to redeem his white identity, in a pattern that parallels the efforts described in this article.

While Cobb was pressured into taking his GAT, many in the white nationalist and newly emerging alt-right have flocked to the tests and encouraged others to take them (Reeve, 2016; Zhang, 2016). Online discussions of this trend exhibit a spirit of ironic provocation as alt-right users, for example, challenge each other to use a GAT to prove they aren’t secret members of the ‘Jewish Internet Defense Force’ (Reeve, 2016). But, not infrequently, individuals post the shocking news that GATs reveal some African, Native American, or Ashkenazi Jewish ancestry or other European groups deemed not white enough (Reeve, 2016). Such results have led some opponents of white nationalism to propose, somewhat tongue in cheek, handing out GATs at their rallies to undermine their claims to white purity (Mangu-Ward, 2017).

An individual problem might also be a collective one for white nationalism, as GATs show that many members of the movement don’t meet the ostensible racial criteria of membership. On Stormfront, the largest white nationalist forum on the internet, senior moderator John Law (2006) offers these definitions of ‘who is white’:

if a person looks White and thinks of himself as White and is the kind of person our other members wouldn’t mind their sisters marrying - and if we know he is no more than one-sixteenth non-White, we consider him White’

and ‘no discernable trace of non-White admixture’, and, decisively, ‘non-Jewish people of wholly European descent. No exceptions. And if you tell us you’re not, we will believe you.’ Law’s social definition of looking and acting white and having a white family pedigree is underwritten by an implicit genetic claim that true whites have no ‘admixture’ and are ‘wholly’ of the preferred ancestry. But John Law wrote at the very dawn of the GAT era, and now for the price of $99 and a mouthful of spit these ‘discernable traces’ are showing up among the initiated.

Science and Technology Studies (STS) scholars have been increasingly concerned with how genetic technologies are disrupting individual and group identities and transforming classification practices upon which they lie (Fujimura and Rajagopalan, 2011; Nash, 2015; Panofsky and Bliss, 2017). The literature has addressed the stakes of identity for the general public of GAT users (Roth and Ivemark, 2018), African Americans and Latinos (Nelson, 2008, 2016; Shim et al., 2018), Native Americans (TallBear, 2013b) as well as homosexuals (Miller, 1995) and disease sufferers (Navon, 2011). Beyond their urgent contemporary political importance, white nationalists, not previously studied, are an interesting case because their simultaneous embrace of a genetic basis for identity and commitment to white purity seems to make them both individually and collectively vulnerable to what Nelson (2008) has termed the ‘genealogical dislocation’ often wrought by GATs.

This article asks how white nationalists respond to what they view as ‘bad news’ or upsetting results that sometimes emerges from GATs, and it situates these responses among those documented in the STS literatures about genetics and identity and lay people’s strategies for engaging unsettling science. Drawing from a novel dataset of discussions of GATs on the white nationalist website Stormfront.org, we trace how white nationalists respond the dilemma of upsetting results. First, we argue that community reactions depend less on what a GAT reveals than on the revealer’s deference to the community’s judgement. Then we show how individual upsetting results are repaired through anti-science strategies rejecting GATs as legitimate knowledge and lay expertise strategies of accepting but reinterpreting damaging results. Though the dominant response to individuals is repair, white nationalists are simultaneously deeply concerned about GATs’ implications for the group, and we track ways they redefine racial definitions and hierarchies in their wake. Finally, the article concludes with some discussion of the broader implications for STS theories of GATs and identity and suggests also that we rethink white nationalism as containing a citizen science movement within it that challenges our intuitions about the democratization of science.

Background and literature

White nationalism and identity

Scholars have emphasized the contradictory and complex character of ‘whiteness’; whiteness is in ‘crisis’ even as it maintains political and cultural hegemony (Daniels, 2009; Winant, 1997). In the broader culture, it is simultaneously viewed as neutral, empty, the absence of race, while assertions of whiteness per se are politically aggressive and supremacist (Daniels, 2009). It is a form of ‘property’ that grounds access to many American rights through structural privilege and the imposition of violence, yet is always at risk of being ‘devalued’ (Harris, 1993). White nationalism is an extreme political response to American racial politics that rejects not only liberal egalitarian ideals but also conservative ‘color blindness’ to assert ‘white pride’, the supremacy of the white race, and social and political policies to secure that position using violence to exclude and suppress of non-whites (Winant, 1997).

As a movement, white nationalism is difficult to define. It is comprised of many groups with differing ideologies, religious commitments, and openness to violence. Indeed, scholars often focus on the different agendas and internecine conflicts among the Ku Klux Klan, Neo-Nazis, Christian Identity, and others (Blee, 2002; Ferber, 1998; Simi and Futrell, 2015; Tenold, 2018). Another is the active disagreement among activists about whether white nationalism is committed to a project of white supremacy, superiority, racial hatred and violence, or is simply a nationalist identity movement that claims to disavow violence, hate and supremacy, seeking only to advance ‘white’ culture and identity, as does ‘any other’ minority identity (Tenold, 2018). For the present study, more relevant than these differences are the shared, taken for granted dimensions of the white nationalist worldview. Primary is the idea that race is the dominant variable in human society, determining hierarchies of culture, politics, and history and authentic collective identity. Crucial to this is a commitment to ‘race realism’ that holds races and the differences among them are genetically given and transmitted through blood (social constructionism is ridiculed as a cultural Marxist conspiracy). White supremacy, white nationalists believe, is due to both the cultural and genetic superiority of the white, Aryan race.¹

In the white nationalist imagination, their racial position and distinctiveness is fragile and always under threat (Daniels, 2009, 2018; Simi and Futrell, 2015; Tenold, 2018). Part of this is political, institutional and cultural, as nationalists see whites as victimized by non-whites and Jews who conspire to promote public policies that ensure their numerical and political rise, while celebrating multicultural diversity to stigmatize heroic white history. More crucial for our purposes is the genetic threat: White blood or genes produce white superiority, but breeding with non-whites undermines those qualities and leaves indelible traces of non-white ancestry for many generations. Cultural and political changes promoting multiculturalism are especially dangerous to white nationalists insofar as they promote interracial sexual relationships that would permanently undermine the genetic purity that is the foundation of white supremacy.

Sociologists have shown that white nationalist ideological and political commitments don’t just float freely but are honed and transmitted in strong community structures. Simi and Futrell (2015) describe the set of ‘safe spaces’ - private communities, music scenes, online forums, etc. - in which white nationalists can find solidarity and support in the face of a hostile world. These spaces inculcate a strong ethos of mutual loyalty and a commitment to the race. Socialization into these communities is intense and emotional (Blee, 2002), and those who leave often describe white nationalism as like an addiction (Simi et al., 2017). The network of safe spaces and the mutual rejection of white nationalism and mainstream society thus help reinforce the sticky, dispositional character of white nationalist identity. GATs intercede into this matrix.

GATs and individual identity

Genetic Ancestry Tests (GATs) work by assaying variations in an individual’s DNA, using a statistical algorithm to compare these to reference databases of samples from diverse populations around the world, and then to make inferences about the individual’s ancestry based on specific links between them. Industry estimates suggested that by early 2019 more than 26 million Americans (~1 in 12) had taken a GAT (Regalado, 2019). GATs are probably the most common direct contact that most people have to the ‘genomics revolution’. Critics have argued that GATs, despite their claims of genetic objectivity, may not be accurate (Bolnick et al., 2007; Lee et al., 2009; Royal et al., 2010) and represent a very incomplete vision of ancestry and relatedness (Nash, 2015; Royal et al., 2010; TallBear, 2013b).

