Saving Her Sorrows in the Cloud
Saving Her Sorrows in the Cloud
Caitlin van Bommel
References
- Derrida, J. (1995). Archive Fever.
- Evans, H.R., Lankford, A. (2024). Femcel Discussions of Sex, Frustration, Power, and Revenge. Arch Sex Behav 53, 917–930.
- Friz, A., & Gehl, R. W. (2016). Pinning the feminine user: gender scripts in Pinterest’s sign-up interface. Media, Culture & Society, 38(5), 686-703.
- Foucault, M. (1969). The Archaeology of Knowledge.
- Haraway, D. (1985). The Cyborg Manifesto.
- Haraway, D. (2016). Tentacular Thinking: Anthropocene, Capitalocene, Chthulucene. In Staying with the Trouble. Duke University Press.
- Johanssen, J., & Kay, J. B. (2024). From femcels to ‘femcelcore’: Women’s involuntary celibacy and the rise of heteronihilism. European Journal of Cultural Studies, 0(0).
- Sontag, S. (2003). Regarding the Pain of Others.
- Stiegler, B. (1994). Technics and Time, 1.
- Tiqqun (1999). Preliminary Materials for a Theory of the Young-Girl
The original research paper and findings were conducted in collaboration with other students at the University of Amsterdam under the supervision of Dr. Idil Galip.
Some field notes on archive media, mourning, and mapping the femcel.
What does it feel like to research the internet? It feels like poking around in the gap between the train and the platform. What does archiving the posts and personas of teenage girls feel like, when that is the object of your research? It feels like scraping small pieces of girl-pain up off the pavement with a chisel, flicking them up into your test tubes, your folders. What does mapping the femcel feel like? It feels like a claustrophobic digital box, being forced to trudge through a hoarder's house of datafied misery, and when you get to press delete for ethical reasons, you sigh in relief.
Earlier this year, I participated in a research project (as a researcher, not a subject, I’d just like to clarify) that aimed to ‘map’ the femcel. ‘Mapping’ here introduces some interesting questions about cartography and the internet: What are the limitations of our measurement? Can we chart a course into femceldom? Is it really mapping if we jump from post to post and image to image in an algorithm that has been trod before by many, many sad girls? The digital world is so complex and omnipresent that it is almost an anti-map. If it were the 90s, I’d lean heavily on early notions of Baudrillardian cyberspace, and Fredric Jameson’s labelling of it as a ‘total’ space, and I’d call what I’m doing a kind of postmodern cartography. But three decades on, dear reader, you and I both know that ‘simulation’ is not enough, and it does not begin to cover what we experience from the moment we wake up. There is no longer a question of logging on.
The research project, which began with an intensive one-week data sprint, was intended to attempt a form of categorization and exploration of the way in which femcels (the way some lonely, insecure, and rejected women choose to label themselves online) create, express, and share their identity visually. The plan was to tentatively map these algorithmic pathways on Pinterest, an image-sharing app whose UX has long been targeted towards female demographics. Over the course of the project we collected some 1,200+ images, and I began to think of myself as a sort of archivist of girl-pain, becoming more lost in what this identity meant the more images I gained. Rather than merely housing the datasets, my folders took on the essence of the misery and insecurity contained in the hyper-ironic, terminally online vernacular of the images they held. You may have, as I have, spent your entire adult life on the internet, but when you consume these posts as a researcher, your usual immunity to the unhappiness of digital strangers degrades rather quickly. My underlying uncertainty of who the femcel was—and why I had a right to archive her pain—began to grate on me.
“Incel” as a term originated on a Canadian woman’s 1990s loneliness-centered website, Alana's Involuntary Celibacy Project. Lonely users connected with each other through Alana’s small forum and mailing list. And bucketloads, of course, has been written on what happened next (spoiler: it’s not Alana’s fault). Countless archived forums can, and probably have, been scoured and scraped in order to chart some kind of course for the violence and disaster since perpetuated under the term incel. Are digital men archived more? Do men digitally archive more? Who’s to say.
