Facebook: Our Contemporary Colosseum
How conversation became blood sport, and what to do about it
“And maybe there’s no peace in this world, for us or for anyone else, I don’t know. But I do know that, as long as we live, we must remain true to ourselves.” — Spartacus, 1960, Dalton Trumbo
I have a confession: I’m addicted to online debate. I’m not exactly sure when it began, though warning signs appeared early in life. Perhaps someone should have thrown me a lifeline after being kicked out of Sunday school for repeatedly exploring heretical lines of reasoning. But of course, that didn’t happen. And so this insatiable need to pit my own representation of the world against those of others continued its growth within. To feed its endless hunger, I rummaged through the familial, social, and educational alleyways of available conversations, always in search of a fix.
Like many addictive substances that possess medicinal value, exposure at controlled doses and used under the guidance of a professional can be beneficial. Such was the case with my early exposure to debate. It was euphoric, almost magical. Using words alone I could probe and explore the mind of another. And from the generic blanks we apply to one another, I carved increasingly high-fidelity models of those around me. With enough practice, it was possible to hold entire conversations with these models. I could stand their perspectives up against topics we’d previously discussed as well as anything else I learned! This was a wonderful skill for an only-child who spent quite a few hours alone in his own head. And beyond that, this newfound ability also afforded some degree of control over my environment and those within it. Debate was not merely a useful coping mechanism, it was a powerful tool for shaping myself and the world around me.
Just as I realized the utility and power of my addiction, an entirely new frontier of human communication burst forth upon the scene. The Internet. It promised a utopian vision built around the decentralization and free flow of information, and it was going to revolutionize the world as we knew it. Overnight the world shrunk to the size of a silicon wafer. Predictions regarding its impact flowed freely, though none could hope to fully comprehend its implications. In many ways we still can’t. Upon this rocket of change we’re riding, I can’t shake the suspicion that merely trying not to fall off has strained our current modes of being to their breaking points. Though we’ve begun to investigate, a comprehensive understanding of how complex human systems — and the underlying values driving their behavior —respond to such disruptions remains elusive. But one shouldn’t be too harsh in their adjudication of humanity. After all, we’ve only just begun to glimpse the troubling imagery emerging from the digital utopia’s rather dystopian underbelly.
But at the time I certainly didn’t know any better. I was hooked. Before me lay an infinite expanse of crystalline digital space, each individual shard unique, a reflection of its creator. Beyond that, my conversational alleyways and backwaters were now linked to an information superhighway, and I felt the need to exercise my lead foot. First stop IRC, then onward to AOL Instant Messenger, followed by detours down the country roads of innumerable message boards advertising sideshow spectacles that promised to educate, entertain, or amuse. And yet, while I sensed a great expanse within this digital universe, the connections and conversations retained a feeling of intimacy. We were pioneers exploring a new frontier. After days spent trekking through the digital wilderness — creating maps, observing the wildlife, and more generally communing with this new and increasingly strange form of human experience — a digital campfire surrounded by other global villagers was only a click or two away.
At the time, contemplating the Internet felt a lot like looking up at the stars. I’d sit and meditate upon the vast distances crossed and varied journeys undertaken by the twinkling patterns before me. And I felt that same sense of wonderment reflected in those I encountered along this new frontier. We shared thoughts and feelings we wouldn’t dare speak amongst our of “real-world” peers. After all, few of us held much power over our immediate physical, social, or geographic surroundings. This was especially true for those of us, like myself, who weren’t yet adults. It was this unique context, this pseudonymous digital milieu of circumstance and mutual respect for fellow travelers in a strange land, that allowed us to convene with trust and bestow upon one another the benefit of the doubt more often than not. Of course disagreements flourished, but most were contained to specific points of interest and the alignment of parties often shifted from one conversation to the next. Simply put, it wasn’t easy to place these people within the over-simplified and unidimensional categories that often lead to conflict.
Of course settlers soon followed. The Internet quickly evolved from a superpower of the technological vanguard to the de facto mechanism for information retrieval and popular communication. As we entered the third millennium, Google became a verb and grandma got an email address. Yet the mold into which the Internet’s milieu would settle remained unclear. At first the social graph of the Internet looked much like the underlying structure of the Internet itself: highly distributed, diverse, and heterogenous. Like many diverse ecosystems found throughout nature, natural cleavages acted as useful bulkheads. And while not very efficient for advertisers who wished to know the location, behavior, and preferences of their targets at all times, this structure made it relatively difficult for rage to travel freely across the landscape.
