“We should restrict immigration to limit labor supply so that the native born can earn better wages” is not the same type of message as “immigrants are cockroaches and we should shoot refugees as they cross the border.”

“We should reject colonial oppression of our country” is not the same type of message as “we must wage religious war against our oppressors, and it is righteous to for young men to be suicide bombers”.

“We should end affirmative action on the basis of race because it isn’t fair to individual members of the majority group” is not the same type of message as “all members of X minority are inherently worthless, and we should purge them”.

“Rape accusations should be handled by law enforcement professionals rather than by school administrators, and the evidence should need to meet a high legal standard to be actionable” is not the same type of message as “X person claims they were raped but everyone knows they are a lying slut and were asking for it.”

Etc.

Why do I even have to try and explain this?

> The rules could say absolutely no racism

This is not what the rules do say. What they “could say” is not relevant here. The rules say: “Hateful conduct: You may not promote violence against, threaten, or harass other people on the basis of race, ethnicity, national origin, sexual orientation, gender, gender identity, religious affiliation, age, disability, or serious disease.”

This is elaborated here if you want to read about it https://help.twitter.com/en/rules-and-policies/hateful-condu...

Specifically disallowed are (1) violent threats, (2) wishing, hoping or calling for serious harm on a person or group of people, (3) references to mass murder, violent events, or specific means of violence where protected groups have been the primary targets or victims, (4) inciting fear about a protected category, (5) repeated and/or non-consensual slurs, epithets, racist and sexist tropes, or other content that degrades someone, (6) hateful imagery

It’s very hard to see it, but these rules facilitate bias.

When I listened to the interview on Joe Rogan with the Trust and Safety officer, one of the examples she gave was the Milo Youannopolis quote “the world would have been better off if your mother banged you on the rock as a baby” or something to that effect.

This is gross, and offensive, but not a threat. It’s not a threat to talk hyperbolically about a hypothetical and ridiculous past event.

Unless you already find a person scary.

In which case it very very much sounds like a threat, or an incitement to violence.

Language needs context and if you believe that people with certain characteristics (including political orientation) are more likely to be violently dangerous, it is absolutely natural to view things they say as more threatening than you might otherwise.

It’s still bias.

It’s bias if a policeman finds “I have a gun in the glove compartment, and a license for it” more threatening based on someone’s race.

If it were based on some real or perceived fact about the world, still bias.

Actually definitionally bias.

Humans use bias (context) to make sense of the world. And that context (bias) is necessary for communication to makes any sense.

Which means those rules can only be implemented in a way that reflects the bias of whomever is implementing them.

> Milo Youannopolis quote “the world would have been better off if your mother banged you on the rock as a baby” or something to that effect. This is gross, and offensive, but not a threat. It’s not a threat to talk hyperbolically about a hypothetical and ridiculous past event.

As false positives go, if this is the one you're worried about, you can tell that your case isn't strong by asking exactly what a platform that accidentally threw out that sort of content-baby with the actual-threat bathwater would be losing. Even if "it's not a threat" were inarguably true, it doesn't make much sense as a response, because there's no discourse in there, it's simply abuse, and likely runs afoul of other rules, certainly of other sensible rules that could be established.

But to say it's not a threat really stops short of thinking about what was said. “the world would have been better off if your mother banged you on the rock as a baby” is a statement that bypasses any particular criticism -- the kind of thing that might be a valuable feature in speech and discourse -- and instead simply asserts a total lack of value in someone's life. If you think that discourse is valuable, part of the reason for that has to be that you think ideas matter. What are the consequences if people believe in that idea -- the idea that some specific person's life has a total lack of value because they said something Milo doesn't like?

"Not a threat" is a technical judgment, perhaps even a distinction that matters. But there's absolutely the implication of violence, and hyperbole is not an adequate defense, nor would any sensible person find it so if they were told that someone should have done the world a favor and set them on fire as a child.

> rules can only be implemented in a way that reflects the bias

Which is another way of saying that rules will always be about values, both of those who wrote them and those who carry them out. Whether that's a bug or a feature depends on how fit the values in play are.

>If you think that discourse is valuable

Who exactly is going to decide? Rules and values committee?

> What are the consequences if people believe in that idea

So people are stupid, and the all ideas should be first vetted by non-stupid people. Nice.

So what's the alternative? "The rules aren't perfect, so let's get rid of all of them" is not improving things over "the rules intent is good, but humans implementing them will be biased". The solution would be a way to reduce bias.

Ironically, one true solution to the problem of bias is diversity. The thing the groups who most rail against banning of misgendering or racist language despise. Get a a "good conservative" into the group which enforces the rules, as an ombudsman or a watchdog whatever, the same way we'd do for polticically other, and then it can play out.

