4 ms·

My preferred operationalization of free speech is "minimum intimidation". Slurs don't qualify for protection under this operationalization, because they intimi

My preferred operationalization of free speech is "minimum intimidation". Slurs don't qualify for protection under this operationalization, because they intimidate the listener without providing any information content. Same goes for protesting a talk so loudly that the speaker is drowned out, or doxxing. If your speech serves only to intimidate others into silence, I don't think it should be protected. Curious to hear everyone's objections to this high-level approach. (Note: I would want to add a bunch of clarifying details before making this Twitter's official policy, but it seems more interesting to provide those details in response to objections, instead of writing a long boring comment.)

komali2 4y ago
Imo it's too simple a ruleset you've got here, though I get why we want to do this because in the end what we're all trying to do is find out how we could write down something enforceable, like a TOS or law. Those things can't be very vague... Sort of. Laws seem to allow for some discretion on the part of the justice system. Anyway to me it's similar to saying "violence is bad," yet we probably both believe people should be allowed to emoloy violence when violence is being done to them. I think similarly applies to hate speech. If a nazi shows up to campus to deliver their message of hate, it's ethical to drown them out and prevent any spread of the hateful message. "But both sides but hypocrisy but free speech" Imo tolerance is not a universal ethical position but instead a peace treaty. Peace treaties should not be suicidal. If you make a peace treaty with someone and they attack, you should abandon the treaty. Hateful speech (slurs etc) are a form of attack, they're a form of violence. They don't have a place in civil society, civil society should not allow this sort of speech or behavior in it. Someone who is saying these sorts of things is telling you clear as day that they don't tolerate some part of you. So why should you tolerate them? They've abandoned the treaty. What that doesn't mean is that civil society should go out of their way to hunt and kill anybody that likes to say slurs: that's the opposite of a peace treaty. More like they should be rejected utterly from participating until they're willing to again subscribe to a peace treaty (by being civil, not attacking people). Deplatforming vocal racists is thus ethical. Their language is attack on civility and our ability to maintain peaceful society. It's an important distinction.
- ShredKazoo 4y ago
  >Anyway to me it's similar to saying "violence is bad," yet we probably both believe people should be allowed to emoloy violence when violence is being done to them. Yep, there's an important parallel. I think a good frame is "protected speech". If someone says something to try & intimidate you, and you respond by trying to intimidate them, then nothing said by either party should be considered "protected" -- Twitter admins should feel free to delete the entire exchange. I don't think a policy of "intimidating speech for self-defense is protected" is workable, because people will intimidate others and claim it was in response to something the person said to intimidate them, and adjudicating such claims gets messy. If you want to say something nasty to someone else because they were nasty to you, go ahead. Just don't expect your rejoinder to be protected. >Hateful speech (slurs etc) are a form of attack, they're a form of violence. I don't think we should use the word "violence" this way, because reserving the word "violence" for physical harm helps us avoid upward spirals of antagonism. It's good if a conflict stays in the realm of words and doesn't go beyond that. Calling speech "violence" blurs the boundary between those domains.
  - komali2 4y ago
    Chain of effect - using words to call for someone's beheading is surely making them complicit in the physical harm that follows, particularly if the person is a figurehead that can feasibly assume their requests will be followed out. Thus we grant that there's at least one case where speech could be considered violence, can we not consider others that might be involved in causal chains of physical harm? The general air of racist rhetoric in the usa meant that black people were effectively dehumanized which meant that calls for lynching led to actual lynchings. With this context I feel it's reasonable to call the use of a racial slur, "violence". I should hope it's obvious I don't believe it's as harmful a form of violence as physically harming someone. We use the word violence to describe many remarkably different in nature things, so I don't think we risk devaluing the word.
    - ShredKazoo 4y ago
      >Thus we grant that there's at least one case where speech could be considered violence, can we not consider others that might be involved in causal chains of physical harm? As a reductio ad absurdum, suppose you and I are medical researchers. We're trying to figure out the best way to treat Disease D. You favor Treatment A, I favor Treatment B. We participate in a public debate defending our respective preferences. Suppose as more data comes in, it becomes abundantly obvious that Treatment B works way better, and Treatment A actually exacerbates Disease D. I acknowledge that reasonable people could've favored Treatment A at the time we had our debate, and you acknowledge that the new data favors Treatment B. Nonetheless, some people in the audience of our debate felt you were persuasive, used Treatment A on their patients, and their patients died. Does that make you a murderer? The scenario can be progressively modified: Maybe there was recent research showing Treatment B worsens Disease D, but you hadn't read it yet. Maybe you made a mistake in your statistical calculations. Maybe we decide you are a murderer, in which case no one wants to debate medical questions anymore lest they be declared a murderer, and the medical field languishes instead of developing new best practices. >I should hope it's obvious I don't believe it's as harmful a form of violence as physically harming someone. To play devil's advocate -- it wasn't obvious to me, and I think people with less context will find it even less obvious. So from the exact point of view you're arguing for -- creating a causal chain to physical harm -- I'd say overusing the word "violence" does have the potential to create harm, through the escalation of antagonism as I described previously. (BTW, although I think you're mistaken about the use of the word "violence", I would consider your claim to be protected speech, and don't think censorship of your claim is justified -- e.g. if the HN moderator stepped in and deleted your comment, I'd call that an overreach.) >We use the word violence to describe many remarkably different in nature things I haven't noticed this.
sooheon 4y ago
There is speech that is intended to directly harm without intimidating, no?
- ShredKazoo 4y ago
  Can you give an example of harmful speech that doesn't also intimidate? I think most speech intended to harm also ends up intimidating people, but there are probably exceptions that aren't coming to mind.
  - sooheon 4y ago
    An extremist discussion board, say dedicated to harming some other group. They may have no interest in advertising their plans, their speech is used to coordinate attacks.
    - mcv 4y ago
      Also: libel. It does harm, to the point that damages can be awarded, but it's not necessarily meant to intimidate the target, just to misinform others about the target.
    - ShredKazoo 4y ago
      I think that counts as conspiracy to commit a crime, so it's already covered in the criminal code? I'm focused on trying to develop speech norms for speech that's not a crime. See https://news.ycombinator.com/item?id=33998327 https://news.ycombinator.com/item?id=33998327
      - sooheon 4y ago
        Gotcha
anigbrowl 4y ago
Same goes for protesting a talk so loudly that the speaker is drowned out By this standard the speech of people booing Musk in a stadium the other day was not protected and they could be subject to legal sanction. I doubt this was what you had in mind.
- ShredKazoo 4y ago
  I don't think legal questions related to free speech and Twitter moderation questions related to free speech need to have the same answers. I'm happy with the legal framework we have in the US. The question I'm trying to answer: Suppose a large group of hecklers attend a public lecture. They disrupt the lecture and prevent the audience from hearing the speaker's message. Is there a sense that they are interrupting the lecturer's exercise of free speech? Or alternatively, is it the case that anyone who scolds them is interfering with the hecklers' right of free speech? (Does the scold also have a right to free speech? How deep does this rabbit hole go?) I think it's possible to thread this needle by defining the thing we want as "transmission of information", and the thing we don't want as "intimidation". You could argue that boos are also transmission of information. I saw a short clip of the boo thing and it didn't seem like they were loud or numerous enough to overwhelm what was spoken into the microphone. So based on my current operationalization, those boos basically qualify as protected speech. If audience members preplanned a large disruption which made it impossible to hear anything Elon said, that would be anti-free-speech according to my current operationalization. It intimidates and prevents information transmission. I still have to think about the case where they spontaneously disrupt the event, unintentionally making it impossible for him to speak. It seems analogous to the case where people are killed because they're in a big crowd where everyone is squeezed into a tiny space.