But the market’s disdain for a vacuum approaches that of nature’s, and so densely-connected social networks were born. From an explosive Cambrian landscape populated by all manner of creatures, the Leviathans emerged. Facebook, Twitter, and Reddit ruled the day. Given their scale, it’s helpful to consider the ecological niches these networks fill:
- Facebook represents a digitally connected social fabric extending that of our evolutionary social landscape. It places one in the psychological center of a social graph. One’s connections reach outward toward those with whom one shares real-world experiences, and further outward toward those one might increasingly call strangers. Messages flow relatively unbounded in form, though most definitely shaped by social mores and the pairing of one’s words and images to their real world identity.
- Twitter also represents a novel manifestation of behavioral patterns common throughout human history. It has evolved to fill the digital gossip niche, and thrives upon the rapid, dopamine-driven, spreading of infectious sound bytes between otherwise disconnected groups, often pseudonymously. For this reason it’s perfect for unleashing memetic pathogens upon a populous yet to evolve robust antibodies. Unfortunately presidents also use it.
- Reddit’s heritage traces back before the social networking explosion. Its topical organization retains echoes of the frontier’s campfire conversations, though the addition of voting dynamics and the influence of the news cycle on content-sourcing dramatically shifts the incentives of its users. Additionally, the notion of labeling a piece of content as either good (up-vote) or bad (down-vote) adds polarity to any discussion, no matter how inoffensive the content in question.
Despite their differences, they share a common feature: we’ve migrated from our frontier campfires, conversations within relatively secluded digital enclaves, into something that more closely resembles the public spaces of pre-industrial cities. This systemic shift carries with it far-reaching consequences given the scale to which these sites have grown. It’s these consequences I’d like to unpack and discuss here, as I feel they bear some responsibility for both my own addiction and the increasingly volatile polarization of our society.
First and foremost, communication on these platforms turns us into psychopathic autists. Before you jump down my throat, consider that I myself have a sister who suffers from severe autism, and do not use that term lightly. With respect to psychopathy, modern interpretations converge around two main dimensions. The first is labelled “selfish, callous, and remorseless use of others”, and is associated with narcissistic personality disorder. The second is labelled “chronically unstable, antisocial and socially deviant lifestyle”, and is associated with antisocial personality disorder. Does “narcissistic and anti-social” sound like anything you’ve seen online lately? So why might these platforms draw out such ideally dormant inner spirits? A convincing hypothesis, which also ties in the notion of autism, posits that online interaction short-circuits our evolved mechanisms for interpersonal empathy and emotional regulation. We don’t fully understand these mechanisms, but it looks like entire classes of neurons and neurological circuits have evolved to help regulate interpersonal communication. And because these mechanisms evolved long before the Internet was created, they remain inactive or less-active when communicating online. For example, as Key and Peele demonstrate here, we’ve all argued with someone via text only to realize that, had they been sitting in front of us with their full range of emotions and expressivity on display, we could have avoided the argument entirely. Of course it’s not fair to lay blame entirely at the feet of social networks, as this phenomena may emerge within any solely textual communication platform. Yet the scale of such networks, working in concert with the mechanisms described below, acts as an amplifier — it both fuels and fans our inner flames.
Secondly, social networks have grown into a spectator sport in which the many watch the few compete, clapping or jeering with their digital thumbs and arrows. Spectatorship tends to orient groups around the most readily available and polarizing dimension, and also masks personal accountability for one’s actions within the herd. This lowers the threshold at which one will express the types of psychopathic, emotionally-detached behavior described above. Consider how quickly conflict emerges when groups congregate and polarize around an emotionally charged topic like a sports game, political rally, or even a family gathering. But at least when these situations are grounded in physical reality one must confront the consequences of violence head on. In the extreme, those who don’t believe themselves capable of such violence but still engage in it often end up with some degree of PTSD. This is to say that despite our worst tendencies, we’ve evolved mechanisms to help us regulate our violent impulses so long as they occur within the physical world in which we evolved. The online ecosystem is largely freed from such constraints and can amplify the crowd’s roar to the tune of billions of voices.
Third: rage gets attention, and attention sells ads. To put it succinctly, our collective dependence on platforms with an incentive to profit from anger may prove the greatest moral hazard of our time. It strikes me as deeply disingenuous that leaders like Mark Zuckerberg stand before the world and tell us they have no knowledge of their technology’s contribution to the breakdown of social trust and rapid increase of ideological polarization. I’ve discussed this matter with friends who work at FB for years now, sharing my recommendations as an engineer and designer with a background in the behavioral sciences. And guess what: they were aware of the issue. Guess what else: solving it would decrease engagement and therefore profitability. Like the cigarette companies and sugar lobby executives before him, Zuckerberg has now engaged in the age-old American tradition of denying your company’s obvious externalities. Of course I understand where he’s coming from, though it’ll also be rather difficult to increase shareholder value in the midst of a civil war. That’s a problem for tomorrow, I suppose...
