The sea lion: A “friendly” harasser

In 2014, two notable video game developers and one feminist video game critic fell victim to a large-scale online harassment campaign. In what is now known as Gamergate, loosely organized groups and individuals sent death threats, rape threats, slurs, and otherwise demeaning or aggressive messages to the address of women active in the gaming industry. Gamergate has since become a paradigm example of two things. On the one hand, it was one of the first large-scale online harassment campaigns. On the other, Gamergate illustrated the misogynistic backlash in the video game community. Though these two elements of Gamergate are both of interest to the argument of this paper, I will focus first on Gamergate as an instance of harassment.

A clear-cut philosophical definition of harassment proves hard to find, as it is mostly conceptualized in the context of specific forms of harassment: racial harassment, workplace harassment, or perhaps most notably, sexual harassment. Harassment in the broadest sense, which I take from Anita Bernstein, is “unwelcome verbal or physical conduct … that has the purpose or effect of creating a hostile environment” (Bernstein Reference Bernstein1994, 1241). Note that, in this definition, Bernstein already points out that the creation of a hostile environment need not be the intent of a harasser for certain behavior to count as harassment. It suffices that the hostile environment is the effect of that behavior. Think of the man catcalling a woman in the streets or grinding against her on a dancefloor (it's just a compliment! I simply find you beautiful!). Bernstein writes that (sexual) harassment “can cause harm even when causing harm is not the offender's purpose” (Bernstein Reference Bernstein1994, 1241).

Gamergate is a nice example of harassment because it offers us insights into the many different shades of (online) harassment. Most obviously, Gamergate offers examples of vicious, overtly aggressive, and hostile forms of harassment. Death threats followed upon rape threats. In a practice known as “doxing,” private information about the female gamers was purposely exposed online. The victims’ social media accounts were swarmed with misogynistic and abusive messages. One of the victims was falsely accused of having a relationship with an influential male game critic. These elements of harassment are what we might call paradigm examples of harassment: they are intended to harm the victim, and they are overtly aggressive and hostile. The intent of the perpetrators is unmistakably malicious towards the victim. Such forms of harassment are generally easy to spot, and relatively easy to condemn as bad and harmful. This is not to say that overt aggressive harassment is easy to digest, or not hurtful to the victim. On the contrary. My point for now is that it is relatively easy to recognize and categorize overt aggressiveness as bad.

Gamergate however also gives us insight into a less overtly bad form of harassment, now known as “sealioning.” One of the main strategies Gamergaters used was to swarm the victims online with seemingly innocent comments or questions, as if to suffocate them with “genuine” questioning. Andy Baio writes about the phenomenon: “Anyone who's mentioned the #Gamergate hashtag in a critical light knows the feeling: a swarm of seemingly random, largely-anonymous people descending to comment and criticize” (Baio Reference Baio2014). In a short comic dating from 2014, Wondermark (alias of David Malki) introduces the figure of the sea lion to capture this strategy (Malki Reference Malki and Wondermark2014). The Victorian-style comic depicts a woman discussing the behavior of sea lions with a friend. She is pictured saying: “I don't mind most marine mammals, but sea lions? I could do without sea lions.” We see, in the same frame, her friend warning her: “don't say that out loud!” A sea lion appears in the second frame, saying “pardon me, I couldn't help but overhear.” In the next frames the sea lion follows the woman into her house, and intrusively demands her to answer the questions cited as the opener to this paper. He begs for proof for her statement, even though the woman repeatedly insists that she is trying to do other things, like have breakfast or sleep.

“Sealioning” has since its introduction in this comic become a well-known term to describe a specific kind of on- and offline harassment or “trolling,” consisting of repetitive demands for evidence or argumentation with the goal of making the interlocutor seem either unreasonable or unwilling when they refuse to engage with these multiple demands (Poland Reference Poland2016, 144). In this paper, I focus not so much on the comic that coined the term but rather on the more general phenomenon it tried to capture, simply because there are elements of humor in the comic that are importantly dissimilar to the way sealioning is used more broadly, most obviously the absurd triviality of the claim about sea lions that sets off the discussion. I also want to look beyond the particularities of Gamergate, in which case the use of chatbots as well as the added layer of online anonymity raise specific additional questions that lie beyond the scope of this paper, and are, to my mind, not essential for the phenomenon of sealioning.

What, then, is essential to sealioning? Common definitions of sealioning range from specifically addressing sealioning as an online tool, to broader definitions of the phenomenon in real life. Amy Johnson describes it as follows:

A few elements stand out from her definition. First, Johnson states that sealioning is a rhetorical tool. She explicitly calls sealioning a “rhetorical strategy” (Reference Johnson2017, 13). This seems correct: sealioning involves engaging in a debate. Secondly, Johnson addresses the characteristic fact about a sea lion, as we might naturally call the one engaging in sealioning, that they portray themselves as reasonable while asking for information that is readily available elsewhere or already granted. Johnson distinguishes two possible goals of this behavior: to exhaust the target on the one hand, and to make them seem unreasonable on the other. Sealioning is thus intended to both affect the target (in the sense that they are pushed to “give up” or “give in”) and the onlooker (in the sense that they should be convinced of the target's flawed reasoning). It seems to be important for the sea lion that the discussion take place in public, as making the target seem unreasonable to others is part of the intention of the sea lion. A last element from Johnson's definition is that sea lions work intentionally. Their strategy is willful and malicious.

