Auto/biographical Avatars

It is impossible to address the dynamics of voice and identity online without raising the issue of avatars. When users of digital technologies express themselves they frequently borrow others’ bodies, faces, and/or words to do so, employing celebrities, strangers, cartoon characters, animals, or inanimate objects as their mouthpieces or proxies. Such gestures can be seen as auto/biographical acts: autobiographical because these individuals are deploying mimicry, citation, and ventriloquism in order to convey their own life experiences; biographical because, in so doing, they help to shape the stories, “star images,” and public personae of the figures being mimicked, cited, or spoken through.

To some extent it has ever been thus, and theorists of auto/biography have been particularly attentive to the “range of relational others evoked and mobilized within life writing for the purposes of self-narrating and self-knowing.”¹ If Alison Light argues that biography has always had a “tendency toward ventriloquy,”² Sidonie Smith and Julia Watson go one further, insisting that in works of life writing “no ‘I’ speaks except as and through its others.”³ Provocative in the context of classical autobiography, this statement feels considerably less controversial when applied to online culture, whether it’s Facebook and Twitter users deploying memes and reaction GIFs to render their personal experiences more relatable, YouTubers role-playing as videogame Vikings, or anonymous “alt-right” trolls co-opting cartoon frogs as vehicles for hate speech. Meanwhile virtual Instagram influencers like lil Miquela and Shudu Gram proselytize tirelessly for the brands who fund them, staying on message in a way that flesh and blood counterparts like Essena O’Neil⁴ apparently cannot; chatbots pose as humans while “ghost workers”⁵ pretend to be AI software, performing menial tasks that it is still cheaper and easier to underpay real people to perform; “deepfake” videos take advantage of breakthroughs in machine learning and voice synthesis to translate Hollywood stars' faces onto porn performers’ bodies or turn heads of state into photorealistic digital puppets.

Digital culture, in short, is rife with aliases and alter egos, (mis)identification and dissimulation, projection, heteroglossia, and prosopopoeia. Picking up where my essay on the networked voice left off – and structured, like that essay, around a set of propositions intended to draw attention to key trends in digital culture – this piece attempts to illuminate the role of avatars in digital identity work, situating my research in relation to that of other researchers exploring this field.

In a mediascape where it seems to be getting harder to trace voices to their sources, Sarah Kessler and Karen Tongson have observed a tendency to turn to metaphors of ventriloquism and karaoke.⁶ Ventriloquism, Kessler notes, offers a framework for talking about “situation[s] in which one individual acts as the communications medium – usually the speaking or singing vessel – for words, songs, and other ideological formulations that originate or originated with someone else.” As conditions of digital “infoglut” catalyze a resurgence of conspiracy theories,⁷ tropes of ventriloquism witness the desire to identify a culprit manipulating things from behind the scenes – while also serving to describe the very real use of “sockpuppet” accounts online for purposes of pranking, fraud, and propaganda. Karaoke, meanwhile, is seen by Tongson “as a kind of shorthand for ‘the unoriginal,’ the debased copy, the amateur re-enactment.” Where this implies that there is a clear distinction to be made between ur-texts and tribute acts, critics like Michael Moon would caution against such an assumption; he reads scenes of karaoke and lip-synching in films like David Lynch’s Blue Velvet (1986) as reminders that identity always involves a mimetic dimension, that insofar as we ever find “our own” voices, we do so by parroting others.⁸ And if karaoke still carries a whiff of the ersatz, this fact does not seem to have deterred users of TikTok, which incorporated the lip-synching app Musical.ly in 2018, and has become one of the world’s most downloaded apps by fostering novel forms of singalong self-presentation.

In a different key, Erica Scourti’s performance “Think You Know Me” is pitched somewhere between a karaoke performance and a ventriloquial séance, with Scourti allowing herself to be possessed not by a spirit, but by predictive text software trained on a corpus of her e-mails and social media posts.⁹ Sometimes the joke is on the software, which struggles to distinguish between the metaphorical and the literal, space and time, business and pleasure, facts and emotions. But sometimes the joke is on Scourti, who struggles to keep up with her device as it lays bare her preoccupations, pet phrases, and linguistic dodges for the audience. A Bergsonian might say that if we’re invited to laugh at software failing to be fully human here, we’re also laughing at the human tendency to slip into machinelike routines and habits – a tendency that Bergson argues is fundamental to comedy, which is for him a means of mocking our failure to be dynamic, flexible, and agentic.¹⁰ Sianne Ngai, however, proposes that Bergson’s theory, while undoubtedly useful in unpacking the comic burlesques of the mechanized Fordist era, may need to be overhauled for the post-Fordist age, an age marked by “the cross-coupling of play and work” and “an increasing extraction of surplus value through affect and subjectivity.”¹¹ Ngai contends that in contemporary comedy we are often invited to laugh not at characters’ inflexibility but at their overflexibility, the frenzied elasticity with which they negotiate a working world of precarious short-termism and always-on connectivity. This reading certainly chimes with Scourti’s autobiographical artworks, which underscore the plight of the zanily hyperactive artworker perpetually chasing grants and residencies, ingratiating herself with curators, and mining her own feelings and relationships for material for her work.