Survey research has shown that GATs prompt people to understand human traits and racial differences as genetically determined (Phelan et al., 2014), even though GATs are often marketed as a means of celebrating diversity and overcoming differences (Bliss, 2013; Bolnick et al., 2007; Lee, 2013). Such claims have led to expectations that individual users of GATs will interpret results as having deterministic implications for their own identities (Bolnick et al., 2007; Nash, 2015; Nordgren and Juengst, 2009). But research on users has not borne this out. In her study of African American roots-seekers, Nelson (2008) found that surprising GAT results often produced initial ‘genealogical dislocation’, but that individuals did not absorb results unquestioningly. Rather they incorporated results into their identities partially and selectively, through a process Nelson calls ‘affiliative self-fashioning’. Likewise, Shim et al. (2018) found that a group of Latina and African American women who received genetic ancestry information incidentally as part a broader genetic health study were unmoved by previously unknown information because most considered their ancestry to be thoroughly ‘mixed’ anyway.

Roth and Ivemark (2018) have proposed ‘genetic options’ theory to explain the conditions under which GATs lead to identity change. Individuals are likely to use new GAT information to change their identity when it accords with a pre-existing aspiration for identity change and when the individual perceives it would pass social appraisals among one’s peers. They find that whites were most likely to be dissatisfied with the supposed blandness of their identity and to have a pre-existing desire to burnish it with new racial or ethnic ties. Whites were thus most likely to change self-conception after a GAT, a propensity Roth and Ivemark interpret as a form of racial privilege (c.f., Waters, 1990).

We can adapt this framework to analyze the white nationalist case. Roth and Ivemark (2018) assumed that identity would remain stable unless GAT data, identity aspirations, and perceptions of social appraisal pointed toward change. White nationalists’ identity aspirations are to maintain or deepen their ties to whiteness and (non-Jewish) Europeanness, and they reject whites who seek to demonstrate non-white ancestry. But, consistent with their ideologies of threatened whiteness, white nationalists’ default in the face of anomalous GAT results is not to assume their whiteness but to fear for its loss. The situation is the inverse of that described by Roth and Ivemark: Instead of identity remaining stable without an aspiration and community permission to change, white nationalists assume a change to non-white identity unless the white nationalist community offers permission to remain white.

Such an appeal might seem puzzling at first, given promulgated membership rules - ‘if you tell us you’re not [white], we will believe you’ - but it makes sense when we recall the ways white nationalism functions simultaneously as an ideology and a community. The ideological centrality of racial purity plus the constant fear of white despoilment (Daniels, 2009, 2018; Ferber, 1998) animates the sharp sense of identity destabilization evident in many GAT user appeals to their white nationalist peers. But Stormfront functions as one of white nationalism’s ‘hidden spaces’ (Simi and Futrell, 2015), so it represents one of the few places where a distraught white nationalist could hope for a sympathetic ‘social appraisal’ to maintain their claims of whiteness.²

A main purpose of this article is to understand the social processes by which white nationalists interpret GATs. We show two general cultural repertoires constituting these appraisals, which enable rejection or repair of an individual’s troubling GAT result. The first develops a white nationalist counter-knowledge (Borkman, 1976; Holton, 1992) to reject GATs’ legitimacy by claiming family genealogy to be superior, claiming the capacity to see race directly, and suspecting the tests to be part of an anti-white conspiracy. The second, in contrast, develops a form of white nationalist lay expertise (Epstein, 1995; Wynne, 1992) that, through detailed explication of the epistemological construction and functions of GATs, wedded to a white nationalist understanding of history and social relations, allows social appraisals that dispute the accuracy or interpretation of particular GAT results.

GATs and group identity

A major lesson of scholarship on GATs is that their meaning is not set in the abstract but given by the particular identity dilemmas of the groups using them (Davis, 2004; Nelson, 2016; Smith, 2016; TallBear, 2013a, 2013b). A root of this research is Rabinow’s (1996) influential essay on ‘biosociality’ which considers how group identities and boundaries are being affected by GATs and genetic tests more broadly. On the one hand, Navon (2011) describes ‘genomic designation’, a process where a genetic discovery (such as a chromosomal microdeletion) leads to the creation of a new type of people that may be so phenotypically heterogeneous as to be previously unrecognized as sharing anything in common. Once an identity is genomically designated, processes of institutional recognition and collective action can solidify it. Scholars have also discussed the geneticization of pre-existing medical or social categories. Homosexuality is an interesting example, because many activists have seized onto the ‘gay gene’ as legitimating their claims that homosexuality is natural, not a choice nor a sinful lifestyle (Miller, 1995).

On the other hand, TallBear (2013a, 2013b) describes what might be thought of as ‘anti-biosociality’ on the part of Native Americans and other indigenous activists. She describes how efforts to use GATs to posit Native ancestry function as explicit or implicit assaults on tribal groups’ sovereignty to define membership. Further, the rights of indigenous people around the world are often linked to their claims to ancient occupation of particular territories and distinctiveness as a people from other groups. Thus, geneticists’ efforts to sample indigenous DNA for ostensibly neutral purposes of tracking historical migrations and population relationships can seriously endanger such claims. Native and indigenous resistance to genetic technologies can thus be a matter of collective survival.

Somewhere in the biosocial middle is Nelson’s (2016) description of uses of GATs in African American reconciliation projects ‘to repair ruptures caused by fractious social and political struggles or efforts to (re)unite communities’. For example, activists have used genetic ancestry information to establish the legitimacy of claims forcing universities to acknowledge and compensate for the ways they historically benefitted from slavery. Thus, in reconciliation projects, GATs can serve neither to constitute or solidify identities, nor to undermine the identity claims upon which sovereignty is based, but instead to open up new political agencies or possibilities (Panofsky, 2018).

GATs intersect collective narratives about white nationalism in ways that combine all three of these biosociality paths. For one, white nationalists’ ideological commitment to genetic understandings of race might suggest openness to some version of genomic designation or genetic legitimation of whiteness using GATs. But second, their emphasis on the value of genetic purity and distinctiveness suggests a vulnerability parallel to indigenous groups, and thus reasons for anti-biosociality. And finally, white nationalists’ self-image of their politics is a reconciliation project and GATs may help them realize those politics. White nationalism is animated by hate, exclusion and violent ideation, but its ideologists are eager to justify these as whites’ only recourse against a society that has turned against them. Every group but whites, they believe, benefits from politics, policies, and culture that celebrate multicultural diversity, yet ‘white pride’ is an anathema (Tenold, 2018).

As we show below, GATs stimulate white nationalists to engage all three of these kinds of biosociality. Even as they seek to minimize and deny disruptive implications of specific GAT results for particular individuals, they expend considerable energy on the biosocial implications of GAT for white nationalism as a whole. Though GATs pose a risk to white nationalism that bears some cultural resemblance to the situation Native American and indigenous groups face, they generally don’t reject its collective relevance but instead debate how whiteness might be genomically defined, whether some racial hierarchies should be rethought, and how white nationalists might pursue historical and political narratives akin to reconciliation projects.

Research setting

We examined posts from Stormfront.org, the largest online discussion forum dedicated to ‘white pride worldwide’ (Daniels, 2009). Founded in 1996, the Stormfront site boasts one million archived threads and over twelve million posts by 325,000 or more members.³ Caren et al. (2012) found that from 2001-2010 Stormfront had 50,000 unique individuals contributing about seven million posts. In 2010 alone, there were 400,000 visitors and four to five million page-views per month. This amount of traffic is roughly equivalent to the popular US left-wing politics site DailyKos.org (Hara and Estrada, 2003). By these numbers, interest in (if not necessarily deep commitment to) white nationalism is a massive public phenomenon. Users are from all around the US and the world, but are slightly concentrated in states in the American West (Caren et al., 2012). Stormfront users are probably somewhat older on average and have slightly different concerns than those who have recently driven the flurry of alt-right online activity (Daniels, 2009, 2018) And while Twitter, Reddit, and other social media outlets coupled with backchannel communication have allowed white nationalists (and other) movements to reach new audiences for recruitment and mobilization (Donovan, 2016), Stormfront is still the most popular far right site, and its longevity makes uniquely valuable for tracking white nationalist opinion.