Though I want to question this idea of willful intent as a necessary condition for sealioning, Johnson does get at the heart of sealioning. Unlike the death threats and rape threats also present in Gamergate, it is characteristic for a sea lion to present themselves as reasonable. A sea lion presents himself and his questions as primarily neutral, innocent, reasonable, or even friendly and concerned (either with the victim or more generally with the truth). The sea lion presents himself as a helpful, caring, and honest inquirer. What constitutes sealioning is precisely that it is not prima facie hostile or aggressive. Importantly, this is precisely what makes sealioning effective, because why would the victim want to turn down reasonable conversation? Does the unwillingness to argue not prove the instability of her argument? If she does engage, however, the victim risks either losing her temper, and being deemed unreasonable on those grounds, or channeling all her energy towards the sea lion instead of furthering her own goals.

Though not yet meeting their criteria for official entry, the Merriam Webster dictionary describes devotes a blog post to the word, describing the characteristic double bind a sea lion forces on his victim as follows:

Again, the friendly-presenting character of sealioning is highlighted as the core of the practice. Urban Dictionary (2016) follows the same lead, stating that sealioning is:

Urban Dictionary highlights that the sea lion places a burden on the target to share knowledge repeatedly, even though it already exists elsewhere. This might exhaust the target endlessly, but because the question is framed as a reasonable one, she is left with the choice to either respond (and tire herself needlessly) or seem either unreasonable or unwilling (and therefore prove the point of the sea lion). We will come back to this double bind at length in a later example.

Sealioning is not, as is the case with other forms of harassment, aggressive or overtly hostile. On the contrary, the effectiveness of a sealioning lies precisely in its deceivingly friendly face. The sea lion presents himself much like the honest philosophical inquirer, his main concern being the truth. The sea lion seems to care about (the learning curve of) his victim, too, and wants them to find the truth together. Sometimes a sea lion appears helpful, even, because all he wants to do with his questions and remarks is make his interlocutor a better reasoner. Ostensive friendliness lies at the heart of sealioning.

Despite its ostensive friendliness, sealioning is a form of harassment. It is not surprising that the term sealioning was coined by someone addressing online trolling: sealioning has the purpose or effect of creating hostile environments for its victims, in which they are not heard, are silenced, and the validity of their experience actively denied. The nature of sealioning is that it is repetitive and intrusive. The Wondermark comic does a good job at showing how sealioning works by giving a victim the impression of a safe space, even entering the victim's most valued spaces of rest and calm. The bedroom in the comic could symbolize a victim's social media account, or her enjoying a nice meal, simply minding her own business, her feeling safe and heard, until she isn't any more. Sullivan et al. (Reference Sullivan, Sondag, Rutter, Meulemans, Cunningham, Speckmann, Alfano, Lombrozo, Knobe and Nichols2020) describe sealioning as “Incessant, bad-faith invitations to engage in debate.” They elaborate: “the rise of trolling as a form of engagement means that many invitations to debate are not made in good faith. By good faith we mean, roughly, that one is willing to change their views considering some conceivable evidence or argument that one's opponent might offer.” Note that the sea lion is not truly intent on listening to his victim's testimony, but rather has other goals.

We can discern two goals of sealioning: there is the political goal of forestalling action on the one hand, and the epistemic goal of undermining the victim's authority on the other. A particularly successful sea lion manages to fulfill both goals at once, but either one could be considered a successful outcome of sealioning. This might need some unpacking. I have mentioned already the double bind in which a victim of sealioning finds herself. The two choices she can make—either drain her energy on the debate or seem unreasonable for not wanting to engage—correspond to the two goals of sealioning. The first one is political: sealioning often has the effect of forestalling action, and “wasting” time by engaging in tiresome and repetitive debate. As an illustration, we might think here of Sara Ahmed's analysis of diversity work at universities (Reference Ahmed2017, 94). She describes how diversity workers at institutions like a university rarely get anything done, because “being appointed to transform an institution does not necessarily mean the institution is willing to be transformed” (Reference Ahmed2017, 94). However, the appointment itself, the apparent “willingness” to talk, often serves as enough of a deflection from doing anything substantial.

