Abstract
In 2004, awash with the hope for a public sphere reinvigorated by the popular internet, the online arms of many U.S. newspapers opened their websites for comments. Now, nine years into this experiment, many newspapers have abandoned the practice of allowing comments. Online news sites have adopted a variety of strategies to deal with offensive comments, including turning “comments off,” not archiving comments, and adopting aggressive comment moderation policies. These strategies present researchers who wish to understand how racism operates in the new public sphere of mainstream news sites with a set of methodological dilemmas. In this article we (1) lay out the methodological pitfalls for the systematic investigation of the prevalent pattern of racism in online comments in the public sphere and (2) suggest steps by which scholars may deal with these methodological intricacies. We conclude by pointing to the broader implications of online content moderation.
The rise of many-to-many communication networks initially offered great promise for a “new public sphere” of citizen journalists and, at long last, an overthrow of corporate agenda-setting in political news gathering, reporting, and framing. In this new public sphere, the people—rather than news media pundits or news corporations—would (re)shape democratic discourse through direct participation (Perlmutter, 2008). Simultaneously, the advances toward racial equality achieved through the political struggle known as the Civil Rights Movement, along with the disembodied nature of online communication, has led some scholars to speculate that the web may serve as a democratic digital utopia—a world in which minds are privileged over bodies (Bleeker, 1995; Hansen, 2006).
In 2004, awash with the hope for a public sphere reinvigorated by the popular internet, the online arms of many U.S. newspapers opened their websites for comments. Nearly a decade later, many newspapers have abandoned the practice of allowing comments (Washington, 2010). Online news sites have adopted a variety of strategies to deal with vulgar and offensive comments, including turning “comments off,” not archiving comments, and adopting aggressive comment moderation policies. These strategies present researchers who wish to understand how racist discourse operates in the new public sphere of mainstream news sites with a set of methodological dilemmas.
Perhaps most puzzling is the way in which racism seems to “erupt” in discussions of events not overtly about race. For example, in the aftermath of the 2007 shooting rampage at Virginia Tech, poet Nikki Giovanni gave a moving eulogy qua original poem. Within minutes, a video clip of her reading appeared on YouTube where viewers added comments to the site. Many attacked Giovanni in extreme racist language characteristic of the pre-Civil Rights (or perhaps the antebellum) era. The presence, indeed the preponderance, of racist comments in the public sphere suggest a problem that Howard Rheingold has referred to as a “classic tragedy of the commons dilemma,” in which “flamers, bullies, bigots, charlatans, know-nothings and nuts in online discourse take advantage of open access to other people’s attention” (Rheingold, 2004: 121). This article interrogates the methodological dilemmas of studying racism in online comments and suggests a possible way forward.
Background
Racism in the press is not new (van Dijk, 1991, 1992, 2000). Yet the advent of the comments sphere of online news outlets has significantly transformed the discursive shape of racism. Given this momentous turn that is not even a decade old, it is striking that more research has not better interrogated either the nuances of racism’s online news manifestation or the perils and promises of social scientific inquiry into the matter. While some research has examined the presence of racism amidst online news formats (Harrison et al., 2010; Meyers, 2004; Tateo, 2005; Teo, 2000), this article aims to challenge previous assertions that the web is a utopian space free of social distinctions like race (e.g. Beckles, 1997; Miller, 1995) and gender (Plant, 1998). Indeed, the editors of Race in Cyberspace (Kolko et al., 2000: 5) found it significant enough to write: race matters in cyberspace precisely because all of us who spend time online are already shaped by the ways in which race matters offline, and we can’t help but bring our own knowledge, experiences, and values with us when we log on.
The internet provides no escape route from either race or racism. Instead, race and racism persist online in ways that are both new and unique to the internet, alongside vestiges of centuries-old forms that reverberate both offline and on (Daniels, 2009a).