Turning the artist into her own avatar, “Think You Know Me” also points to the changing temporality of the voice. Technologies like the gramophone allowed for the voice to be played back, but also played backwards¹² – not to mention slowed and sped up, chopped and screwed. As Mark B. N. Hansen has argued, however, one of the key characteristics of twenty-first-century digital media is their predictive and probabilistic orientation,¹³ which turns recordings of the past into means of predicting and shaping future behaviors. If this principle is already familiar from, say, search suggestions and results tailored to reflect the user’s browsing history, the promise now is of using these capabilities to abstract the voice, and even the personality, from the mortal individual: the start-up eterni.me has courted investors with the prospect of “artificially intelligent 3-D avatars… that will look, sound and, most important of all, act like individuals who are no longer with us.”¹⁴

For another example of how metaphors of karaoke and ventriloquism overlap with the language of avatars we might look to the recent trend for “NPC memes” on the notorious imageboard 4chan. The abbreviation NPC comes from the world of single-player videogames, in which player-characters – vehicles for living, thinking humans capable of understanding and changing the state of the gameworld – interact with “non-player characters” who are little more than clockwork puppets, reacting predictably to stimuli and parroting the same old lines. The meme attained popularity as a means of poking fun at individuals whose fondness for platitudes and catchphrases is taken as a sign that they lack the ability to think for themselves – whether it be “normies” leaving jejune comments on friends’ selfies, blinkered Trump supporters uncritically regurgitating Fox News propaganda, or doctrinaire liberals scrupulously employing what anons¹⁵ disparage as tiresomely “politically correct” terms. NPC memes attained wider visibility, however, as a means of dismissing the rhetoric of “social justice warriors” as rote posturing – furthering, in so doing, the reactionary right’s policy of portraying of its own ideological positions as logical and rational (if superficially unpalatable) accounts of objective realities.

NPCs, or so the meme goes, are distinguished by their lack of an “inner voice.” In some instances of the meme their values and opinions are formatted as code, akin to the artificial intelligence routines that govern videogame characters’ behavior. Thus (putatively) knee-jerk opposition to Donald Trump is expressed as

if (man.Color = = Color.ORANGE)

Man.Bad = = true;

While the tendency to give credence to women who accuse men of predatory behavior becomes

If (accusation.GetAccuser () .Gender == Gender.WOMAN)

Accusation . Believe ();

in a computer science–inflected update of the tactics used to portray women testifying to abuse as “tainted witnesses.”¹⁶

Other iterations of the meme bemoan the banality of phatic communication. One sees male NPCs commenting on a female NPC’s selfie with generic placeholders,¹⁷ mocking the inanity of what Ego Media’s Alexandra Georgakopoulou calls “ritual appreciation”¹⁸: NPC#100842123_2 responds with “Positive Response #54,” NPC#100234001_2 with “Humourous Remark #2.” As in other iterations of the meme, it is possible to detect a mixture of contempt and ressentiment towards those capable of exchanging commonplaces and compliments in good faith. There is a kind of compensatory mythology at work here, whereby the socially marginal position of the anon doomed to be “forever alone” (as memes popular in the late 2000s and early 2010s had it) becomes a sign of mainstream society’s inability to accommodate those who truly “think for themselves.” Beyond an investment in the political controversies often referenced in these memes, then, we see a horror of the reality gestured towards in Blue Velvet’s karaoke scenes: the idea that our opinions and identities are not entirely our own, that language forces us to resort to clichés and borrowed phrases to communicate. One can imagine why this idea might be particularly distasteful to libertarians, rugged individualists, and self-proclaimed free thinkers.

The NPC meme’s trajectory – from anonymous gamers on 4chan to a wider uptake on Twitter and other social media – is, by this point, a thoroughly familiar one. In her 2012 study of social media discourse, linguist Michele Zappavigna considers a number of jokes and “phrasal templates” that were initially shared on anonymous message boards by “communities of online gamers” before “spread[ing] more generally to the point where the inside joke is a joke shared throughout the ‘internets.’”¹⁹ Many of the examples she considers (including “the internets,” an allusion to one of George Bush Junior’s famous slips of the tongue) involve coding oneself as a member of the cognoscenti by posing as a member of the “illiterati,” speaking through or as an avatar whose grasp of language is considered comically eccentric or deficient.²⁰ Readers are expected to be knowing enough to know that you’re actually in the know. From the use of so-called Engrish (“a parody of grammatical errors in non-native English, most often Japanese English”²¹) to the sympathetic commonplace “I know that feel bro” (which supposedly originated from a German image board), such instances of what the Merriam-Webster dictionary terms playful use of disjunctive grammar often involve citing or mimicking nonnative speakers.²² This mimicry not infrequently shades into xenophobia and racist caricature. Given how fertile spaces like 4chan’s “politically incorrect” board /pol/ would become as breeding grounds for white supremacy, it should not surprise us that in two of the memes Zappavigna considers the “illiterati” enlisted as comic mouthpieces are rappers Kanye West and Xzibit, whose use of African American vernacular English is more or less implicitly mocked.