The Ahmed illustration points to one reason sealioning may forestall action. The ostensive friendliness and willingness of the sea lion might indeed excuse him from doing other, perhaps harder work that requires real change. Sealioning becomes a kind of filibuster. As long as someone is talking, there's no need to act. But the sea lion not only excuses himself from taking action, he also prevents his victim from doing so. As Sullivan et al. mention in their definition of the phenomenon, “it's important to recognize that engaging in debate often brings with it significant opportunity costs in terms of time, effort, and emotional engagement” (Sullivan et al. Reference Sullivan, Sondag, Rutter, Meulemans, Cunningham, Speckmann, Alfano, Lombrozo, Knobe and Nichols2020). The victim of sealioning must put on hold her own projects in order to engage in the discussion. In the case of Gamergate, for instance, this might mean that the feminist game developer gives up time to write critical pieces or develop her own work when she chooses to engage with her commentators.

The second harm involved in sealioning is an epistemic harm. Sea lions seek to undermine the rationality of the target, by questioning them so extensively that they at some point fail to answer coherently or can be caught making some sort of rational error. Oftentimes, the victim might prefer silence over taking this risk, which effectively makes them seem unwilling to engage, further proving the point of the sea lion. On the other hand, sea lions might also intend to undermine their target morally, by pushing them until they lash out, upon which a sea lion can rebut that they have remained friendly and calm throughout, as the sea lion in the comic also mentions. Once the target loses their temper, the sea lion has proven their point, and led any onlookers to doubt the moral authority of the target (“They are so emotional! They are unwilling to have a discussion! They treat us unfairly!”). Though the presence of an onlooker seems not necessary for epistemic harm to occur, as the victim might also come to doubt herself and her own credibility, the performative nature of sealioning is often prominent. Sea lions wish to show somebody (a real or an imagined audience, or the victim herself) the flawed reasoning of the victim.

The good-willed sea lion: Effect without purpose

We have focused so far on the most malicious forms of sealioning. The Gamergaters knowingly and intentionally employed this strategy to tire or undermine their opponents. When turning to the Bernstein definition of harassment, the Gamergaters had the purpose of creating a hostile environment. But what happens when there is effect without purpose? Is there not a difference between the sea lion who knows what he is doing, and one who is ignorant of the effects of his questioning? I propose that there is, but it is not a difference that exculpates the sea lion either way. Let us turn to a more subtle and complicated example of sealioning.

In the summer of 2020, the Belgian-Iranian actress and playwright Sachli Gholamalizad was the invited guest in a talk show called Alleen Elvis Blijft Bestaan (VRT 2020). The premise of this show is that one important person in the public eye is interviewed in depth. The guests are also invited to show relevant clips or bring otherwise interesting media that they would like to discuss during the interview. Though Gholamalizad tries to steer the interview in a different direction, the larger part of the interview revolves around her roots, her childhood as an Iranian in Belgium, and her experience with racism. She is asked multiple times whether her experiences with racism are real, and to what extent she may have imagined them. The interlocutor states to Gholamalizad that no Flemish person would ever exclude her, has she not simply encountered some bad apples? It is an open invitation for Gholamalizad to give evidence of the systemic nature of her experiences, evidence that by 2020 existed in undeniable amounts and can be readily found anywhere online. The interlocutor also asks her whether perhaps she has cultivated this image of being “the other,” by choosing to show “exotic Iranian” clips in the talk show. A bit later, the interlocutor points to an alleged incoherence in Gholamalizad's thinking. Is it not incoherent, he asks, to dream of borderless nations and still hold on to an Iranian identity and emphasize the importance of role models?

These are but fragments of the televised interview. I think we might convincingly call what happened here a form of sealioning. The interlocutor repeatedly asks her to give evidence that is broadly available, even when she addresses how tiring it is to answer these questions. She is asked, in the public eye, to defend her experiences and doubt is cast on her rational capacity. The interlocutor is friendly and calm, and seemingly speaks with the voice of reason. Gholamalizad gets visibly tired with these questions and would prefer to speak of other topics. We can easily imagine the conversation taking a dark turn, had Gholamalizad not remained calm. The viewer knows that for Gholamalizad it is key to not lose temper, become angry or frustrated, or to avoid the questions, since this would result in her being perceived as uncooperative or irrational. She knows that to ignore these questions about her experiences with racism is to admit, in the eye of the public that is, that she has no proof for them. She is caught in a double bind in which many people of color have found themselves before: show up and have her experiences be publicly doubted yet again, or stay away and be blamed for not engaging in the debate. Targets of sealioning face exactly this double bind: there is no right choice to make.

In many instances in which someone is exhausted, made to defend themselves and their experiences, expected to stay calm or else she is not credible, and asked for evidence of a well-researched fact (like racism or sexism), these questions are asked intentionally to exhaust or undermine the other, as was the case in Gamergate. What I want to draw the attention to with this example is the absence of overt malicious intent on the side of the interlocutor. He is hardly set on undermining Gholamalizad, making her appear irrational, or silencing her. On the contrary, he invited her to his show not with the intent to ambush her, but to shed light on the underrepresented experiences of an Iranian-Belgian woman. He seems eager to learn, and even thanks her for her insights. The interlocutor is unlike the Gamergaters, in the sense that he is convinced that he acts with (or even out of) care for his subject, not with hostility. Yet, as Bernstein notes in her definition, harassment lies not only in the purpose of creating a hostile environment, but also in the effect of doing so. In this example, though the purpose may have been absent, the environment of this TV show was hostile to Gholamalizad.