Fourth, the companies in question have hired armies of researchers to maximize the amount of time and energy spent inside their arenas. Imagine what this might look like in terms of fans within a physical arena. It goes beyond the tricks employed by casinos and into the realm of Orwellian coercion. Can you ever really leave? Perhaps, if you’ve turned off your notifications, prohibited emails, and haven’t yet been trained to compulsively open the app when unlocking your phone. It’s impossible to overstate the impact of this at scale. In the most conservative of thought experiments, imagine if only half of Facebook’s users felt they were spending an hour more than they they’d prefer on the platform each day. That’s 1 billion hours of human attention per day being converted into fractional ad revenue, or 365 billion hours of human potential every year. Beyond studies demonstrating the psychological damage done by such prolonged exposure, imagine what might be possible if we instead directed that precious time and energy toward improving our own lives or the lives of those around us. It would be one thing if their robbery of our time produced a net-positive return for humanity; instead, the opposite seems more likely by the day.
Fifth and finally, the feedback loop between social media and for-profit media creates the psychological environment of a crowded theater. At any moment, someone’s words may trigger increasingly virulent ideological outbreaks. Therefore the stability of the ecosystem rests upon a hair-trigger, much like a forest after an extended drought. This fragility attracts parasitic pyromaniacs seeking to score points by appealing to the worst instincts of the crowds on either side of any issue. Pundits bound by this unholy marriage between politics and the media may never wash all the blood off their hands.
If we zoom out we can see all five points working in unison to tear apart our social fabric of trust and decency, both invaluable cultural technologies upon which we’ve built our civilization:
- A spark of psychopathic digital combat triggers crowd polarization and a loss of inhibition.
- The crowd goes wild and catches fire.
- At this point algorithms pick up on the emerging conflagration and blow the embers far and wide, hoping to draw more moth-like spectators to the flames, and perhaps fresh meat to take the place of fallen soldiers on either side.
- Those who try to leave are roped back in by the army of PhDs and their pushy notifications.
- And finally, as Rome burns, one begins to hear the notes of the media’s fiddling, tuned finely by their rapt observation of the cataclysm in progress, signaling that it’s time to begin the cycle anew.
By combining these factors, I believe we’ve created an arena for a new kind of blood sport, unlimited in its capacity for violence and unrivaled in its reach. Within its walls we encourage the emotionally-detached, psychopathic combat of addicts for the purposes of profit and political gain. We expend our time and energy watching, cheering, and jeering our fellow citizens as they do battle. And perhaps most tragically, while trapped inside we act out battles relating to the most polarizing and intractable problems outside the arena while doing little to address the issues around which we might otherwise unite. Sound familiar? To me it sounds like the dynamics of the Roman Colosseum.
And as in Rome, those of us doing battle are the most enslaved to these dynamics. While we might relish the short-lived glory of an opponent’s blood coating our intellectual weaponry, we remain addicts. It’s our need to pursue the next fix that provides the energy upon which this Ouroboros of chaos feeds. Therefore our best chance to stop this lies with us, the addicts, to cut off the supply of rage and replace it with something more productive. If you find yourself in the audience, a spectator watching addicts cut one another to the bone, consider either how you might help us break the cycle of escalation, or at the very least how you might otherwise spend your time. If the other gladiators out there are anything like me, they’d probably prefer the fireside conversation of the frontier, replete with its infinite freedom, to the slavery of addiction.
Beyond this, responsibility lies with those courageous enough to read the writing on the wall and lead initiatives to create more psychologically and socially sustainable digital ecosystems. It’s not so hard to envision a future in which our communication tools help guide us toward healthier and more productive conversations and away from our more Mephistophelean instincts. Of course, working against the short-term economic incentives of a company from within often proves futile. Though humans sometimes buck the trend, evolution tends to give rise to systems that optimize for immediate survival. And while at some level we understand the concept of sacrificing to improve our chances for a better future, we often only sacrifice appropriately after experiencing immense pain in the face of mistakes bequeathed to us by forebears who refused to do so. But if any such courageous individual wants to do something about these systemic issues in a non-partisan fashion, here are some suggestions.
Above all else, please don’t think you can solve the issue by restricting your users’ speech. Being Internet companies, I’d think you’d take the Streisand Effect to heart, but apparently not. As evidenced by Facebook, Twitter, Reddit, and YouTube’s responses to date, the consensus seems to be: block content we don’t agree with, or find NGOs aligned with our beliefs so that we may proxy the blame for such decisions to biased organizations with nice-sounding names. Congratulations, you’ve now used your own privileged position to punch down at users whose opinions you don’t share. And furthermore, you’ve increased the migration incentives and market value of any alternative platform offering safe haven from your policies. You may say “to hell with those bigots, let them go”, but that only serves to permanently entrench the current level of polarization and further balkanize our ideological landscape.