Not much research examines the new digital saga of racial and racist discourse in, and in response to, online news formats. A few lines of inquiry have emerged to stress the salience of race in online representations (Melican and Dixon, 2008; Mitchelstein and Boczkowski, 2010; Ronkin and Karn, 1999; Steinfeldt et al., 2010). For example, Hughey (2010) investigated how online film critics, both professional and amateur, make active meaning of “white savior” cinema, as well as how members of black fraternities and sororities reproduce their racial and gender identity amidst online chat forums (Hughey, 2008). Daniels (2008a , 2009a, 2009b) studied the discourse of avowed white supremacists online, in both overt expressions of racism and in more pernicious “cloaked” sites. Daniels (2008b) also scrutinized how notions of hate speech and civil rights are re-interpreted when expressed in digital formats, and Daniels and LaLone (2011) examined the racism in online video game communities. Together, these studies suggest that the internet allows for the acting out of “nonconsensual racial fantasies” (Sharpe, 1999: 1094). Sharpe (1999: 1094) writes: It is not simply that we can be more “honest” about race issues on the Internet because we are not accountable but that the personas we create (even if they begin as “us”) start to take on lives of their own …
The racist comments at online news sites are, in some ways, the realization of such “nonconsensual racial fantasies.”
Despite the emergent arena of research into online racial identity, only a handful of studies have systematically examined the way comments on the online forums or news sites intersect with racial invective. For example, Hughey (2012) examined how racism and xenophobia, in the particular form of anti-Obama Birtherism, was constrained and enabled by the format of online news commenting in the New York Times and the Wall Street Journal. Through a random sample culled from a population of 12,539 comments left at the Times and the Journal, Hughey (2012) found that commenters constructed several “narratives of belonging” that outlined where differently racialized people belong and what traits they supposedly share. In their study of over 1000 posts related to University of North Dakota’s Fighting Sioux nickname and logo used for their athletic team, Steinfeldt and colleagues (2010) found a critical mass of online forum comments represented disdain toward American Indians by providing misinformation, perpetuating stereotypes, and expressing overtly racist attitudes toward Native Americans. What is needed at this critical juncture in this field of research is a roadmap for how social scientists can navigate the increasingly complex and fast-paced intersection of race, news reporting, and the internet with a coherent and replicable methodology. Specifically, we need guidelines on how to recognize and study increasingly disguised forms of racist discourse in public and virtual contexts.
Methodological dilemmas
During the mid 2000s, many news sites saw the advent of the modern internet as potent instrument for the dissemination of news (Washington, 2010). Given the skyrocketing growth of social networking sites and users at the time, many saw the online arms of their papers as a way to attract new readers. It is fair to say that many thought such user- and comment-communities would manifest, at best, as benign discussions of the news at hand or, at worst, as heated discussions surrounding hot-button topics covered by the story. Such an assumption was motivated by the notion that the web is, at its core, a neutral space (Bleeker, 1995; Schwartz, 1995) free of the confines of racial identity in which minds are privileged over the “politics of the skin” (Fanon, 1967). Newspapers were unprepared for the ways in which race performs like a “ghost in the machine.” 1 Soon, the comment pages of newspapers exploded with racism of the most virulent type (Washington, 2010).
As a response, the online arms of newspapers adopted a variety of strategies. The first was simply to abandon commenting or to allow commenting on certain stories. For example, papers like the New York Times, Boston Globe, Buffalo News, Raleigh News & Observer, San Diego Union-Tribune, and the Reuters news service no longer allow anonymous online comments (Falkenthal, 2011; Pérez-Peña, 2010). Second, papers began requiring registration at their sites before posting and, along with registration, many online news sites require some form of identity verification connecting users’ real names and addresses to their online identity before allowing comments on their sites (Curley, 2010). A third approach has been to adopt aggressive comment moderation policies in which discourse deemed offensive is either removed or pre-screened. This third approach is of interest to social scientists because, while some comment moderation is automated, human actors are generally required to monitor comments. Such processes are both time-consuming and labor-intensive, so some newspapers hire subcontracting firms to “sweep” (remove) offensive comments. We are interested in the comments – supposedly tame and free of racist intent and/or content – that find their way to the public, past the gatekeepers.