It might be helpful to understand part of what is hostile about this situation by describing it as a form of epistemic exploitation. Nora Berenstain describes epistemic exploitation as follows: “Epistemic exploitation occurs when privileged persons compel marginalized persons to produce an education or explanation about the nature of the oppression they face. Epistemic exploitation is a variety of epistemic oppression marked by unrecognized, uncompensated, emotionally taxing, coerced epistemic labor” (Berenstain Reference Berenstain2016, 570). Gholamalizad is effectively asked to do the work of educating the TV show host, even when she expected to be on the show to talk about her work. She also mentions in the interview that she is used to explaining the matter of racism to her white fellow citizens and that this is draining to her. We might further make the case that she is subject to what Miranda Fricker (Reference Fricker2007) calls testimonial injustice, since even after doing the work of explaining her experiences as a Belgian-Iranian woman, the TV host casts doubt on them again. I will come back to this point later. For now, note that effectively the continuous questioning of the TV host creates a hostile environment for Gholamalizad.

Culpable ignorance

For completeness, let me draw out a third case. We discussed how the feminist game developer might have to neglect her own work to engage with commenters’ remarks or questions. Now imagine that the feminist game developer receives a message from a co-feminist uttering her concerns about unethical practices in the gaming industry. This is not far from the truth. Several commentators noted that Gamergate had the potential to address real malpractices in the industry, but its entanglement with sexism and bigotry made it impossible to separate these significant concerns from trolling.

These borderline cases might be harder to assess in real life, but for the sake of argument, imagine that we can discern the concerns of one commentator as sincere and true. In this case, engaging in the discussion might actually benefit the work of the game developer, for she might gain new insights, or sharpen her own ideas. The commentator plays the role of an honest inquirer. She furthers the goals of getting at the truth, and eventually bettering the state of the gaming industry.

I want to say that this last case is significantly different from that of the malicious Gamergater. That might come as no surprise. Both engage in conversation, but only in the case of the concerned feminist do we find a true concern both for the truth and for justice—two key concerns for the responsible hearer, as Miranda Fricker notes (Reference Fricker2007, 99).

Now look at the case of the TV host and Gholamalizad. I want to say that what is happening here rubs closer to the Gamergaters than to the concerned feminist case. This might come as more of a surprise, because wasn't the TV host concerned with the truth, even if his sincere concern resulted in a hostile environment for his guest? Surely, the TV host is unlike the ill-willed Gamergaters, because he lacked the intention to harm his guest. In his experience, he was indeed concerned with both the truth and with justice. But we cannot reasonably maintain that he is like the concerned feminist either. The TV host, I want to say, is culpable of harassing his guest, despite his good intentions.

What makes his behavior harassment is that we may reasonably expect the TV host to have known better. Referring to the work of Sher, who argued that there can be responsibility without awareness, José Medina argues that in some cases, we might be culpable precisely because of a lack of awareness. In other words: we might be culpable for and because of our own ignorance. The idea of culpable ignorance has been discussed and developed by many different authors in a plethora of complementary ways, prominently in the area of critical race theory (see, e.g., Mills Reference Mills1997; Wekker Reference Wekker2016; Moody-Adams Reference Moody-Adams1993; Jungkunz and White Reference Jungkunz and White2013; Gordon Reference Gordon1999). It is no wonder, then, that Medina explains his concept of culpable ignorance using an example of racial ignorance:

As Medina emphasizes, the culpability of ignorance heavily relies on what we might, given the context, reasonably expect someone to know. The TV host appears ignorant about several things: he asks Gholamalizad to explain experiences of racism, revealing that he does not know about them, he is surprised when she mentions racism in Belgium, and he seems ignorant of the effect of his questions on Gholamalizad, that she is tired, that being asked about racism is a recurring event for her. I think we may reasonably expect from someone preparing an interview for national television about topics concerning racism and migration to be informed about the realities of racism. Furthermore, we may reasonably expect them to know about interview techniques that make a guest feel at ease. The TV host's ignorance, I think, is culpable enough to hold him accountable for the hostile environment Gholamalizad was in, and thus to understand this as a case of harassment, despite the good will of the interviewer.

Would it have been different if we tweak the example slightly, and replace the TV host by a friend of Gholamalizad's and change the setting to her own living room, without cameras? Gholamalizad's friend might not have the responsibility of a TV host preparing for a live interview, so this releases her from some of the culpability. But we might argue, as I would, that even though the TV host has a special responsibility, we might reasonably expect most people to know some basic things about racism and race. It is perfectly reasonable, I would say, to expect that people know, for instance, what the meaning and connotation of certain slurs are, that people of color still get called these, or that racial discrimination in the workplace is a reality. Racism as a topic has been amply treated in relevant media outlets for years, and especially in the wake of growing Black Lives Matter movements and the racial awareness these movements have repeatedly called for, it has become hard to avoid articles, testimonies, and statistics on racial discrimination. To remain ignorant about these topics is arguably a willful act for which one can be held accountable even if one is not an interviewer on national television.

