There are many interesting formations that might be called networked phenomena. Homophily and the tendency towards triad closure. Scott Feld’s Rule (I’m more likely to make friends with someone who has more friends than me). Small world phenomena (those 6 degrees of separation). “The Strength of Weak Ties” (reportedly the most cited sociology paper in history). In all, a series of social forms that complicates typical binarisms like individual versus group.
All of these have their positive and negative sides, but few networked phenomena have been met with more ambivalence than that of contagion, the idea that things (memes, viral videos, fashion) spread from person to person in a way that is similar to an epidemic; that is, people believe certain things or participate in certain behaviors without necessarily having “decided” to do so. Instead, the chances of “contracting” an idea, a fashion, or a new technology come down to the structural position in a network—a question, for example, of k-threshold models, where the chance of contagion depends upon the topology of connections vis-à-vis other infected nodes.
Given its identification with epidemiological contagion, it is not surprising that social contagion brings with it a negative valence, conjuring up fears of loss of autonomy, of being reduced to “hosts” for the “viral” propagation of information in a network. Contagion is at the heart of the fear and fascination of the zombie. It is also part of the latest panic in politics, one that centers on a vision of an electorate easily manipulated through fake news propagated through social media.
In 2012, Facebook conducted an emotional contagion experiment involving nearly 700,000 of its user accounts. By manipulating the news feeds to include more “positive” or more “negative” content, researchers demonstrated that affect was contagious—negative content on the feed would influence users to post more negative content on their own walls (Sampson 2017). Of course, Facebook researchers defended their research, but, in their own words, the experiment demonstrates the capacity of Facebook to lead “people to experience the same emotions without their awareness” (Kramer, Guillory and Hancock 2014: 8788). And just 2 years after the publication of the Facebook research, the U.S. elections were decided through just this sort of emotional contagion—along with the exploitation of other network effects.
One of the more pernicious factors enabling (and compounding) the impact of social or affective contagion is what Lerman, Yan and Wu (2016: 9) call the “majority illusion”: “Local prevalence of some attribute among a node’s neighbor networks can be very different from its global prevalence, creating an illusion that the attribute is far more common than it actually is. In a social network, this illusion may cause people to reach the wrong conclusions about how common a behavior is, leading them to accept as a norm behavior that is globally rare.”
Combined with the contagion of fake news, “majority illusion” may results in decidedly pathological behaviors, including “Pizzagate” and white terrorism, among other things. These networked, viral acts have led to new college courses and guides to detecting “fake news” in order to disrupt the cascade of misinformation.
However, the term “contagion” suggests more than networks or even epidemiology. While the etymology of the word is grounded in touching and contact (from the Latin contagion), it also emerges out of earlier understandings of maladies attributed to connection—including connection-at-a-distance. In particular, it echoes nineteenth century anthropological theories of magic in J.G Frazer, the “contagion” that could bring misfortune down upon the victims of witchcraft through the theory that what had once been together was henceforth forever linked. For example, the hair, fingernails or (more recently) the photograph could be mobilized to influence people for evil ends.
As Frazer explained in his 1890 Golden Bough, “Thus the logical basis of Contagious Magic, like that Homeopathic Magic, is a mistaken association of ideas; its physical basis, if we may speak of such a thing, like the basis of Homeopathic Magic, is a material medium of some sort which, like the ether of modern physics, is assumed to unite distant objects and convey impressions from one to the other” (37). But with his reference to “ether” (and this comes up several times in The Golden Bough), Frazer suggests that connection itself is not enough, that there needs to be a medium of contagion for, say, the malefic influence of the mistreatment of a footprint on the person who made it. Here, he joins other late 19th century scholars who sought in such a medium a material basis for the universe, a connective tissue to unite all things (Raia 2007).
In the age of social media, our “selves” are literally spread through linked networks of online documents and media that connect to everyone through short paths. It is that dual character of online interaction—the way it involves multiple “documentary selves” spread across social media platforms, on the one hand, and the way that these are embedded in connected networks, on the other—which ushers in new fears of contagion, ones that suggest a future “loss of self” and that seem to accompany other panics over the loss of individual autonomy, including the replacement of human workers with AI and robots. In her critique of social media in Alone Together, Sherry Turkle charges that social media is inauthentic and dissociative, leading both to the dissolution of identity and to the engendering of ultimately shallow and unsatisfying relationships with people with whom we’ve networked. “As we distribute ourselves, we may abandon ourselves” (Turkle 2011: 12).