Stormfront thus functions as one of white nationalism’s key ‘hidden spaces’ out in the open (Simi and Futrell, 2015). It is comprised of forums and blogs driven by usually pseudonymous member discussion, making it likely the world’s largest racist public sphere. Stormfront’s (2017) mission states:

We are a community of racial realists and idealists. We are White Nationalists who support true diversity and a homeland for all peoples. Thousands of organizations promote the interests, values and heritage of non-White minorities. We promote ours. We are the voice of the new, embattled White minority!

Earlier descriptions (e.g. Stormfront, 1998) emphasized the forum as a space for planning of militant action, but this description mimics the language of left identity groups.

Stormfront aims to sidestep two kinds of internal conflicts. One is the endemic infighting among different factions and movements on the racist right - neo-Nazis, KKK, skinheads, Christian Nationalists, and now the alt-right (Ferber, 1998; Tenold, 2018). Stormfront discourages these conflicts in the name of white pride. Another is conflicts or hierarchies among white nationalities or ethnicities: ‘White people are the descendants of all historically European peoples … [we celebrate] the cultural diversity of the White race …. Our watchword is no more brothers’ wars’ (Law, 2006).

Many prominent Stormfront statements downplay white supremacy and instead emphasize white nationalism, with the implication that this is exactly what other civil rights movements for social inclusion have always been. They argue that white Europeans ‘are a distinct and unique people with our own culture, history and destiny’ whose ‘biodiversity’ is as important to preserve as any endangered species (Stormfront, 2004). Yet their sense of white preservation is dependent upon race consciousness, racial separation and purification, the sense that interracial mixing puts whiteness at risk, and constant denigration of ‘egalitarians’, Jews and non-white people. The buttoned-up invocations of nationalism cannot conceal the white supremacist and Nazi symbolism and rhetoric, the racism and antisemitism, and the threats of violence both offhand and impassioned that ooze off of nearly every webpage. Many Stormfront posters clearly perceive themselves to be victims of history engaged in ‘reconciliation projects’ (Nelson, 2016) akin to those pursued by African Americans, but the violence and terror that underwrite these visions is never far from view.

Methods and data

To address how white nationalists interpret GAT data that challenge their individual and group identity aspirations, we analyzed discussions of GAT reveals on Stormfront. Using a variety of keyword searches on Stormfront’s native search engine, we downloaded every instance we could find of individuals posting their GAT results and subsequent discussions of those results. We identified 639 posts where users revealed their GAT results in some way between 2004 (roughly when GATs first became available) to May 2016 (when data collection ended). These posts occurred within 70 different discussion threads; all of the posts within these 3,070 posts were downloaded, read and coded by the research team.

Though the data offer remarkable insights into white nationalists’ concerns, debates, and styles of thought, they also have some specific limitations. First, posting GAT results is completely voluntary and there are no checks on veracity. Given the pressures about purity and membership, we might expect people to withhold results revealing ancestry considered problematic or to report ‘whiter’ results than they actually received. This makes the problematic results that people actually do post all the more interesting.

Second, the data structure is complicated, which makes most kinds of quantifications and generalizations difficult. It is tempting to think of each GAT reveal as a distinct contest - a poster reveals the GAT result and an identity aspiration, there is a discussion, and a community appraisal results that celebrates or ostracizes the poster. But in reality, these are not discrete contests, but complex, overlapping, meandering conversations that might cover multiple result reveals (disclosing different kinds of personal information) and other issues simultaneously. It isn’t always possible to tell which earlier post a response is addressing, and conversations may occur across multiple threads, or may lay fallow for a long time only to be later revived. Any particular result might draw a range of responses from supportive to hostile, and thus ‘outcomes’ of debates can be difficult to determine. Further, Stormfront discussions sometimes concern whether a post is authentic or that of a ‘troll’ trying to stir conflict.

We use these data, with their limitations, to analyze what Roth and Ivemark (2018) would call white nationalists’ ‘identity aspirations’ and especially their ‘community appraisals’ about GATs. We do not attempt detailed analysis of the conditions under which certain individuals receive different responses to their results or the ‘careers’ of particular topics or individuals over time. The results do not lend themselves well to precise quantification and the few numbers we present here only serve to illustrate general trends or comparisons within the corpus. Finally, the analysis in this article is largely synchronic, and we leave it to future research to show how white nationalist views have evolved over time as the GAT technologies themselves have developed.

Of the 639 posts where individuals revealed their own GAT results, in 153 the precise GAT technology could be ascertained, and for 102 of those an identity aspiration could be ascertained. These were the most complete, direct, and unambiguous reveals - though not necessarily representative and unreliable for establishing rates, they are lower bounds in terms of counts. Table 1 shows that about two thirds of the detailed results were revealed with an identity aspiration, but about half of those were expressed as potentially upsetting results.

Table 1.

Detailed GAT reveals: Identity aspirations.

Identity aspiration challenged	49
Unwelcome surprise	28
Uncertain of how to interpret results	21
Identity aspiration confirmed	53
Confirmation of prior knowledge	28
Welcome surprise	25
Detailed results without identity aspiration	51
Total number of cases	153

We see that many GAT users are willing to appeal to fellow Stormfront posters for interpretive help or moral support in the face of problematic news despite the ostensibly absolute membership criterion (‘non-Jewish people of wholly European descent. No exceptions. And if you tell us you’re not, we will believe you’). None of these represented a public conversion moment where the user personally renounces or publicly criticizes white nationalism after learning of personal non-white ancestry - if GAT caused any such conversions, they did not occur publicly on Stormfront.

Table 2 reveals the counts of our top-level codes for responses to posters’ results for any of the 639 cases. The main finding here is that there are vastly more reactions about the interpretation of GATs themselves than reactions to the individuals posting the results. The wide range of discussion suggests GATs don’t have a clear meaning and represent a problem to be worked through by Stormfront users. Furthermore, emotionally supportive responses roughly balance out responses that take the results literally (as opposed to suspiciously) and shame, exclude or denounce the poster as not white.

Table 2.

Community appraisals.

Personal responses
Seek better data analysis or move data	224
Emotionally supportive statement	111
Suspicion that the original poster is a troll	72
Shaming or exclusion of original poster	65
Suggests that poster misinterpreted test results	44
Calling the original poster ‘not white’	42
Impersonal responses
Provides an educational or a scientific interpretation	1260
Delegitimize genetic tests in general	224
Reject testing company	135
Suggest there are technical/statistical error in tests	114
Racist claim/rant not targeted at person or test	98
Refute the test results on scientific grounds	45

In the following analysis, we describe how Stormfront posters handle problematic GAT revelations. First, we consider the conditions under which community appraisals end in support or denunciation of the poster. Then, we look at the strategies for repairing problematic results for individuals - anti-scientific rejections of GATs and scientific reinterpretations of GATs based on lay expertise. Last, we consider how Stormfront discussions accommodate the collective implications of GAT results by rethinking racial definitions, boundaries and hierarchies.

Community appraisals attacking revelations of upsetting results

Our coding shows that personally supportive messages in response to GAT posts are more common than shaming, denouncing, or excluding messages. What provokes these different responses? First, as mentioned earlier, the discussions are polyphonic, sometimes chaotic, and any GAT reveal might draw both supportive and denunciatory responses. The substance of the GAT results matter, but almost any result can draw supportive or denunciatory responses. More important is the way a poster responds to a community appraisal, whether with deference or defensiveness.

