Pepe the frog: A cartoon turned meme turned renegade.

Questions of scale, variability and trajectory of an image.

Disclaimer: Some of the links in this article will lead to the online-imageboard 4chan. The dynamics of this website result in ever-changing content, some possibly offensive and inappropriate and “not safe for work.”

Exhibit 1: Donald Trump’s twitter 2015

On 13 October 2015, not long after Donald Trump had declared his candidacy for the presidential elections, the above image was retweeted on his official account @realdonaldtrump. It is unlikely that Trump himself created the image, yet the source remains unknown. The tweet links to four other accounts, including @BreitbartNews, and redirects to a video on YouTube, titled “You Can’t Stump the Trump (Volume 4),” consisting of a compilation of short clips in Trump’s favour and his immunity to criticism.

Trump, known for his erratic posting, nonetheless raises questions with this uncommented tweet of an image showing a humanoid cartoon frog as the president of the US. This text will look closer at  by looking at the trajectory of an agglomerate image, a meme, and will cover memetic theories to illustrate its elasticity in meaning and scale. It will look at the conditions of the internet, in which the meme thrived, and focus on key events when it became visible. It will look at transgression, appropriation and reclamations, at acclamation and at convictions of the mainstream media and academic and subcultural belief, ranging from indifference to delusion.


“In 2016, a cartoon frog named Pepe, mascot of white supremacy, makes Donald Trump President of the United States.”

What sounds like a conspiracy theory or a strange joke at best was reported not only by partisan sources, but also by the mainstream media. Most notably, even Donald Trump and Hillary Clinton actively contributed to this debate.

The story starts in 2003 with an obscure and anonymous online community and ends with Donald Trump being president. Its main character is Pepe the Frog, its setting the internet that had been evolving from a hard-to-navigate digital no-man’s land (its contents oscillating between hilarious, disturbing and illegal), to a politically charged, yet mainstreamed battleground: a war between liberals and the alt-right, fought out with viral images and the media.

As the main body of this text has been completed before May 2017, it will cover events up to that day, including the depicted statistics. In regard to current events, a short update will be found at the end of this article. 

From Boy’s Club to “Sad Frog” – Pepe becoming a meme

feelsgood man
Feels good man, 2008

When Matt Furie posted his comic “Boy’s Club” on his myspace in 2005, or when he published his first book a year later, he, as stated in an 2010 interview with, did not expect the scale one of his characters would reach. His humorous narrative is simple if not infantile, and involves foul language, bodily fluids and weed. It is mostly conveyed on single pages depicting the weird–yet relatable–everyday life of four college friends, “energetic Brett, charming Andy, hedonistic Landwolf, and unassuming Pepe.” “energetic Brett, charming Andy, hedonistic Landwolf, and unassuming Pepe.” According to knowyourmeme, in 2008, one of the stories, in which Pepe completely pulls down his trousers to urinate, proclaiming “Feels good man” in the last panel, found its way online. Someone had scanned the page and uploaded it onto the /b/ (random) section of the imageboard 4chan, where it was cropped and coloured in.

pepe furie official comic feels good man
The original, ca. 2005

Now, this platform stands out in the sea of today’s social-media-dominated internet with its simple architecture and old-fashioned design. An anime figure embedded in warm colours greets the visitor with the choice which specialised board they want to access, ranging from anime over fitness, politics and Not-Safe-For-Work Content (NSFW, usually involving pornography or gore)–/b/ being the most popular, “notorious for its rude, and aggressive rhetoric and content” (Shifman 2014, 178). The platform’s FAQ explains how the website works: a new thread is started with posting an image; any reply to this thread has the option to post one, too. All participants are anonymous, without the need to sign up for an account. The thread’s life spans are short and expire after a few hours or days into automatic deletion, being replaced by another of the around 200 threads of every board. Threads getting many responses are “bumped” into visibility to the top of the list, until even popular threads are forced to move to the bottom of the board, where they eventually vanish. Since posts are constantly threatened to disappear, participation to “bump” the thread is required and a strong selection process emerges “by requiring content the community wants to see be repeatedly reposted, and potentially remixed,” which hints that longer-remaining threads mirror the general interest of the anonymous crowd (Bernstein et al. 2011).

graph 1
Peak in 2008 – “Feels Good Man,” looking at the time from 2004 until April 2016

“Feels good man” was of interest. The image started to appear in other threads to communicate feelings where words would not suffice, like so often in digital realms. He was becoming a staple, a visual manifestation of these words. To trace this success, Google Trends is a useful tool to show power-law distributions for searches of specific key terms even after deletions of the threads because it “measures and graphs the relative popularity of Google searches for a key phrase” (Milner 2016, 38). , “Feels good man” first peaks in 2008 when it began circulating on 4chan.

“Feels bad man,” 2009

The image started to evolve, when users recognized the exploitability of it, and first variations appeared, such as  “Sad Frog,” showing a depressed Pepe, often textually referring to “Feels good man” with a speech bubble reading “Feels bad man.” It was mostly used “to denote feelings of failure or disappointment,” that emerged around 2009 on 4chan.