As Kessler remarks, ventriloquism often “overlaps with minstrelsy.” Lauren Michele Jackson has written extensively on how images of black bodies and burlesques of black speech are used to generate cultural capital on social media. Discussing Shudu Gram, an African “Digital Supermodel” (as her Instagram bio has it) created by a white British man, Jackson cites Eric Lott on how “blackface minstrels, who originated during the antebellum period, allowed white audiences to indulge their intense fascination with blackness without having to interact with actual black people.”²³ In her writing on the use of catchphrases and reaction GIFs online, Jackson notes how frequently the labor of manifesting affect is delegated to black performers who are co-opted as avatars through forms of what Jackson dubs “digital blackface.”²⁴

Similar concerns came to the fore in a recent dispute between actor Alfonso Ribeiro, famous for playing the preppy fall guy Carlton in the 1990s sitcom The Fresh Prince of Bel-Air, and Epic Games, makers of the cartoonish online shooter Fortnite (2017). Fortnite, which has become phenomenally popular with children and teenagers, encourages players to create and customize distinctive avatars, and Ribeiro’s “Carlton dance” is one of the “popular dance routines [Epic] has cannibalized to make its Emotes, digitized dance moves which, along with weapon colors, outfits and character designs, can be purchased in-game using real currency.”²⁵ As Yussef Cole notes, Epic has also “borrowed” moves from hip-hop artists like BlocBoy JB, whose “Shoot dance” first featured in a video that shows a large group of young, mostly black men at a basketball court, “leaning on the trunks of cars, throwing dice, holding guns, laughing and dancing with abandon.”²⁶ Watching these moves being copied by Fortnite’s cartoonish avatars, one is reminded of Robin James’s analyses of how cultural forms and modes of performance associated with “non-bourgeois black masculinity” have often become appealing to mainstream audiences only once they have been adopted by artists who are not black men.^{27 28} Given this, Chance the Rapper’s call for Epic to share its profits with the stolen dance moves’ originators is understandable. Such calls, however, must also be recognized as perpetuating logics of privatization and individualism; as Martin Zeilinger notes of a proposal by Kodak to track the trajectory of digital photographs using blockchain technology and remunerate photographers via a proprietary cryptocurrency, such “solution[s]” entrench the idea that a cultural work should be seen “first and foremost, [as] a valuable asset that we must protect, monetise, and commercially exploit… an ownable, tradeable artefact.” ²⁹ Such systems also, of course, depend on a model of “creatives” as atomized entrepreneurial individuals, dependent on platforms and with limited bargaining power – the same model of “platform capitalism”³⁰ that Uber has so successfully pioneered and that drivers are now organizing to oppose.

If Fortnite represents another instance of the videogame industry “borrowing” from black culture, my Ego Media research on grime culture looked at how musicians like D Double E, Skepta, and JME have reversed this polarity, using samples and similes to enlist videogame characters as their avatars.³¹ While it is seldom framed this way, grime is full of autobiographical accounts of what it’s like to grow up playing digital games, and its portrayals of inner-city life are peppered with invocations of virtual dragons and dinosaurs, karate champions and ninjas. Just as gamers will pick a character on the basis of certain qualities (dexterity, durability, aesthetic appeal), so, on the track “Street Fighter Riddim,” D Double E deploys the cast of the Japanese fighting game Street Fighter IV (2008) as his lyrical avatars, using them to figure what he presents as his defining traits – from his eloquence and entrepreneurial drive to his formidable weed habit – in a bravura sequence of pun-laden similes. This fondness for simile over metaphor is typical of grime and reflects a notable tendency in digital identity work more generally. Where metaphor asserts the identity of tenor and vehicle, similes highlight likenesses that are partial, provisional, and situational. Like the social media user who speaks as a particular memetic avatar for the duration of a post or tweet, the grime artist morphs from bar to bar. Often MCs will use similes to align themselves with stereotypes of black masculinity; in other cases they knowingly subvert expectations, as when Dizzee Rascal describes himself as “going on Black / Just like Cilla.”³²

Such moments wring drama and comedy from what Ricoeur calls “the dialectic of otherness and internalization underlying the process of identification.”³³ Discussing the enduring “dispositions” that enable us to assert the continuity of identity over time, Ricoeur contends that “to a large extent… the identity of a person or a community is made up of… identifications with… models and heroes, in which the person or community recognises itself.”³⁴ Through the process of identification, “otherness [is] assumed as one’s own” and integrated into one’s identity.³⁵ But where Ricoeur emphasizes ongoing “fidelity” to these “models and heroes,”³⁶ the grime MC reeling off similes at a tempo of 140 beats per minute is promiscuous and protean, “like Cilla” one bar, “like Godzilla” the next. This is not to say that consistency and continuity are sacrificed completely: as Joy White notes, the most successful grime artists have proven adept at crafting coherent and compelling personal brands. True, those artists present identities that are, to some extent “contingent and fluid,” foregrounding different aspects of their identities in different contexts and locations (“Stormzy is an English MC in the UK but introduces himself as Ghanaian at [a] Christmas concert in Lagos… Skepta is Nigerian in Lagos and an English MC in New York”³⁷); but their fondness for similes is perhaps best understood as a kind of rhetorical cubism, a means of shuttling listeners between different perspectives on the same subject. These are not deconstructive ripostes to the idea of a stable identity, so much as they are recognitions of the way that, under post-Fordism, we are expected to remain faithful to flexibility, incorporating mutability as a core character trait. Cannily selecting from the available avatars, shifting aspect and affiliation as circumstances demand, the grime MC makes a show of skills we are all expected to possess in what Celia Lury has called the age of the “experimental individual,” vocalizing a “flexible, ever-shifting self.”³⁸

