The Impact of Toxic Influencers on Communities

16 minute read

There’s a common character that can be often spotted in many online communities. It’s the “toxic influencer” (sometimes referred as “intellectual bully”). A toxic influencer is a person that covers a useful role in a community, but at the same time creates a toxic environment around them that can become bullying in some instances — yet they never over-step the line of proper abuse, so they can’t be easily managed.

These are the three traits of a toxic influencer:

  • Needed — they have a level of knowledge on a specific topic, or they have time, or both, and as such they found a spot in the community that isn’t covered well, and they took it for themselves, in practice providing a needed service to the community. 
  • Toxic — they create a very tense environment around them, usually with a mix of passive aggressive takes, constant poking on people’s nerves, acting condescendingly, and general low-key abuse. Note however that this doesn’t mean they are antisocial: they are likely to be very good at talking, possibly even very charismatic.
  • Lawful — This is a key difference of toxic influencers: they operate within the rules very well, both implicit and explicit ones. They use their intelligence and awareness, so they are very careful in never over-stepping the community rules in terms of abuse and mistreatment. Even more, in some cases they weaponize the existing policies for their own gain.

Note that this can happen in any community, but has a stronger tendency to happen when the community isn’t purely for socialization but there’s a shared goal with technical skills involved — like an open-source community, or a do-it-yourself community, a sports association, and so on. 

“Bullies are generally defined as people who intimidate or control others to achieve their aims. They may collaborate when their goals are being met, but they lack fairness or honesty. Workplace bullies generally manipulate or terrorise those with status below or equivalent to themselves. They may also intimidate superiors, such as threatening to resign at a critical point”.

Dr M Lamia (2017) ”The psychology of a workplace bully”.

A Story

I used to be a community manager, in a couple of large online communities around 2005, and a few stories come from these days. In one of them specifically I was also an “admin”, so while I was directly moderating some forums, I was also part of a smaller group with higher access. For instance, I co-authored the rules for the forum that have been active for over 10 years until the forum shut down, with minor changes.

I take a story from these days because it’s deep in the past, as I’ve no interest in calling out specific people in more recent years. 

One of the forums was the Tech Area, where people discussed computers in a broader sense, but more often asked advice on something not working or on something they wanted to do. One day, a guy started to show up. He went through all the threads on the main list, and replied to each one of them. The technical part of the reply was good, it’s just that the general message was “You’re an idiot, and because you’re so stupid, I’m coming here to save you, here’s how you solve the issue”. But to be clear, never directly insulting, never that explicit, never that obvious.

At first the moderators started to ask him in a few threads to tone it down, we were a friendly community, and we wanted for people to feel supported.

The reply was always “but I’m helping them, see?”.

And he did, sure. Over time, some people started also liking him: “but he’s always here, he replies fast, and provides solutions, he’s a good person”, “I had a problem and he solved it in just a few hours!”. Moderators had a hard time keeping up as they were all volunteers, while the guy was seemingly very frequently checking the forum and replying. When asked about “but you don’t have a job”, he replied that he had his own company, was rich, and owned a Ferrari.

The problem was that technically there was no explicit violation of our guidelines – this was before I co-authored the new ones, which made it easier for moderators to take action. Unsurprisingly these rules were partially based on the experience we had with this guy.

So over time what happened is that the Tech Area stopped being a small community, because he was always there first, nobody could help. Also there was no conversation possible, because his replies were always framed as absolutes — even when he was obviously and demonstrably wrong.  It was a one-man show. A lot of people stopped participating, coming in just asking when they had problems, and moving to other forums.

The few people remaining split into two groups: the people that were accepting him as superior, and thus he rewarded with compliments and protection, and the people that didn’t, who started becoming antagonistic and trying to one-up him by spotting mistakes and replying before he could. 

The community was basically gone, until one day he disappeared. The community started getting better, some people started being more skilled at replying and helping… but a few months later he was back again. This cycle repeated a few times.

In the end, it was a combination of him making a mistake, and him walking away: he got a temporary ban (a few months), and he never came back. We got lucky, but this happened only after years, and the forum was basically dead after that. Some people started mirroring the behaviour of the toxic influencer that left. It took months for that forum to recover.

How they come to power

There’s a common dynamic when these people arrive. Usually it starts with a gap in community needs. This could take many forms, and doesn’t have to be a big need. Sometimes it can even be a need that wasn’t even perceived, but when the person arrives, they do good, and the community responds well.

That person is helpful!

