Don’t feed the trolls, and other hideous lies

The mantra about the best way to respond to online abuse has only made it worse

According to the conventional wisdom of the internet, there’s one simple guideline for responding to trolls: don’t feed them. Ignore them, don’t react to them, don’t give them the attention they want. The axiom has become such a reflexive piece of advice and assumed knowledge that it can often be difficult to see the misperceptions and dismissiveness at its heart, the four hideous lies that perpetuate the cycle of misunderstanding, and the grim, cruel reality it has helped to enable in online culture.

The first great lie is about the sanctity of the past.

Recently, I tweeted about the pervasive nature of trolling and how people have always excused online behavior that is largely inexcusable. Almost immediately, a professor in London chastised that idea. He insisted that there was indeed a golden age for trolling, especially “for those of us who can actually remember the “eternal September,” the month in 1993 when a huge influx of America Online users began to overwhelm the online culture and norms of Usenet.

Reader, I laughed. It is unsurprising for a distinguished professor to engage in this kind of “gatekeeper” behavior. After all, his esteem rests on the fact that he knows certain things that others do not. Like all gatekeeper behavior, it was ostensibly a check on the credibility of the target. Also like all gatekeeper behavior, it wasn’t really about whether or not someone passes the test, but rather the gatekeeper feeling like they can control what is true and not true about the subject. Alas for him, I was there in 1993, too, equal parts young, naïve, and shy, but so damn excited about the idea of suddenly communicating with people around the world. This was a new thing, after all. And I will never, ever forget my first reaction to dealing with strangers on the internet: “Why is everyone so mean?”

From Usenet to early online forums, it was all the same: insults, flame wars, secret languages and inside jokes, and the glad-handing justification that always accompanied it: they were simply “trolling.” It was all just a joke. When my confused little brain bumped up against this notion, there was an immediate pushback with the general sentiment of “forget it, kid. It’s internet town.”

Some other gatekeeper might come along and say, “Well, you just had to be there for Arpanet or [insert whatever period comes just before your own].” But it doesn’t matter. Whether we’re talking about AOL, AIM, early 4chan, or the early days of Twitter, there has always been a myth about the time and place where things were more innocent, when trolling was all in good fun. But what everyone really remembers about these proverbial times isn’t their purity. It’s how they didn’t see the big deal back then. They remember how they felt a sense of permission, a belief that it was all okay. But that was only true for those who were like them, who thought exactly like they did. All the while, someone else was getting stepped on and bullied while others laughed. The story of the internet has always been the same story: disaffected young men thinking their boorish and cruel behavior was justified or permissible.

And it was always wrong.


The second great lie is that trolling is harmless.

Trolling is broadly defined as “a deliberately offensive or provocative online post with the aim of upsetting someone or eliciting an angry response from them.” That can imply a lot of different things. Typically, it speaks to insincere motives, like saying something you don’t mean in a political discussion just to upset someone. But trolling can also encompass any kind of willfully obtuse nonsense that’s designed to confuse people. For instance, I know someone who posed as a major fan of Creed (the band) and started an online petition to get the name of Creed (the movie) changed so it wouldn’t confuse the fan base. It seems laughable, particularly because Creed (the band) is a popular punchline. But it also makes you consider the actual point: was it poking fun at the Creed fandom? The band itself? Or just people who easily believe in absurd things?

The truth is that all trolling, whether we admit it or not, has a meaning and a target. You are inherently saying, “This subject is worthy of mockery,” which is exactly why John Oliver’s specific brand of trolling stunts have such laser-targeted focus. He takes on bureaucratic institutions, high-powered tyrants, homophobia, and social issues in an approach that embodies the very definition of “punch up” in comedy. It also reveals the core problem of trolling that so much of the online world wants to ignore. It is inherently an act of satire, something that comes with real targets and real responsibility. But the core intent of trolling is the opposite: it’s not just to provoke, but to run away from the responsibility of the joke itself.

A Twitter follower reminded me of a line in the famous parable from Bion of Borysthenes: “Boys throw stones at frogs in fun, but the frogs do not die in fun, but in earnest.” Defenders of trolling insist it’s all just a joke, but if trolling is inherently designed to get a rise out of someone, then that’s what it really is. In many cases, it is designed to look and feel indistinguishable from a genuine attack. Whether you believe what you are saying or not is often immaterial because the impact is the same — and you are responsible for it, regardless of how funny you think it is. It is a lesson kids learn time and time again on the playground, and yet, it is ridiculously difficult for people to accept the same basic notion in online culture, no matter their age. Why is that so? Because those are the social norms that develop when you create a culture where everything is supposed to be a joke.