Grime’s love affair with gaming sheds light on how, by putting players in command of proxy bodies, digital games have opened up novel expressive possibilities. We should be careful, however, not to overstate or misrepresent the implications of the player-avatar relationship for conceptions of identity. As Adrienne Shaw insists, “Playing as a character that is ostensibly ‘other’ to you (in terms of gender, race, or sexuality) is not necessarily transgressive or perspective-altering” just as “playing as a character that is like you (in terms of demographic categories) does not necessarily engender identification.”³⁹ In fact, as Lisa Nakamura long ago argued, such forms of “identity tourism” can be decidedly “regressive.”⁴⁰ Despite the protestations of scholars like Shaw and Nakamura, certain pundits insist on the potential of digital “empathy simulators” to foster greater understanding, helping users to see things from alternative points of view. Such thinking has been particularly prevalent in corporate discussions of virtual reality technologies: witness Mark Zuckerberg affirming that “one of the most powerful features of VR is empathy”⁴¹ or Chris Milk piously asserting that “through this machine we become more compassionate, we become more empathetic… we become more human.”⁴² Such notions have received richly deserved criticism from academics^43,44 and game designers alike. In “The Road to Empathy,” her 2015 show at New York’s Babycastles gallery, Anna Anthropy (whose autobiographical game Dys4ia (2012) has been lauded for its witty and candid portrayal of her experiences of hormone therapy⁴⁵) exhibited a pair of her old shoes and a pedometer, inviting “players” to literally walk a mile in her shoes.

As Ms. Anthropy drolly implies, putting players in charge of surrogate bodies is not a quick-fix solution to anomie, ignorance, and intolerance. But videogames and VR remain potent and dynamic media nonetheless, and (as the selection of titles I discuss here illustrates⁴⁶) game developers are already probing the influence of emerging technologies on modes of understanding and enacting identity. Using forms of representation and narration familiar from “old” media, videogame designers have also pioneered novel forms of procedural expression. If they often struggle to meld old and new gracefully, this only renders their works more compelling as commentaries on the difficulty of reconciling traditional notions of identity with those ascendant in digital culture.

Requiring users to elaborate possibilities immanent in the code, videogames beg our complicity even as they leave themselves vulnerable to unexpected forms of subversion, from everyday “counterplay” to elaborate mods and hacks to the kinds of metagames played by speedrunners and theorycrafters.⁴⁷ The rise of social media, vlogging, and streaming has fostered a move towards games that are less concerned with telling predetermined stories than they are equipping players to construct and circulate their own narratives.⁴⁸ Often these stories are winningly silly; sometimes they are pointedly political or bleakly misogynistic.⁴⁹ Meaning, here, emerges from the terms on which specific players activate the ludic and semiotic potentials of particular avatars. Nina Freeman’s autobiographical games Cibele (2015) and Lost Memories Dot Net (2017)⁵⁰ reflect on the forms of ambiguity and ambivalence that can flourish in the gap between the individual and her virtual proxies. In the former Nina explores the online world of Valtameri through her avatar Cibele from a college bedroom plastered with images of the Japanese pop culture heroines who have inspired her personal style; in the latter we decorate the younger Nina’s blog with selfies, photos of celebrities, and illustrations of favorite anime, manga, and videogame characters. As I note in my article on Cibele, there are parallels here with

the modes of “brand[ing] the self” practiced by the young women [Amy Shields] Dobson studies, whose online profiles are often decorated with what she dubs “‘dream girls icons’... cartoons, animations, and digitally rendered or digitally altered” images of “heterosexy” female bodies.⁵¹ ….on the one hand, the presence of such images on a personal profile might suggest that the figures depicted are to be understood as “standing in for the profile owners themselves” in a gesture of “identification with objectified women”; on the other, these images can be interpreted as “imagery consumed or liked by the profile owners,” serving as sexualized objects in relation to whom those profile owners can manifest a kind of sexual subjectivity, performing “complicity and self-alignment with a historically male, heterosexual gazing subject position.”⁵² [For Dobson] these images offer a means… of “agentically engaging with the aesthetics of femininity and heterosexy consumer culture in a way that is signalled as both playful and authentic.”^53,54

The dynamics of metonymy, relationality, identification, and desire can be complex, cryptic, and rather queer in cases like these – and this, we are invited to conclude, is in part why such forms of avatar play appeal to young women who are experimenting with their identities. When Nina poses for a sexy selfie lying on her bed with a “love pillow” that depicts a submissive, scantily clad anime girl, she is at once knowingly objectifying herself and projecting a certain kind of sexual agency.

In understanding such forms of self-presentation it can be useful to look back to earlier phases in media history. In other work I have argued that “Gaiety Girls” like Ellaline Terriss and Constance Collier – performers who helped to establish the blueprint for mass media celebrity in the 1890s – can be seen as part of the prehistory of the digital avatar. This was the conceit at the heart of Moving Past Present,⁵⁵ a collaboration with artist Janina Lange in which we “reanimated” Terriss and Collier as digital marionettes. Both women became stars under the stewardship of the impresario George Edwardes, who made the Strand’s Gaiety Theatre famous as the home of musical comedy in the 1890s. Edwardes’s “girls” were already avatars of a kind, tasked with embodying forms of femininity that were modern in some respects, but posed little threat to patriarchal privilege. Described by Stephen Gundle as “the world’s first branded showgirls,”⁵⁶ Edwardes took full advantage of popular media and merchandising to keep the them in the public eye.