Sad frog, 2009

These copies somehow still related back to the original image and were embraced as much as the ur-Pepe, an attitude towards images typical of the boards that were created in 2003 by a teenager called Christopher “moot” Poole by copying the code of the Japanese forum  2ちゃんねる, read: 2 Channel or 2ch, and 4chan, too quickly accelerated into a cultural hub (Phillips 2015, 18). Built to give western fans of manga and anime a forum, the early community exhibited strong connections to the geeky Japanese otaku-subculture, made up of social reclusive individuals that exhibit, according to Mizuko Ito, “a postmodern sensibility expressed through arcane knowledge of pop and cyber culture and striking technological fluency” (Ito 2012, xi). Users of these websites are media-natives, equally consuming original stories and fan-made derivative works, even parodies, based on characters rather than narratives–it is irrelevant whether the visual material is created by the original author or a secondary source (Azuma 2012, 49). Benjamin’s aura is no longer meaningful, as long as the derivative is recognizable; the myth of authorship declines as the creator becomes indistinguishable in the anonymous mass where vigorous evaluation and consumption of media “certainly seems to move at the level of simulacra where there are no originals and no copies” (Azuma 2012, 32). For Limor Shifman, copies are “the raison d’être of digital communication” (Shifman 2013, 373).

What is a meme?

From this context, early memes emerged.

They are defined as collectively generated visual content, jokes and parodies of cultural goods circulating online. As a means of communication they are native to our mass-mediated world (Shifman 2014, 2). Not as individual images, but as agglomerates whose items always refer to one another, they create internal networks (Milner 2016, 2). Memes spread and reproduce “by various means of copying and imitation […]. In this environment, user-driven imitation and remixing […] have become highly valued pillars of a so-called participatory culture” (Shifman 2014, 4).

Meme-scholar Limor Shifman names three points of analysis by which memes can relate to each other and by which they can be transformed: content (the ideologies within it), formal aspects and stance, with “which relates to the information memes convey about their own communication,” the position of the addresser towards the text, a tone “with funny, ironic or serious dimensions (Shifman 2013, 367).

Pokémon-Pepe, ca. 2009

Ryan Milner describes early memes adapting the reductive aesthetics of the internet’s programming, often as image macros inherent to computed information, deliberately celebrating the vernacular in anti-elitist sentiments. Memes are bricolage, images crudely put together from material at the creator’s disposal using heterogenous, limited tools. “Bricoleur’s contributions to culture come from reappropriating what others have already created”; mainstream culture is influenced through their reappropriated products. (Milner 2016, 61) Often, images, like Pepe, are used together with text, putting the original image in new contexts to convey emotion (Milner 2016, 66).

Hito Steyerl talks about “ poor images,” small-file copies circulating at the expense of quality and resolution–by-products of digital culture. There are parallels to memes–subject to the aesthetics of the internet, requiring literacy to see the original in the copy. Like poor images, they are low in a hierarchy of images where sharpness and an abundance of pixels are valued, even fetishized. Memes seem to be stuck in the infant stages of image processing–Microsoft Paint versus Photoshop.

Despite them seeming trivial, Shifman recognizes that “they actually reflect deep social and cultural structures. In many senses, Internet memes can be treated as (post)modern folklore, in which shared norm and values are constructed through cultural artifacts such as Photoshopped images or urban legends.” (Shifman 2014, 15)

When the first memes were created on 4chan, users connected the concept to evolutionary biologist Richard Dawkins pre-digital 1970’s memetic theory, where ideas of spreading and evolving genes were applied to cultural change–memes as “units of culture and behaviour which survive and spread via imitation and adaptation” (Metahaven 2013, 31). These cultural ideas depend on longevity, fecundity (their appeal and ability to catch on and spread), and their copyingfidelity (traceability to its origin during circulation). The internet seemed to provide ideal conditions for derivative media, as they can be accurately copied, “since digitization allows lossless information transfer. Fecundity (the number of copies made in a time unit) is also greatly increased. […] Longevity may potentially increase, as well, because information can be stored indefinitely in numerous archives” (Shifman 2014, 17). Still, Dawkins biological theory is not always applicable to internet memes: “Too much fidelity to an original idea or text […] may actually undermine the longevity and fecundity of a memetic idea” (Milner 2016, 69): Deviation from the original, like the derivative Pepes displays, is key: “a shared premise has to remain just recognizable enough to allow for creative innovation within a social vernacular” (Milner 2016, 73).