What I can tell you is that there’s very little archiving being done out there for, of, or due to the femcels. The term scarcely even has a Wikipedia page in the shadow of all this male rage. “The gender politics of unfuckability” and a new aesthetic of heteronihilism arranges itself upon a logic so similar to other catalogues of the digitally female: the visual (Johanssen & Kay). Only the output of very fresh, recent research is even able to make a basis for this claim. Enter our research on Pinterest. Although perhaps Pinterest was feasible because no one has built a particularly good, free quantitative scraping tool for Tumblr. Since the late-2010s mass exodus from the Tumblr platform following a change of ownership and drastic adult content restrictions, all that nerdy feminized angst has pillowed outward to sad girls on Twitter, to teenage Whisper users, and of course, to join Narcissus at his pool in the form of girls’ Pinterest boards. So this is where we chose to make our stand. Susan Sontag, in Regarding the Pain of Others, writes that “narratives can make us understand. Photographs do something else: they haunt us” (Sontag, 71). What she means is the affective power of images, and the nature of affect in the context of remembering—if an image is connected to a time, a place, and a lived experience, collective remembering is “an ethical act, has ethical value in and of itself” (90). What then, of femcel memes on Pinterest?
Fig.1 - Sometimes, the femcel maps herself. A helpful infographic from user ‘gabriela’. Perhaps the pathway is forged by living on the internet while being witchy, psychologically disturbed… and a rabid consumer.
When confronted with the overwhelming archival task of ‘mapping’ the femcel, I very quickly discovered that as researchers, we could not navigate the internet as if we were, let me use the term, “civilians”. We were hunting for a very particular thing (a good juicy ‘femcel’ post), and when we found it, the urge was to swoop in and screenshot it like a Victorian pinning a butterfly. In the interests of objectivity, this could not happen. We had to be patient. The issue also, as is the case for so much research in digital culture, is that in-groups are rarely directly self-referential. These girls are not pasting ‘femcel’ onto every post they make. Some members of the larger vibes-based media aura may have never even heard the term. So we couldn’t just search ‘femcel’ and take a discerning selection of the top five hits. We gathered posts like a deep-sea trawler gathers fish.
The trawler in question was a browser extension designed to capture and file all images on a loaded webpage, with adjustable parameters for grabbing as many images at a time as your processor can handle. When we had our 1,200+ images, we ran them through 4CAT (a web-scraping dataset tool) and analyzed them with a CLIP model. This model sections images into the categories of romantic, euphoric, melancholic, nostalgic, bored, and rage, based on the color-makeup of the image dataset. The exact process of this feedback loop is detailed below in a mock-up diagram we developed for the report on our research (Fig. 2).
Fig.2 - The Algo-loop method used in the project. We had to start with a fresh account and fresh algorithm each time.
What you may notice from this diagram is that we did not, indeed, simply slap the term ‘femcel’ into Pinterest and start scraping. We had to gently prompt it, you see. We were the omniscient, diligent archivists and cartographers, of an organizational aesthetics of woe, rejection, self-hatred, obsession, horniness, and isolation. We had to think like a femcel. What algorithmic pathways would she take into the viper’s nest? (and believe me, some of it got quite nasty). You start with Lana del Rey, misandry, radfem ideology, the pinkpill. You gaslight yourself into a kind of Alice in Wonderland, hoping to fall down just the right rabbit-hole for the ‘recommend’ (next) algorithmic loop to grab you by the hair ribbons and say, “We know what it is; you hate men. We hate ‘em too. Now, why don’t you stay on the platform a little longer? We have ads to serve.”
We all know (I am making a grand presumption here) what Haraway was banging on about in the 90s. Yes, we get it. We are wonderful cyborgs, shimmering with the potentiality of assimilation into the machinic, against the pulverizing forces of digitalized capitalism. We make ourselves weapons as we too are weaponized. This period of technological critique was concerned with the dissolution of boundaries; at that point in time the internet held infinite possibilities for disrupting every imaginable status quo. The Cyborg Manifesto (1985) was text at play, delighting in irony and provocation. Haraway’s more recent work is instead colored by grief. In Tentacular Thinking (2016), she makes her case for a new direction, a sympoietic ethics of care. Sympoiesis, to put it simply, is the entanglement of all things, interlinked. Instead of the destruction of borders, Haraway’s later work reflects on the need for sympoietic alliances or worldings of multispecies beings in systems that range from the smallest to the largest possible structures in the universe. The cyborg, the machine-girl integration, has been shown up by three decades of digital advancement; she fails to connect with the wider, planetary-scale computation that rests upon extractivism of every kind. The future mode of being must instead be tentacular.
What researching the internet will not teach you is how great the gap really is between mainframe (vastness of the digital) and node (little old you). When you immerse yourself in the vibe-soup of digital culture now, (and I emphasize now, because just like Heraclitus’s river, it will not be the same culture you step in tomorrow) your affective sense-making capabilities are singing with light; you understand it all now! Every inference! Every intangible clue is yours for the taking! And on to the next post, and on, and on. This beautiful one-day situationship you find yourself in with the interface is soon interrupted by any attempt at external explanation or categorization, let alone an academic one.