Moderation policies
The moderation polities of flagship U.S. papers demonstrate varied, but relatively unified, approaches toward civil discourse and race. For example, the New York Times states that one should: use respectful language that does not abuse or discriminate on the basis of race, religion, nationality, gender, sexual preference, age, region, disability, etc. Hate speech of any kind is grounds for immediate and permanent suspension of access to all parts of this service. (New York Times, 2011)
USA Today states that: “Hate speech or stereotypes that attack or disparage an ethnic, sexual or religious group or a member of any such group may result in comment removal” (USA Today, 2011). The Wall Street Journal states that users will: not upload, post or submit anything that is obscene or contains profanity or that may be hateful or offensive on racial, ethnic, sexual or any other grounds; is harmful, vulgar or distasteful; or is defamatory, libelous, or invades another person’s privacy or proprietary rights. (Wall Street Journal, 2011)
Many of these papers once allowed anonymous commenting. Now, users must register and provide “real life” contact information about their identities, ways to verify their identity, and how they may be contacted if need be. For example, in August 2010 the Buffalo News did away with anonymous commenting, which itself stirred a debate. Naysayers expressed that such a move curtailed First Amendment rights to free speech. Supporters of the policy, the Buffalo News wrote: are relieved that the astonishingly hateful and venomous commentary on news stories (“It’s as though when you order a sirloin steak, it comes with a side of maggots,” writes Gene Weingarten in a column that ran today) will likely be restrained once people have to identify themselves. (Sullivan, 2010)
However, moderation policies such as these carry methodological implications for discourse analysis. In a pilot study, the authors found that the most popular and widely used papers simply moderate-out racist commentary. The result, in effect, is “whitewashed” data. While such moderation strategies may be good for either newspaper business or civil public discourse, they represent a challenge for potential discourse analysis because they construct an ideal image of participation in the public sphere that hides the racism that frequently emerges.
A methodological way forward
A diverse cadre of scholars has documented the linkage between racism and discourse across the globe (e.g. Reisigl and Wodak, 2001; Riggs and Augoustinos, 2004; Steyn and Foster, 2008). Specific to a U.S. context, a substantial amount of scholarship examines how college students navigate politically correct discourse related to race (Bonilla-Silva, 2002; Bonilla-Silva and Forman, 2000; Foster, 2009; Myers and Williamson, 2001). Simply put, many now avoid public discourse that even hints at racist intent or content. This research demonstrates how racist rhetoric has changed form. What once was said in overt Jim Crow-style epithets, now finds expression in subtle, symbolic, and polite forms (Bonilla-Silva and Forman, 2000; Gallagher, 2003).
Racist discourse has not only shifted form, it has also changed location (Daniels, 2009a; Hughey, 2011b; Picca and Feagin, 2007). Given that racist language has (1) changed form to become more subtle and (2) moved backstage to private areas, one might view the intersection of racist discourse and the internet as a third space betwixt and between the public and private spheres. It allows for intimate discursive interaction and the formation of an “imagined community” (Anderson, 1991) between users who may entertain the notion that they are conversing in the private “backstage,” when, in fact, their comments are on a public “frontstage” for all to read. Here, we interrogate how such discourse may fly under the noses of moderators, moderation policies, as well as scholars engaged in discourse analysis, and suggest ways to study such phenomena. We are not so concerned with whether or not these policies and actions have in fact changed the nature of racial discourse from overt to subtle. While it is clear that these virtual spaces engender a place for new forms “race talk” to take place (see Myers, 2005; Nakamura, 2002), its possible causal role in generating that discourse is anything but clear or settled. Hence, we are interested in the racist language that gets past these reviewers and moderation policies, and how analysts might become savvy to the presence of racist discourse when it takes a disguised form in contexts declared “post-racial” because of institutional guards.
Coded language
Coded racial language has long been used to convey subtle, yet potent, racial meanings in ways that appear well reasoned and focused on the common good (Mendelberg and Oleske, 2000). Newspaper comment fields engender specific types of coded language to fly well under the radar of moderators, like in other surveilled virtual formats. For example, young Facebook users often employ coded language or “social stenography” to hide the intended meanings behind posts and status updates from parents and moderators (Boyd and Marwick, 2011). In respect to online news sites, subtle and coded language is often employed. For example, in speaking about Barack Obama, a Wall Street Journal comment from 2 March 2011 read: If this POS [Piece Of Shit] current White House occupant isn’t the product of the soft bigotry of low expectations/equal opportunity/race based initiatives then why can’t he find his birth certificate, produce one written piece he contributed to the Harvard Law Review (where he served as editor) and stop fighting all attempts for information?
So also, in reference to Islam and the conflation of “Muslim” with “terrorist” a comment from the New York Times on 10 August 2006 states: Please explain to me just one more time why we must value diversity, sensitivity and political correctness more than National Security, The ‘religion of peace’ is at it again! Let’s try profiling. Who knows? It might work. I’ll gladly help pay for the ACLU lawsuits through my tax dollars.
In the same thread, another comment read: Not all Islamic people are terrorists, but all the terrorists are Islamic! Stop the political correctness and ban Islamic people from airports and aircraft. maybe the non-terrorist Moslems will finally stop being so indifferent and start policing their own flock.