Note that the culpability of one's ignorance grows proportionately to the amount of evidence that is readily accessible to counter that ignorance. The TV host and the friend might be excused for some amount of ignorance, because they live in a society that obscures and denies experiences of some more than others, but the more evidence is within their reach, the more culpable they are for not resolving their ignorance. Medina writes: “The collective ignorance may not be of one's choosing, but one cannot inhabit it comfortably and without making any effort to combat it (even when opportunities to do so present themselves), and legitimately use this inherited ignorance to excuse one's actions.” (Reference Medina2013; 140). In other words: culpable ignorance involves resistance to hearing the evidence when it is presented. And this so happens to be one of the characteristics of sealioning: the sea lion is not intent on listening, or having his mind altered by relevant counterproof. We can tell sealioning not just because people have failed to do very basic learning beforehand, but that when they are confronted with clear and obvious information, they refuse to acknowledge it and keep repeating the same or nominally varied questions.

If the TV host had asked one of his questions about racism, listened to the answer, and then moved on, perhaps we could generously interpret this as a successful attempt to combat ignorance, either his own or that of the audience. I would not be inclined to call this harassment. What makes his behavior harassment is the fact that he continues to question whether racism is as real as Gholamalizad sets it out to be, even after she has testified with sufficient detail and argumentation. Also, though he states that she changed his views on racism and expresses how happy he is to hear her story, later in the interview he makes some ignorant remarks about other forms of discrimination, which leaves the viewer skeptical about how well he listened to Gholamalizad's points about the experience of discrimination.

We can now understand the innocence maintained by the good-willed sea lion as a self-image nurtured by ignorance. Gloria Wekker describes, in the context of race in the Netherlands, how racial ignorance allows white people to maintain a self-image of innocence (Wekker Reference Wekker2016). Her view gets at precisely the kind of innocence we find in the TV host, and Wekker's argument dismantles the idea that such ignorance exculpates one from knowing and doing better. Much in line with Medina, she argues that ignorance is not innocent. The sea lion, however, exemplifies how ignorance presents itself as innocent. This innocence in turn works to put the responsibility with the victim: “I'm sorry that you felt hurt,” “I did not mean to corner you,” “I didn't know you felt that way,” “I'm just asking because I'm curious, why would that upset you?” In other words, the good-willed sea lion gets to decide what is hurtful or hostile and what is not. This gets us at the importance of calling the TV-host's behavior harassment. Denying the existence of a hostile environment to Gholamalizad, because the TV host didn't intend harm, places the experience of the good-willed TV host above hers and starts from the assumption of white innocence. “Friendly” harassment plays into and cultivates that innocence, which is what makes it so deceptive and tricky to spot. However, I argue that we should pierce through the performative innocence of “friendly” harassers and hold them accountable for not knowing better.

Note that sealioning need not be successful to count as harassment. Even if Gholamalizad does not catch the bait, or calmly and eloquently answers the question, she has undergone some amount of harassment. The environment created is no less hostile because the victim is somehow able to deal with the threat quite well. The sea lion nonetheless succeeds in the two relevant goals as I defined them. First, the political goal of forestalling action is furthered by spending less time on discussing Gholamalizad's movies and other artistic works. Imagine what ground she might have covered for the TV audience if they'd simply discussed her films instead of rehashing the fact of racism? Secondly, Gholamalizad's authority on her own experiences is undermined, as well as her authority as a director and artist, by reducing her story to her experience of racism. This furthers the second, epistemic goal of sealioning.

To reiterate, we discussed three cases of friendly-presenting inquisitive dialogue. The first, the Gamergaters, is a clear instance of sealioning, which is a form of harassment, because the explicit goal of these commenters adheres to the double goal of sealioning: to forestall action and to undermine the victim's credibility. The second, the TV show example, is also a form of sealioning, and thus also of harassment, because even though the double goal of sealioning is not the explicit purpose, it is indeed the effect of the interview, and, importantly, the interviewer is culpable for this effect because the conversation has this effect due to his culpable ignorance. The third form, the concerned feminist example, is not a form of sealioning nor of harassment, because the explicit goal and the effect are joint better understanding and a bettering of the gaming industry.

Bad faith and performative goodness

It might be helpful finally, particularly for the purpose of understanding the culpability of the good-willed sea lion, to zoom in further on a concept used in some of the definitions of sealioning above: bad faith.

The existentialist notion of bad faith has been successfully used to illuminate the social dynamics of race and racism, most notably by Franz Fanon (Reference Fanon1967) and Lewis R. Gordon. Gordon describes bad faith as an attitude entailing “a flight from displeasing truths” (Reference Gordon1999, 29), lying to oneself (8), and “the effort to hide from responsibility for ourselves as freedom” (8). Gordon's use of bad faith gets at similar points as those made by Medina and others about ignorance, though from a different angle, namely that looking away from ourselves or the realities around us does not exculpate us from responsibility about them.

