Friday, February 1, 2019

On humanity, videogames, and resisting operationalized logic.

Loyola University, New Orleans

In October 2018, just prior to the November midterm elections, Twitter banned close to 1,500 accounts that all featured the same gray, expressionless cartoon avatar. “NPC,” a version of the Wojack or Feels Guy reaction image, is a meme generated by the right-wing internet as a representation of liberals. Accounts featuring NPC as an avatar, Twitter alleged, had violated Twitter’s terms of service when, while mockingly impersonating liberals, they spread “intentionally misleading election-related content,” such as the incorrect election date.

The name of this icon—“NPC”—is a reference from videogame culture. Shorthand for “non-playable character” or “non-player character,” an “NPC” is any character a human user does not control. Without a user to direct their movements, dialogue, and interactions, the NPCs are directed by the game’s programming. Depending on the specific character and the type of game, an NPC can be as simple as a still image next to text bubble dialogue or as complex as a photo-realistic, voice-acted, three-dimensional model guided by adaptive artificial intelligence that reads, reacts to, and learns from the player’s interactions with the gamespace. Some NPCs have well-drawn personalities and players can get to know them over the course of multiple hours of gameplay. Most, though, are the equivalent of movie extras. They mill around town in a simulation of daily life to make the virtual world seem vibrant. They fall in with the attacking enemy horde only to be dispatched unceremoniously by the heroic player character.

When NPCs aren’t important to the story, they usually aren’t designed for sustained player interaction and attention. Their animations are on short loops, their movements follow set, predetermined paths, their dialogue is scripted and repeats frequently. If the player lingers with them for a moment, the veil of immersion breaks down and their computer-driven behavior reveals itself.

It’s this quality of the NPCs—their pre-programming—that became the seed of the right-wing meme. The insult, summarized by Kevin Roose of the New York Times, is that liberals are “programmed” to repeat the same limited dialogue and action scripts, like NPCs. They are thus unresponsive to the assumed sound, logical arguments of the right and therefore incapable of independent thought.

Beyond characterizing liberals as poor interlocutors, however, as many commentators like Cecilia D'Anastasio of the videogame culture blog Kotaku observe, the NPC insult is dehumanizing and potentially threatening. Given all manner of virtual suffering and death some videogames allow players to dispassionately inflict on these characters with no human behind them, coupled with the right-wing’s tendency to dehumanize and suggest violence against political rivals and minorities, the potential harm of the meme goes beyond spreading misinformation on social media.

At the same time, there are a few fundamental ironies at play in the NPC meme. I want to set aside for now the most obvious irony of using a meme to criticize opponents for repeating received arguments, but that’s part of this phenomenon as well. Instead, I’m interested in how the meme misunderstands our relationship with digital characters and environments. In brief, the NPC meme describes computer-driven, non-playable characters as incapable of responding to logical arguments, when in reality the computer code comprising an NPC is itself nothing but a logical argument parsed by a computer.

At a material level, NPCs are a bundle of scripts called to perform when the game state meets certain circumstances. The scripts may not be very complicated, depending on how important it might be to the game. Nonetheless, they are basically just a string of “if this, then that” functions. For example, if the player selects dialogue option “A,” then, the NPCs plays dialogue line “X,” which had been designated as the response to option “A:” if input A, play line X. That means, literally the only time NPCs change their behavior is when presented with new logical argument, such as when a programmer or modder injects new bits of code.

It is not just the NPC who is limited by programming. The playable character similarly has only a particular set of available animations, dialog, and abilities. It receives input from the player and so can engage the virtual environment as the player chooses, but only through the range of affordances defined by the game’s if-then statements. For example, no matter how much a player of Grand Theft Auto: San Andreas might want the protagonist CJ to go around hugging pedestrians, it simply isn’t an available in-game action. There is no changing the player character’s mind, perspective, or responses either. He—and it is almost always “he”—may tend to have more compelling reactions to the player’s input than the NPCs, but, without rewriting the game’s code, it too can only express what it was programmed to express.

The NPC and PC both exist within a protocological environment and can only ever enact the environment’s programmed logics. Down to the level of computer programming, protocol allows for creative operation of the functions, but not breaking protocol. As Alexander Galloway explains in Gaming: Essays on Algorithmic Culture, it is this flexibility that enables the digital system to maintain stability. In this sense, both PC and NPC function within the same operational logic, neither capable of differentiated response to player inputs regardless of who pushes the buttons or their political and philosophical leanings. 