Plenty has also been written about the practice and systemic outcomes of archiving itself. Foucault considered archival practice to be a paradigm that governs what can be expressed, known, and remembered. But whereas Foucault was interested in the ways in which state power resulted in knowledge, Jacques Derrida later expounded on the archive as a psychoanalytic, ontological problem. For Derrida there is an aspect of ‘hauntological’ power; the archive is challenged and subsequently haunted by what it cannot preserve—the silences and gaps. Equally it represents a desperate attempt to cling on to what has already slipped away. Derrida characterized ‘archive fever’ as both a death drive and a survival drive (thanatos and eros—Freud’s mutually opposing instincts), a compulsion to preserve and protect while in the process distorting, selecting and betraying knowledge.
But what do we do with the theories of Foucault and Derrida when self-archiving is now an autopoietic digital process? The subjects of our archival practice were themselves archivists. Women and girls creating, storing, preserving, touching up, and remediating each piece (an antinatalism meme, or perhaps a hazy image of Lana del Rey) of their personal collection and performance, their little archive in their corner of the internet. Almost the entire internet is archived in some form or another, and most of the processes are automated and entirely non-human. The internet is an autopoietic monster. We must be sympoietic to survive within it.
Bernard Stiegler, late director of the Institut de recherche et d'innovation (IRI), wrote about this in the typical prescient verbiage of a French philosopher turning to face the early digital age. He proposed technology as a grand pharmakon (the cure and the cause, the poison and the antidote), and that in all use-cases, it is pharmacological in its capacity to harm and to salve wounds, both material and immaterial. No more is this true than when applied to tertiary retention, the notion that human memory can be externalized—in writing, photographs, audio—and continue to shape how we recall and perceive the world as well as how we develop our culture intergenerationally. We have primary retention, the immediate moment, and secondary, which is the remembrance of what has already occurred. Tertiary is Stiegler’s addition; in externalizing the primary and secondary, it structures the collective memory with the technics (the tools and methods that come from and drive our technologies) we have available to us.
Technics relating to tertiary retention have almost always been constructed on human terms. Our language is our own, we make the paper pages of books by hand, we choose when to record our own voices or to sing by the fire to the next generation. But when the vehicles of tertiary retention and our own inbuilt bio-computational archiving meet an outward, external computational logic of archiving, the pharmakon begins to take on a wicked form. Humans are inseparably technic beings, but our tertiary retention now relies on the digital. The digital wailings and collective understanding of the cultural output of the femcel are shaped exclusively by digital media. Our primary and secondary remembering of femcels comes part and parcel with platforms who make money from directing our attention; we consume the sadness of a thousand digital girls through a device hell-bent on extracting as much from the material world—whether it be rare earth minerals, data labelling, or affect—as possible. Our collective memory is shaped by increasingly standardized, commodified pieces of information. The desire to act is bent towards the will of the market. Stiegler warns of a state of symbolic misery, in which the symbols we consume no longer support us to individuate, as in the past, but rather encourage us to consume more. This is the poisonous end of the pharmakon.
Many of the image results we found represented profound contradictions, not least because I am not even sure if ‘femcel’ has its own validity as a term. An incel, as a home-grown internet signifier, is not a gendered involuntary celibate. A girl may be an incel. So the shortening of ‘female incel’ to femcel came with its own curious eddies in the stream. Was it inceldom in pink flavor? Like male incel forums and subreddits, it has been noted that femcel communities contain a general tension between nihilistic resignation and participants’ attempts to "ascend" out of involuntary celibacy. A study from 2024 found that “although some femcels referenced their anger, hatred, or desires for revenge, this antipathy may have been rooted in their concerns about how to find a suitable intimate partner while avoiding the threat women often face from violent men” (Evans & Lankford). The difference in the undercurrent of sexual dynamics, and therefore the type of negative feelings that result, cannot be understated. The relationships to both desire and violence for those participating in these communities are marked by very different offline, material realities.