2
Such coded language represents a challenge for scholars savvy to the polysemy of language and to the resilience of racism and its ever-morphing properties. Too often, arguments over the frequency and form of racism bifurcate into vulgar relativist stances about the impossibility of proving racist meanings versus critical approaches that posit racism in every syllable (Daniels, 2011). To avoid overly sophistic and structuralist approaches, scholars of discourse analysis must develop what we call Racial Internet Literacy (RIL). Just as scholars have outlined how racial codewords reproduce “real life” racial segregation and inequalities (e.g. when home insurance agents ask: “Do the kids in the neighborhood play hockey or basketball?” Farley and Squires, 2005: 33), scholars must also become acquainted with the slang and language of virtual racial invective and messaging.
Not only must new methodologies be racially literate, but scholarly approaches must outline the grammatical, semantic, and pragmatic strategies for constructing racial in- and out-groups based on subtle insinuations of “authentic” belonging, superiority, and/or normality. For example, Ronkin and Karn (1999) studied how the subtle parodying of stereotypical black vernacular reproduced a discourse of African American intellectual inferiority. In particular, they found: (1) asystematic graphemic representations of phonetic segments; (2) the hyper-use of be; (3) semantic and pragmatic derogation; and (4) the use of vulgar expressions (Ronkin and Karn, 1999: 363). In so doing, they found key instances in which such subtle internet discourse enables and posits racist and essentialist characteristics of African Americans in ways that would be highly offensive in other media contexts, such as call-in radio shows or letters to the editor.
In this vein, new methodologies must consider the context of the story and the dominant political climate of the time. That is, the discourse must be analyzed in relation to the overall racial climate of the era (e.g. “color-blindness”) and the story’s particular context (e.g. the election of president Obama). Given that nothing is a priori racist about discussions of Obama, birth certificates, and presidential qualifications, the larger social and political stage in which this drama plays out must be spotlighted in future research.
“Common sense” racism
When coded language is not used, newspaper commenters often cover their racialized speech in “common sense” appeals to supposedly race-neutral principles and/or by appealing to historically dominant and well-entrenched racial stereotypes that are collectively shared and rarely challenged. This defense occurs in three ways: (1) abstract arguments that invoke the individual’s right to engage in “free speech,” (2) accusations of victimhood that appeal to “political correctness,” and (3) seemingly matter-of-fact statements that are based on implicit racial stereotypes and myths. Such rhetorical strategies evade moderation because they shift focus from the specifics of the racialized content to questions over abstract principles of civil and democratic discourse or to supposedly scientific or “obvious” racial differences thought natural or innate. For example, a comment on a New York Times article about “Illegal Immigration” (15 August 2006) illustrates one way that this “common sense” racism plays out.
Immigrants have and continue to change the US landscape. As a 3rd generation CA resident imigrants have effected our lives. The violence and low level of education are the main concerns. Over crowding, over reaching government policies such as rent control to public smoking bans are not what I consider effective government. The CA school system is closer to the bottom of 52 than the top. This is due to immigrants whose mother tongue is not english. As a father of two preschool children, we decided in January 2006 to move from Los Angeles to Munich Germany to raise our children.
Here, the comment combines his dislike of “politically correct” government policies (e.g. smoking bans and rent control) with his displeasure regarding “immigrants whose mother tongue is not english.” Finally, in a move that defies logic, the commenter concludes with his decision to engage in a rather extreme form of white flight—moving to Munich, Germany—thus aligning himself and his family with the category “immigrant” in a nation that has a much more developed welfare state infrastructure than the U.S. This comment locates the sources of racialized social problems in the bodies of darker-hued people, while avoiding the systemic and patterned relations that differently allocate resources and which affect immigration rates. By blaming “illegal immigrants” (a coded term nearly synonymous with darker-skinned people south of the U.S. border), the racist discourse slips by and is reproduced in this online space (cf. van Dijk, 1997: 31).
Another comment from the New York Times (4 August 2009) illustrates how the racist Birther ideology found its way through moderation by appealing to free speech: That there’s a PR spin going on with respect to the birthers’ claims — that much, at least — is quite obvious. In my mind, the jury is still out and MoveOn and other organizations that are attempting to shut down CNN/Lou Dobbs and others who would dare cover this matter need to realize that while the allegations may prove false, the First Amendment protections ought to stand strong. Attacking someone’s First Amendment right to be a nut — or a truth teller, whatever the case may be — is the greater evil, in this reader’s opinion.