It’s no accident that the corners of the internet that subscribe most deeply to this idea are also the most openly miserable. While some clearly use “joking” as a justification for abuse or even violent threats, there’s little larger comprehension or interest among huge swathes of internet culture about how satire, irony, or intent actually function, much less in the distinction between what they consider “trolling” and actual abuse. Drawing such lines would be against both the protocol and intent behind the creation of internet culture at large — a culture that was designed to escape the responsibilities of the social order. In that pursuit, internet culture subconsciously turned itself into a calloused nub, a place where so many “jokes” are the equivalent of running and shouting “fire!” in a movie theater, and a place where the biggest joke of all is the idea of caring about anything in the first place.


The third great lie is about what fixes it.

One of the most popular solutions that arose in online culture was, again, the mantra of “don’t feed the trolls.” This meant that any time a troll popped up in an online situation making inflammatory remarks, you were supposed to ignore them because responding would derail the thread and give them the attention they wanted. What no one seems to remember is it never worked, practically on any level. There was always someone who wanted to troll back in the opposite direction, someone who genuinely got offended for a personal and valid reason, or someone who wanted to try to be reasonable. Instead of solving anything, “don’t feed the trolls” became a motto for people who want to act above it all or regale us with stories about how much harder it was to troll back in their day when they had to troll uphill, both ways! But most of all, it became the mantra of how to ignore online abuse completely.

The premise of “don’t feed the trolls” implies that if you ignore a troll, they will inevitably get bored or say, “Oh, you didn’t nibble at my bait? Good play, sir!” and tip their cap and go on their way. Ask anyone who has dealt with persistent harassment online, especially women: this is not usually what happens. Instead, the harasser keeps pushing and pushing to get the reaction they want with even more tenacity and intensity. It’s the same pattern on display in the litany of abusers and stalkers, both online and off, who escalate to more dangerous and threatening behavior when they feel like they are being ignored. In many cases, ignoring a troll can carry just as dear a price as provocation.

It all harkens back to Cliff Pervocracy’s analogy of the “missing stair,” where everyone works around the obvious dangers of a situation because they are so used to “dealing with it” by outright ignoring it. If someone speaks up about the danger, they are dismissed. Why complain when you can “just hop over” the missing stair? But on a systemic level, it all adds up to something so much more than a mere missing stair. For many people on the internet — especially women, people of color, and the LGBTQ community — it is an entire broken staircase, full of loose nails, jutting floorboards, and impossible leaps. And there are so many others who don’t notice it because they either get to use the elevator or are already on the top floor.

Not only does this sort of ignorance function as a kind of tacit permission, but it also ignores the inherent threat of the troll’s true intent. What the troll, the stalker, and the abuser really want out of the situation is to feel powerful and in control. And they will not stop until they feel it. Therein lies the most horrible aspect of the “don’t feed” mantra: rather than doing anything to address the trolls, the more tangible effect is to silence the victim and the reality of their abuse, or worse, to blame them for it. For far too many who promoted this idea, the true goal was silence, to avoid facing what is happening and the impossible responsibility of it.

“Don’t feed the trolls” also ignores an obvious method for addressing online abuse: skilled moderation and the willingness to kick people off platforms for violating rules about abuse. At one website I used to write for, everyone constantly remarked that we had the most amazing, thoughtful commenters. How did we achieve this? Easy: a one-strike policy. Complete zero tolerance. Did people complain? Of course they did. But it stopped people with bad intentions from being a part of the community, and it kept all the well-meaning people on their best behavior. It wasn’t perfect, but it was good.

It also a took a ton of effort on the part of the entire writing team. We had to ignore the other popular sentiment of “don’t read the comments” (which is largely about trying to maintain sanity while staring at the void) and embrace a jaw-droppingly obvious fact: what truly derails any given thread or conversation lies not in a given response to trolls, but the very troll who is trying to derail in the first place. The second you treat them as a “constant” or inescapable part of your community, you have given them permission. You make them a missing stair. And the impact of doing so is only exacerbated when you scale up.

Facebook, Twitter, and Instagram are now so large that they are considered “unmoderatable” communities. We like to pretend this was a pure facet of their size, but it is inescapably a part of their ethos. They are platforms forged in the fires of troll culture, founded and operated by techno-libertarians who didn’t understand why they had to care about any of this. They set out with no intention to moderate at all. Zuckerberg just wanted to rate hot girls, after all. But in 2018, the staggering effects of non-moderation are just starting to hit them, and they have little idea how to address or even intellectually engage with the idea.