Gordon's ideas on bad faith can further illuminate the possibility of good-willed badness. The harassment of the TV host can be attributed to him having the attitude of bad, rather than good faith. According to Lewis, bad faith need not be ill-intended (though it most certainly can, and as such is also applicable to the ill-willed sea lion), because the crux is precisely that we unknowingly hold false realities as truths, because we neglect to investigate and question them. We can spend our whole lives living in bad faith without having realized it. To speak in these cases of malicious intent would be a stretch. How can we have bad intent or be ill-willed if there is no consciousness of the lies we tell ourselves to begin with? It is better, then, to understand these forms of bad faith as a type of neglect to inform ourselves, as the absence of taking one's responsibility rather than the presence of ill-will. There is the lack of intention to listen, to entertain the other's perspective, rather than the presence of intention to undermine. So far bad faith illuminates the same points about the TV host case as did Medina's ignorance.

But there is another merit to the notion of bad faith in the context of “friendly” harassment that motivates my dwelling on it a bit longer. The original Sartrean example of bad faith sheds light on precisely what is so interesting about sealioning, namely that ostensive friendliness and good intent can sometimes obscure badness.

Sartre introduces the reader to bad faith with the famous example of an over-eager waiter:

What is interesting about this example for our purposes is that the exaggerated efforts of being helpful, of attending to the customer, are precisely what makes this waiter bad-faithed. His whole persona collides with his role as a waiter, but this overt preoccupation with being friendly and coming across as friendly prevents the waiter from attending to himself and others truly and authentically.

It might appear as a bit of a stretch, but I think it is worth drawing the parallel with the TV host. The TV host, much like the waiter, knows what it generally takes to be a good host. You ask questions, you seek to illuminate, you try to embody the questions an audience might have, and so forth. But this preoccupation with the formalities of being a TV host obscure the particularities of the situation the TV host is in. His guest is an Iranian, with a rich socio-political history, living in Belgium, making film. His guest belongs to a group that is repeatedly met with skepticism and doubt (more on this in the next section). He fails to attend to himself outside of his role as well, neglecting to address his ignorance and prejudices. Lastly, he fails to take the responsibility tied to his role as a public authority. Instead of repeating what happens in chatrooms and comment sections on social media, he could have set an example. He could have chosen, freely, to step outside of the paradigm of inquisitive TV host. Following Sartre, this would have been good faith, a TV host realizing his freedom to make authentic decisions. Instead, we find a TV host adhering to a script, insensitive to the particularities of each new situation.

Consider the following scene from Phoebe Waller-Bridge's series Fleabag (Reference Waller-Bridges2019). A family goes out to dinner, and everything turns sour quickly—dad asks insensitive questions, stepmom is condescendingly passive-aggressive, one of the daughters has a miscarriage in the restrooms and refuses to tell anyone or go to the hospital, the other daughter smacks the husband of the first daughter in the face for his flippant opinions, and at least three people end the night in the bathroom stalls with bloody noses. Amidst all of this, a waitress (“needy waitress,” the main character says in characteristic Fleabag style directly to the camera) is constantly looking for ways to attend to the table. She responds disappointedly when another waiter had already filled their glasses, she kindly forces drinks on her diners by stopping by every two minutes and finally gets herself in the middle of a fight between the table guests. Needless to say that she is one of the people to end up with a bloody nose.

The Fleabag character is, I think, a prime example of the Sartrean bad-faithed waiter, who moves like “some kind of automaton” (Sartre Reference Sartre1956, 59). So concerned with her function, we can imagine her overt helpfulness turning pushy, intrusive, misplaced, and unwelcome. The waitress is so concerned with her performative goodness that she fails to read the situation properly. Much like with the Sartrean waiter, the ostensive goodness of the TV host is not rooted in authentic concern for the guest, but rather in a concern for his own, performative goodness. Sartre notes, the waiter “is playing at being a waiter in a café” (Reference Sartre1956, 59, emphasis in original). Similarly, we could say the TV host is playing at being a good host, rather than authentically finding out what that means. As such, his preoccupation with goodness can, in bad faith, take the form of harassment.

Sealioning and testimonial injustice

Now who poses the real threat to good interaction: the ill-willed Gamergater or the bad-faithed, culpably ignorant, yet good-willed TV host? Who is the real sea lion? I want to say both are, but they are dangerous in different ways. The Gamergater is dangerous in the most obvious sense: from the safe space of anonymity, he inserts venom into the debate, willingly and knowingly causing harm to his victims. These are the festering depths of the internet that have the potential to explode in physical violence. As Kate Manne says in her analysis of aggressive incel culture, the bad-willed sealions are the “overachievers” (Reference Manne2017, 62).