Also part of this prehistory are the “poets and lovers”⁵⁷ Katherine Bradley and Edith Cooper, who wrote together under the alias Michael Field. In 1892 (the year before Collier and Terriss made their debuts at the Gaiety) Bradley and Cooper published Sight and Song. A collection of ekphrastic verse, it saw them recruiting figures from renaissance paintings to act as their surrogates. From Saint Jerome to the goddess Venus, the poems use these figures to explore the treacherous discursive terrain of gender, desire, and faith. For Animating Sight and Song⁵⁸ I worked with Ana Parejo Vadillo on an illustrated online edition of the poem “Antonello da Messina’s Saint Sebastian,” a text that exemplifies the queer forms of avatar play found throughout Sight and Song. Forged during a bout of scarlet fever, Edith Cooper’s passionate identification with the martyred saint depicted in da Messina’s canvas was consolidated when her hair was cropped, bringing out the gamine good looks that her co-author and lover so admired. The poem suggests desire for Sebastian, but also empathy – a word introduced into English by Field’s contemporary Vernon Lee, an art historian, novelist, and (as Max Saunders notes) To-Day and To-Morrow contributor who also favored a male pseudonym (she was born Violet Paget) and whose aesthetic theories were developed in collaboration with her companion Clementina “Kit” Anstruther-Thomson.

Framing these individuals as part of the digital avatar‘s prehistory may sound deterministic or reductive, as if I am demoting earlier texts, forms, and technologies to wayposts on a road leading inevitably towards the media landscape of today. This is not my intention. Rather, I want to highlight the changing terms on which proxy bodies have been made available for purposes of self-expression and identity play at different points in media history, and to foreground the queer questions such practices pose. From mass-produced halftone postcards to imported anime merchandise, Quattrocento canvases hanging in the Dresden Gemaldegalerie to JPEGs downloaded over a dial-up connection, it is important to attend to the forms and infrastructures through which avatars are accessible. As Julie Rak reminds us, to be “on-line” used to mean being connected to the railway,⁵⁹ and, as Ana has argued, rail networks were crucial to the brand of queer cosmopolitanism that Michael Field cultivated.⁶⁰ Based in suburban Reigate, Bradley and Cooper relied on trains to carry them to and from the cultural hotspots of London and (via Dover) to the various European galleries they visited in the course of composing Sight and Song. Ana has argued that their experiences as railway passengers profoundly shaped their writing and thought, informing theories of spectatorship that remain suggestive in our age of branching hypertexts, digital screens, and augmented reality overlays.⁶¹

Sight and Song is also a product of such pre-digital networks as the postal service and the telegraph, which connected the Fields to correspondents like the art historian Bernard Berenson, who was sending them photographic reproductions of Botticelli paintings long before Wikipedia and the Google Art Project would place high resolution color reproductions of these works at any web user’s fingertips. Contemporary works like Tabitha Nikolai’s Ineffable Glossolalia (2018)⁶² – a virtual reality piece that incorporates reproductions of etchings, collages, and paintings by artists like Dierick Bouts, Cornelisz Anthonisz, Max Beckmann, and Otto Dix alongside imagery drawn from contemporary gaming culture – underscore just how dramatically things have changed. And yet, as a work about piecing together a trans identity in a transphobic culture, Nikolai’s game also harks back to Sight and Song’s strategy of using borrowed bodies to think about sex and gender.

Of course, sex and gender cannot be adequately understood without a consideration of class, ethnicity, and nationality – as the careers and personae of “girls” like Collier and Terriss attest. The cigarette cards and postcards that Janina and I harvested from ebay for Moving Past Present show how Collier, whose Portuguese heritage meant she was legible as nonwhite, was frequently cast in “exotic” roles that pandered to Imperial Britons’ fears and fantasies in the early stages of her career. By the time she was playing Mrs. Atwater in Alfred Hitchcock’s Rope (1948), however, she was more often asked to incarnate whiter-than-white British hauteur. Blonde ingenue Terriss, meanwhile, starred in comic dances and skits that were often inspired by US minstrel shows. These performances, which played on the incongruity of a quintessential English rose performing what was sometimes euphemistically called “plantation music”^63,64, look a little different (albeit no better) once we know that Terriss was born in the Falklands to a father who tried his hand at farming sheep before returning to the UK to find fame as an actor. These women’s biographies and careers, in short, were both profoundly shaped by colonialism and migration.

These histories help to contextualize the contemporary phenomena I have been studying, fromthe global circulation of grime⁶⁵ to the forms of avatar-based neo-minstrelsy and “digital blackface” discussed by Jackson. They also show how myopic and disingenuous constructions of a traditional British identity under threat from migration and “globalism” are, underlining how, in Ricoeur’s words, once discourses of nationality become “separated from history and geography… traits are solidified and lend themselves to exploitation by the most harmful ideologies of ‘national identity.’”⁶⁶ This is not, however, to say that we should be uncritically lionizing marginalized historical figures; as feminist media historian Laine Nooney puts it in her profile of pioneering game designer Roberta Williams, “the common practice of ‘adding women on’ to game history in a gesture of inclusiveness fails to critically inquire into the ways gender is an infrastructure that profoundly affects who has access to what kinds of historical possibilities at a specific moment in time and space.”⁶⁷

In a similar vein, Sarah Banet-Weiser has recently argued that the online “economy of visibility” tends to favor forms of feminist expression that are upbeat and celebratory.⁶⁸ In the realm of popular history and biography this can result in the recuperation of pioneering icons whose stories are used to demonstrate how enlightened and tolerant we have become. Such accounts often betray the affective dynamics of what Heather Love calls “emotional rescue,” whereby the recovery and retelling of stories about individuals victimized during their lives and since forgotten is presented as a “heroic” act on the part of the historian.⁶⁹ As Tavia Nyong'o argues, putting Love’s work in dialogue with Saidiya Hartman’s research into the Atlantic slave trade, such retellings can, at least in some cases, be seen as “reinscrib[ing] the violence” they ostensibly deplore through a “second violence of retelling.”⁷⁰ Imposing a false coherence on the past, they deny the capacity of historical subjects to disturb our complacency and evade our capture, too often reducing them to dummies through whom we can give voice to “progressive” platitudes.