After a string of comments were deleted in that same comment thread, a new line of rhetoric emerged to frame the Birthers and their supporters as victims of political correctness. In blending antiblack rhetoric with claims of politically correct “reverse racism” against whites, racist discourse again found expression: His staff are doing a great job avoiding the real truth behind is illeged US Citizenship as well as other serious topics, bravo. Most of you want to play the race card for Mr. Obama in sympathy well that is a crock. Im a white male that was born and raised in this country ans my parents were and thier parents etc.. and I cannot find a job I’m qualified for, nor can I get a free education because someone elses accent or skin tone needs to be addressed before mine and other whites like myself. And the funny part is is that I’m the minority now not Mr. Obama and other non caucasian racial people of this country. […] Caucasians have passed laws, and rules that try to be fair to every other race in the country which is more then ok with me but we excluded ourselves, why is this? We will fail as a nation, regretibly because of soft sympathizers like you. (4 August 2009)
Comments in the Wall Street Journal (26 May 2009) demonstrate how a story about Sonya Sotomayor’s confirmation to the Supreme Court quickly attacked her as unqualified, as occupying a seat meant for a white (man), and as representing a supposed double standard in which whites are the real victims of the racial order.
I know it’s naive to think that political and racial correctness doesn’t play some part in her nomination and probably just as naive to assume that the person best suited for the job will be chosen regardless of gender or race. What really bothers me the most is the fact that if she were a white male and her comment regarding her experience had the terms “wise Latina” and “white man” revesed so she said that she hoped the decisions made by a wise white man would be better than those of a Latina there would be the greatest uproar of acusations of racism you have ever heard! That person would be castigated as a racist and unfit for the job! But when she says essentially the same thing, it all is accepted as OK. So it really does reinforce the idea that racism and condescention is never acceptable unless it is directed at white men, then it is.
In the same thread, commenters began to evidence strands of white nationalist thought: that whites are oppressed and should recognize and act upon that oppression as a racial collective with their “allegiance, taxes, and blood.”
I think this is a simple example of the fact that there is a double syandard of acceptable behavior—one for “whites” and another for people “of color” or “non-Northern European ethnic diversity.” Most, yes most, “white people” are not racists nor do they want anything more than the most qualified person to fill the position regardless of age, race, or gender. Once we “white people”, more specifically “white men”, finally come to the conclusion that it isn’t about the pursuit of that goal, it brings into question why we continue to support the system as a whole with our allegiance, taxes, and blood.
Such comments evade moderation because they appeal to principles of unfettered free speech, or by simply claiming that their rhetoric, or the rhetoric of other commenter, was not racist because “one of their best friends is black.” For example, a 22 April 2011 comment from the Wall Street Journal reads: “I keep hearing all these charges of ‘racism’ despite no evidence. […] it isn’t because I’m a racist. My best friend is black […].” Another comment on the Wall Street Journal on 9 November 2009 reads: The problem with PC is that it is intended to flatter the speaker and not the subject. It makes us all feel warm in our genitals to talk about giving equal opportunities to all people equally. We are free to view ourselves as such trusting and benevolent souls when we allow Arab Muslims into our military while we fight two separate wars against a Muslim enemy in Arab nations. […] I’ve got no problem with Muslims. I have friends who are Muslim who are just wonderful.
Claims of inter-racial friendship and camaraderie supposedly insulate the commenter from the charge that they are “racist,” because a “racist” would never possess inter-racial affinities in the first place (Bonilla-Silva, 2010; Hughey, 2011).
To analyze the modern incarnations of racist discourse in these virtual spaces, scholars must attune themselves to the common uses of appeals to the principles of political liberalism, fear-mongering over “reverse racism” or “political correctness,” and expressions of implicitly held racial stereotypes.
Allowing incivility for sake of balance?
