Behaviour and Community in Open Science

By Os Keyes

In response to the latest incident of bro misbehaviour in open science, Jon Tennant has written up a very polite plea for civility and good behaviour in open science advocacy. Despite my general love of good behaviour (I’m an ethicist and former community manager) I disagree with pretty much all the well-intentioned advice it contains.

There are a lot of individual bits I could dig into (you embed the tweet of a woman in STEM already complaining about being harassed by open science bros into the tweet you broadcast to ten thousand people, including open science bros? really?!) but the big thing is those final three paragraphs:

What we have to remember is that Open Science is a social movement. This means that we have to be sure that all communities and demographics feel welcomed and invited to participate in discussions, and not excluded from them. It can be easy to let our emotions and frustrations and interfere with this, that much is understandable. But we can be better and stronger as a community by actively preventing this. I’m not innocent either, and have in the past said some things I regret to others on social media. What matters, though, is using those experiences to become better for ourselves and others.

When we see bad behaviour, from and towards anyone, it is each of our responsibilities to call it out. Preferably via a civil note in private, rather than any sort of public shaming. If we see this sort of behaviour happening and don’t try and stop it, then we’re complicit in allowing it to happen. This is especially the case when it is happening within our own communities or social circles. Don’t turn a blind eye, as that never solves anything.

And it doesn’t matter that this sort of behaviour is actually fairly widespread on social media. What matters is being better than that. Taking the moral high ground, engaging with those who you don’t agree with in a courteous manner, listening instead of insulting. These are all things that we should be practising on social media, because that’s just doing this stuff right. At the end of the day, we want a social environment where people feel welcomed to share their views without fear of retaliation.

This is bad advice. Not the idea of ‘we should be nice to people’, that’s great - but specifically, the argument that if you spot bad behaviour happening you should politely call the person out in private, explicitly avoiding public callouts.

Not only is it bad advice, it’s avoidable bad advice, because every longstanding community - including the (non-open) scientific one - has been dealing with this problem for years, and any halfway decent community manager or simply any marginalised person could look at it and go ‘ahhhh’.

To break it down: misbehaviour, normally by white dudes you could buy twenty times over for what they’re worth long before you hit what they think they’re worth, is not a novel thing. It’s a wider pattern replicated in pretty much every field of western endeavour due to a hell of a lot of social conditioning. Because of this, communities much older than the open science community, and the people they contain, have already tried solving it. They’ve been doing so for kind of a while, and members of those spaces (particularly but not exclusively marginalised members) have a good idea of what works, what doesn’t, and what the consequences of common attempted solutions are.

The consequence of emphasising that callouts be civil and private…is that harassment and abuse continues. Within your community, and outside it. If someone is harassing or abusing a community member, and the emphasis is that callouts be private, the person called out is free to keep harassing more just so long as they do it where the original complainant did not see so that nobody spots it’s a pattern. If the emphasis is that callouts be private, even if people do put 2 and 2 together and boot the harasser, they are free to just mosey along to the next community along and continue their pattern.

We know this is what happens because we’ve seen it, in a ton of spaces, scientific ones included. If you look at the harassment-in-STEM-departments controversies over the last few years, you’ll notice a common marker: there was always more than one victim and incident, and people didn’t take it seriously as a problem to be solved until it was public, which is what let the harasser cross the line so many times.

Furthermore, private callouts invite newer members of the community to internalise the entirely wrong lesson. If your callouts are private, the pattern that a new community member sees is: people are talking, someone misbehaves….and nothing happens to them. The lesson from this is: that kind of misbehaviour is okay in that space. And so you go from having one asshole to having two, and four, and eight, and it doesn’t matter if you call out individual assholes privately because they’re reproducing memetically.

On the other hand, what happens if you do a public callout? Well the first thing that happens is that the harasser suffers actual, genuine consequences for their behaviour - which is a good way to ensure they think twice about doing it again. The second thing that happens is other people see those consequences. Marginalised people - perhaps people the harasser has messed around before, who have been reluctant to say anything given the lack of any suggestion complaints would be met positively - know this is a place where they can expect at least some kind of shielding and support. Non-vulnerable people, particularly those who are inclined to assholery, know that this space is one in which assholery will not be tolerated.

In other words, you get a space where marginalised people feel supported, non-marginalised people feel like there are rules, and assholes feel like they should shape up or ship out. That seems pretty much ideal for a set of community norms, as far as I can tell. And the alternative - private callouts - does none of this. Instead it actively protects the people causing the problems by allowing them to do whatever they want as long as they be sure to be caught by different people each time. It, and its emphasis on ‘civility’, prioritises the feelings and reputation of the person causing the problem above the individuals suffering.

Again: this isn’t new. This is community dynamics 101. This is about missing stairs. If you want to create a comfortable and open environment, that means you need to very visibly deal with those who instead want it to be their own private gentleman’s club. Picking a process that emphasises their reputation, their feelings, does not do that. Picking a process that makes damn clear misbehaviour won’t be tolerated does.