Unfortunately this person also starts doing — either consciously or unconsciously — things that create a very tense environment. This can happen in multiple ways, here’s a few behaviours:

  1. Polarization —this is probably the most dangerous. This person will make sure that people, especially new people, are inoculated with the idea that there’s a good group (the one led by them) and a bad group (everyone else). They might emphasise every normal human misunderstanding to reinforce this idea, until people just naturally side with them. This basically uses normal social group behaviour to their advantage, by creating a fictional divide and reinforcing it.
  2. Passive Aggression — everything they don’t like, gets a negative note from them, sometimes even taking things happening in one discussion, and mocking them in another, in a way that only the two people involved will likely notice, and everyone else… will just see one annoyed person snapping back seemingly for no reason. Unsurprisingly, this reinforces very well the us-vs-them behaviour.
  3. Swarming — they reply every (or most) of the discussions. They are there, they feel everywhere, so people start being on alert about getting an answer from them. If they are smart, most of the replies will be positive, constructive, and harmless — which again unconsciously contributes in creating a positive halo around the person.
  4. Repetition — issues are constantly brought up, even if they seemed settled in the past. Sometimes they might even take some time to prepare the ground, get people to agree with them first telling their part of the story, before raising it again, so they have reinforcements ready. There’s no closed case for them, until they get it their way.
  5. Judgemental — they know, and they make their knowledge visible and loud. Even more they make people feel the emotional weight of being judged by them as inadequate or not knowledgeable enough. This ends up making people either avoid learning and going away, delegate everything on the topic to the person (an outcome that reinforces their role in gap-filler), or challenge their take directly (thus creating an even stronger polarization).

These behaviours — and more — might be mixed differently, but the outcome is always the same: they acquire status, sometimes even formally, they get a group of people protecting them, and everyone else is just a bad person against the good they are doing — remember, they are filling a community gap!

New people coming in either accept their behaviour, and become part of their group accepting their abuse, or walk away. You can imagine how harmful is this to the community as a whole, because the only people that join at this point are… people like them, or close to that kind of behaviour. No diversity, no growth, nothing is possible. There’s no community anymore.

You can see another example of this in Linus Torvalds, even if in that case it was his own project, not one he joined. It’s important to capture the essence of these behaviours, not the exact sequence of events, so we aren’t skewed by how this plays out, as it can be different from community to community.

How to spot these people?

It’s hard, there’s not a shortcut here. It’s hard specifically because they represent a duality between doing good, and doing bad. And in abstract, that’s what all humans do. So why is this a type, and how can we identify it?

First of all, it changes a lot if you’re part of their group, or outside. And if outside, if you’re close to them, or not.

If you’re part of their group, your perception is going to be biased in their favor, and it’s likely you’re one of the people defending them. It’s hard if you’re in this position, because of course you like them, and there’s a reason why you’re part of the group. The most important thing to notice here is that yes, there’s a group you’re part of, and yes, there’s a dynamic at play that is hurting others. Also notice if you keep protecting someone, regardless of the context.

If you’re not part of their group, and you are not close to them, then it’s likely impossible to even see that something is going on. The behaviour is very subtle, and a lot of the passive-aggressiveness happens across many discussions. For example what you’ll likely see is suddenly someone snapping for a “normal” comment… that however is not normal, as it was referencing a previous, separate discussion. The only way to see, is to get close. Check their interactions, see how it happens. 

If you’re not part of their group, and you are close to them — either because you need their skills, you’re working together, or else —  then you are likely feeling the impact directly. Unfortunately, what you notice is that when you try to speak up, this person is immediately protective, and protected. And nobody else seems to be saying anything, so you end up being gaslighted into thinking it’s you.

The problem is… none of these signs are universal and a sure way to tell what’s going on. There are many other scenarios where these descriptions can fit. We can try to use the breakdown of traits and behaviours into some questions — keep in mind however that answering positively to a question doesn’t necessarily mean there’s a problem. It’s the combination that makes the type, so it needs to fit the whole profile, not just that single question.

Trait questions:

  • Needed — is there a gap in the community that only one person is filling? is anyone considered irreplaceable? are they taking the spotlight for themselves?
  • Toxic — are people pushed away from the community? does this person leave a bad feeling after interacting with them? do multiple people feel the same? do they act very defensively to any criticism and blame others? do they show their skills but never to educate, just to argue?
  • Lawful — is this person manipulating the rules to their own personal advantage? do they use written and unwritten rules for their personal gain instead of community gain? are they more interested in the rules than in the goal people are trying to reach?

Behaviour questions:

  1. Polarization — do they keep reinforcing the idea that there’s a “us” and a “them”, usually through generalizations? do they try to bring new people on their side? do they keep blaming others?
  2. Passive Aggression — are their insults subtle? are they stubborn? are they irritable or very defensive when criticized? do they complain about not being appreciated? do they keep bringing up personal attacks, in a way that is not actionable?
  3. Swarming — are they present in many different spaces at the same time?
  4. Repetition — do they keep bringing up old issues and asking to resolve them again and again? are they able to accept a decision taken by others if they disagree?
  5. Judgemental — do people fear the judgement of this person and are relieved when they approve?