In framing the Gaiety Girls and Michael Field as players in the prehistory of the digital avatar, my collaborators and I tried to remain attentive to the conditions within which these individuals were able to articulate identities – and the sometimes troubling or puzzling terms on which they did so. If Terriss’s offhand use of racial slurs in her otherwise fluffily innocuous memoir is jarring today, Collier’s discussions of her own ethnic identity in the autobiography Harlequinade: The Story of My Life betray some decidedly dubious ideas about “breeding” and heredity.⁷¹ And, as our article on Michael Field argued, while the pair may have been reclaimed by queer social media users as “inspiring historical lezzies,” such framings smooth over aspects of their relationship, and of their respective sexual identities, that remain difficult to parse.⁷²

My practice-based research projects have also sought to dramatize and reflect upon questions of voice and agency. Moving Past Present culminated in a performance at which actor Meghan Treadway recreated scenes, skits, and dances featuring Collier and Terriss for a live audience. Thanks to a DIY motion capture rig, her movements were in turn mirrored by the digital Ellaline and Constance avatars Janina had created. The 3D models and motion data were then uploaded to Sketchfab,⁷³ where they are currently available for game designers and digital animators to download. Tracing this trajectory (live performances, captured on film, recreated in front of an audience a century later, and recorded by digital gesture-tracking technologies for submission to an online archive) we wanted to highlight how the same movements register differently as they pass from body to body, medium to medium, platform to platform, moment to moment. In a sense, Meghan was in control of the Ellaline and Constance avatars, continuing a history that has seen Terriss and Collier serving as avatars for a succession of playwrights, directors, advertisers, photographers, impresarios, and historians; in another sense, however, these performers might be said to have been puppeting her, having laid down kinaesthetic patterns that Meghan strove to faithfully replicate.

Similar questions underpinned An Archive of Tingles,⁷⁴ the podcast about ASMR culture that I made in collaboration with two YouTubers who create ASMR video content. As part of the podcast I asked my collaborators to recite short essays that I had written, essentially having them ventriloquize academic perspectives on ASMR culture. By putting words into their mouths and then giving them the opportunity to express their own views, I wanted to foreground the way that scholarly accounts of online culture often use particular users and communities to illustrate claims or corroborate theories with which those users and communities might not agree. In the case of ASMR, researchers like Emma Leigh Waldron have argued that the relationship between viewer, ASMRtist, and media assemblage should be construed as sexual.⁷⁵ This view goes against the opinions of the majority of the ASMR community, which generally considers the idea that “the tingles” are sexual to be both inaccurate and stigmatizing. Waldron’s work is nuanced, theoretically sophisticated, and in many respects convincing, but for ASMR devotees unacquainted with the theories on which her claims are founded, it may also be difficult to accept.

As Paolo Ruffino observes, in such instances academic researchers can appear to be “parasites,” feeding off of the cultures they study while apparently giving little back.⁷⁶ Charges of parasitism also dog biographers, who are frequently accused of turning a profit from others’ stories and secrets. For Smith and Watson, Gertrude Stein’s Autobiography of Alice B. Toklas – a 1933 text written as Stein’s eponymous lover and companion – highlights biography’s ethical stakes in a particularly pointed way: “Some critics have suggested that this might be a fraudulent act, an act of ventriloquism of Alice’s voice. Is it parasitic, an act of appropriation of Alice’s experience? Or is it an act of dedicated speaking through the other that commingles the boundaries of identity into a shared subject?”⁷⁷ Here Ruffino’s recommendation that academics should accept their status as parasites, “find[ing] the most hospitable way to play our role,”⁷⁸ is pertinent. Digital technologies have made it easier than ever to appropriate others’ words, names, voices, likenesses, and stories, turning us all into potential avatars. While these shifts have (understandably) fostered fears of deception and violation, it is worth exploring the new forms of intimacy and hospitality they might facilitate too.

This is the approach taken by CoH Plays Cosey, a 2007 collaboration between musicians Cosey Fanni Tutti and Ivan Pavlov, alias CoH. In the resulting LP’s liner notes Tutti describes how the project “grew out of our email discussions,” after “Ivan suggested I create a diary of voice sounds that represented and exorcised the clusters of emotional responses to events over a period of time. These were then surrendered to Ivan to be given new form.” The album is entirely made up of “recordings of air passing the mouth of Cosey Fanni Tutti” – which is to say of Tutti speaking, whispering, groaning, muttering, howling, sighing, singing, and, on occasion, playing the cornet. In her statement on the project Tutti appeals to traditional understandings of the relationship between voice, identity, and artistic expression. The recordings she sent to Pavlov are described as expressing “the internalised and the private via a pure source, my ‘voice,’” while her own practice is characterized in terms of “exploring the inner senses and communicating the inner voice.” Tutti mobilizes understandings of the diary as a privileged, ordinarily private vehicle for personal truths (an understanding also central to many of the diary songs Clare Brant has analyzed⁷⁹), and of the voice as bearer of the individual’s unique essence. She also emphasizes how much was at stake for her in a project that marked the first time she had “relinquished control of how [her] utterances are... interpreted and presented” musically, an experience that she says “displaced me from my familiar viewpoint of my essential self” in ways that were “uncomfortable but also liberating.”