Given the moderation policies of these news sites, it is a strange occurrence that a severe incivility is often found to manifest in the backlash against antiracism. Perhaps these comments are allowed in attempt for papers to seem “democratic” and “fair and balanced” in an era in which many neoconservative voices attack media for a supposed “left-wing bias.” For example, in the aforementioned 10 August 2006 story in the New York Times, two comments read: Left-winged conspiracy theorists are as ignorant and close-minded as their proverbial right-winged war-mongering enemies. Could it be possible that this was vigilant police work that saved lives, you doubt junkies? Come on people. Quit being so cynical and opportunistic about sharing your anti-Bush/anti-Blair hatred. We get it, you feign an ‘enlightened,’ albeit simple, worldview and scoff at any facts to the contrary. Get over yourselves. People’s lives were saved. No matter what your political persuasion or opinion on Bush, we should be thankful. If you seriously think that this fresh breath of good news in the media is an evil PR stunt by the anti-christ George Bush, your shameful and clouded conscience is just as ignorant as the people you so hate. Long live freedom. It amazes me at the number of ignorant, left winged liberals in the world! Is there anyone who will call a spade a spade and realize that we are in the midst of a crisis and not a finger pointing game? I bet all you nose-thumbing democrats would lay your life on the line for your country, right? Oh no, you’d just whine about how someone else should do it for you! You’re fortunate to have the right to an opinion!
So also, a 21 April 2011 comment in the New York Times read: “Not American Enough”? Has anyone noticed the obvious? I guess it will take 10 years going by. Then some really smart guy will figure out that Obama was raised Islamic in Indonesia. I’m not saying he’s Islamic now. […] Sure, Obama’s got the mental maladies one catches in the left wing fever swamps. It’s anyone’s guess what really goes on in this clown’s mind. He never had his academic background vetted - thanks to the likes of Taranto and other gutless “journalists”. The cesspool we call the South Side of Chicago has included a politically influential gang that slaughtered everyone who wouldn’t convert to Islam. Then there’s the Siamese twins - Farrakhan and Rev. Wright - both of them Qaddaffi pals.
In a 2007 discussion about illegal immigration agencies in different states increased their searches for illegal workers. Comments on this story took many traditional paths (e.g. jobs no one wants, cheap labor because big business is greedy); however, one comment in the discussion on the USA Today piece on illegal immigration stood out in opposition to the rest: Now Now none needed we have already proven that we will stay here and we are a force that will not be ignored, so what no AMNESTY LOL, we do not need it we are taken this country and making it ours, the question is should we give you AMNESTY when we are in control? Becareful we are the new face of america and with the changing faces so to shall the power pass.
This comment provoked a series of responses.
I have a job and legally too..I am a citizen, how about you? If you are not legal this is not your country. And calling Americans cowards certainly does not harness any good will for your cause…besides, I am quite certain I could stand up to you real men and beat the living salsa out of you. Keep on talking though…you definitely are helping to encourage my fellow citizens to stand up and fight against people like yourself.
And this poem: Chicksh!t Mexicans. Too afraid to fix their own country. Run to America lazy chicken Mexicans. Running away is so much easier than fixing YOUR country. Run to America scared little Mexicans. Run away from your governmment. Its so easy to run to America than stay and fight for your righs in YOUR country. Run away chicken little Mexicans. You are such wimps.
Many news outlets seem to allow racism and racial invective when combined with political discourse. In the context of a politically bifurcated and antagonistic environment, an intimate synthesis of political rhetoric and racial othering may occur. When overt progressive attitudes over race are expressed in these virtual spaces, it appears that moderators allow for more overt forms of racism for sake of balancing the argument. Hence, discourse analysis must separate the wheat from the chaff so as not to conflate openness to political rhetoric with the allowing of racist discourse. Failure to examine the mechanisms by which overt political rhetoric on the internet (perhaps the site of the most virulent and abusive forms of today’s discursive political battles) is mapped onto nonwhite racial groups, would make invisible the ways in which nonwhites are often labeled as the catalysts and carriers of social dysfunctions and pathologies, while whites are deemed normal or neutral participants in an otherwise color-blind society. Nuanced discourse methodologies could then detail the ways that white supremacist logic employs political rhetoric as a Trojan horse amidst the internet landscape. Given that the internet is becoming the major means for the production of public opinion and the dominant consensus on ethnic and racial affairs, this matter is more than academic.