Memes, agglomerates for Shifman, cannot go viral, but one manifestation of a meme can, being passed on from person to person, enhanced by social media platforms until its visibility reaches over several networks (Shifman 2014, 55).  For Steyerl, their crudeness makes images accessible and circulate quickly through sharing platforms, forums and image boards, until they reach absolute visibility, making them a new form of commodity: downloadable, editable– “And the results circulate.” Mirroring this thought, Shifman sees the spread of ideas, memes, copied and remixed “through competition and selection” (Shifman 2013, 364f). And Milner concludes that human agency is key here, as memes are only shared if they trigger an emotional response in an individual. “Media texts become memetic when they connect with enough participants to inspire iteration after iteration from a fixed premise.” (Milner 2016, 29)

Smug Pepe, 2011

The meme as a corpus is “alive on the Web: it only takes a couple of mouse clicks to see hundreds of versions […]; the visibility of memetic variability–as well as the timeline of the meme’s appearance–is an integral part of contemporary internet memes” (Shifman 2014, 30).  Focusing on Pepe’s spread, by 2010, he had become relatively famous in pertinent subcultural communities of 4chan. Matt Furie did not seem to mind when asked in an, but he wondered about Pepe’s vectorization, his taking a life of its own detached from the original comics.

“Sad Frog” spawned “Smug Frog,” as another emotional reaction image.

Mainstreaming Pepe

A completely new level of visibility was reached three years later–through celebrities’ social-media: Katy Perry posted a Pepe in November 2014, lamenting her jetlag, and  Nicky Minaj followed a month later, again with a reaction-image. Their fans, respectively, became familiar with the frog by googling it and subsequently circulating it further with likes and reblogs, likewise expressing emotions. Pepe, as “the frog meme,” had left the subcultural realms of 4chan and taken over social-media. The general public started looking.

In an article on  Motherboard on 9 April 2015, Roisin Kibert reports Pepe’s status quo of having developed from an in-joke to a meme recognized by the mainstream, which was met with disdain by the former in-group, declaring him dead, “a commodity” unable to evolve further. The aggression against “normies,” non-internet dwellers who were “invading” the popularizing platform, thus changing its userbase, posting without being familiar with the behavioural norms online, spawned an “Angry Frog ” in mid-2015.

Angry Pepe, 2015

Feeling that “their” meme had been taken away from them (a common behaviour of 4chan is claiming collective ownership of a meme, according to Milner 2016, 105), 4chan tried to counter the mainstreaming by voraciously flooding the boards with “rare pepes” to crash the “market” with authentic products, even grossing a high price on ebay for them. Speculating on Pepe’s future, Kibert assures that his inappropriateness would protect him from advertisers, though, and, most profoundly, wonders whether Pepe “with the right partner, […] could transform into a prince.” In light of the following developments, it is surprising how right she was.

Memetic power both ways

Memes are elastic. They can be jokes, insults or raise serious arguments, depending on which elements are added to it and in which context it is employed. Shifman even emphasizes their political dimension (Shifman 2014, 122f). This showed especially in the wake of Occupy Wall Street (described in the book Can Jokes Bring Down Governments by design studio and research collective metahaven) when 4chan’s and memes could be traced to the Occupy Wall Street movement–by partially incubating hacker-collective Anonymous, a catalyst for media-native introverts to channel small-scale political acts into a social-networked physical movement and shaping a new anonymous protest-culture. Memes were a visual means of communication and identification and served to protest the normalization of financial crisis, countering the establishment’s reality-management-strategies by removing it from its framework since their inherent humour hits institutional politics by declaring every of their realities nonsense, disobediently enabling alternatives. Jokes (or memes) are powerful, because they “perform what everybody knew but couldn’t say. […] The meme has escaped the confines of internet forums, and is becoming a tool useful to targeted political struggles” (Metahaven 2013, 56f).

Anonymity, and the alt-right

The anonymous condition of the internet is suspected by many scholars to disinhibit behaviour, which might help sharing personal information to create a communal identity (Hollenbaugh and Everett 2013, 283). Parmy Olson observes these behaviours in her book We are Anonymous: 4chan maintains the anonymous feature; users believe it to promote logic and inhibit vanity that would lead to the development of elitist hierarchies; likewise, the only hierarchy within the anarchy existing on these boards is the difference between the original posters–OPs–and the community (Olson 2013). On 4chan’s Japanese predecessor 2ch, the anonymous and uncensored condition enabled brutal honesty and irony that accentuated “the boundary between those who “get it” and others who don’t,” as observed by Akihiro Kitada (2012, 70). On the other hand, there were media-bashing and bullying, and even reactionary posts in political sub-forums, opposing the initial sincerity of the boards and instead deliberately taking up unpopular opinions (Kitada 2012, 81). This suggests that even though these platforms didn’t generate reactionary sentiments, they nonetheless enabled their expression–as 4chan and 2ch exhibit a similar trajectory, just a few years apart. The community had changed.

Around 2014-2015, a new ideology had begun to become of medial interest : The alt-right, a movement mostly incubated online, opposed to traditional conservatism and propagating nationalism and spreading their ideas through media outlets like Breitbart News. On 4chan, meanwhile,  /pol/ (politically incorrect) was gaining popularity–a board with adult-content known for its “endorsement of extremist politics,” and seen, with /b/, as one of the most notorious boards. In July 2015, a drawing attributed to a Malaysian artist appeared there, showing smug Pepe in front of Trump’s announced border-fence to Mexico.