Digital media serve to connect Tutti to Pavlov here, but also to alienate her from herself. Pavlov’s processing of her recordings (which are manipulated, looped, and spliced so that, say, a rhythmically recurring plosive becomes a techno-style kickdrum, or a single vowel is stretched out into a metallic drone) twist her words into strange new forms. In his statement Pavlov expands on what it might mean to “play Cosey,” explaining that the album’s title is meant to evoke “an actor performance… an amusing computer game, as well as a sex play of sorts.” While Pavlov represents himself as ‘identifying with the emotional content of the recordings,” he also describes the production process in coolly technocratic terms, likening “twisting and turning” these audio files to manipulating “the shapes in a Tetris game” and boasting of “being able to fully control the outcome” thanks to audio technologies that have moved beyond mere playback to facilitate more complex kinds of play.

An email excerpted in the sleevenotes links this play to Pavlov’s interest in “how I would think of trying to identify my ‘self,’ if I was willing to think… about discovering (or eventually being) oneself in/through other people.” For some these musings might ring alarm bells, echoing the naïve idealism of “empathy games” discourse. Elsewhere, however, it becomes clear that Pavlov and Tutti are alive to the risks and ethical complexities inherent in this kind of play. To “play” someone can also, of course, mean to betray them – to play them for a fool, tricking them into acting against their best interests. The CD’s sleevenotes stress the role of gendered power dynamics in determining who gets to play (with) whom, as Tutti invokes the history of “male artists” using “the image of woman… to establish ideals and expectations… drawing attention to a particular point that they want to make.” Pavlov (knowingly?) assumes this role in his own statement, dropping tantalizing allusions to the realm of salacious female secrets that he, the male artist, has traversed to bring us this recording. Describing Tutti’s audio diary entries as “so raw and emotional in [their] honesty that it made me blush as I listened,” he explains that he wanted to “convey the rawness I witnessed yet without the difficult discomfort in facing the most private, the most sincere, of someone else.” The phrasing is ambiguous here; is Pavlov tactfully protecting the privacy of his collaborator (a collaborator vilified by the Daily Mail for shamelessly/shamefully exhibiting her used tampons at London's Institute of Contemporary Arts in 1976)? Or is he sparing the listener’s blushes? In what sense can music self-evidently cooked up with sophisticated production software “convey rawness”? As with many of the digital vocal artifacts I have analyzed here, the tracks on CoH Plays Cosey privilege sonic texture over semantic sense, mood over meaning (indeed, the close mic’d coos of “cos(e)y with me” on opening track “Closer” make it another forerunner of ASMR videos). Tutti’s words, where they are discernible at all, tend towards the poetically abstract and opaque rather than the concretely personal – though whether this is because intimate details were omitted from her audio diary or merely reworked beyond recognition is impossible for the listener to know.

2007 already feels like a very long time ago in digital culture. It was, lest we forget, the year of the iPhone’s release (but three years before Siri’s), the year after Twitter’s launch and Google’s purchase of YouTube. But the questions that animate CoH Plays Cosey have only become more pertinent. In many ways the last five years have been a deeply dispiriting time to be thinking about avatars, the voice, and online identity. I joined Ego Media around the time of the #GamerGate movement, as self-identified gamers struck out at critics and designers attempting to rethink who videogames should cater to and represent. Often singling out women, queers, and people of color, the movement enacted and coordinated its harassment and abuse via forums, IRC channels, and social media. I write this in the wake of the Christchurch mosque shootings, a white supremacist massacre informed by the grammar of gaming livestreams, accompanied by a manifesto littered with memes. This week Buzzfeed reported that Sargon of Akkad, an antifeminist YouTuber who made his name during GamerGate before being recruited by the UK Independence Party, and who has defended his flagrant misogyny as a form of “satire,” was coordinating a campaign to run for the European parliament via Discord (a chat application popular among gamers) on a channel found to be rife with xenophobic invective and far-right propaganda.⁸⁰

These events have been viewed as symptoms of a more widespread breakdown in the neoliberal consensus. And indeed, while the framings of identity and subjectivity that have underpinned neoliberalism undoubtedly retain considerable purchase, it is also true that they are being robustly challenged on a number of fronts. For the left, those framings were always too individualistic, occluding structural inequalities and downplaying the significance of class. As we have seen, recent developments in digital culture hardly show these charges to be ill-founded; efforts to foster diversity and extend representation often go hand in hand with the commodification of difference and the expansion and refinement of cybernetic control mechanisms, and are frequently ineffective where they are not counterintuitive, masking the fact that while a few privileged “others” may now be admitted to powerful institutions, “neoliberal capitalism maintains structural asymmetries of power” through other means.⁸¹ Inclusion, in such a context, is fraught with risk, whether we are talking about the inclusion of minorities in voiceprinting and facial recognition data sets that render them vulnerable to discriminatory surveillance, or about efforts to increase the visibility of “women in tech,” which often end up exposing their newly visible subjects to abuse.⁸²