Disclosure: I am an european left of center, so from american perspective I'm basically communist. I use conservatives as an example because the seem to be the largest group complaining about things like enforcing harassment policies. To communicate my view on this debate with an example, insisting on misgendering people when that causes them pain is worse than a mere insult, it's equivalent to walking up to veteran with PTSD, pulling out a toy gun to their face and going "BANG". That shit would not offline, and it should not fly as a performative act online to normalize it either.

The alternative is to follow the law.

Clearly isn't the same

I'm not sure “the world would have been better off if your mother banged you on the rock as a baby” isn't a threat.

For example if you remove the colourful language from it "the world would have been better off if you were dead" sounds rather threatening.

Context is important "Who will rid me of this troublesome priest?"

“The world would be better off if you were dead” is an opinion, not s threat: it states no intent of future action.

> Context is important "Who will rid me of this troublesome priest?"

That's not a threat, it's a solicitation; it requests other to engage in rather states an intent to carry out future action. But “the world would be better off...” is neither, on its face.

Just because something is an option doesn’t stop it from being threatening. A reasonable person could become afraid that they would be harmed by a person that expressed a preference for their death.

Have you heard of the StackOverflow CoC controversy where a larger part of the community, possible the majority, disagree on the exact definition of harassment of other people on the basis of gender identity?

Did you also hear about the Wikipedia Fram controversy where a large party of the community, including the highest democratically appointed authority (arbcom) disagree with the wikipedia foundation on what harassment is?

To take a random sample further from the past, a guild wars 2 developer accused a fan of harassment after criticism of the game story line got posted. The controversy lead to two developers being fired and the company posting a huge apology which said that criticism of the game story line is not harassment and any fan should be free to express constructive views about the art.

A rule that forbids harassment only has meaning based on how harassment get defined, and with current cultural environment that definition is all over the place.

An even more political loaded area, hateful rhetoric is quite common. Here in Sweden there were a brief suggestion to forbid it on the national broadcast channels, but then the idea had to be basically scrapped as the definition could never really be agreed on. If you have one political party using immigrants as the target demographic for all problems, and an other political party using males as the target demographic for all problems, should we ban both? That is not going to happen.

Protected category is interesting but a extremely US centric concept where the federal government specify a list of protected groups and anyone not on it is a free target. It is a list not even all the states agree on, as political views is sometimes added by specific states and others don't. The concept of an "official religion" is also one which is constantly debated, with groups like copyme trying to get included in the protected category. EU has a bit better and broader definition in Article 10, but the enforcement is mostly up to individual nation states.

You still don't get it. People will claim what you're saying violates some rule if they don't you or your position. Doesn't matter if YOU consider it to be reasonable or not. I don't disagree with your distinctions at all but people are shitty and will not be as benevolent as you.

Regarding your examples. If you say "We should restrict immigration to limit labor supply so that the native born can earn better wages", you will be read by many people as if you were saying "immigrants are cockroaches". These people will group you with people who say the latter thing, and with people who imply the shooting part.

I know that for a fact. I live in Europe. We had a migration crisis recently, and no matter what you believed, how nuanced your position was, you always ended up being called either "a compassionless monster who doesn't care about other people dying", or "a traitor supporting parasites in taking our jobs and destroying our economy".

“We should reject colonial oppression of our country” very much implies “we should wage a war against our oppressors” (not sure why you bring religion into this). Suicide bombing didn't enter the western Overton window just yet, but it surely will if things ever get bad enough. For evidence, just skim [0].

> “We should end affirmative action on the basis of race because it isn’t fair to individual members of the majority group” is not the same type of message as “all members of X minority are inherently worthless, and we should purge them”.

That tells me you haven't been on the Internet recently. It absolutely means the same thing in the eyes of a certain vocal minority that currently dominates all social media, and has an uncanny ability to summon Internet mobs in order to ruin people's lives.

> “Rape accusations should be handled by law enforcement professionals rather than by school administrators, and the evidence should need to meet a high legal standard to be actionable” is not the same type of message as “X person claims they were raped but everyone knows they are a lying slut and were asking for it.”

Except in practice the former is understood by many as the latter, and can end up with you in the news headlines and having your life's work taken away from you.

--

Point being, rules like Terms of Service or Codes of Conduct are easily weaponized and used as tools of oppression if they're not backed by impartiality, rationality, a firm commitment to uphold the spirit of those rules and to nip all attempts at rule lawyering in the bud. Otherwise, either the service itself will abuse the rules, or the users that are best at escalating minor disagreements will. I have zero trust in Twitter on this matter.

--

[0] - https://en.wikipedia.org/wiki/Suicide_attack#History,_pre-19... One of the interesting things I've learned here today is that apparently, the US trained people to perform suicide bombings with nuclear weapons. The Cold War was an insane time.