It starts by acknowledging that these systems are so large and pervasive and such an important part of people’s forward-facing lives that it is intrinsically necessary to protect the well-being of the people on it. For many, social media networks are a huge part of not just how they socialize and connect with other people, but how they do their jobs. These platforms have succeeded in making themselves indispensable to many users, which renders absurd the suggestion that the abuse festering there is something people can easily “opt out of” by not participating.

When Zoe Quinn pursued legal action for the horrors of Gamergate, she was frequently confronted with this so-called solution by police officers and even the judge who decided not to issue criminal harassment charges against the man who orchestrated an online harassment campaign again her: just get offline. But as Quinn wrote in her book Crash Override, “The internet was my home, and treating it like a magical alternate dimension where nothing of consequence happens was insulting. Telling a victim of a mob calling for their head online to not go online anymore is like telling someone who has a hate group camped in their yard to just not go outside.” The consequences of this attitude are very real. In today’s online world, people can claim the power of a threat with none of the consequences of actually making a threat. Just last week, Milo Yiannopoulos called for the shooting of journalists. Then, when someone did exactly that, he quickly insisted that “he wasn’t being serious.” This is the heart of trolling, especially when it’s built around the intent to terrorize.

But this is all really happening. And the large-scale internet needs the figure out the way to guarantee the same protections as smaller communities by moderating with a sense of decency and displaying the same basic sense of judgment as a damn open mic night. (There is a reason Michael Richards is not asked back to The Laugh Factory.) The powers that be in social media can’t just make it about who is saying bad words, try to algorithm their way out of the problem, or play every side in the name of “fairness” when it leaves so many of us to the wolves. They have to make an ethical choice about what they really believe and what ideology they want to represent moving forward. Because they cannot reap the reward of what they have built without taking on the responsibility and the cost of it, too.


The last lie is the one that says any of this is simple.

It is not. Online abuse is infinitely complex and human, and there are no easy life hacks for solving it. No one can “stop” it as though it’s a singular entity careening down a street. Nor can they address it without addressing larger societal problems (especially toxic masculinity). It is part of a systemic reality, and as such, it needs large-scale systemic solutions. But any solution needs to start with honesty and identifying the given problem as it exists: we are simply too permissive of “troll culture,” and we always have been.

The toll it has taken is already enormous. For nearly a decade, I approached Twitter with the idea that I would try to be kind and understanding to people no matter what. I was far from perfect at it, but I genuinely tried. And year after year, I became more and more frayed from the constant stream of abuse. If we’re going to speak in superhero code, the only “joke” that the Joker is telling is that of sadistic terror. And to say “I’m burned out” on that particular joke is as colossal an understatement as I can make. I feel like a husk, a walking short nerve. And compared to many, I haven’t even gotten the worst of it.

It’s easy to confuse this sobering despair with cynicism, but as Quinn and so many others put before, we cannot “cede the internet to those who scream the loudest.” You have to fight to claim space for decency. But like democracy itself, that means fighting a war with one hand tied behind your back. Occasionally, it means acknowledging the most horrible truth of all: that once in a blue moon, talking works. Whether with a troll or those that give permission to them, sometimes you can actually get through. But this work also comes with its own grinding weight, which is why we can never make it “the job” of the abused to defend their humanity or explain over and over why decency should be the norm — or even simply why people should care about other people. That’s why it’s so critical to step up and do what you can to defend or empower the most marginalized people in our society. But in the end, the power to change online culture is not reliant solely on our ability to engage or explain, but on something outside of our control: whether or not the other person can open their heart and listen.

The biggest mistake we ever made with trolls was making the question of abuse about how to placate and fix them instead of how to empower the people they hurt or manage your own well-being in the face of them. Like so many abused people, we thought the solutions involved walking on eggshells and not provoking them back. But instead, we must acknowledge “that we are what we pretend to be, so we must be careful about who we pretend to be.” And that means acknowledging the awful, terrifying power of jokes and the immunity we seek in “not being serious.” This is exactly why people troll in the first place. Because deep down, they know it’s serious, and that’s exactly why it makes them feel powerful.

This brings us to the only thing I know for sure in all this: often, the online abusers win because the game is set up for them to win the moment they decide to play. They have the power to hurt, deeply and profoundly. They always have. And don’t worry, they’ll discover where this game takes them with time. But when it comes to how we respond to them, our tactics can and will vary, and they may involve anger, humor, love, tolerance, blocking, or maybe even some productive discussion. But ultimately, if we care about abuse, we cannot care most about whether we have comforted, converted, or even fed them.

We have to care more about the people they hurt.