Nearly any result can be understood as a problem. For example, KayTee3147 (10-09-2012) revealed that her mother’s GAT showed ‘PersianTurkishCaucasus 11%’ - literally Caucasian - ancestry, and she appealed, in a panic, to the community for interpretive help. In a different example, auswhite (08-28-2013) revealed receiving a result that included 0.25% ‘Middle Eastern and North African’ identity. In response, demines (08-28-2013) wrote, ‘As per Stormfront’s rules, you have to be of wholly European descent to be white’ with a link to John Law’s membership rules. In this example, even a miniscule fraction of genetic ancestry (if taken seriously eight or nine generations in the past) and in the other a strange sounding, though assuredly ‘white’, ancestry can pose a problem for some community members. Though both received comforting replies: bioprof reassured auswhite, and Enzed told KayTee3147, though the Caucasas ‘population is muslim’ today, it was ‘white originally’ and the ‘Persians are Aryans’ (10-09-2012).

Even evidence of more sharply denigrated ancestries might receive supportive responses. aniia (04-09-2010) reported that her white-appearing adopted sister was told she had an L3 MtDNA haplogroup, which is characteristic of Sub-Saharan African populations, and all of the dozen or so responses were in the vein of explaining away or minimizing the results. A separate thread initiated by Herja (12-21-2013) included about a half-dozen posters discussing Ashkenazi Jewish ancestry in their test results. In no case was an individual’s identity denounced during the discussion of how such results might appear in one’s test.⁴ Thus, signs of black and Jewish ancestry, the two most despised among Stormfront members, are not less reparable in principle.

Rather than the content of the results, whether community appraisal turns largely supportive or denunciatory depends on personal and interactive factors. Though posters’ gender is impossible to know, revelations that seem to be by women (either in their content or the connotations of the pseudonym) tend to draw supportive rather than denunciatory responses.

More crucial, though, the tenor of response depends on whether a problematic revelation is made with proper deference to the community. Witness this exchange:

Orion22 (11-13-2014)

Hello,

Has anyone received their DNA result from saliva DNA test?

I received my results today, and I am 58% European, 29% Native American and 13% Middle Eastern.

I am pretty sure Middle Eastern is Caucasian too, as well as European, so it means I am 71% Caucasian?

…

Gargoyle (11-14-2014)

Looks like you won’t be a member here anytime soon.

Orion22 (11-15-2014)

I am pretty sure you are NOT 100% pure either. Good luck with your DNA test, if you are North American, you will be surprised how much Native American DNA you have.

TommyGunOrange (12-01-2014)

you are seriously retarded and ignorant about genetic studies

…

no, you’re just jealous because you want to be white but you cant and you hate the fact most white americans really are 100% white

…

its not our fault your ancestors were stupid fcking race mixers and race traitors don’t get mad at us just because you’re stupid dirty dog

and for your info, the vast majority of genetic studies and results show that yes, the typical white American is 99%−100% white/European

…

go troll somewhere else, There must be a La Raza website out there somewhere. you’d feel more welcome among your half breed cousins.

…

Orion22’s problematic revelation draws a wry and fairly gentle rebuke from Gargoyle. But then Orion22’s defensive rebuttal, challenging Gargoyle’s ancestry, draws a savage personal attack from TommyGunOrange (edited down by about 3/4) which itself drew a rebuke from a moderator.

Another exchange with a similar starting point has a very different outcome. RogerOne reveals (12-31-15):

Hello, got my DNA results and I learned today I am 61% European. I am very proud of my white race and my european roots. I know many of you are ‘whitter’ than me, I don’t care, our goal is the same. I would like to do anything possible to protect our white race, our european roots and our white families.

The response from FadingLight (01-01-16) was quick and harsh:

I’ve prepared you a drink. It’s 61% pure water. The rest is potassium cyanide. I assume you have no objections to drinking it. (You might need to stir it first since anyone can see at a glance that it isn’t pure water.) Cyanide isn’t water, and YOU are not White.

FadingLight suggests RogerOne kill himself, invoking the common idea that 39% non-European ancestry would be clearly visible and also that non-European ancestry is like a poison. However, Beorma246 (01-01-16) responded a bit more mildly: ‘If you do care about the White race, don’t breed with any White women. Therefore not polluting our gene pool.’ Thus, RogerOne might not be white, but he could potentially be an ‘ally’ to white nationalists. Neither Orion22 nor RogerOne is embraced by the community, but the difference in their reception - savage attack vs. ally-ship at the margins - is linked to differences in the demeanor and defensiveness of their posts.

Thus, whether the community appraisal of GAT results tends toward acceptance and repair or denunciation and ridicule has less to do with the content of the results themselves than the gender and especially the interactive demeanor of the poster.

Community appraisals repairing upsetting results

We’re now in position to address directly how Stormfront members interpret GATs so as to enable the repair of endangered identities. As we showed in Table 2 above, there were more supportive than denunciatory responses to individuals’ GAT revelations and more general discussion of GAT meanings than interpretations of individual results. What do community appraisals look like that help individuals maintain their original white identity in the face of a problematic GAT revelation? First, we show one set of arguments that aims to deny the legitimacy of GATs as a valid tool for saying anything about identity. Then we demonstrate an alternate framework that accepts the validity of GATs but interprets them so as to undermine inferences that endanger particular identity claims.

Rejecting the validity of GAT

First, we turn to claims about GATs that reject the idea that they might tell white nationalists anything about their identities. These claims are anti-scientific in their outlook, expressing skepticism about both the scientific worldview and scientists themselves while elevating a white nationalist counter-knowledge as superior (cf. Holton, 1992).

One basis for such arguments is the idea that traditional genealogical knowledge based on family histories and archival records is superior to GATs. For example, in response to a GAT result that contradicted the poster’s family history, one reply was:

My advice is to trust your own family tree genealogy research and what your grandparents have told you, before trusting a DNA test. These companies are quite liberal about ensuring every white person gets a little sprinkling of non-white DNA (we know who owns and runs these companies). Rather, these tests can be used to affirm what you know about your own European ancestral groupings, deep origins, etc.

It’s also very unlikely for whites to be mixed if their genealogy shows all European ancestors 5 or more generations back. Rampant race mixing wasn’t going on back then the way it is today. (Bellatrix 06-21-2015)

Genealogical research is a popular pursuit among Stormfront posters and they frequently describe it as the best way to learn about one’s identity. Genealogy is more specific about identity - linking one to people from particular groups and places rather than vague populations from some unclear historical past - and it is under the identity seeker’s control.

A second justification for rejecting GAT results is that race or ethnicity is directly visible. In response to a poster’s distress that GAT identified Jewish ancestry, Gladiatrix responds, ‘I wouldn’t worry about it. When you look in the mirror, do you see a jew? If not, you’re good’ (05-26-2014). CatchTheInnocence mocked the form of the GAT by posting his own ‘mirror test results’: ‘golden/reddish beard, bright blue/green eyes with a beautiful yellow circle and a White mans nose’ (07-07-2015). The idea behind the ‘mirror test’ - a play on the GAT data reveal (Nelson and Huang, 2011) - is that a true white nationalist consciousness is capable of seeing race directly; or that non-white or Jewish ancestry will leave visible traces that the race-conscious will be able to see. From this perspective, if someone does not ‘look white’ or if their appearance leaves room for doubt, then that person is not white. GATs thus add nothing.

Indeed, the role of racial visibility within white nationalist racial epistemology sheds additional light on the tendency discussed above to attack posters who reveal substantial (~30-40%) ancestry they consider nonwhite. Such a revelation is prima facie evidence that the poster is a ‘troll’ because, if such a result were true, ‘anyone can see at a glance’ (FadingLight 01-01-2016) that they aren’t white. Thus, they believe it unlikely such revelations were honest inquiries or genuine identity confusion, but rather a provocative invasion of their space.