Of course, these right-wing memesters are not considering the how videogames participate in the functioning of power in a control society. The NPC insult is simply supposed to delegitimatize and dehumanizes people who are seen to be uncritically repeating arguments and buzzwords by, ironically, reproducing and spreading a meme. But this irony is precisely the trick of protocol that makes it so effective. It offers an agency that appears like self-determination, but within constrained, consistent, repeatable perimeters, much like memes themselves.

Logical argumentation is presumed to be an alternative to NPC-hood, when the logic of protocol is what makes possible the NPC in the first place. Videogames invite players to take on an objectivized logic—everything is permitted, all is expendable, as long as the mission is completed—which in its very design devalues not only the NPCs but also the player’s own interiority in order to function universally. We have to ask, then, is this kind of cold rationality the avenue to human dignity and independence as the meme supposes or a means to its suppression?

The same week these stories about NPCs came out, the President of the United States equivocated over the killing of a journalist allegedly by Saudi Arabia. He argued the life of one non-US citizen isn’t as valuable as maintaining a lucrative arms deal. The logic of this argument relies on the dehumanizing assumption that human life can be weighed against a trade agreement, that it has a calculable, comparable exchange value.

This gross utilitarianism in international politics is not new, of course. The so-called “virtuous war” military strategy defines success as reducing the number of casualties suffered by one's own side. While reducing casualties is an admirable goal, it also treats human life as a benchmark statistic, sets parameters for “acceptable” losses and “high value” targets, and removes the human cost from combat. It can even justify more warfare and more casualties. After all, if the lives of one’s own soldiers aren’t at risk, there may seem to be less to lose from a military response.

The virtuous war is exemplified by the expansion of war-at-a-distance technologies like drones, which allow the US to strike targets without exposing soldiers to the battlefield. A 2010 United Nations report on targeted killings, however, warned against pilots potentially assuming a “playstation mentality.” The suggestion characterizes piloting a drone as akin to playing a videogame. The two technologies—drones and videogames—share a development lineage (some drones are even piloted using controllers similar to those sold with the Xbox360). The concern raised by the UN here is that the drone pilot might assume a similar attitude toward the silhouettes on their targeting display as they might take toward NPCs in a typical videogame.

In Mixed Realism: Videogames and the Violence of Fiction, I explain that NPCs are so expendable that games often have to inject them with human value simply so that the plot doesn’t fall apart and fictional events happening to virtual characters have a sense of meaning, gravity, and consequence. The Call of Duty series, for instance, which just saw the release of its fifteenth installment, often does this by having players look an NPC in the eyes. Call of Duty: Modern Warfare 2 uses this technique numerous times, even having previously unrecognizable NPCs gaze directly into the first-person camera just before they are dispatched. Despite the fact that the player will shoot thousands of identical NPCs over the course of the game, looking one in the eyes as they die lends it a degree of humanity.

I optimistically suggest that our capacity to see human value in NPCs, to feel anything when we dispatch an arrangement of pixels, is an opportunity to resist operationalized logic inherent in digital games. Doing so requires a different mentality toward play, though, one that sets aside the strategic efficiency implied by game designs and incentives in order to pursue other values such as sympathy and community, ambiguity and complexity, or revelation and awe.

The NPC meme may stick around for a while or it may go the way of so much other internet detritus. Regardless, the dehumanizing calculus at its core endures. The contemporary manifestation has found verdant soil in digitally mediated cultures, spreading on social media sites and in videogame communities. Responding to it will likely require not more logical arguments, but the ability to recognize the human in mere digital representations.


Timothy J. Welsh is author of Mixed Realism: Videogames and the Violence of Fiction. Welsh is assistant professor of English at Loyola University, New Orleans.

"In Mixed Realism, Timothy J. Welsh proposes a fresh approach to understanding digital games and contemporary literature that is essential, relevant, and engaging."
—Zach Whalen, University of Mary Washington



Crogan, Patrick. Gameplay Mode. Minneapolis: University of Minnesota Press, 2011.

D’Anastasio, Cecilia. “How The ‘NPC’ Meme Tries To Dehumanize ‘SJWs.’” Kotaku (blog), October 5, 2018.