To create humane systems, apply the lessons of evolutionary psychology and behavioral economics to the mechanics of communication on your platforms. I know for a fact that companies such as Facebook and Google use such knowledge to increase the bottom line. Why not use it to improve the quality of communication as well? Ideas abound on this front. For example, perhaps use a bit of the machine learning budget to train systems that can recognize and point out apolitical biases such as confirmation bias or cognitive dissonance. How about an autocomplete that suggests less incendiary phrases and helps to re-unify language used to discuss controversial topics, which has been drifting toward linguistically separate-but-equal territory for the past 20 years? Or what about a system that can show, based on what you’re typing, the varying emotional responses you’ll likely elicit amongst friends? Perhaps this could use actual photo-adjustments (or eventually VR) a la FaceApp? Hell, emojis would be better than nothing. Of course, in isolation these might be used with malicious intent, as every technology has potential downsides. But I’d argue it’s possible to hedge against such abuse with responsible use of the suggestions to follow.
Another set of features ripe for improvement are the mechanisms by which users communicate the value of others’ contributions. Facebook has moved toward increasing the range of reactions over the past year, adding to the like button a small palette of emojis. However, without using these reactions to visibly update the user experience, these change strike me as window-dressing. Also, why can’t we map our reactions to a more diverse set of values? Doing so could enable the creation and evolution of a visual values indicator on our peers’ profiles. For example, do they spend all their time trolling, or do people generally find their comments to be balanced and insightful? Yes, it’s harder than a smiley face, but that’s why software engineers make the big bucks, right? Even if you fail trying, it’ll be a better story for your grandkids than “grandma spent her life getting people to click ads and compartmentalize rage until it boiled over violently in public”, no? Humans possess a wide range of value systems that intersect in myriad patterns. Perhaps task some of the brilliant data visualization specialists with creating new ways to visualize why different value systems combine violently versus peacefully, and foster public discussions regarding the implications by live-streaming for anyone who wishes to display them at the office, in their home, or within a classroom? Visualizations of this nature and the insights that flow from them can help us to re-frame conversations from stalemates to explorations.
Also, speed matters for more than just uploads. The time horizon over which we interact with information impacts our ability to engage in meaningful dialogue as opposed to knee-jerk reactions. Anyone who’s written something quickly and clicked submit, only to regret it seconds or minutes later, understands this. Engineers are familiar with the idea of rate-limiting requests and increasing the wait time between responses accordingly; why not do the same for messages we’re sending one another? Much like Netflix’s “perhaps you’d like to stretch your legs” messaging to politely disrupt binge sessions, rate limiting responses to conversations with increasingly angry language would give us time to cool off before responding, or realize it simply isn’t worth it. At the very least a notification might help. Furthermore these privileges and responsibilities could be mapped to a meritocratic reputation-building system that unlocks greater power within the platform as users demonstrate greater responsibility. StackOverflow uses just such a strategy to great effect. And to avoid internal ideological bias, such features could be implemented in a manner that’s at least partly open-source and subject to review.
Finally, given that these platforms know us so well, why aren’t they giving us concrete suggestions about how we might shape the world outside their walled gardens or expose ourselves to ideas at the boundaries of our comfort zones? Whatever value system one holds, it’s preferable that they go out into the physical world and test their ideas against their peers before becoming increasingly radicalized within a digital cauldron of resentment and anger. This applies to even the most unpalatable of value systems, as without the ad-driven engine of spectacle and blood such ideologies tend to manifest far more mundanely. Of course we’ll never rid the world of conflicting opinions, and that’s a good thing. But we can at least find ways to encourage the interpersonal re-humanization of those we demonize online. It’s easier to have a conversation before each side has donned their masks on the way to the flash mob, and it seems appropriate that those who have benefitted from the stoking of online conflict might help facilitate the reconciliation.
While I do believe the above suggestions would dramatically improve online discourse if implemented even-handedly, I’m not not holding my breath. The fact remains that the suggestions above would likely decrease engagement to some degree (good for society, bad for quarterly earnings). I’d actually bet that many of the above suggestions have been raised internally given the recent political climate. And it’s equally likely that they’ve been turned down by those who know they’d be risking a massive stock hit and a job search if they made changes that negatively impacted user engagement. Perhaps by following their current paths these platforms will sow the seeds of their own disruption. Or perhaps the pain will rise to intolerable levels and they’ll find within themselves the courage to enact meaningful change. In either case they should have enough respect for their users to take responsibility for the Colosseum they’ve built and the drama unfolding before our eyes.
Until then I’ll live the life of a recovering addict, pouring every ounce of myself into resisting the siren songs echoing across the digital ocean, pursuing this seemingly futile quest to starve the beasts of their rage, and knowing full well I’ll likely come up short, only to begin the count once more from zero. One, two, three…
But for now I’m free from their chains and off to explore new frontiers.
