This is one of the essays I delivered to my patrons last month. If you want to support more work like this, and see it earlier, you can sign up here.

There are plenty of reasons that discussions around codes of conduct, their goals, and what should be included in them can be frustrating. Among people talking in good faith, however, the primary reason for frustration is that people are using different theoretical frameworks for codes of conduct and they’re not making those frameworks clear. It’s obviously difficult to come to any agreement about what should be contained in a code if you don’t agree on why you have one.

So here are several common models people use when discussing codes of conduct. Hopefully, this guide will make it easier to explain where you’re coming from when advocating for a particular code and to recognize both what model someone else is using and where the strengths, weaknesses, and unspoken assumptions of that model are.

Free Speech Model

In the wild, this is almost never a code of conduct policy model. It’s much more likely to be an argument against having a code of conduct rather than a framework for building or enforcing one. One of the better examples of this is this email sent to the mailing list for Sasquan, the 2015 Worldcon, under the title “Harassment Policy -> Freedoms Policy”:

Thinking about a ‘harassment policy’, I am wondering:

We’re a diverse, creative bunch, and many of us are excellent and exacting writers.

Rather than constructing a Soviet-like, PC-nightmare policy, what would happen if we took the approach of enshrining members’ individual and collective freedoms of speech, beliefs, activities and associations?

Could we forge a policy that sounds and feels more like the US Bill of Rights, and less like the USSR’s Article 58?

(Reference: http://www.cyberussr.com/rus/uk58-e.html)

“Notice: you may encounter divergent political opinions, and verbal and written statements in print and on clothing at variance with your religion or irreligion. You may encounter any manner of expressions or artworks that you personally consider profane, or reverent of concepts you may personally consider repugnant. Too bad, this event takes place in a free country.”

(Would love a 2nd amendment equivalent here too…)

(3rd amendment might not be necessary, but an equivalent might admonish convention staff from entering members private rooms, or be a reminder that convention policies have no effect in private rooms or outside of convention space.)

“All members have the right to be secure in their persons and personal effects, and free from unwanted physical contact.”

More ideas?

The free speech model views codes of conduct as restrictions on people’s free behavior, which is correct, and antithetical to life in a free society, which is incorrect. There are two major misconceptions that go into this model.

The first misconception is that having these rights enumerated under the Bill of Rights (or any other country’s declaration of rights) means setting out rule-free spaces. These rights weren’t written down in our foundational documents, then magically existed because we said they were important. Every right claimed requires that we restrict other people’s behavior that infringes on that right. In fact, sometimes these rights even come into conflict with one another, requiring the courts to settle which right has priority in a given situation.

Rights require rules and limits.

The second misconception is that there aren’t already restrictions on speech and other constitutional rights at the events adopting codes of conduct. They may be implicit instead of explicit, like the “Shut up and take it” buried in “Too bad, this event takes place in a free country” above, but they exist.

Don’t think so? Consider what would happen if a panel discussion were interrupted by a group of people taking turns telling a panelist that their views were repugnant for the full panel time, even after a moderator told them to stop. Not only would everyone understand that this was outside the bounds of accepted behavior, but many of the attendees would be upset if organizers allowed the behavior to continue.

Rights are already not absolute in these contexts.

Given all that, however, there are still good reasons to think about protecting speech when creating a code of conduct. There are topics we want to carve out space for, as sex activists did when they suggested modification of the Geek Feminism wiki sample code of conduct or as atheist groups did when they carved out space to criticize religion without allowing harassment of believers. Free speech may not be a good theoretical framework for a code of conduct, but as a consideration, it can be a useful framework for thinking about the limitation we may want to build into our codes.

Criminal Model

There are two versions of the criminal model of thinking about codes of conduct. They may occur together or separately.

The first version of this model can be distilled down to “Harassment is already a crime. This should be a matter for the police to handle, not event or group organizers.” For an example of this in the wild, you can see the original policy adopted by the 2015 World Fantasy Convention.  It basically says, “We’re not trained to determine what harassment is, but the police are, so we’ll get the local police in to handle any harassment complaints.”