Fence-Pepe, 2015

Later that summer, Trump tweeted his Pepe. Somehow, the frog had turned from a mainstream meme to a Trump-supporter.

Milner observes that often, when a meme starts spreading in “participatory media, traditional media outlets” would eventually join the conversation (Milner 2016, 187). Attention to Pepe being used in reactionary discourse was therefore delayed in the offline world, until on 26 May 2016,  Olivia Nuzzi delivered an explanation for what was happening on (at times sensationalist) The Daily Beast. Based on twitter-conversations with anonymous insiders, she writes that Pepe-as-Trump was an attempt to reappropriate the frog from the “normies” that had ruined the in-joke with their participation. @JaredTSwift, an “anonymous white nationalist” claimed that mixing Pepe with Nazi propaganda to make white nationalism mainstream was a campaign by the alt-right, originating on 4chan and was now taking over twitter (with the hashtag  #frogtwitter) to make their agenda visible to the media. Another of her sources,

@PaulTown_, claimed that the campaign had even bigger goals–Pepe had just been a test and had worked, surprisingly.

Trolling culture.

Platforms like 4chan facilitate subcultures sporting transgressive behaviour, but they don’t create it. In her research, Whitney Phillips describes it as trolling, varying from pranks to distressing anonymous attacks on non-anonymous individuals, often vulnerable minorities. They are attacked for their affects from a privileged position of asymmetry. This is rewarded with “lulz,” joyless laughter triggered by the victim’s reaction to the attack. Through anonymity, the troll is emotionally detached and does not take trolling seriously–everything online seems unreal. Mitigating even serious harm, the conditions are decided by the troll–which “justifies the attack. If the target hadn’t been so oversensitive […], he or she wouldn’t have been trolled; therefore it’s the target’s fault” (Phillips 2015, 97).

Milner agrees, stating that trolling is often conducted through memes, making it hard to distinguish “extremism from satire of extremism” (Milner 2016, 142). But even irony “can perpetuate inequality just as much as earnest racism and sexism can” (Ibid., 184). Further, memes mirror society: “If racist cultural associations did not exist, these images would not resonate” (Ibid., 123). For Phillips, trolling emphasizes dominance and generalizations and fit the trope of “white supremacy” (2015, 42). But ideologies are not created online. “Condemning these symptoms without addressing their ideological roots is unlikely to yield meaningful and truly transformative answers […];” without denying troll’s agency, one has to consider that mainstream media discourses, too, have to be held responsible for spreading hate (Ibid., 158f). To understand contemporary (American) media, studying the content made by trolls helps extract cultural information “and in fact can be quite predictive of approaching trends” (Ibid., 135).

The alt-right was thriving both on these boards and the sensationalist mediated landscape.

Exhibit 2: A year later on multiple channels

In a heated presidential race, the rise of the alt-right did not escape Trump’s opponent Hillary Clinton. At a rally in Reno, Nevada on 25 August 2016, she took the stage and condemned Trump for “ taking hate groups mainstream.”

As Clinton is speaking, a (now externally archived) thread appeared on /pol/.  The OP of the thread, recognizable as an American by the flag on top of the post (/pol/ allocates users ID’s to identify posts in a thread made from the same individual, as well as a flag identifying the country the    individual posts from), announced: “At a hill rally in the top left corner of the balcony. Look for me!”.Of significance is the photograph attached to the post: its background shows the venue – proving by custom that they are, indeed, there – while the front is dominated by a paper printout showing Pepe.

The infamous OP, 25. August 2016

The thread gained immediate attention. Mirroring the general interest of /pol/’s users, Anons, many of them appearing to watch the speech’s official broadcast, suggested the OP disrupt the event. While Clinton spoke about Trump’s agenda, 4chan’s excitement about this opportunity reads like a classroom situation: the antipathy against the candidate is as obvious as the support for her opponent and everyone is holding their breath in silent anticipation of the prank.

While the anonymous crowd was rooting for the OP online and Clinton’s speech continued in real life, a third channel opened, this time by an individual who discloses his identity:  Sean Lewandowski tweeted a link to his profile on the livestreaming platform Periscope, where he was broadcasting the rally in real time from his phone. Clinton goes on: “These are racist ideas, race-baiting ideas, anti-Muslim, anti- immigrant, anti-woman, all key tenants making up an emerging racist ideology known as the alt-right.” Taking a dramatic break to let the words sink in, the online channels erupt as a scream echoes through the venue.


screenshot video
A timed juxtaposition of the three channels

Lewandowsky’s camera shook upon him yelling. And while Clinton unperturbedly picks up her speech, 4chan’s thread displays a deafening storm of replies (“DUDE YOU JUST YELLED PEPE DIDN’T YOU. DIDN’T YOU!;” “HE SHOUTED PEPE THE FUCKING MADMAN” “I llove [sic!] you anon!”). The bully–and by now all signs point towards Lewandowski being the OP–had succeeded, the energy, the empowerment of being the in-group, and the glory of trolling is felt and eternalised on the official, mainstream-media accounts of the speech. Lewandowski is being escorted outside, and it is easy to imagine him grinning at the crowd a last time.