komali2 4y ago

Imo it's too simple a ruleset you've got here, though I get why we want to do this because in the end what we're all trying to do is find out how we could write down something enforceable, like a TOS or law. Those things can't be very vague... Sort of. Laws seem to allow for some discretion on the part of the justice system. Anyway to me it's similar to saying "violence is bad," yet we probably both believe people should be allowed to emoloy violence when violence is being done to them. I think similarly applies to hate speech. If a nazi shows up to campus to deliver their message of hate, it's ethical to drown them out and prevent any spread of the hateful message. "But both sides but hypocrisy but free speech" Imo tolerance is not a universal ethical position but instead a peace treaty. Peace treaties should not be suicidal. If you make a peace treaty with someone and they attack, you should abandon the treaty. Hateful speech (slurs etc) are a form of attack, they're a form of violence. They don't have a place in civil society, civil society should not allow this sort of speech or behavior in it. Someone who is saying these sorts of things is telling you clear as day that they don't tolerate some part of you. So why should you tolerate them? They've abandoned the treaty. What that doesn't mean is that civil society should go out of their way to hunt and kill anybody that likes to say slurs: that's the opposite of a peace treaty. More like they should be rejected utterly from participating until they're willing to again subscribe to a peace treaty (by being civil, not attacking people). Deplatforming vocal racists is thus ethical. Their language is attack on civility and our ability to maintain peaceful society. It's an important distinction.