Take these as a sort of scoring system. If all the questions score high, then, you might have a case of a toxic influencer — it’s still not certain, but at that point you’d want to look into it more closely.

Someone that is just a negative person, but doesn’t have a damaging effect on  the community, isn’t a toxic influencer per-se. Also, we all make mistakes, we all try to create our own group, we all have some form of passive aggressiveness, we all have that old issue we bring up. It’s the combination of factors, and the effect of the community, that is different.

The Psychology of the Toxic Influencer

First of all, this is not a diagnosis. I’m not going to attempt to state something about their inner thoughts, as that’s work for a therapist, which they should definitely go and see.

However, when dealing with abusers and bullies, I think there’s value in recognizing that it’s a consequence of some form of their own suffering. I’m asking to empathize with them, as it’s useful. I know this is not easy if you’re part of a community that has this character, so it’s ok to just do it briefly. It’s also not meant to be an attenuating factor. Why they do something doesn’t take away anything from the harm caused.

One of the possible dynamics at play is that they are using knowledge gain to ease their own suffering. In their past, they found out that every time they gained knowledge and expressed it, they were able to get positive feelings from others. This became a virtuous circle based on their own smarts combined with their own fears. This unfortunately also made them avoid working on their suffering itself, so they just have their coping mechanisms, and that’s why it’s dangerous to question their knowledge: it’s like asking them to face their suffering.

You can also see how knowledge and social recognition then also fuel their ability to socialize, because that’s where they get validation. So that’s another self-reinforcing loop, where they acquire more and more social skills.

“People wrongly assume bullies have low self-esteem, but their behaviour is actually a response to internalised shame. Although some people who live with shame have low self-esteem, those who behave like bullies tend to have high self-esteem and hubristic pride. They attack others to take away their shame – which allows them to remain unaware of their feelings”.

Dr M Lamia (2017) ”The psychology of a workplace bully”.

This means it’s hard to make them “understand” or “see” their own behaviour, because they are driven from a deeper kind of fear. This is also why they will likely perceive any attempt to stop them as aggression: when an abuser is given a boundary, to them it will feel like abuse.

What to do?

Taking action is very hard: they will always have a group of people defending them. These are selected people that over time joined because they tolerated their behaviour, and now they enjoy their protection. They can’t see the toxicity, they just see the benefits to the community.

How can anyone try to attack someone that does so much good to the community?

It’s also likely that everyone else will agree with you, and they would appreciate it if that person wasn’t there anymore. However, it’s likely that they aren’t going to speak up. They have better things to do, especially after they have tried already to do something in the past.

This is often the cycle:

  1. A new person comes in.
  2. They notice the abuse.
  3. They try to do something.
  4. The clique protects the abuser.
  5. Everyone else is too tired to support, so they stay silent.
  6. The new person, unsupported, retreats.

By retreating they either now abandon the community, like all the others that didn’t even try to fix it and just left, or they stay and build their own defense mechanisms to tolerate the behaviour: “I’ll just do my part, and try to ignore that person as much as I can”. It’s unlikely they will join the clique because they recognized the toxicity in the first place.

The options here are limited, exactly due to the protective dynamic outlined above:

  1. They make a mistake clearly against the rule.
  2. They decide to walk out on their own.
  3. Someone else comes in challenging them, and is able to take their spot.
  4. The community leads make a very unpopular decision to remove the person even if no explicit rule was broken.

Some people might think: isn’t there also an option to make them understand, and “fix” them? Yes, in theory, there is. Yet, it’s likely to take someone with the skills of a therapist, the will of the person involved (it’s impossible to change someone that doesn’t want to change unless cornered, and even there, might change in the wrong direction), and multiple years of work.

“Asking nicely has its limits, eventually you must take action.”

J. Atwood (2015) “Handling an Intellectual Bully”.

If your community has a toxic influencer, there’s no easy way out. Review the above, integrate with some readings on the psychology of bullyism and community management, and prepare a long term strategy.

If your community doesn’t have a toxic influencer, note that some research identified how prevention is usually the most effective strategy. You really want to be prepared to manage them before the community is impacted.

One important note: this can’t be fixed with just a policy.

They will use it to their own advantage.

Good luck.

Further readings

Thanks to Tammie Lister for providing a sounding board while writing this, and some insightful ways to say difficult concepts.  Thanks to Andrea Middleton and Anne McCarthy for a review and smoothing some sharp edges.