If you’ve been in online public spaces for any amount of time, you may have noticed a strange phenomenon. Anytime politics are mentioned, people turn straight up evil; they become racist, radically anti-establishment, dooming, conspiratorial, etc. This is shocking because it does not reflect reality, most people don’t behave like this. Then, why does it happen online?
This topic is important and is covered by a lot of literature, research, and personal opinions. Common reasons include:
- Bots from state actors meant to sow political divide
- Algorithms rewarding outrage
- Anonymity shielding users from consequences
- Selection bias, we notice toxic comments more than neutral or positive ones
- Silent majority, most civil people don’t interact online
- Echo chambers with toxic environments developing toxic cultures
These are all true to some degree, but they don’t paint a full picture, and, in my opinion, miss the most important color.
These do not explain certain behaviors, most notably YouTube comments from news videos. This is a fascinating phenomenon I recommend you see for yourself. The comments of popular news outlets (NYT, BBC, etc.) often only contain from 2-30 comments since these networks post so many videos. Many, and sometimes the majority, of these comments are being completely deranged and toxic while not meaningfully interacting with the video. This seems routine on the web, except that almost none of the factors above apply well to explain this:
- YouTube comments have little to no algorithm, they’re just sorted by recency and likes (irrelevant for short comment sections)
- Users are shielded from consequences but those comments often aren’t interacted with and see no consequence to shield from
- They’re easy to count and obviously disproportionate
- No toxic culture since news networks cover extremely broad topics in a tame, neutral and objective way
Although, news networks are often the target of bots, and there are way less comments than views. But still, I don’t think those two factors explain this consistent phenomenon perfectly.
I think the real culprit is something more fundamental: the credibility problem. When you interact with someone in person, you don’t start from zero. Before anyone speaks, we’ve already assigned them some baseline credibility based on appearance, how they carry themselves, their accent, whether they seem like part of the community. This social capital means they can afford to be measured, admit uncertainty, or explore nuance without immediately being dismissed.
Online, especially anonymously, everyone starts at zero credibility. Or worse, they might start below zero since there’s an assumption that random internet commenters are probably idiots or trolls. This creates a desperate situation where users need to establish that they’re worth listening to within the comment itself, alongside making their actual point. Combined with the universal human need of feeling valued, the result is rhetorical escalation, like bold claims, absolute certainty, provocative language, tribal signaling. This is exactly the kind of rhetoric-over-truth that Socrates warned about, but it becomes almost rational when you’re fighting for credibility from nothing.
This explains why those YouTube comments are so weird. These commenters have political opinions, but they know that if they just state facts, nobody will bother to listen to them, they will get shadowed by more rhetorical comments, and they might be the target of more rhetorically active users.
There is also the phenomenon of Redditors, those who do not know much about a topic, but want to appear smart and knowledgeable, so they mask their ignorance with rhetoric.
This same dynamic explains something else that’s rarely connected to toxicity: why humor dominates so much of online interaction. Being funny is one of the most efficient ways to gain social capital, online and in real life. Make someone laugh and they immediately see you as intelligent and worth reading, and it’s one of the most attractive traits in romantic relationships. Online, it’s easier and low stakes compared to making a serious argument, and it works across tribal lines in a way political takes never could. I’d argue that humor-driven comments vastly outnumber troll comments on most platforms, though you wouldn’t know it from the discourse about online behavior. On a typical YouTube video about gaming or music, the top comments are almost always jokes, references, and witty observations. Or in other words, people engaging in jolly cooperation.
I never liked the anonymity shielding and silent majority theories because they assume an universally bad inherent human nature. So, for this reason, I think the credibility problem best explains most behavior in online public spaces, not just toxicity, and is more rooted in real-world logic than all of the previously listed reasons.
This leaves me to wonder why the credibility problem is one of the lesser-discussed reasons for toxicity online.