But especially with sealioning, there lies a wholly different harm in the good-willed inquisitor who neglects to address his ignorance. As we saw, ostensive friendliness is at the heart of sealioning, and what is more friendly than a good-willed person? Because it is so hard to discern what one might reasonably expect another person to know, it often feels like the duty of the victim to engage in the debate, educate, and stay strong. The double bind really finds its expression with the good-willed sea lion.

I have said at the start of this paper that Gamergate was a prime example of two things: of harassment, which we have discussed in depth, and of the misogynistic backlash. Now, it is no surprise that these go hand in hand. As my examples of sealioning showed, the most pertinent cases of sealioning are hard to divorce from racialized and gendered sociopolitical contexts. This, it seems to me, has everything to do with our tolerance for ignorance and the social penalties we get for not knowing. That the TV host is not really penalized for his ignorance is not surprising, because the ones with the power to penalize him are as deeply rooted in a system that rewards racial ignorance for the white people maintaining it, as Charles Mills explains in his influential work (Reference Mills1997).

It follows then that those about whom most ignorance exists are most likely to be victims of sealioning, and feel obliged to do the emotional and educational work, instead of spending their time and energy on other projects that are important or enjoyable to them. Holding people to higher bars with regards to which ignorance we forgive and which we hold them accountable for would mean that sea lions are less likely to succeed, because their ignorance would not be confused with innocence, friendliness, and curiosity, as it is now.

As such, there is a close bond between the Frickerian concept of testimonial injustice and sealioning. Stating the two goals of sealioning, I argued that the second goal involves undermining the authority or credibility of the victim. Testimonial injustice, as Fricker has it, “occurs when prejudice causes a hearer to give a deflated level of credibility to a speaker's word” (Reference Fricker2007, 1). Testimonial injustice occurs in situations of testimony or knowledge exchange governed by socially engrained prejudices about who has the epistemic authority. It takes place in what Fricker calls “an economy of credibility” (30), in which some have more epistemic credibility than others in virtue of their perceived belonging to a given social group.

Now in principle, sealioning need not involve social prejudice or credibility deficits. Sealioning can occur outside of these sociopolitical dynamics. In a fictive example of sealioning Robert Bloomfield asks us to imagine a website aimed at cat lovers, intended to be a space to discuss their shared love for felines (Reference Bloomfield2018). Imagine now that dog lovers swarm the website and flood the different conversational threads with questions about why cats would be superior to dogs in any way. They demand from the cat lovers that they make a reasonable case for having this particular interest in cats, effectively steering them away from actual productive conversation among the cat lovers in which they might exchange tips and tricks or relevant experiences with cats. This, too, is a case of sealioning. However, there is arguably no hierarchy of credibility between cat and dog lovers. It is not the case that, generally, we trust dog lovers more than cat lovers based on prejudice about either group of people. There is also no deeply entrenched communal ignorance about either group.

It is tempting to say that the Bloomfield case is the “pure” case of sealioning. I think we should resist that intuition, because most cases of sealioning do not exist outside of the sociopolitical contexts of marginalization and oppression. The pure, paradigm case should precisely be one that cannot easily be divorced from the sociopolitical context of marginalization, the power dynamics and history, and the status in the hierarchy of credibility of victim and harasser.

The bond between testimonial injustice and sealioning is twofold. First, the hierarchy of credibility gives rise to sealioning, and serves as a basis to it, for sealioning starts with an attitude of doubt, question, skepticism. As said, these attitudes are not bad ones to hold in general, as exemplified by the concerned feminist. But in the case of sealioning, I hope to have shown, the ground for skepticism and inquiry is often culpable ignorance (which could take the form of prejudice). Different people are approached with different levels of skepticism and inquiry based on the social group they are identified with.

Secondly, sealioning further increases testimonial injustice, because it has the effect of pushing groups of marginalized people who are already associated with the emotional, the irrational, and the untrustworthy, to eventually confirm these prejudices. Pushing long enough, the victim might lash out emotionally, become impatient and irritated (all the while, the sea lion remains so friendly!), fail to debunk an accusation of inconsistency, or simply refuse to talk further, which is in turn interpreted as a sign of untrustworthiness (“is she out of arguments?”).

Both the case of Gholamalizad and the Gamergate case are neat illustrations of this dynamic. Both are cases of sealioning of which the victim is, not coincidentally, a member of a marginalized group asked to defend or speak on behalf of that group. Note also that, in both cases, to be an advocate for a group was not the main intention of the victim. Gholamalizad is a filmmaker, and even though some of the Gamergate victims were openly feminist, other victims were merely harassed for being women in the gaming industry. Their responses, however, were taken to be exemplary of a group: Iranian immigrants in Gholamalizad's case, women in the case of Gamergate. If the sea lion succeeds to undermine the credibility of his victim, the onlooker will likely come to doubt not only the authority of the victim, but also of those she is taken to represent. And just like that, more sealioning behavior is assured in the future.

How do we break this cycle of testimonial injustice? In the last section, I narrow in on what it might take to do this.

Where to look for possible interventions?