Der Derian, James. Virtuous War: Mapping the Military-Industrial-Media-Entertainment Network. Boulder, CO: Westview Press, 2001.

Galloway, Alexander R. Gaming: Essays on Algorithmic Culture. Minneapolis: University of Minnesota Press, 2006.

Infinity Ward. Call of Duty: Modern Warfare 2. Activision, 2009.

Rockstar North. Grand Theft Auto: San Andreas. Rockstar Games, 2006.

Roose, Kevin. “What Is NPC, the Pro-Trump Internet’s New Favorite Insult?” The New York Times, October 19, 2018, sec. U.S.

Triple Zed and Y.F. “Wojak / Feels Guy.” Know Your Meme. Accessed January 21, 2019.

Wagner, John. “Trump Says Curbing Arms Sales to Saudi Arabia in Response to Missing Journalist Is ‘Not Acceptable.’” Washington Post, October 11, 2018, sec. Politics.

Welsh, Timothy J. Mixed Realism: Videogames and the Violence of Fiction. Minneapolis: Univ Of Minnesota Press, 2016.

Friday, January 25, 2019

Fashioning Feminism: On Bodies of Information.

William & Mary

What does a bulletproof dress prototype have to do with the digital humanities?

A lot actually, according to artist micha cárdenas. Such a garment, which was crafted from Kevlar airbags scavenged from a junkyard, could be capable of stopping a 9mm bullet. It’s one of the objects featured in the latest addition to the Debates in the Digital Humanities series, Bodies of Information: Intersectional Feminism and Digital Humanities.

As a piece of apparel, the dress dramatizes the higher risk of mortality that people of color face in confrontations with law enforcement. Of course, communities allied with #BlackLivesMatter are also deploying statistics, metadata (like hashtags), and even information visualizations to quantify how the inequities of state power do violence to black and brown bodies, as well as how activists can mobilize in response. Nonetheless, the metallic clothing created by cárdenas represents a critical kind of “embodied gesture” that she argues is as essential as big data number crunching, if not more so.

Others in this new collection, such as Marcia Chatelain – creator of the #Ferguson Syllabus – and Beth Coleman of the City as Platform Lab, similarly make the argument that #BLM should present central rather than peripheral concerns for digital humanities practitioners in the academy.

Furthermore, digital humanities scholars “can extend their work to be more accessible to low-income people,” cárdenas writes, “and to considerations of nondigital technologies, by abstracting the concept of algorithms to include recipes and rituals.”

Bringing DIY craftivism to the digital humanities is a commitment for Kim Brillante Knight as well. This scholar of “viral media” uses an unusual form of data visualization to depict the frequency of the use of the #prolife hashtag on Twitter. Rather than show a word cloud or network graph, Knight uses five pink LEDs as a meter to measure the occurrences of the relevant tweets.

“The medium for the visualization is a black T-shirt,” Knight explains, “onto which I have hand embroidered reproductive organs: a uterus, fallopian tubes, cervix, and part of a vagina.” The project also uses microcontroller technology and conductive thread.

Other evocative objects – such as yearbook photos – become artifacts of critical reflection in the new volume. Texas A&M professor Amy Earhart describes the unintended consequences of digital humanities projects that reveal sites of institutional shame. For example, she includes images from a project digitizing college memorabilia that reveal photos of student organizations with members proudly “wearing their Klan robes, with typical cross insignia, hoods, and brandishing swords.”

Rather than merely digitizing archives without reflecting on their design – who is included, what is excluded, and why some histories are deemed not worth preserving – this collection encourages digital humanities researchers to question what gets privileged in a library of rare materials and how digital archives can foster different perceptions of the historical record.

Brandeis medievalist Dorothy Kim, who has been a lightning rod for alt-right abuse online, invites us to consider what gets lost when we only experience the digital copy of a text. Kim notes that solely its visual elements are captured, and its other sensory features become lost. “Medieval reading practices were not linear,” Kim asserts, “often required vocality to read out loud or sing out loud, ideally required slow and repetitive rereading, were emotive, and involved sound, smell, touch, taste, visual, and even bodily calisthenics.”

The epistemological rethinking that digital technologies make possible is highlighted in many of the groundbreaking essays in the volume, including “Toward a Queer Digital Humanities” by Bonnie Ruberg, Jason Boyd, and James Howe.