Do you have an example of Twitter using its discretion to shut down a participant who (in your opinion) did not violate the site’s rules? (Where they didn’t acknowledge it as a mistake and later re-instate that person? Mistakes are obviously not completely avoidable.)

We are talking about what will get you banned from Twitter, not what will get you called a racist jerk or a terrorist sympathizer or whatever by other participants on the site.

Pessimistic view follows, but:

All of your acceptable examples most definitely fall under "inciting fear about a protected category".

Also, wouldn't "references to mass murder, violent events, or specific means of violence where protected groups have been the primary targets or victims" prohibit Jews from speaking about the holocaust?

The explainer part of the terms (actually reading then clarified a lot):

> We prohibit targeting individuals with content that references forms of violence or violent events where a protected category was the primary target or victims, where the intent is to harass. This includes, but is not limited to sending someone:

> media that depicts victims of the Holocaust;

> media that depicts lynchings.

Ah, then my earlier point is not correct. Thanks

> will you pay me a dollar for each time I'm called a racist, white supremacist, xenophobe, etc.?

I'll pay you a dollar every time your account gets blocked, because I don't give a crap if other users call you mean names for expressing bigoted views. That's not what anything in this thread is about.

> Do you think, for example, Twitter is going to limit engagement on tweets from political opponents calling Trump a racist?

If they break their terms, sure. But calling someone racist doesn't fall under any of these: https://help.twitter.com/en/rules-and-policies/hateful-condu...

> ... I don't give a crap if other users call you mean names for expressing bigoted views.

Are you trying to prove my point here, or was that inadvertent? I listed a set of political and legal discussions around sensitive topics that happen to involve race and national origin. I didn't even express any personal viewpoints at all. But in your mind, a "carefully and precisely represented right-leaning position" is enough to warrant being slandered for bigotry. Which is precisely the problem here.

> those rules, exactly how you have stated them, can be--and are--used to clamp down on valid political speech.

As far as I can tell Twitter has not “clamped down on valid political speech”, but only on targeted harassment, threats, and so on.

You are complaining that supporting a racist worldview gets someone called out as a racist by other users of the site. But that’s expected and healthy discourse for a democracy. There’s nothing in Twitter’s terms of service that says people can’t make fun of someone or call them racist.

This is entirely different from getting banned by the platform for e.g. inciting violence.

Claiming that a right-leaning political worldview is racist is a nasty ploy, it’s slander to suppress a dissenting opinion. It’s a dodge from someone unwilling or unable to defend their position by engaging in meaningful debate.

However, repeated degrading attacks against someone are against Twitter policy.

> Repeated and/or non-consensual slurs, epithets, racist and sexist tropes, or other content that degrades someone. We prohibit targeting individuals with repeated slurs, tropes or other content that intends to dehumanize, degrade or reinforce negative or harmful stereotypes about a protected category.

When these policies are applied with an uneven hand, they betray the political bias of those applying the policies.

If you have an opinion (any opinion really) about a race or a national origin, that makes you a bigot, yes. Non-bigots don't have opinions on a color of a skin, nor a place of birth.

Those things are not chosen by an individual and therefore should not be discussed nor should the individual be generalized based on them. You can criticize an individual based on their action regardless of where they're from or how they look like. As soon as you start generalizing everyone who looks like them / is born in the same area as them, you become a bigot.

There, you were already so close to being self-aware that I hope I helped you achieve awareness.

I think you misread what I wrote. Nothing in there was an opinion about a race or a national origin.

No, you haven't expressed any opinion here. You've alleged that you do have some though:

> around sensitive topics that happen to involve race and national origin.

I’ve been careful not to, because the point isn’t to argue a side, it’s to discuss how valid political arguments might be suppressed due to wrongful accusations of, e.g. racism or alleged violations of terms / codes of conduct.

To be clear I agree with your first two paragraphs. It’s only the third which alleges that you’re correcting a belief I have, that I do take issue with.

So I think again you’re reading into what I wrote too much. Why beat a dead horse, but the point there is to recognize that e.g. “immigration policy” is necessarily going to impact populations that are predominantly of a certain race and/or national origin. But crucially, it’s not racist to think that a country should have a border or debate how set immigration policies are enforced along that border. It’s not racist to examine sub-populations which are moving illegally across the border, e.g. to examine the prevalence of drug running, human trafficking, or gang activity within that sub-population, or to examine the share of that sub-population that is currently charged or has been found guilty of committing a violent crime versus other populations such as legal immigrants or natural citizens. This is just basic sociology.

So that’s what I mean by “sensitive [political] topics” that happen to involve race and national origin.