In the case of “friendly” harassment like sealioning, there are three (groups of) agents involved who have the potential to intervene successfully and halt the harasser. Some, I want to argue, have a responsibility to act, respond, or change, others merely have the option to do certain things, but do not bear the responsibility.

Let us work our way up roughly from least to most promising potential to intervene. First, there is the victim. Some people are quite good at avoiding the bait and calling out the strategy behind the incessant questioning. But ultimately, as discussed, the point of sealioning is precisely to put the victim in an impossible position. As is characteristic of sealioning, there is no good response, since the choice presented to the target is a double bind: either you respond but get exhausted and your energy is diverted away from effective solutions, and you run the risk of eventually losing your temper and being deemed rude or crazy, or you refuse to respond, and appear unwilling or at a loss for argumentation. The better way to respond then is to lay bare the motivations of the sea lion. Having a well-defined term for this behavior can help with this strategy, as it enables targets of sealioning to call out this deceptive friendliness.

This brings me to one of the aims of this paper. With the conceptual clarification of sealioning, and broadly “friendly” forms of harassment, I hope to have added to the hermeneutical means victims have to recognize and name the situations they find themselves in. Because sealioning and “friendly” harassment broadly construed is more likely to happen to people who are already marginalized, finding ways to address and call out these sea lions is therefore a means to the goal of hermeneutical justice, as described by Miranda Fricker (Reference Fricker2007). Being able to understand seemingly friendly behavior as harassment, and have words like “sealioning” to describe those experiences, might indeed fill a “gap in collective interpretive resources” that constitutes hermeneutical injustice because it “puts someone at an unfair disadvantage when it comes to making sense of their social experiences” (Fricker Reference Fricker2007, 1).

It seems however quite unfair to have to cultivate appropriate responses to unfair situations, rather than to have the sea lions stop their behavior. Ending this paper with practical tips for victims calls to mind the strategy of offering self-defense classes to women to address sexual harassment: we are teaching victims to be stronger and more resilient, instead of addressing the problem at its core. The question of justice in the end should not be how to best respond to unfair situations, but rather, as Joel Anderson and Axel Honneth aptly put it, “the question of justice is whether the burden is fair” (Anderson and Honneth Reference Anderson, Honneth, Christman and Anderson2004, 131). The victims of sealioning should have to bear the burden to address the wrongs against them as much as victims of sexual assault should: not at all.

That brings me to the second agent, who without a doubt has the greatest potential and responsibility here: the sea lion. To combat the harassment, it would of course be easiest if the sea lion would simply refrain from acting this way. Now if the sea lion is ill-willed, and knowingly wants to undermine another, it is not likely he will be susceptible to the arguments made in this paper or find a call to stop his behavior compelling. But the case is different for a good-willed sea lion who is simply ignorant about his behavior. There is some clear low-hanging fruit to be picked here: truly listening to the other minimizes the risk that we are sea lions to others. Also, as sealioning (especially the good-willed kind) is so intertwined with testimonial injustice, ignorance, and prejudices, we do good by addressing these in ourselves, and by being skeptical about our own skepticism. We can practice what Miranda Fricker calls being a virtuous hearer, who is concerned with truth, but also with that which is conducive to truth, such as the minimization of prejudice in one's credibility judgments (Fricker Reference Fricker2007, 99). It is worth noting that being a virtuous hearer might involve more than simply a volitional effort, as Linda Martín Alcoff has argued in response to Fricker, since credibility deficits and testimonial injustices are also nourished in pre-cognitive frameworks (Martín Alcoff Reference Martín Alcoff2010). We may nonetheless find in Fricker's work, but undoubtedly also in the work of others, many ways in which to sharpen ethical sensibility for our epistemic and moral relations to others (Fricker Reference Fricker2007, 81), and this indeed seems to be the obvious task at hand.

Now there is one last party involved in sealioning. We discussed earlier the performative nature of sealioning. The sea lion wants to show someone that his victim is wrong, should be doubted, cannot debunk the criticism, and so on. Though sealioning might occur in private as well, it has a performative aspect to it, for we could say the sealion imagines an audience. In some cases, the audience is arguably the victim herself, which admittedly complicates this last point I want to make. Let us say, then, that when an audience is present, and in many cases there is, the audience has the greatest potential to make the sea lion's attempt unsuccessful.

The sea lion finds confirmation in others. ‘See!’, he says with his questions; ‘she cannot defend herself!’ And indeed, she cannot, for she is caught in the double bind. But the onlookers can choose to question the inquisitive stance of the sea lion, see through his strategy, and name it. Imagine, in the TV studio, that someone in the live audience or another guest had spoken out, insisted that Gholamalizad that she need not answer these questions, and pointed out the previous answers Gholamalizad had given and to which the TV host had failed to listen. Surely, it would have caused a scene, and it might have appeared as ostensively unfriendly to the host. But it would have been an example of someone acting in good faith, seeing their responsibility, resisting this host's performance of innocent ignorance, reading the context right, and realizing their freedom to speak authentically, keeping both the goals of truth and justice in mind.