Ruberg, Boyd, and Howe articulate basic principles: “If queer knowledge always resists completion, it becomes clear that queering metadata means more than adding new vocabulary to existing taxonomical systems. Queerness also points toward a shift in the very methodologies of metadata collection. To queer metadata, queer thinking must be brought to bear on the conceptual models and tools of object description as well as its content.”

The collection even includes Deb Verhoeven’s “Be More Than Binary” challenge to the international digital humanities community, as well as a number of essays that question what it means to speak of “community” in the digital humanities at all.

In emphasizing the importance of feminist digital humanities, this collection does much more than merely highlight digital archives that commemorate the previously hidden accomplishments of women. In addition to acknowledging transgender and nonbinary forms of digital humanities, these essays consider what is feminized as well as what is female. For example, Sharon Leon acknowledges the many professional roles that disprove the “Great Man” myth. And Julia Flanders encourages her audience to interrogate assumptions about all technical systems of knowledge production as they think about both print and digital publication processes. Flanders reflects upon how her own Women Writers Project “mirrored a shift in feminist theory from a second-wave attention to the visibility and rights of women . . . to a third-wave focus on how the structure of discourse enacts and reinforces cultural power dynamics of gender, race, class, coloniality, and other differentials.” There is also a wonderful essay by Susan Brown, who celebrated the 20th anniversary of Orlando recently, that deconstructs aversions to tropes of delivery and service associated with the “handmaiden” position in the digital humanities with an incisive reading of Atwood’s The Handmaid’s Tale.

As we performed our own informational labor as the editors/handmaids of this book – collating comments from the peer-to-peer review process or indexing the key terms in the volume – we found ourselves marveling at the sophistication of the feminist thinking modeled in this collection and the fundamental questions that it explored. Sadly a single blog post can’t do justice to the dazzling array of ideas in a table of contents that concludes with two essays about why videogame design and analysis of its player community practices might rightly belong with the growing corpus of digital humanities scholarship.

Readers are likely to appreciate how this book challenges existing attitudes and stereotypes about a rapidly expanding field. As an added benefit, with its affordable cover price and open access launch in a few months, Bodies of Information also offers a rich set of resources for students who are interested in exploring how digital technologies can promote activist scholarship, community alliances, and public engagement in the academy.


Elizabeth Losh is associate professor of English and American studies at The College of William & Mary with a specialization in new media ecologies. She is coeditor, with Jacqueline Wernimont, of Bodies of Information: Intersectional Feminism and Digital Humanities; author of Virtualpolitik and The War on Learning: Gaining Ground in the Digital University; and coauthor of Understanding Rhetoric: A Graphic Guide to Writing.

Friday, January 18, 2019

Frankenstein and anonymous authorship in eighteenth-century Britain.

University of Wisconsin–Madison

Having celebrated its 200th anniversary in 2018, Mary Shelley’s Frankenstein is perhaps one of the most well-known novels of the early nineteenth century. While many are familiar with Shelley’s classic novel and can immediately picture some version of the work’s iconic monster, few are aware that when Frankenstein was first published in 1818 it was an anonymous novel. Nowhere on the title page does the name “Mary Shelley” appear. The novel that began its life as an exercise in writing a ghost story during the particularly cold and wet summer of 1816 at Lord Byron’s Villa Diodati in Geneva was conveyed to the British reading public with no indication of its author’s name.

Indeed, when in 1817 Mary Shelley’s husband Percy tried to help her sell the completed novel to his publisher, Charles Ollier, he did so without disclosing the author’s name. Percy Shelley wrote to Ollier: “I send you with this letter a manuscript which has been consigned to my care by a friend in whom I feel considerable interest.” The manuscript was rejected. Acting again as Mary Shelley’s agent, Percy Shelley would eventually find a publisher for the novel in Lackington and Co., but Mary Shelley’s name was withheld. Percy Shelley referred to the novel in a letter to the publisher as “not [his] own production, but that of a friend…”