A third way to dismiss problematic results is to reject GATs themselves as produced by companies whose leaders or scientists have an anti-white bias. In the response above, Bellatrix accuses companies that offer GAT of having a pro-multiculturalism agenda in which they try to confuse whites about their identity by attributing to them some nonwhite ancestry. Others link this bias to a specifically Jewish conspiracy advanced by GAT companies, in particular 23andMe, which they believe are owned by Jews and thus one more aspect of the Jewish controlled media (for a full articulation of such claims, see Blut and Boden, 2016). For some, the conspiracy goes beyond ideology and misrepresentation to a fantasy about biowarfare:

I would be interested in taking a DNA test to explore my ancestry, but one thing prevents me. That is the fact that 23 and Me is Jewish controlled and it would not be surprising if all the others are too … it IS possible to develop synthetic diseased that will kill only whites with a 100 percent death rate …. I think 23 and Me might be a covert operation to get DNA the Jews could then use to create bio-weapons for use against us. (ErikTheWhite 10-15-2016, quoting Volodyamyr)

For many Stormfront posters, one can discount GAT results because the putatively Jewish company ownership is invested in sowing racial doubt and confusion among whites. Furthermore, one should not participate in testing because it also empowers Jews and governments to surveil and ultimately to attack whites - GAT users are not only fooling themselves but may be betraying the cause.⁵ Though cast in implausible anti-Semitic and conspiracy theory terms, the critique here echoes those of GAT critics who have noted the technology is between science and the market and, lacking an objective standard, companies may tell people what they believe they want to hear (Bolnick et al., 2007).

Thus, the first strategy for Stormfront users seeking to repair the upsetting identity news from a GAT is to deny the legitimacy of GATs as a valid source of knowledge. Such a rejection is based on the one hand on the claim that GATs are shaped by ideologically driven companies and on the other on the articulation of a counter knowledge of identity - based on traditional genealogy and the capacity of white nationalist consciousness to see race and ethnicity - that is autonomous from interests hostile to white nationalism.

Accepting but reinterpreting GAT

A second general strategy for building community appraisals that seek to repair damaged white nationalist identities accepts GATs as legitimate knowledge, but reinterprets their genetic, statistical or historical frameworks in order to dismiss an individual’s particular result.

The most common tactic is to dismiss low levels of anomalous ancestry as ‘statistical error’. Posters often respond that ancestry percentages in the low single digits are not significant and can be discounted: ‘Every single White person I have spoken to who has taken this test ends up with less than 1 percent of some obscure region in their DNA. It’s not worth stressing over’ (SaxonCelticPride 09-15-2013). Others put the threshold of non-concern at about 5%, sometimes citing GAT companies’ own guidelines that error of that magnitude in such estimates is expected.

But there is a lot of obsessing and questioning about these small portions of upsetting results, so they are clearly considered dangerous. One individual wrote, ‘Mine says 98% Europe and I will not be revealing the others that make everything add up to 100%. I view the rest as noise DNA that’s probably there from many hundreds of years ago during conquests between Europe and other places’ (greengoddess 09-14-2013). It is ‘noise’, so the poster feels justified in not revealing it (likely to preempt negative comments), but then she gives the noise an explanation in historical processes. Despite assurances of their insignificance, low percentages can be hard to ignore. As one posted:

They had me at 5% [non-European] and the rest European decent, whether it’s legit or not I did the smart thing and went on the forum where they racially profile people by traits and race features all the time, they classified me and said I looked of Germanic, Celt stock and Borreby, Alpine like was the trait. (Ghostofwar1119 07-31-2015)

Concerned about the genetic results, this poster went to a crowdsourcing site where a person can submit his or her picture to be rated by peers for its European ethnic membership. Again, echoing the claimed capacity to see race, Ghostofwar1119 was gratified that the suspicious DNA fraction was invisible.

Likewise, tiny percentages of non-white results are viewed as part of the multicultural conspiracy of 23andMe and other GAT companies trying to sow confusion among whites:

EVERY single American’s results that I have seen ALWAYS have this 0.1% non-white garbage….[results from 23andme are ‘rigged’] for the very reason and cause of trying to spread multiculturalism and make whites think that they are racially mixed … 23andme has been called out for it’s new method of determining ancestry, this whole 0.1% or 0.2% african or native american (or whatever non white it may be) garbage is 100% falsified and inaccurate. (Herja 02-21-2014)

This poster doesn’t dispute the overall scientific legitimacy of GATs, but rather the scientific basis of disclosing very small portions of ancestry, and thus attributes GAT companies’ disclosure of the information as expressing anti-white motives.

A more sophisticated strategy for reinterpreting problematic GAT results to be less damaging is to make a version of the argument that correlation does not equal causation in the results. For example:

the main flaw is that they base your results on common dna segments.

For instance if a significant number of Turks had a certain segment resulting from the Greeks who used to live there, a Greek taking the test might come up as part Turk, not because he has Turkish ancestry but because some Turks have Greek ancestry.

Bingo. This may even account for face-value White Americans who come up with a smidgen of Amerindian. It’s not that the White folks necessarily have an Injun in their woodpile…. it’s that the base population of Injuns from whom they analyzed the markers had some White in theirs. (Skyrocket 06-01-2014)

The poster is describing how GAT companies estimate an individual’s ancestry by comparing DNA with that of a reference group. But all reference groups are defined by fiat, since geneticists must generally sample from contemporary populations rather than historical populations believed to be their origins. This poster’s critique, that an ancestry inference could be a spurious artifact, when shorn of its racist overtones echoes those of academic critics of GAT (Bolnick et al., 2007; Royal et al., 2010) and is one of the basic problems for population geneticists.

Other versions of this critique are made less on logical grounds, but in terms of elaborate historical theories that account for the mixture of white or European genes with non-white or non-European populations. For instance, one poster understood his/her ancestry to be Italian. ‘However, the DNA tests show that on both sides, I belong to hap-logroup U5a1a, which means all my ancestors come from Northwestern Europe!’⁶ A responder explained the result in terms of the heterogeneous mix of peoples constituting Italy:

It’s not really all that surprising, Italy is a fairly new country and before that there were quite distinct regions to it, it used to be Celts and Etruscans in the North, Greeks in the south Italics in the middle etc and later on the South became the Kingdom of Sicaly, also Byzantine Greek occupation happened in large portions of it, in the Central and North and the South it passed into the hands of Lombards, Franks, Goths, etc, so it’s not really all that surprising that your ancestry is from Northwestern Europe. (MaxVictory 01-04-2008)

In a separate attempt to make sense of a Native American result in his GAT, Spiralsun1 said ‘We do know that the first Americans were Europeans, and that vikings came here too’ (01-06-2006), then referenced several TV documentaries. In another example, FadingLight explained away anxieties about receiving results linking one to Persia: ‘these are OLD strains of White genetic material that turn up in odd places. Remember that Persia was a WHITE civilization to start with, and all of that surrounding area was White, too, until the Semites came’ (10-09-2012). Several years later, in a response to a person concerned about a small portion of ‘Senegal’ ancestry, FadingLight’s patience for GATs and the anomalous results had worn thin:

See, THIS is why I don’t recommend these tests to people. Did they bother to tell you that there were Whites in what is now Senegal all that time ago? No? So they led you to believe that you’re mixed even though in all probability, you are simply related to some White fool who left some of his DNA with the locals in what is now Senegal. (07-01-2015)

With this strategy, GAT results that seem to discredit an individual’s claims to white European ancestry are reframed via alternative histories (often understood to be suppressed by the mainstream). Rather than evidence of contemporary impurity, these are traces of deep histories of whiteness including its ‘heroic’ conquests, ‘foolish’ mistakes of whites, and also sometimes ‘tragic’ genetic incursions through sexual aggression of non-whites.

A final strategy is to seek out an alternative analysis for one’s genetic data. Many GAT companies allow customers to download their genotype data in a digital file. Several members reported sending their data to a sympathetic professor who, until he became overwhelmed with requests, would reanalyze people’s data (e.g. Herja 2-21-14; Nick Dean 10-18-2015). Others advocate uploading one’s data to GED Match, an online website that offers several analytical tools that give more detailed ancestry information than many GAT companies. Interestingly, GEDMatch is very opaque about what it does with the data. Unlike their skepticism of mainstream GAT companies, Stormfront posters seem to view GEDMatch straightforwardly as tool for pursuing an autonomous reinterpretation of GATs independent of the mainstream’s multicultural biases. Subsequent to our data collection, it became widely reported that GEDMatch has been used widely by police to solve cold-cases, most famously the Golden State Killer (Zhang, 2018).