The right, meanwhile, has long dismissed even the most perfunctory attempts to address prejudice and inequality as “identity politics,” even as it has borrowed terms and tactics from its opponents (witness the rebadging of misogyny and anti-feminist propaganda as “men’s rights activism” by the online “manosphere”⁸³). Here Lauren Berlant’s analysis of how “conservative cultural politics” has “dilute[d] the oppositional discourses of the histor­ically stereotyped citizens – people of color, women, gays, and lesbians” remains all-too relevant.⁸⁴ Writing in 1997, Berlant observed that “against these groups are pitted the complaints [of] white and male and heterosexual people of all classes who are said to sense that they have lost the respect of their culture, and with it the freedom to feel unmarked”; these “formerly iconic citizens… sense that they now have identities, when it used to be just other people who had them.”⁸⁵ Where, in the Clinton era, one common response to this state of affairs was the “desire that the nation recommit itself to the liberal promise of a conflict-free and integrated world,” in 2019 that prospect seems decidedly remote. In the wake of 9/11 and 2008, the Arab spring and the Syrian civil war, many “formerly iconic” Westerners appear to favor the other reaction that Berlant identifies: that of “forg[ing] a scandal, a scandal of ex-privilege: this can include rage at the stereotyped peoples who have appeared to change the political rules of social membership, and, with it, a desperate desire to return to an order of things deemed normal… setting up that lost world as a utopian horizon of political aspiration.”⁸⁶

Today, digital media are being used to manufacture all manner of scandalous forgeries – from spurious news stories to baroque conspiracy theories. The resulting sense of cacophony and confusion has made some receptive to promises of high-tech solutions and/or more extensive surveillance: the development of new modes of translation, transcription, identification, and analysis; the collection of bigger vocal data sets on which new tools can be trained; the monitoring of formerly private or encrypted channels. Such measures, we are told, will secure the link between speaker and voice, sort lies from truth, and render cryptic idiolects transparent. But these, I would argue, are the wrong lessons to take from the current situation. In fact, as data become a new kind of lingua franca, a “language of domination” more powerful than French, English, or Mandarin, there is all the more reason to heed Glissant’s warning that “opacities must be preserved; an appetite for obscurity in translation must be created; and falsely vehicular sabirs must be relentlessly refuted.”⁸⁷ As Nyong'o insists in his work on fabulation, the production of “factual fictions… anexact in their intensifications and manipulations of the truth” has long been an important survival mechanism for marginalized communities, providing a means of circumventing the “hegemonic demands for legibility and transparency that so often simply expose and endanger minoritarian lives.”⁸⁸ Game designer and cybersecurity activist Zoe Quinn makes a similar point in their memoir Crash Override. As the primary target of GamerGate’s largely anonymous or pseudonymous hordes, one might expect Quinn to reprise calls for a “real-name Internet.” But if they believe tech corporations could be doing much more to combat abuse, they also assert that while “one might see the relative anonymity of the online world as something that allows people to do heinous things to one another without accountability… anonymity is also what can give isolated teenagers like I was the ability to talk about their queerness without fear of being outed… trying on new identities... seeing if anything really fit.”⁸⁹

These texts show that while digital technologies may multiply possibilities for confusion, dissimulation, and role-play, this is not always a bad thing. While there is certainly a need to regulate and moderate online platforms and to offer support and redress to those who need it, the answer to fascistic forgeries is not to automate listening and extend surveillance, and nor is it to insist that web users delete their alt accounts and aspire to scrupulously pellucid literalism. While there may be, as Michel Serres asserted in The Parasite, no escape from the “new regime of noise,”⁹⁰ this does not absolve us of the responsibility to listen closely and contextually; quite the opposite. While it is neither possible nor desirable to completely eradicate uncertainty and obscurity, the kinds of hermeneutic and historical skills that have been devalued in the drive towards datafication have never been more important. They are not proof against all forms of disinformation and deception, nor do they provide an infallible method for decrypting forms of expression saturated with irony, ambiguity, and ambivalence – but only the most misguided of tech solutionists would believe that this could or should be our goal.

In many of the case studies addressed in this and the preceding essay, we can see a disregard for the content of speech and the meaning of what is being said, whether it be ASMR communities searching for affecting voices, YouTube turning a blind eye to the contents of its “engaging content,” or biometric pattern recognition systems that abstract statements into acoustic data. While these examples suggest a troubling disinclination to listen, they also make it abundantly clear that the verbal and semantic dimensions of speech are not and never have been the be-all-and-end-all. If the liberal establishment’s response to current political crises has been inadequate, we must recognize that its failures have to do, on the one hand, with the fetishization of a model of “free speech” that obfuscates underlying power differentials, and, on the other, with a similarly uncritical commitment to affirming the efficacy of reasoned debate and rational argument – a commitment that has legitimized spurious “both sides” binaries and obscured the fact that (Enlightenment ideals notwithstanding) shoppers in the “marketplace of ideas” are as likely to make ill-advised impulse buys as they are sound investments. Listening closely means attending to the conditions under which it is possible to speak and be heard – and acknowledging that some highly “engaging” ideologies are not worthy of a hearing.

Ultimately, the new modes of speaking to, for, and as others that digital media are opening up should be seen not as unprecedented threats to the integrity of the speaking subject, but as reminders that the process of articulating identities has never been straightforward. Only by situating these shifts in relation to longer-term developments in forms of self-expression and conceptions of subjectivity can we truly understand how digital technologies are changing the status of the voice – and, with it, the nature of identity.