Unlike the terminology that arises as a result of male incel subculture (such as the use of ratings systems, and the terms ‘foid’, ‘roastie’, or otherwise), many of the femcel image results we collected made little or even no reference to men. Instead, there existed a vast, inward-facing world, directed at the interiority of one’s misery. Some of these posts lamented the very existence of human attachment, let alone that of men to women, or women to men. At the same time, many of them resent their own apathy, and simply desire to desire something, anything, at all. You may have seen references or perhaps read of the first mainstream incel murderer, Elliot Rodger. A femcel equivalent is perhaps Jodi Arias (Arias herself not associated in any way with femcel-ism, but rather simply a young, conventionally attractive blonde woman who murdered her boyfriend). The ways in which femcels treat and deify Arias through their Pinterest archives is rather fascinating. There is no coherent ideology, only spectrums of affect. It is not what Jodi did (that is merely the source of clout, or the hook), but what emotions and opportunities for self-identity and expression that her act elicits.
Fig.4 - A Jodi Arias meme from user ‘lukin’.
In true Margaret Atwood fashion, many of the Jodi Arias memes have performative appeals to outward desire contained within their outwardly projected dreams of violent liberation and control. She’s crazy and murdered her male partner? Well then, her pixelated courtroom images will be steamrolled into a larger, coquettish trend of romanticizing one’s own mental illness, because men love crazy girls. Let your misery drive you into insanity; guys love that, guys want to fuck crazy pussy. Your unhappiness and abnormality is your pharmakon, it is the cause of and the solution to your loneliness. But does archiving and remediating images of Jodi acquire power, in a Foucauldian sense, if femcels represent the disenfranchised?
Tiqqun, an anonymous, radical post-Marxist collective, produced a series of journal articles between 1999 and 2001. One of these articles was Preliminary Materials for a Theory of the Young-Girl. The Young-Girl is not necessarily a female figure, but she is the result or the subject of participating in, being commodified by, and thereby upholding contemporary consumer society. The ‘Girl’ aspect comes as a result of her characteristics being typically associated with womanhood and taken from the commodification of femininity itself through vanity, performance, and self-image. The Young-Girl sustains herself, others, and society through the flattening of self-expression, and instead only performs acts of self-valorization. Her body and existence become manageable and submissive to commodification. In this sense she represents a form of biopolitics. The unfettered practices of freedom, intimacy, and individuality become alienated. Even her rebellion is a commodity. You see where I’m going with this.
Of course, in some sense, femcel desires, (and equally, subjugated, queer, or absent desires), are themselves alienated from patriarchal capitalism. They can all be said to foil the Young-Girl. Their archive is haunted by the gaps and silence of meaningful human connection. Unfortunately, we don’t just live under patriarchal capitalism. It has perhaps been superseded by platform capitalism, and no vehicle is more crushing in its ability to commodify digital expression than the platform. Pinterest’s archival logic is the ultimate weapon against rebellion. As shown in Fig. 1, femcel identities inevitably come to revolve around material aesthetic signifiers, and once the algorithm catches the slightest whiff of those signifiers, it will throw them at you again and again and again and again. As Tiqqun wrote, "However vast her narcissism, the Young-Girl doesn’t love herself; what she loves is 'her' image, that is, something foreign and exterior to her, but that possesses her."
Black Swan. Mitski. Lana del Rey. Anorexia. Kate Bush. Sylvia Plath. Girl, Interrupted. This is what our image scrapers turned up, again and again. White. Pink. Black. Blonde. Grey. Skinny. Coquette. White. If you’ve been any kind of girl on the Western internet, you probably know exactly what I mean. This is what femcel archives, if indeed the algorithmic pathways themselves can be said to be archives, contain. This unfortunate aesthetic has the equally unfortunate affective trait of being an extreme info-hazard, primed to turn a young girl into the Young-Girl. You cannot properly format your pain without using a platform architecture to overlay your text onto a faded image of Lily-Rose Depp, which you must then upload onto Pinterest. Maybe the image is cherished, fetishized, on the image-wall for a moment. The Young-Girl requires her self-valor through an endless stream of external images and signifiers. Resistance to the Empire of commodity is brief, fleeting, made with irony and vitriol—before it falls into the soup of movies, music, celebrities and visual motifs that must uphold some kind of universal notion of the term ‘femcel’. This was the map I charted.
Over the course of the project, I did begin to think that every archive is, in a Derridean sense, a cemetery. The body of the Young-Girl is buried in every Pinterest board. But instead of merely preserving the past, the true labor of digital archiving is actually the act of scripting the future. In the end, we are not archiving ourselves for ourselves, but for the machinic architectures that will survive us. If every archive is a cemetery, then above them are not clouds, but the Cloud; not just the site of mourning, but the grand-scale digital necropolis where we are able to luxuriate in our own disappearance. Is that bleak? Sorry. I might have spent a bit too much time online.