The interruption shows clear signs of harmful trolling: the lulz harvested from the community, the exploitation of Clinton’s affect, picked up when infamous  Breitbart interviewed Lewandowski and legitimized his behaviour, calling the speech “low energy” and “deceptive in her attempts at branding a broad movement as an evil, racist terrorist group”. More importantly though, they define the yelling-Pepe-incident (whom they called alt-right’s mascot) as “fun.” This aggressively mitigates the effect the interruption had.

The incident shows the scale Pepe had reached by then, appearing on three platforms simultaneously, invoked by different actors pursuing different agendas: Clinton’s event being physically infiltrated by Lewandowsky, who circulated the speech online, and disembodied 4chan, detachedly present. Borders are transgressed. A polemic warfare is waged online, not only between politicians, but with anonymous actors for whom memes function as a “trojan horse,” when they become visible: “Anyone even remotely connected to 4chan (or online culture generally) immediately knew that trolling had been afoot” (Phillips 2015, 66f).

A meme as weapon

Another combat could not come quick enough: On September 11, Donald Trump’s eldest son raised a controversy by instagramming a photoshopped image of the action-movie “The Expendables” on Instagram. The image, captioned “THE DEPLORABLES” by an unknown creator, depicts himself, his father, a group of voices of the alt-right movement–and Pepe the Frog. It reacts to a remark Clinton had made not much earlier: “[…] you could put half of Trump’s supporters into what I call the basket of deplorables. Right? The racist, sexist, homophobic, xenophobic, Islamophobic — you name it.”  The “deplorables”, though, reclaimed the term, creating another branch of the increasingly reactionary Pepe-meme that was, again, widely acclaimed and shared, as the image demonstrates. This resonates with Shifman, according to whom “we live in an era driven by hypermemetic logic, in which almost every major public event sprouts a stream of memes” (Shifman 2014, 4).

“The Deplorables” were followed up a day later by posting an article titled Donald Trump, Pepe the frog, and white supremacists: an explainer on  Clinton’s official website, in which Pepe is described as “a symbol associated with white supremacy,” recognized by Trump through his tweet a year earlier. With the article, Pepe was now officially endorsed as a Trump-mascot and Google Trends peaks for “The Deplorables” and “Pepe the Frog.”

pepe deplorables
Peaks for “Pepe the Frog” and “the deplorables,” looking at the time between May 2016 to April 2017

How had a comic figure, now acknowledged by both candidates as associated with the right, become so politically laden while still being traceable to the original image?

The media at war

Most certainly, online-media were now, right after the event, anxious getting an immediate statement from Furie. In many of these accounts, does not seem to have changed his indifference about the memetic status: in an  interview on 13 September, he sees it as a joke, but also acknowledges Pepe’s elasticity and him being out of his control. In The Washington Post, he dismisses his swinging right as a phase and distances the alt-right-meme-Pepe (made inappropriate and therefore “theirs alone to use”) from his’, the author, worries that “[a] Pepe avatar on Twitter is now generally read as a sign of an alt-right affiliation.”  In The Atlantic, Furie calls Pepe’s meme-ness inspiring and states that the alt-right association was “a reflection of the world at large.” Miles Klee, in The Daily Dot, claims that the meme has been co-opted by the alt-right, but relies Furie’s claim that pepe is not a white supremacist but inherently good and unpolitical and belonging to everyone. 4chan, now being ruled by neo-Nazis, as Olivia Nuzzi had also stated  is, according to him, likely to adapt memes to any politically incorrect cause anyway.

Elsewhere, on  Vice’s Motherboard, Jason Koebler corresponds with Ryan Milner, according to whom the meme is not being racist per se, but since the alt-right is using it, the connotation is certain. Whitney Phillips forebodes that Clinton’s article would keep the radicals in the news without effort, all being part of their game. Koebler acknowledges memes changing meanings like symbols and eventually predicts that Clinton’s explainer is a “self-fulfilling meme prophecy. If Clinton says a meme is associated with white supremacy, surely it soon will be. This prediction eventually proved right–the media had taken the bait, unintentionally placed by Clinton, or, when looking back at Nuzzi’s article, precisely not:

Trying to connect to either of Nuzzi’s sources that had revealed plans of the alt-right to reclaim Pepe from the “normies”, is a dead end–both of the accounts have since been deleted. Further, on 14 September, right after Clinton’s explainer, that even quoted part of @JaredTSwift’s account in Nuzzi’s article, the right-leaning news-site The Daily Caller  claimed that the “panic is more or less a complete troll job,” Nuzzi, like many other journalists, had been “fed a narrative–there had never been a plot to make Pepe a white nationalist.