There are several problems with this train of thought. First of all, while criminal harassment certainly exists, it is not the sum total of harassment. The fact that an individual act of harassment may not rise to the level of a crime doesn’t make it not harassment. In fact, the type of harassment we most frequently think of–workplace sexual harassment–is frequently not a crime in the U.S. It’s illegal under Title VII of the Civil Rights Act, but redress for harassment under that statute is civil not criminal. “Illegal” and “criminal” are not synonyms. You can sue your employer if they don’t fix the problem, but no one is coming to arrest your boss or coworker(s) in most cases.

The same is true of the kind of harassment people have been advocating banning at conferences and conventions. The vast majority of it does not rise to the level of criminal harassment. Stating that your organization will put all harassment complaints in the hands of the police is effectively saying that any harassment that doesn’t rise to the level of a crime will be dismissed.

Moreover, this kind of policy formulation discourages people from reporting harassment. People came to your event to learn, socialize, and/or be entertained. Harassment takes away from that already. Reporting harassment takes more, even with a streamlined administrative process. Filing a police report (or dealing with some groups of hired security) takes yet more away, particularly for members of groups (black people, people with mental illness) who are put at risk in their interactions with police.

And that’s before we even get to the question of what people want out of reporting harassment. For the most part, people who consider reporting harassment do so simply because they want the harassing behavior to stop. They want someone who will say, “No, really. Those are the rules”, or keep an eye on things to make sure this isn’t part of a pattern of predatory behavior that will affect more people. They’re not out to have someone punished, much less severely, unless that’s what it takes to end the behavior.

Telling them they have to have the matter treated as a crime is a huge disincentive to most of them. That’s great if your goal is no reports. It’s not great if your goal is no harassment.

There is a related criminal model of conceptualizing codes of conduct that is sometimes advocated for when developing processes. You will recognize this when someone starts talking about proof “beyond a shadow of a doubt”. In this case, people are suggesting that, because some harassment is criminal, any complaints of harassment should be subject to the standards required for a criminal conviction.

The problems with accepting this as a framework should be obvious. Criminal convictions significantly limit people’s civil rights, which is why they alone must meet that standard of evidence. Civil trials have a lower standard of evidence, and governmental administrative actions a lower standard still. While organizations that run groups or events should endeavor to be as fair as possible, asking them to meet standards of evidence meant to keep people from being locked up, when the actual stakes are generally attendance, is not reasonable.

Overall, criminal models have very little to teach us about how to construct codes of conduct. From here on out, however, we’ll look at models that all have some significant utility for shaping these codes. These models aren’t even necessarily in conflict, though discussions between people using different models can still sometimes be frustrating because the base goals and values either aren’t clear or aren’t shared.

Great Community Model

Often, you will hear groups considering codes of conduct say, “We have an amazing community/group of people/dynamic here. We don’t want to spoil it.” That isn’t a good reason to avoid implementing a code of conduct, but it can be a great position from which to build one. It’s always nice to be able to prevent a problem rather than have to react after one has happened.

The thing to remember under this model is that great communities don’t happen accidentally. They may grow out of the actions of many individual members rather than the decisions of leadership, but a good community is still actively doing many things right. If you can figure out what those are, you’ve got the start of a good code. What do you encourage? What don’t you tolerate? What do you do to welcome people? To resolve conflicts? If you have a community that functions well, you have an agreed code of conduct, even if you haven’t written it down.

You must write this down if you want to keep it. Why? Well, the first and most obvious reason is that new people who enter your community should have clear expectations about how they can do their part to keep it great. This is both for you, so you don’t find yourself one day overrun by heathens, and for them, so the barrier to entry is lowered. Writing things down also means your freeloaders–people who have enjoyed the great atmosphere without contributing to it–can no longer claim not to know what’s expected of them. They need to become as great as the rest of the community.