What we have shown here are means of coping with upsetting results that accept the reality and relevance of GAT results but attempt to offer alternative explanations than those seemingly on offer from the testing companies. In this framework, the repair strategy is not to reject scientific or historical knowledge but offer education about the construction of GAT results and to explain those results in alternate terms. This perspective is often supplemented with counter-historical knowledge that emphasize race as the driving force in history.

Rethinking the boundaries of whiteness and the project of white nationalism

While much of the GAT discussion on Stormfront concerns helping individuals maintain their white identity by rejecting or reinterpreting damaging results, the very same results are seen as having important implications for white nationalism overall. After the exchange mentioned earlier in which demines invoked John Law’s rule to denounce auswhite based on his 0.25% nonwhite ancestry, bioprof responded

That rule will have to be updated in the face of genetic testing. A new standard will have to be set based on modern technology. The OPs genome showing some tiny degree of admixture is going to show up in a massive number of individuals of European descent. Strict adherence will result in very few individuals qualifying for Stormfront. (08-28-2013)

Law’s rule for whiteness, ‘wholly European … no exceptions’ becomes a problem as GATs become more ubiquitous and more accurate over time. Non-white ancestry, once hidden in the vagaries of people’s usually obscure family pedigrees, suddenly becomes visible. Some argue that genetics cannot solve a membership problem that is ultimately cultural and political:

Most WN’s do not hold to a ‘one-drop’ rule. If you look White, live White, identify White, if your grand-parents and great-grand-parents looked White/lived White/identified White - that is often sufficient…. Not to mention that many WN’s distrust the DNA services’ (Thomas Stuart 01-09-2014).

Despite such misgivings, there is extensive discussion about how to define membership genetically.

One line of discussion concerns whether it is possible to identify genetic markers of legitimate whiteness or European-ness, along the lines of genomic designation (Navon, 2011). In particular, posters discuss the haplogroups that differentiate among Y chromosome and Mt DNA linages and debate whether particular ones are white and European. For example, jvpski3 (09-25-2015) asked whether the Y chromosome J2 haplogroup is authentically European, and posters debated its supposed Mesopotamian Semitic origins, relation to other haplotypes, and distribution into Europe by Neolithic peoples (What is the difference between Y and MtDNA haplogroups?). In a different thread, ‘Which is the pure white haplogroup?’, a poster referred to a Y haplogroup map of Europe: ‘I see that R1a, R1b and I are the prominent European haplogroups’ (HaplogroupQuestions11, 06-21-08). Semitic-Arab responded that those are the Indo-European/Aryan haplogroups (06-23-08). And SabreWolf, picking up the thread after several dormant years, explained ‘I, J, R, L and their subclades are the major Caucasoid haplogroups …. There are African-Americans with R1b, but that does not make them White. Haplogroup is only useful for tracing the migration path, not to confirm race of individuals’ (04-04-2014). In one thread, the discussion of specifically European haplogroups led to a forum devoted to policy discussion: ‘Should people be discriminated against based on MtDNA and YDNA?’ While some haplogroups revealed on those non-recombining types of DNA were discussed as characteristically European, there was disagreement about their utility for designating whiteness - ‘by your argument Hitler is not “White” since his E1b1 YDNA is more common in North Africa’ (TennVol88, 05-15-2013).

The implicit question in these debates about MtDNA and Y-DNA haplogroups is whether whiteness should be defined only as the lack of non-white genes or whether there might be a positive genetic definition. Could there be a gene or a gene complex for whiteness, as there is for Huntington Disease or Down Syndrome? Could ‘acting white’ be genetically justified, as homosexuality has been for some activists (Miller, 1995)? While questions about what genes ‘make you’ white may raise these possibilities, much more discussion about the genetic definition of whiteness focuses on the negative question of how to identify ‘non-white’ genes. Discussions about the percentages of non-European ancestry that are individually worrisome are at least implicitly about the group definition. Though we observed no consensus about the percentage threshold of acceptable non-white ancestry, most of the handwringing about 5%, 1%, even 0.25% is far ‘stricter’ than the great-grandparent threshold mentioned in Thomas Stuart’s rejection of DNA above, which would demand results below 12.5%. Furthermore, ancestry percentages are a different logic than pedigrees: In the latter, the presumption is of discrete problematic ancestors, but in the former the source of problematic ancestry could be diffused as traces from several ancestors who had been recognized as white.

Others have tried to genetically define membership but to sidestep the fraught question of a quantitative threshold. For example, Peace Through Stormfront suggested, ‘If their point on a PCA [principle components analysis] diagram does not fall within the white cluster, they are not white. If it does, they are’ (07-09-2015). Ancestry fraction estimation works by comparing a sample to DNA of pre-defined reference groups, but PCA simply clusters individuals based on DNA similarity without prior information about group membership. PCA clusters often map onto racially defined groupings relatively well. Race realists on Stormfront and elsewhere often invoke PCA to justify their views and mock social constructionism (Harmon, 2018). While this proposed membership definition seems to sidestep the necessity of defining authentically European haplo-groups or agreeing on an ancestry percentage, the problem becomes how to identify individuals at the margins of cloud-like clusters.

Others sought not to debate tests of whiteness, but to think of new post-genomic ways to redefine it. For example, a post by sparrow attempts a complicated solution to the problem of membership criteria.

I don’t think there is going to be one giant super-nation, I expect there to be multiple smaller nations, possibly confederated in some way. I predict that each nation will have its own unique definition of ‘White’, each nation having its own standard of what constitutes appropriate genetic compatibility. So in one nation having Ghengis Khan as your ancestor won’t disqualify you, while in others it might. Hypothetically, I might take a DNA test and find that I don’t qualify for every Nation and every Nation’s Standards, though I’m sure that at least one of those nations (and probably many of them) will have standards that would include me, because I’m pretty sure that whatever Genetic mix I have is probably shared by a certain % of the White Population where that particular mix is actually the baseline normal. I don’t have to be granted ‘Status’ everywhere, I ‘m pretty sure I will be granted ‘Status’ somewhere. That of course is speculative future that we’re not really at yet. (01-09-2014)

Here the monolith of ‘whiteness’ is disaggregated into a complex set of overlapping, locally defined groups with an unclear hierarchy. This is an effort that puts ‘genomic designation’ at the core but seeks to avoid collapsing under overly rigid definitions of purity.

In another example, the danger racial mixture poses to whiteness was retheorized. The argument concerns non-recombining Y and Mt DNA as a potentially indelible source of racial pollution, in contrast to autosomal DNA, which recombination changes over generations. AngryGoy explains the idea:

HOWEVER, When it comes to direct maternal and paternal lines, I’m a strict ONE DROP fanatic! In particular the direct maternal line for females and the direct paternal line for males. The reason why I’m more liberal with autosomal DNA is that non-White autosomal DNA can be cut in half every generation from 25, 12.5, 6, 3, 1.5, .75 and so on to the point were the non-White admixture is irrelevant. On the other hand, I am more strict with Y and mtDNA haplogroups because these haplogroups are passed from father to son, mother to daughter, and remain virtually unchanged indefinitely for 10 to 20 to 30 generations!

…

I will be somewhat relieved if I find out the bi-racial female has a White Mother or the bi-racial male has a White father. I don’t applaud or condone any bi-racial person with a White partner, BUT, the bi-racial female with a White Mother or the bi-racial male with a White father are the lesser of two evils when it comes to potential assimilation. (11-23-2012)

This is an updating of a one-drop rule that measures the long-term ‘danger’ of non-white ancestry in terms of the possibility that it can be ‘diluted’ in subsequent generations of white-only interbreeding. A non-white father of boys or a non-white mother of girls is a problem because the non-white chromosome is transmitted unchanged down lineages, persisting no matter how much the rest of the non-white DNA is diluted through the generations. While non-white fathers of girls or non-white mothers of boys are less dangerous because non-white Y and MtDNA will not be passed on to children.