ShredKazoo 4y ago
>Anyway to me it's similar to saying "violence is bad," yet we probably both believe people should be allowed to emoloy violence when violence is being done to them. Yep, there's an important parallel. I think a good frame is "protected speech". If someone says something to try & intimidate you, and you respond by trying to intimidate them, then nothing said by either party should be considered "protected" -- Twitter admins should feel free to delete the entire exchange. I don't think a policy of "intimidating speech for self-defense is protected" is workable, because people will intimidate others and claim it was in response to something the person said to intimidate them, and adjudicating such claims gets messy. If you want to say something nasty to someone else because they were nasty to you, go ahead. Just don't expect your rejoinder to be protected. >Hateful speech (slurs etc) are a form of attack, they're a form of violence. I don't think we should use the word "violence" this way, because reserving the word "violence" for physical harm helps us avoid upward spirals of antagonism. It's good if a conflict stays in the realm of words and doesn't go beyond that. Calling speech "violence" blurs the boundary between those domains.
- komali2 4y ago
  Chain of effect - using words to call for someone's beheading is surely making them complicit in the physical harm that follows, particularly if the person is a figurehead that can feasibly assume their requests will be followed out. Thus we grant that there's at least one case where speech could be considered violence, can we not consider others that might be involved in causal chains of physical harm? The general air of racist rhetoric in the usa meant that black people were effectively dehumanized which meant that calls for lynching led to actual lynchings. With this context I feel it's reasonable to call the use of a racial slur, "violence". I should hope it's obvious I don't believe it's as harmful a form of violence as physically harming someone. We use the word violence to describe many remarkably different in nature things, so I don't think we risk devaluing the word.
  - ShredKazoo 4y ago
    >Thus we grant that there's at least one case where speech could be considered violence, can we not consider others that might be involved in causal chains of physical harm? As a reductio ad absurdum, suppose you and I are medical researchers. We're trying to figure out the best way to treat Disease D. You favor Treatment A, I favor Treatment B. We participate in a public debate defending our respective preferences. Suppose as more data comes in, it becomes abundantly obvious that Treatment B works way better, and Treatment A actually exacerbates Disease D. I acknowledge that reasonable people could've favored Treatment A at the time we had our debate, and you acknowledge that the new data favors Treatment B. Nonetheless, some people in the audience of our debate felt you were persuasive, used Treatment A on their patients, and their patients died. Does that make you a murderer? The scenario can be progressively modified: Maybe there was recent research showing Treatment B worsens Disease D, but you hadn't read it yet. Maybe you made a mistake in your statistical calculations. Maybe we decide you are a murderer, in which case no one wants to debate medical questions anymore lest they be declared a murderer, and the medical field languishes instead of developing new best practices. >I should hope it's obvious I don't believe it's as harmful a form of violence as physically harming someone. To play devil's advocate -- it wasn't obvious to me, and I think people with less context will find it even less obvious. So from the exact point of view you're arguing for -- creating a causal chain to physical harm -- I'd say overusing the word "violence" does have the potential to create harm, through the escalation of antagonism as I described previously. (BTW, although I think you're mistaken about the use of the word "violence", I would consider your claim to be protected speech, and don't think censorship of your claim is justified -- e.g. if the HN moderator stepped in and deleted your comment, I'd call that an overreach.) >We use the word violence to describe many remarkably different in nature things I haven't noticed this.

sooheon 4y ago

There is speech that is intended to directly harm without intimidating, no?

ShredKazoo 4y ago
Can you give an example of harmful speech that doesn't also intimidate? I think most speech intended to harm also ends up intimidating people, but there are probably exceptions that aren't coming to mind.
- sooheon 4y ago
  An extremist discussion board, say dedicated to harming some other group. They may have no interest in advertising their plans, their speech is used to coordinate attacks.
  - mcv 4y ago
    Also: libel. It does harm, to the point that damages can be awarded, but it's not necessarily meant to intimidate the target, just to misinform others about the target.
  - ShredKazoo 4y ago
    I think that counts as conspiracy to commit a crime, so it's already covered in the criminal code? I'm focused on trying to develop speech norms for speech that's not a crime. See https://news.ycombinator.com/item?id=33998327 https://news.ycombinator.com/item?id=33998327
    - sooheon 4y ago
      Gotcha

anigbrowl 4y ago

Same goes for protesting a talk so loudly that the speaker is drowned out By this standard the speech of people booing Musk in a stadium the other day was not protected and they could be subject to legal sanction. I doubt this was what you had in mind.

ShredKazoo 4y ago
I don't think legal questions related to free speech and Twitter moderation questions related to free speech need to have the same answers. I'm happy with the legal framework we have in the US. The question I'm trying to answer: Suppose a large group of hecklers attend a public lecture. They disrupt the lecture and prevent the audience from hearing the speaker's message. Is there a sense that they are interrupting the lecturer's exercise of free speech? Or alternatively, is it the case that anyone who scolds them is interfering with the hecklers' right of free speech? (Does the scold also have a right to free speech? How deep does this rabbit hole go?) I think it's possible to thread this needle by defining the thing we want as "transmission of information", and the thing we don't want as "intimidation". You could argue that boos are also transmission of information. I saw a short clip of the boo thing and it didn't seem like they were loud or numerous enough to overwhelm what was spoken into the microphone. So based on my current operationalization, those boos basically qualify as protected speech. If audience members preplanned a large disruption which made it impossible to hear anything Elon said, that would be anti-free-speech according to my current operationalization. It intimidates and prevents information transmission. I still have to think about the case where they spontaneously disrupt the event, unintentionally making it impossible for him to speak. It seems analogous to the case where people are killed because they're in a big crowd where everyone is squeezed into a tiny space.