Shortly after the novel was published in January 1818, with a print run of 500 copies, reviews of the novel began appearing in periodicals. Some reviewers, noting novel’s anonymity, hazarded an attribution. Walter Scott in Blackwood’s Edinburgh Magazine wrote: “It is said to be written by Mr Percy Bysshe Shelley, who, if we are rightly informed, is son-in-law to Mr Godwin; and it is inscribed to that ingenious author.” (Mary Shelley would later write to Scott to correct this error). An anonymous reviewer in The Literary Panorama, and National Register reported in its review of Frankenstein: “We have heard that this work is written by Mr. Shelley; but should be disposed to attribute it to even a less experienced writer than he is. In fact we have some idea that it is the production of a daughter of a celebrated living novelist.” The British Critic was even crueler in its dismissal:

The writer of it is, we understand, a female; this is an aggravation of that which is the prevailing fault of the novel; but if our authoress can forget the gentleness of her sex, it is no reason why we should; and we shall therefore dismiss the novel without further comment.

Given the harsh reviews Frankenstein endured from many, though not all, critics and the obvious antipathy to women writers held by some critics we might not be surprised that the novel was published anonymously. Indeed, we may wish to attribute some causal relationship between the expected reception of Frankenstein and its woman author and the decision to publish the novel anonymously. As Susan Eilenberg notes, however, “there was nothing peculiarly feminine about anonymity, nor anything very uncommon about it, either.”

Indeed, the novel in English emerges over the course of the long eighteenth century as a largely anonymous form. As James Raven asserts, “it is clear that the overwhelming majority of the English novels of the eighteenth and early nineteenth centuries were published without attribution of authorship on the title page or within the preface or elsewhere in the text.” In 1818, the year Frankenstein appeared, 62 new novels were published in Britain and Ireland. 41 were published anonymously (66%). Of the 21 novels that appeared with their authors’ names attached, five were attributed to male authors and 16 to female authors. These figures upend assumptions we might make about gendered authorship as well has how atypical authorial anonymity was for the novel.

In Everywhere and Nowhere: Anonymity and Mediation in Eighteenth-Century Britain, I argue that because anonymity was typical of texts published (and performed) in the long eighteenth-century, we must rethink both how we approach anonymous texts and how we attribute motives to authors to account for that anonymity. I suggest we move from approaching anonymity as a product of an individual author’s choice to understanding it as an aspect of textual production. We tend to assume that anonymity is a choice made by an author and that named authorship is the default state. Publication history, however, suggests otherwise – anonymity was the default state, particularly for new novels like Frankenstein. We might understand the anonymity of a novel like Frankenstein along the lines that we understand Lackington and Co. issuing the novel in three volumes – the typical physical form of novels in the late eighteenth and early nineteenth centuries. That is, the anonymity of Frankenstein (and those other 40 novels published anonymously in 1818), while informed by the author’s choice and individual motives, is as much a product of the forces of generic expectation, publication practices, and the collective actions that bring a literary text to be in the world.


Mark Vareschi is author of Everywhere and Nowhere: Anonymity and Mediation in Eighteenth-Century Britain. Vareschi is assistant professor of English at the University of Wisconsin–Madison.

"This is fresh, compelling, detail-rich scholarship and essential reading."
—Brad Pasanek, author of Metaphors of Mind: An Eighteenth-Century Dictionary

"Everywhere and Nowhere is that rare thing: a genuinely interdisciplinary study, capacious and illuminating, of how anonymous authorship impacts meaning across genres and media. In Mark Vareschi’s hands, anonymity is transformed into a lens for reexamining the most fundamental literary concepts (authorship and intention, medium, textuality) and renovating them—not just in the domain of print but across the rich media ecologies of the eighteenth century."
—Michael Gamer, University of Pennsylvania


-Percy Bysshe Shelley, The Letters of Percy Bysshe Shelley, ed. Frederick L. Jones, 2 vols. (Oxford: Clarendon Press, 1964), Vol. I, p. 549; p. 553.
-Blackwood's Edinburgh Magazine 2 (March 1818): 613-620
-The British Critic, N.S., 9 (April 1818): 432-38
-Susan Eilenberg, “Nothing’s Nameless: Mary Shelley’s Frankenstein,” in The Faces of Anonymity, ed. Robert J. Griffin (New York: Palgrave, 2003), 171.
Raven, 143.
-James Raven, “The Anonymous Novel in Britain and Ireland, 1750-1830,” in The Faces of Anonymity, ed. Robert J. Griffin (New York: Palgrave, 2003), 143; 164.