But bioprof posted this rejoinder:

What?! I don’t get this. mtDNA and Y-DNA have little influence on genetic expression when compared to autosomes. I do see your ‘washing out’ of autosomes reasoning, but I also see long-term admixture of those genes more problematic. Once that admixture begins in a population it just builds and builds until you have Brazil. (11-23-2012)

The response is that the idea of differential genetic dilution might make sense in the abstract, but that the non-recombining parts of DNA are not responsible for racial phenotypes. Furthermore, once non-white DNA is admitted into the gene pool it cannot be controlled and the practical outcome will be a racially heterogeneous population. What we can see in this debate is a tension between bioprof’s phenotypic and pragmatic understanding of the genetics of race and AngryGoy’s logical and essentialist conceptualization, both of which have genetic connections but different practical implications.

bioprof’s comments reflect both white nationalists’ obsession with the idea of genetic mixture as pollution (with Brazil as the dystopian endpoint), but also the anticipation of a future where GATs might reveal ‘genetic expression’ of racial traits. Lurking around the edges of some discussions is the question of what genes are responsible for looking and acting white, and that this information might be more important than that about being white. This possibility is invoked in a discussion about acceptable thresholds of Jewishness, Blomov writes ‘SE Europeans have quite a bit of Middle Eastern admixture so 12% Jewish is technically not a deal breaker. The problem is there is no genetic test that measures quality’ (5-3-18). Discussions anticipate the increased availability of genetic links to physical, behavioral and cognitive traits. Consumer genetic tests are increasingly offering genetic trait prediction - from hair and eye color to tasting preferences to IQ - which will soon provide white nationalists ample material for adding notions of ‘genetic quality’ to their evolving identities, boundaries and racial theories. Keel (2010) has shown how white nationalists use GAT revelations of partial Neanderthal ancestry (which geneticists also link to trait differences in contemporary populations) to explain purported white physical traits and intellectual superiority. White nationalists have dramatized their genetically influenced adult lactose tolerance as a sign of whiteness by chugging milk - the whitest food - at rallies (Smith, 2017). Both examples contain incongruities: The first inverts a staple of scientific racism that black people are supposedly closer to ‘primitive’ non-sapiens human species (Gould, 1996); the second ignores that genetic lactose tolerance evolved in several cattle-herding African and Middle Eastern populations, as well among some Europeans (Gerbault et al., 2011).

Discussion

We began with a puzzle: Given their absolutist rules of racial membership and a strong genetic component to their ideology of racial identity, how do white nationalists deal with GAT information that seems to indicate evidence of nonwhite ancestry? Using discussions from the white nationalist website Stormfront, we showed first that GAT results, be they ‘good’ or ‘bad’ news for the poster’s sense of whiteness, can draw a wide range of responses, but harsh attacks depend less on the results themselves than on the comportment of the poster: Cooperative and humble posters are likely to receive helpful responses, while aggressively defensive responses draw attack.

Most responses helped individuals repair the upsetting results through one of two general strategies. The first rejected the legitimacy of GATs as a source of identity knowledge by asserting the superiority of white nationalist counter-knowledge which holds family genealogy more accurate than genetics, claims the capacity to ‘see race’, and believes GATs to be part of a multicultural, Jewish conspiracy. The second strategy accepted GATs as legitimate science, but through an assertion of lay expertise offered a set of sophisticated statistical, logical, genetic and historical critiques in order to argue against this or that particular piece of identity disrupting information. At the group level, Stormfront posters debated how GATs might be tools for identifying who is white and also how white nationalist views about racial definition might be reimagined to accord with GAT revelations.

This study ironically shifts Nelson’s (2008) analysis of affiliative self-fashioning as the logic by which African Americans engage GATs. Nelson argues that African Americans’ cultural knowledge of scientific racism and exploitation and exclusion from biomedical science enabled them to navigate GAT results, selectively appropriating elements that promote particular narratives of community attachment. It turns out that white nationalists also engage in practices analogous to affiliative self-fashioning, but from a background of enacting hatred and racism, not suffering and resisting them. This suggests that affiliative self-fashioning is a widespread practice that can be animated by almost any cultural starting point.

Furthermore, our study complicates Roth and Ivemark’s (2018) theory of genetic options, which itself extends Nelson’s argument. They argue that ‘genetic determinism’ is the wrong model for thinking about GATs’ identity effects, and that identity stasis is people’s default reaction unless surprise results are accompanied by identity change aspirations and an accepting community. For white nationalists, the situation is almost inverted: The aspiration is to ‘remain’ white, but, in a nearly deterministic fashion, even a minor anomaly can be profoundly disruptive to an individual. White nationalists with upsetting or uncertain results can appeal to Stormfront for a community appraisal to help them maintain whiteness, which they are likely to receive, on balance, so long as they can endure some roasting and deferentially engage peers. Further, white nationalist GAT users may find themselves suspended between two communities’ appraisals - the nonracist community that wants to use a white nationalist’s anomalous result to attack their racist commitments, and the white nationalist community that serves as the ‘hidden space’ (Simi and Futrell, 2015) of generally supportive appraisals. Ironically, posters on Stormfront have an interest in flouting the rigid membership rule to maintain the community, while it is in the offline world where a poster’s more liberal friends and family are likely to insist on its rigid application to be able to point out white nationalist hypocrisy and deny the possibility of identity repair.⁷

Both Nelson and Roth and Ivemark are clear that GAT interpretation occurs in a social context, but both theories black box that context to a considerable degree. For Nelson, this context is largely African Americans’ general history and cultural inheritance; for Roth and Ivemark, the context of community appraisals is implicitly treated as a perceived generalized other (Mead, 1934). But our study has shown how the social context for interpreting GATs is complex and evolving, polyvocal and contradictory, and related to ideological tensions about the relationship between individual identity, group boundaries and genomic selves.

Our study also shows how white nationalists combine multiple, contradictory repertoires of meaning in the collective construction of their biosociality. First, we should note that Rabinow (1996) introduced ‘biosociality’ as the democratic, creative and liberatory modern logic of the life sciences, which would leave elitist, determinist and racist ‘sociobiology’ in history’s dustbin. The white nationalist case is certainly a democratization of knowledge, but a racist one that combines the racism of sociobiology and the populism of biosociality. But white nationalist responses to GAT are polyvocal: On the one hand, narratives of white purity and distrust of ‘Jewish’ science lead some to invoke ‘antibiosociality’, which has ironic and inverted parallels to native and indigenous groups’ skepticism of DNA as neo-colonialism (TallBear, 2013b). But white nationalists also have a strong genetic essentialist conception of race and some imagine complex ways of genomically designating (Navon, 2011) whiteness. GATs also serve what white nationalists, viewing themselves as victims of multicultural America, would see as the pursuit of ‘reconciliation projects’ (Nelson, 2016). This works first, to make visible and thus to claim diversity as a feature within whiteness without the presence of people of color. Additionally, through GATs they can rhapsodize about white nationalist deep histories and origin stories (about Aryan lineages and heroic white conquests), both as evident in the results and as explanation for anomalous incursions of seemingly suspect ancestry.