Yet it’s not clear that this “distribution” is an even process. Does posting a video on Twitter mean the same thing as posting it on Facebook? Do 500 followers on Twitter mean the same thing as 500 friends on Facebook? But what if we take the medium of the connection more seriously? In terms of contagion, what if the people connected through the contagion of ideas were less important than the structures in which they were connected? Sampson’s extension of his arguments about virality and contagion to advertising on social networks takes up the importance of the social platform: “From this perspective, marketers don’t need to infiltrate the self via the mirrors and mimicry of ideology, but instead they tap into the contagious social medium in which consumer beliefs about products and brands are readily passed on as affective contagions. As follows, marketing is not the creation of self-identity, but rather the production of sensory environments in which the contagious social medium can be encouraged. The social medium becomes the producer” (Sampson 2017: 68). That is, the idea and the “host” may themselves not be primary agents of contagion; it is the connective medium which proves conducive to magic of contagion. In other words, pace Frazer, the connection between, say, an umbilical chord and the child to which it was once attached is less important than the rituals, practices, and material culture through which that connection is observed, maintained, and manipulated (Frazer 1890: 40). Or rather, we find “contagion” swapping places with varied practices that we could deem “contagious.”
What happens when more and more of our personal and social lives are organized as networks? In a sense, our behaviors are simply networked–explicable through linked nodes,not as an ‘individual, but not an amorphous, superstructural group. These have all kinds of implications for social action, cognition, identity and feeling. As Sampson (2012: 168) writes, “Decisions are not, as such, embedded in people, or in the voluntary exchanges with others, but in the very networks to which they connect. It is, like this, the network relation that leads the way.” It is these networks themselves that have become contagious, and not just any networks. Yes, online social networks themselves have become more popular, but, more than that, highly localized social networks–SNS platforms that support networks characterized by high density and centrality. These are social networks for cliques (or complete graphs) where everyone knows everyone else, and everyone talks to everyone else. To what extent are these densely networked small worlds generative of contagion?
These questions, and the way they seem to anticipate anthropology’s “ontological turn,” are at the heart of Steve Toutonghi’s 2016 novel, Join, a near future where people “join” with each other, taking on a new, collective identity that is an amalgam of all of the individuals (the “drives”) who have been incorporated. The main characters, “Chance” and “Leap,” are further subdivided into the people who make up the join: Leap One, Leap Two, Leap Three, and so on, amalgams of people that are combined (they have all of the memories and experiences of their joins) but still individual (they are aware of themselves as individual entities separate from their joins).
Joins are different from ordinary humans (“solos” or, more pejoratively, “ferals”) in at least two respects. First joins are never alone: “His parents used metaphors—it’s like being more attuned to all of who you are, all your different desires and fears; it’s like remembering who you were ten years ago, before events changed you. They said the awareness of being more than one person included a comforting sense of companionship” (Toutonghi 2016: 9). Second, joins are theoretically immortal. With each drive possessing the memories of its joined alters, memories and identity might continue indefinitely. “If Chance Five dies, then Chance–and therefore Javier Quispe–will live on through other drives in the join. That can continue forever. In a perfect join, human beings lose both their existential sense of isolation and their mortality” (10). This constant connectivity and simultaneity makes joins formidable—literally the sum of their parts–and, ultimately, concerned more with their own intertwined lives and thoughts over that of the world around them.
And it also introduces some pathologies that are unique to joined humans, what might be called networked maladies. The first is what Toutonghi calls “meme virus.” In this “reflexive spongiform encephalopathy,” behaviors build up through positive feedback into compulsive repetition: “Infected joins would develop an absolute fixation on an idea or complex of ideas.” The second, a “flip,” concerns the disruption of the join itself, one that cascades into illness and death for the rest of the joins. As a novel, “Join” follows on the trail of both of these maladies in order to explore the implications of “joining” for conceptions of self and other.
These fictional maladies are also useful windows onto network contagion. In both, the problem seems to lie in both being connected and not being connected. “A bad flip, the truly catastrophic kind, is really fascinating, one of the most interesting conditions in all of joins science. The network connection is both established and not established, leaving the join incomplete” (76). Being connected but also not connected is the primary tension in social media. With whom do I connect? How often? And whom do I disconnect? For example, in Korea (where I do some of my fieldwork), users engage multiple social media in their daily lives, and platforms like Facebook and Instagram have proven extremely popular. These accounts are visible (to some degree) by a public, even if (with Facebook) you’ve opted for more privacy. But in Korea, there has also been a concomitant growth in proprietary platforms that support social media sharing among a close circle of intimates: KaKaoTalk is one of those–with nearly 100 percent adoption in South Korea, it is the app of choice for online discussion and media sharing. BAND is another.