The whitening of news commenting and the putative digital divide
Given the synthesis of public/private spheres that characterizes online news commenting, we may be witnessing the whitening of news comment fields. Importantly, these sites should not be theorized as anomic virtual locations free from the requirements of social expectation and accountability. Rather, these news sites can serve as spaces for the reproduction of a certain kind of racial persona—an unfairly victimized white person living under the onus of political correctness where one can’t tell it like it is. Given that people seek community and ways of expressing meaningful aspects of the self, the internet has been found to reaffirm racial identity and racial communities (Byrne, 2008a, 2008b; Everett, 2002, 2008; Nakamura and Chow-White, 2011). Moreover, given offline racial segregation in housing, education, and employment (Brooks, 2011; Logan, 2002; South et al., 2011), the absence of meaningful inter-racial interactions and experiences is likely to transform virtual realms into spaces for a normalized and de rigueur sense of whiteness-as-victim (Nakamura, 2002). Research does indicate that certain online spaces attract certain racial groups, especially when a pattern of racial rhetorical and/or political agendas is attached (Daniels, 2008b, 2009a; McPherson, 2000).
Such dynamics may contribute to the famed “digital divide.” Rather than a divide over resources or knowledge (as it is often framed), the scarcity of internet content relevant to nonwhite interests, coupled with racist discourse, may contribute more to lower internet adoption rates amidst people of color (Brock, 2006). Moreover, when news sites are heavily moderated, those who express racist rhetoric may in fact leave certain news sites to seek out others where their discourse can find expression. Preliminary research, operationalized via interviews with white, middle-class men who frequent conservative blogs/news sites, demonstrates that some retreat to places where they may say what they like without “fear of being called racist” (Sobieraj and Berry, 2011). Given the high probability of these events, the challenge for analysts is to incorporate longitudinal and comparative models into their research, whereby racial discourse and the racial demographics and/or racial avatars of users are noted in order to detect rates, trends, and processes of virtual segregation and/or hostility.
Conclusion
The unexpected (for some) emergence of overt racism in online comments has prompted news sites to change their practices. In large part, news sites now use bots to search for profanity and racial epithets while human moderators judge whether or not discourse goes beyond the pale. Given this attention it is clear that moderation matters and that news sites take seriously their charge to guard the health of the public sphere. However, the moderation of comments certainly gestures toward at least three areas of concern.
First, moderation is costly. In an era in which newspapers are framed as a dying medium (even their online arms), these papers spend considerable time, money, and energy on racial gate-keeping. Moreover, to avoid these costs and the loss of revenue from advertisers who do not wish to risk association with incendiary and offensive comments (Falkenthal, 2011), newspapers are simply disallowing commenting so as avoid potential firestorms of controversy. Given the potency of such an economic impulse, the retention of overt racism and the smuggling of subtle racism into the fields of comment sites are certainly startling. The presence of online racism speaks of a legacy of racial and ethnic othering in which group status, rights, and responsibilities are often questioned in concert with moral panics over very real or constructed social problems.
Second, thorough moderation may provide a false sense of a “post-racial” or “color-blind nation.” Although a quick overview of unmoderated sites like the now defunct site College ACB.com or YouTube.com would quickly disabuse anyone of this notion, it is evident that many conflate the absence of racism with the moderation of racism. Many now incorrectly assert that racism is declining in significance due to lesser manifestations of antebellum-style rhetoric in select public spheres. However, given that the internet now represents a strange synthesis of the public and private, racial discourse is simultaneously a form of backstage discourse (Hughey, 2011; Picca and Feagin, 2007) and politically correct narratives structured by the fear of losing face (Bonilla-Silva, 2010).
Third, moderation is itself a reactionary tactic predicated on dominant understandings of racism. Rather than engendering a focus on interfacing with racist discourse in a meaningful way, these news sites simply attempt to ban the problem, frame the offenders as rogue commenters, and then bury their head in the sand—hoping not to encounter more racism. Such a strategy is indicative of our neoliberal moment in which societal issues are boiled down to one’s supposed individual right to avoid them and the claim that one is unaffected by them. Moderation does not address racism, but rather hides it and passes the problem along to other virtual venues where racism will find expression.
Finally, it would seem that the online arms of newspapers are struggling with how to disentangle racist discourse from “culture war” rhetoric. In the current socio-political moment in the U.S. various cultural issues stand as code words for race (such as meritocracy, gratification deferral, nuclear families, and employment). This shapes the framing of news by mainstream outlets and the contours of online news comments, thus enabling the proliferation of racism in ways that evade moderator bots and human gatekeepers.
Footnotes
Acknowledgements
We would like to thank the Duke University Discourse Project for spurring the initial idea for this article, Nick LaLone for research assistance, and the reviewers for their helpful suggestions on earlier drafts.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.