This study also has implications for how we think about white nationalism. Research has focused on its emotional/dispositional (Blee, 2002; Simi et al., 2017), social/organizational (Daniels, 2009; Simi and Futrell, 2015), and political/cultural/religious (Ferber, 1998; Tenold, 2018) dimensions. But few have emphasized its cognitive and intellectual dimensions. Our study reframes white nationalism as containing within it a citizen science movement and a racist public sphere. As we have shown, white nationalists on Stormfront actively, creatively and critically engage genetic, statistical, historical and anthropological knowledge about human diversity, picking and choosing elements to generate their arguments about racial boundaries and hierarchies. While some argue that genetics and biology are tainted and should be ignored, far more are interested in engaging and manipulating their materials. They read and debate academic articles, download their genetic data and analyze it in resources they consider more informative, and some seek to cultivate allegiances with academics they believe sympathetic to their ideas. Thus, many have worked hard to develop lay expertise (Epstein, 1995) that mobilizes the thought style of rationalist academics but combines it with the counter-knowledge of white nationalist consciousness. While certain aspects of their implicit social imaginary (Taylor, 2002) go unquestioned - such as their idea that race is the dominant variable in human life and history and whites are privileged and special within the racial hierarchy (even if white ‘supremacy’ per se is debated) - white nationalists are flexible theorists of racism. White nationalists participating in GAT debates are organic intellectuals who engage in racist bricolage (Levi Strauss, 1966) combining - sometimes ostensibly contradictory - arguments, materials, ideas, scripts, frameworks and thought styles into new forms of racist cognition (Brubaker et al., 2004). Thus, for example, some might argue GATs are valid tools and tainted by a Jewish conspiracy, that race is both visible and hidden, that the one drop rule is absurd and any racial admixture is an abomination. Their ‘rules’ for who is white are actually tools to engage in interaction and debate.

It is worth stating directly that this emerging citizen science is not devoted to local knowledge of disease, ecology or health - previous targets of similar STS work - but of racial theorizing. Brubaker (2017) has noted the ‘return of biology’ in expert and social constructionist discourses of race and ethnicity, but it also is returning as an issue for white nationalists. This is surprising, given that we understand racist discourses as generally founded in ideologies of biological difference. But scholarship on white nationalism scarcely mentions topics of biology or genetics, as keyword searches or the indices of books on the topic reveal. It isn’t that white nationalists had non-biological theories of race, though their theories incorporate culture and behavior in ways that are not wholly reducible to biology (Donovan et al., 2017). Rather, biology and genetics could, until recently, be assumed as part of the background of their ideologies of race and the political questions at the front of their cultural attention. The genetics revolution, but specifically GATs, have turned biology into a set of problems and opportunities for white nationalists.

More than simply the general rise of public talk about population genetics over the last decade or so, what changed, we believe, is that GATs have forced the question ‘who is white?’ and, less directly, ‘what is our theory of race?’ onto the front stage as practical questions for Stormfront members and white nationalists more generally. Furthermore, it is clear that GATs are structuring the very terms of white nationalists’ rethinkings of race because of what they make visible.

On one side, white nationalists see GATs as confirming and advancing core aspects of their ideology. They believe GATs confirm race realism - that racial groups and differences are a biogenetic fact - and demolish social construction, which they usually see as a ‘cultural Marxist’ conspiracy to define race as pure imagined fiction. White nationalists view the ‘100% European’ results that some receive as a demonstration that the ideal of white, European ‘purity’ is an empirical reality for many, pace the liberal claim that ‘everyone is mixed’. And GATs, which can display fractional membership in different European nationalities, allow white nationalists to demonstrate that a particular kind of white nationalist movement isn’t about ‘white supremacy’ but rather champions diversity by advocating for separation of races. Furthermore, with GATs, such claims are not mere assertions of ideology, but are made demonstrable and endowed with the authority of genetics.

On the other side, white nationalist interpretations of GATs go beyond the confirmation of preexisting ideas. The technical affordances of GATs have literally made visible new relationships of belonging and difference and thus spurred white nationalists toward innovative rethinking of the boundaries and hierarchies of whiteness. GATs have led white nationalists to debate whether a genetic test of whiteness is possible and what it might look like. For example, they have begun to think of particular MtDNA and Y chromosome haplotypes as typically European (and many others as disqualifying). They anticipate a future when GATs will reveal not just abstract ancestry, but the genetic basis for traits responsible for ‘looking and acting white’. Most scholarship on whiteness emphasizes its ‘negative’ character: It is the unmarked category that is defined by its lack of ethnic or racial traits and its (genetic) purity from the ‘defilement’ of other ancestries (Ferber, 1998). In contrast, some of the discussions we analyzed are implicitly articulating a novel genomic designation (Navon, 2011) of whiteness, where particular genes produce the white phenotypic qualities they seek to preserve. However, their belief that miscegenation will lead to an eventual white erasure is unfounded.

Other discussions engage the more familiar negative definition of whiteness. But here, GAT has effected a transition in the debates about membership criteria from, ‘Non-Jewish people of wholly European descent. No exceptions’ to, ‘What is the specific threshold?’ GAT has sharpened the divide between an absolute definition of belonging and one that has to contend with quantitative gradations and different compositions of whiteness. The difficulties of articulating specific thresholds have led both to critiques of the construction of such figures and also the desire to identify other possible representations of genetic belonging such as the idea of assigning membership in terms of location in a principle components analysis cluster. These debates have immediate political stakes: Is white nationalism a club for pure whites or a movement that will have to make compromises to be numerically effective (Tenold, 2018)?

GATs have also helped open up a contest between the idea that all race mixture pollutes whiteness and that some might be ‘diluted away’ so long as it does not affect the non-recombining paternal or maternal lineages - an updating of the one drop rule. Furthermore, it has helped open up debate about what exactly is despoiling about racial mixture: is it specifically racial traits that might be carried on particular genes, invisible essences carried with haplogroups that don’t manifest racially, or the uncontrollability of genetic mixture in general? All these positions can be seen in the exchange above about more and less dangerous mixing.

The basic assumptions about race and white purity that are white nationalism’s starting point may make it particularly vulnerable to ‘genealogical dislocation’ (Nelson, 2008) by GATs. But this study has demonstrated the movement’s ample cultural resources for shoring up the vulnerability. The literature on GATs’ identity implications have generally emphasized the possibility of flexible self-creation at the individual level (Nelson, 2008; Roth and Ivemark, 2018), but at the collective level the reinforcement of rigid, deterministic and hierarchal understandings of race and racial difference (Phelan et al., 2014; TallBear, 2013b). In white nationalism, the pattern is reversed: Individuals interpret GATs deterministically for their own identity, but at the collective level there are surprisingly flexible and fluid applications to racial conceptualization. Through these practices, white nationalism can shore up supposed vulnerabilities, promulgate new racial ideologies and challenge academic experts’ interpretations of race and genetics.

Much of the discussion about direct-to-consumer genetic tests and GATs has concerned the potential democratization of knowledge (Lee, 2013). This study shows that GATs do indeed lead to the democratization of genetics knowledge, but not through individual empowerment. Instead, GATs have made particular kinds of relationships visible and thus particular boundary and identity problems and solutions imaginable to white nationalists. They are changing the ideologies of white nationalism, or at least opening up debates and differences of opinion that might lead to changes. But, crucially, these debates about race and whiteness are co-producing (Jasanoff, 2004) a citizen science of race and racism within white nationalism. Thus, GATs are democratizing knowledge here not because they are empowering individual racists to know themselves, but because they are empowering white nationalists to produce racial genetics knowledges and counter-knowledges. These knowledge practices and interpretations may pose considerable challenges to those parties, from liberals and anti-fascists to ostensibly apolitical professional geneticists, hoping to promulgate anti-racist understandings of human genetic variation and identity.

Biographies

Author biographies

Aaron Panofsky is Associate Professor at UCLA’s Institute for Society & Genetics and Departments of Public Policy and Sociology. His work concerns the social dimensions of research on genetics and behavior; genetics, race and racism; the application of genetics and economics to education; and the reproducibility crisis and metascientific expertise. He is the author of Misbehaving Science: Controversy and the Development of Behavior Genetics.

Joan Donovan is the Director of the Technology and Social Change Research Project at the Shorenstein Center on Media, Politics and Public Policy at Harvard Kennedy School, where she researches media manipulation, disinformation, and adversarial media movements.