Both of these suggest what Ichiyo Habuchi (2005) has called “tele-cocooning,” a term that describes a small group of friends that spend a large portion of each day in intimate, digital communication. Or, as Ito et al. define it in their Living and Learning with New Media: “The practice of maintaining frequent and sometimes constant (if passive) contact with close friends and/or romantic partners” (Ito et al 2008: 16). With KaKaoTalk and BAND, the problems of stalking and government surveillance are (at least somewhat) obviated through the creation of closed networks. However, all social networks beg the question of inclusion and exclusion. As dana boyd discovered in her fieldwork with teens, even the most “public” social media implies both a finite audience and a discrete interpretation (boyd 2007). When that “public” includes those who connect to the media or interpret it in unintended ways, there can be embarrassment, anger, trauma, and even suicide—a tragedy that has disproportionately impacted young social media users.
In any case, all of these networked distinctionsbeg the question at the heart of Toutonghi’s novel. What relationship do I have with the people with whom I share? And how do these alters impact me? With the first networked malady, we have the “viral meme”: the physiological correlate of viral media “infecting” multiple, joined hosts. And with the flip, Toutonghi asks the second question. How do the actions of my network impact me? In the novel, a “flip” cascades when one of the joins entertains doubts about the conjoining: the ultimate unfollow. But we could substitute any number of negative or even pathological behaviors that harm the subject (whether intended or unintended).
The remaining question, though, returns us to Frazer’s ideas of contagion. What, exactly, is the nature of that connection that enables the work of behavioral cascades and social contagion? For Toutonghi, the explanation revolves around a “quantum” gateway linking people together. In social media, the nature of the connections eludes easy explanation. Certainly, questions about the contagious influence of white supremacy or ISIS on social media have prompted numerous scholarly inquiries, but is it the content that makes something contagious, or the structure of the networked connections that support the content?
But for this, it might be worthwhile to return to Frazer. Not because of his insights into ethnology which were fatally flawed with the colonial frames in which he worked, but for the insights we might gain into his own magical thinking—a way of thinking about connection and causality that seems very much at the heart of popular understandings of networked contagion.
“A curious application of the doctrine of contagious magic is the relation commonly believed to exist between a wounded man and the agent of the wound, so that whatever is subsequently done by or to an agent must correspondingly affect the patient either for good or evil. Thus Pliny tell us that if you have wounded a man and are sorry for it, you only have to spit on the hand that gave the wound, and the pain of the sufferer will be instantly alleviated. In Melanesia, if a man’s friends get possession of the arrow which wounded him, they keep it in a damp place or in cool leaves, for then the inflammation will be trifling and will soon subside” (Frazer 1890: 40-41).
Social media itself takes the place of the arrow: it is the connection that enables the flow of contagion, the ligature that connects “things which have once been conjoined.” And like the “quantum gateway” in Toutonghi’s novel, this connective substance works by preserving the tension between things connected within a network, and between what counts as the network’s “inside” and “outside.”
boyd, dana (2014). It’s Complicated. New Haven, CN: Yale University Press.
Frazer, James George (2009). The Golden Bough. NY: Oxford University Press.
Habubuchi, Ichiyo (2005). “Accelerating Reflexivity.” In Personal, Portable, Pedestrian, ed. By Mizuko Ito, Daisuke Okabe, and Misa Matsuda. Cambridge, MA: MIT Press.
Ito, Mizuko, Heather Horst, Matteo Bittanti, dana boyd, Becky Herr-Stephenson, Patricia G. Lange, C.J. Pascoe and Laura Robinson (2008). Living and Learning With New Media. Chicago: The John D. and Catherine T. MacArthur Foundation.
Kramer, Adam, Jamie Guillory and Jeffrey Hancock (2014). “Experiential evidence of massive-scale emotional contagion through social networks.” PNAS 111(24): 8788-8790.
Raia, Courtney Grean (2007). “From ether theory to ether theology.” The Journal of the History of the Behavioral Sciences 43(1): 18-43.
Sampson, Tony D. (2012). Virality. Minneapolis, MN: University of Minnesota Press.
—— (2017). “Cosmic Topologies of Imitation.” Parallax 23(1): 61-75.
Toutonghi, Steve (2016). Join. NY: Soho Press.
Turkle, Sherry (2011). Alone Together. NY: Basic Books.
Samuel Gerald Collins is Director of Cultural Studies and a professor of anthropology in the Department of Sociology, Anthropology and Criminal Justice at Towson University. He researches information society and information and communication technologies in the United States and South Korea, and in particular the formation of multi-agent socialities composed of human and non-human agents.