Pepe depicted on the ADL database, 2016

Notwithstanding the political ideologies behind either of the journalists, the practise described here is textbook for Whitney Phillips: the deception of the media is classic trolling behaviour that attempts “to turn the media against itself” by feeding it stories, engineered on truth and lies, that are semi-verifiable online, yet too sensational to not publish them. The media deliver the desired “exposure and laughs” to the trolls, while benefitting by the attention paid to their story, almost in symbiosis. It is nonetheless important to not disregard these incidents as pranks, because that would “overlook the fact that the trolls knew exactly how to manipulate the news cycle, and in the process forwarded an implicit critique of the ways in which media research and report the news.” (Phillips 2015, 5f) But not only had trolling behaviours been appropriated, proving the power of them–the narratives they had been working with are equally powerful.

The moral panic around Pepe had yet to reach another peak: Not long after the Clinton-post, on 27 September 2016, Pepe, or mainly his alt-right manifestation, then as an image of the Frog with a Hitler-moustache was added to the  database of hate-symbols compiled by the Anti-Defamation-League, monitoring “ those symbols that serve as potential calling cards of extremists and anti-Semites,” for having been appropriated and abused by right-wing  extremists. This official call triggered another wave of media coverage, including the Guardian, Independent and Huffington Post , but also foreign media like the German Tagesschau and Die Zeit.

Subsequently, with the mainstream-media coverage, Pepe reaches maximal visibility, climaxing on Google Trends.

peak pepe
Peak in search interest, looking at the time from May 2016 to April 2017
#savepepe, 2016

Could Pepe, now widely a symbol for right-wing forces, be reclaimed and acquire yet another meaning? After months of indifference, Furie, probably feeling responsible, teamed up with the ADL and launched a social-media-campaign to #savepepe by circulating positive messages, calling the alt-right appropriation a “nightmare,” and bringing back “the light hearted spirit of the original chilled-out champion.” In October, his publisher, Fantagraphics had issued a statement declaring the use of Pepe by the alt-right copyright-infringements damaging Furie, and had tweeted an image by Furie earlier that month, showing Pepe urinating on the Trump-Pepe.

Mainstream-media covered the campaign (including Matt Furie on TIME, the Guardian and The New York Times), but the association stuck, regardless how visibly Furie distanced his Pepe from Nazi-Pepe. He had been right when he said that Pepe was out of his control; his authorship had been replaced with collective ownership, if Milner (2015, 15) were asked. Andrew Keen warns that when the audience is the author in a time “where authenticity is almost impossible to verify, the idea of original authorship and intellectual property has been seriously compromised” (Keen 2010, 23).

Meme magic and collective belief

With Trump’s election on 8 November, something else started to seep into public consciousness: the claim that a practise called meme magic had made him president.

Meme Magic, 2015-16

In late 2015, Pepe had been declared a deity of “meme magic” , a fictional concept involving online communities finding links between random events and making the memes they posted responsible for these events. On 4chan, for example, posts become magically charged and predict the future if the unique ID every post is automatically assigned shows subsequent numbers (referred to as GET). On /pol/, reactionary posters began to claim that their meme magic, often involving Pepe, would eventually make Trump president. Needless to say, that Trump’s Pepe-tweet fed their narrative. This conspiracy-theory-engineering was intensifying, when a satirical religion around the  Egyptian frog-headed deity Kek began to circulate in 2015–making Pepe “a modern avatar” of this deity and during the presidential race, the words “praise kek” were often found on the boards (a notable incidence is Lewandowsky’s thread, where the word kek appears 72 times). A climatic moment was a post reading ‘Trump will win’ getting the 7777777 GET.

7777777 Trump will win, 2016

Clinton’s explainer post demonizing Pepe as a Nazi-symbol had spawned a parody-page titled THE REAL STORY BEHIND HILLARY CLINTON’S “CARTOON NAZI FROG” WILL BLOW YOUR MIND, emulating the style of the post, but with meme magic mixed in, eventually prompting people not to “vote”, as in the original post, but to “meme”. The meme had evolved once again.

this is awesome meme vote.png
Left: Parody, right: Clinton’s explainer, both 2016

Media discussions around the appealing narrative around meme magic surged. Paul Spencer on Motherboard compares this in an interview with occult researcher Théodore Ferréol  with internet-folklore. Meme magic apparently employed actual occult traditions that involve symbol being charged and then circulated to raise visibility to influence human consciousness. God-Pepe had turned from a recognized hate symbol into a magic spell, enforced by the candidates’ legitimations, showing again the elasticity of the meme, an oracle predicting the outcome of the elections and accredited with Trumps victory.

Regardless of whether Pepe influenced the elections, the fact that it is discussed, and the scale it reached confirms the power of the narrative. Pepe had become real, while the question whether the alt-right believed their own story is secondary. They were practising the concept of hyperstition, which is–according to Anna Greenspan on a  blog started in the early 2000’s dedicated to exploring this concept–“recognizing a fiction’s effectiveness, using it and still not believing it;” they realized that their nazi-Pepe “has the ability to produce affects, create concepts and transmit signal. Yet his power to do these things makes it impossible to disbelieve in him” (Greenspan 2004). The original character, as an essential part of early online otaku culture, was appropriated and mobilized to convey a message, making Pepe a hyperstitional carrier. The media picked up on that and circulated the message, becoming complicit, while the alt-right rode the tide of the meme, regardless of them believing in Trump or magic, but recognizing his affective power as a symbol. Opposed to general belief, emotion has not been eradicated in politics.