Yes, you do need to be prepared to hear that your great community isn’t great for everyone, that some people make it inhospitable in ways you didn’t know. This is a good thing, even if it’s painful. It means that people take you at your word that you want the community to be as great as you already think it is. It means they trust you to listen to them and to take them as seriously as you do the members of the community who are telling you everything’s great. It’s one of the more uncomfortable compliments you’ll ever receive, but it is a compliment.

Safety Model

The safety model of codes of conduct is best encapsulated as “We want everyone to feel safe here, and we make sure that’s actually true.” This model is particularly good for two things.

First, safety is basic enough to be something everyone understands. Emphasizing an organization’s responsibility to provide as safe an atmosphere as possible makes the idea of codes of conduct accessible to most people. While there are some people who loudly scoff at people feeling unsafe, or feeling unsafe in any situation short of physical harm, those people are a small minority. It may take work to explain why certain actions would make certain groups of people reasonably feel unsafe, but once you hit that point, empathy typically kicks in.

The other advantage of a safety framework is that it pushes you as an organizer to think about how you would want to handle worst-case scenarios. (In an ideal world, it would also push you to rehearse those responses, but that’s a step frequently left out of planning.) Luckily, events that threaten people’s safety are relatively rare, but they require a more sure-footed response. Thinking drastic problems through ahead of time means you’re less likely to make them worse when and if they happen.

The downside of the safety model is the same as the upside. It’s very basic. To paraphrase a friend of mine, “If all they have to offer me is safety, I might as well stay home. I’m already safe there.”

Business Model

The business model addresses the downside of the safety model, and it’s closest to the workplace sexual harassment model that most of us recognize. If you were to sum it up in one sentence, it would be, “We have great things to offer at our event/in our group, and no one should be excluded from participating based on who they are rather than what they do.”

This is the model that informs the Geek Feminism wiki sample policy and other policies that specify protections along historical axes of marginalization, such as race, sex, religion, disability, etc. These policies recognize that people have long excluded groups of people from full participation in society based on these characteristics, even after direct discrimination has been prohibited, and says that won’t be allowed to happen in these spaces. It says these traditional hostilities won’t be used to bar people from participation in something everyone should be allowed to enjoy.

What are the downsides of using a model like this? Mostly that it exposes you to criticism from culture warriors. Your organization may be accused of caving or pandering to special interests, or you may be “social justice warriors”. How seriously you take all that is up to you, but the facts remain that behavior that would be prohibited under this model is demonstrably alive and well in broader society and it has an effect on who participates.

Also, if your code of conduct prohibits this behavior “needlessly” because no one is doing it, then you’re demonstrating that you do, in fact, have a great community.

Restorative Justice Model

Restorative justice is beyond where most communities implementing codes of conduct are in terms of grappling with harassment. It’s graduate-level community building, in part because there are so few successful models to look at or people trained in its implementation. To sum up this model: “Punishment doesn’t fix the original problem. We should find solutions that meet everyone’s needs.”

What are some of the downsides of a restorative justice model? The major downside is that restorative justice is incredibly rigorous, which means it’s time and resource intensive. It requires more effort than many organizers want to put into it, which can lead to cutting corners. Instead of finding out what someone who has been harassed needs in order to be made whole or what a harasser is willing to do, an organizer may push everyone into a “standard” model of making amends.

On top of that, one of the main problems with harassment at events is that it eats up time and energy that people want to spend on the event. A rigorous response to that harassment has the potential to eat up far more. A restorative justice model may well be overkill for most code of conduct violations.

That said, there are lessons to be learned from such a model. The practice of asking people reporting harassment what kind of resolution they’re looking for comes out of restorative justice ideals, although in practice, it always has to be balanced with a safety outlook. Additionally, restorative justice practices around reintegration of offenders are far more rigorous and considerate of victims than most current pushes to allow harassers back into communities. There’s value in at least evaluating your code of conduct through a restorative justice framework, even if it shouldn’t ultimately be the primary focus of the code.

We gain something from using most of these models to think through both what we want codes of conduct to do and how they can get us there. And as long as we’re clear on which framework we’re using at any given time, and what that framework tells us overall, our discussions among ourselves can be more productive than they frequently are.

Want to see more work like this? Support me on Patreon.