Glorious media

This power of the affect is likewise recognized by Giorgio Agamben in his 2011’s The Kingdom And The Glory, in which he examines the nature of power in the Christian West, divided into governmental power, and the power of the Glory. The latter is a mysterious force materializing in ceremonies, acclamation and belief; it enables stability in a politics where executive power is dependent on acclamation, which paradoxically paralyzes it, makes it inoperative. Still, the glory captures this inoperativity, and ensures the governmental machine’s running, while its throne is empty.

Performative acclamation is an expression of “praise, of triumph […], of laudation or of disapproval […] yelled by a crowd in determinate circumstances” (Agamben 2011, 169); showing “[…]the pure and immediate expression of the people as constituent of democratic power” (ibid., 171). The role of acclamation has endured secularization through evolution, now filling the void left by the decline of theology as public opinion. Simplified acclamation therefore still holds power in modern democracies; “If this is true, the problem of the political function of the media in contemporary society that is so widely debated today acquires a new meaning and a new urgency” (ibid., 255). For Agamben, if “the media are so important in modern democracies, this is the case not only because they enable the control and government of public opinion, but also and above all because they manage and dispense Glory, the acclamative and doxological aspect of power that seemed to have disappeared in modernity” (ibid., xii). We are still a society of the spectacle.

With Pepe, the media played a significant role in dispersing the meme, in scaling it up–from imageboards over social-media and mainstream press. They provided a narrative to create affect, and even, in a way, controlled our feelings, giving us a power to collectively believe in.

Trump as troll?

Dale Beran’s article The Skeleton Key to the Rise of Trump delivers a narrative on why Pepe was appropriated by the alt-right, covering 4chan’s meme-making and trolling. The community, mostly socially reclusive men, cloak themselves in anonymity, yet ironically celebrate their disempowerment and shame–feelings that were met by Trump, a “man obsessed with “losers” who seemingly was going to be remembered as one of the biggest losers in history—until he won.” Pepe symbolizes this sentiment, an ideology about “embracing your loserdom, owning it.” Beran argues that this system where hopelessness is normalized ended in voting for Trump as another “spiteful prank;” Trump is equally 4chan and Pepe.

Trump-as-troll is met with contempt by other voices: in July 2016, Whitney Phillips had  already stated that calling Trump a troll was is dangerous and mitigated his harmful rhetoric, victim-shaming since trolling is but an internet-game that the target could always escape by logging off. No matter how hilarious Trumps campaign was, it was not trolling. As it is never certain whether the sender is being ironic on the internet, we need instead to focus on the “content of a particular message and the impact it has on its audience,” and might want to consider the media’s role in Trump’s ascend, and the cultural landscape that he is part of. Phillips and some colleagues answered also to the later emerging discussions around meme magic, challenging Beran’s article’s narrative: despite its appeal and strength, 4chan and trolling was not the deciding factor in Trump’s victory, although It did raise the ideology’s visibility. But ultimately, “[…] the biggest media story to emerge from the 2016 election was the degree to which far-right media were able to set the narrative agenda for mainstream media outlets.” What is referred to as pro-Trump-trolling is but a symptom of a surge in white nationalism and diminishing trust in objective truth and journalism.

Andrew Keen sees this as problematic: Not questioning online content, which is often anonymous and amateurish, results in a decline of professional news-telling and quality content. For him, media should provide information, and not “become an agent of social engineering” (Keen 2010, vii). Ultimately, the mob chooses the truth by believing in it, the media on the other hand controls this belief. There might not be consent about Trumps actions, but there is a belief in Trumps victory narrated through a meme–the Glory has been evoked.

The voices mirror the problems of underestimating trolls and publicly overestimating the power of memes, leading into actually believing the narrative around Pepe. But regardless whether Pepe had actually been a vessel of the alt-right to influence the elections or not, the fact that there are discussions around it is signficant. Whether we call the alt-right’s behaviour trolling or not, they use the mechanisms of trolling that prove highly successful. The image is blowing up and believed to have power that exceeds what images are thought to have. Meme magic, hyperstition–collective belief might be more powerful than the truth.

From an unassuming cartoon frog sucked into the depths of an obscure imageboard, becoming a medium to communicate feelings on social-media, and ending as being seen as a vessel for alt-right propaganda on mainstream media and exiting the digital realm, the concept of scale attached to an image collapses in the memetic context of an in its meaning versatile image. The symbol is open-ended.

Epilogue: 2017

Photo by the author, early 2017

Early this year, a few stickers showing Pepe appeared on the railings along Regents Canal just north of Shoreditch, not quite blending in with the usual mass of stickers advertising bands and activist collectives. What did the appearance of Pepe in London mean for this research; for inducted sympathisers of nationalism; for the unknowing passer-by? Is Pepe being overestimated, or is this yet another step towards Pepe making himself real?

Zara Skirt, 2017

In late April, the global fashion company Zara, released a denim skirt that had gained attention on social media, especially Twitter, over showing two images of a green frog. Regardlessof whether the link was intended or not, which Zara strongly disputes (the image does not bear much resemblance to Pepe), the fact that the connection was drawn again shows the scale Pepe has reached over the last ten years.

And then, Pepe died. On May 6, a comic drawn by Matt Furie is published on the occasion of the annual Free Comic Book Day. Pepe is depicted lying in a coffin. It is his funeral.

Pepe had become too dangerous, had gone haywire, even. Having a life of his own, absolutely unrelated to what Furie had ever wanted for him, he had hung out with the wrong people, until he eventually became one of them. #savepepe had failed, so Furie eventually terminated the frog. (He claims that he had made this decision when Pepe became a meme, though, but also acknowledges that “These trolls, or whatever you wanna call them, they’re kinda like the loudest voice on the Internet.”

But will officially killing Pepe eradicate the alt-right association? The trajectory of the meme that I have drawn here casts doubts. Having become a meme, the author has been made redundant – an image having become a meme cannot be reclaimed or ended by a single directed action. If anything, it brings about a new surge in visibility (Google trends shows another peak in search interest as mainstream media is picking up this issue).

pepe last
Search interest for “pepe the frog” during the last year

Furie’s Pepe might be dead, but Pepe, the meme, will certainly live on, since it is unlikely that Furie’s intellectual ownership of his comic will affect a meme that is authored collectively. But see for yourself.



Agamben, Giorgio. The Kingdom and The Glory: For a Theological Genealogy of Economy and Government. Translated by Lorenzo Chiesa. Stanford: Stanford University Press, 2011.

Andrew Keen, The Cult of the Amateur: How blogs, MySpace, Youtube and the rest of today’s user generated-media are killing our culture and economy (London: Nicholas Brealey Publishing, 2010), PAGE.

Azuma, Hiroki. “Database Animals.” In Fandom Unbound: Otaku Culture in a Connected World, edited by Mizuko Ito, Daisuke Okabe and Izumi Tsuji, 30-67. New Haven & London: Yale University Press, 2012.

Bernstein, Michael S., Andrés Monroy-Hernández, Drew Harry, Paul André, Katrina Panovich and Greg Vargas. “4chan and /b/: An Analysis of Anonymity and Ephemerality in a Large Online Community.” Paper presented at the Fifth International AAAI conference on Weblogs and Social Media, Barcelona, July 17-21, 2011.

Greenspan, Anna “Hyperstitional Carriers.” Hyperstition blog, 26 July 2004. Accessed 20 April 2017.

Greenspan, Anna “On Unbelief (A Beginning).” Hyperstition blog, 27 June 2004. Accessed 20 April 2017.

Hollenbaugh, Erin E. and Marcia K. Everett. “The Effects of Anonymity on Self-Disclosure in Blogs: An Application of the Online Disinhibition Effect.” Journal of Computer-Mediated Communication 18 (2013) 283–302. Accessed April 5 2017. doi:10.1111/jcc4.1200.

Ito, Mizuko. Introduction to Fandom Unbound: Otaku Culture in a Connected World. Edited by Mizuko Ito, Daisuke Okabe and Izumi Tsuji, xi-xxxi. New Haven & London: Yale University Press, 2012.

Keen, Andrew. The Cult of the Amateur: How blogs, MySpace, Youtube and the rest of today’s user generated-media are killing our culture and economy. London: Nicholas Brealey Publishing, 2010.

Kitada, Akihiro. “Japan’s Cynical Nationalism.” In Fandom Unbound: Otaku Culture in a Connected World, edited by Mizuko Ito, Daisuke Okabe and Izumi Tsuji, 68-84. New Haven & London: Yale University Press, 2012.

Metahaven. Can Jokes bring down Governments?: Memes, Design and Politics. Moskow: Strelka Press, 2013.

Milner, Ryan M. The World Made Meme: Public Conversations and Participatory Media. Cambridge & London: The MIT Press, 2016.

Olson, Parmy. We are Anonymous: Inside the Hacker World of LulzSec, Anonymous and the Global Cyber Insurgency, London: William Heinemann, 2013.

Phillips, Whitney. This Is Why We Can’t Have Nice Things: Mapping the Relationship between Online Trolling and Mainstream Culture. Cambridge & London: The MIT Press, 2015.

Shifman, Limor. “Memes in a Digital World: Reconciling with a Conceptual Troublemaker.” Journal of Computer-Mediated Communication 18 (2013): 362–377. Accessed 10 April, 2017. doi:10.1111/jcc4.12013.

Shifman, Limor. Memes in Digital Culture. Cambridge, London, The MIT Press, 2014.

Steyerl, Hito. “In Defense of the Poor Image.” e-flux 10 (2009). Accessed 9 April, 2017.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s