3 ms·
Part 2. These kind of email threats are overwhelmingly empty, in the sense that the person sending them is almost certainly not actually going to come over and
by tzs 11y ago
Part 2.
These kind of email threats are overwhelmingly empty, in the sense that the person sending them is almost certainly not actually going to come over and do the things threatened [1]. Their power lies entirely in the threat itself.
Consider training your spam filter to recognize these threats and automatically trash them. I would hope your normal correspondence with friends and coworkers, and the correspondence you solicit, is sufficiently free of rape threats and the like that it should be easy for modern filters to separate out the latter.
Actually, don't trash them. Separate them out and make them available to anyone who wishes to study them. If many victims pool their threat data, someone may be able to infer enough to bust the anonymity of some of the senders.
[1] Exception: if you are doing things that will anger the kind of people who are willing to physically harm you even if it means they will likely be jailed or killed, then you need to treat these as real physical threats. For instance, if what makes you prominent on the internet is your work in successfully keeping abortion clinics open in deeply religious states, take threats as being serious physical threats.
- tptacek 11y agoI don't understand why you assume that most threats are empty. Are people who mail you rape threats actually going to rape you? Probably not. Are people who are unhinged enough to send a rape threat capable of harming you in other significant ways? Yes, they are.
- tzs 11y agoTo be clear, what I mean by "empty" is that the harm will be from the effects of receiving the message. The message is not a warning of an upcoming attack--it itself is the attack. I assume this because I haven't seen any stats showing otherwise, and so I'm assuming that those making these threats are similar to those who make similar threats against athletes and entertainers, which we have more experience with. The kind of people who will actually carry through with threats to murder or rape are almost always people who have a close connection to the victim, typically former spouses or former romantic partners or former business partners, or are people who are motivated by some kind of fundamentalist religious belief and believe they are doing God's work. It's possible that this will change. The rise of swatting may be a sign that it is starting to, although swatting has mostly been directed toward non-tech targets so far. Swatting lets you call down offline harm with the emotional disengagement of online. I think you are underestimating how different most people are online compared to offline. A heck of a lot of people who are civil offline turn very nasty online.
- tptacek 11y agoNo, you've missed my point. I'm saying that your analysis is fixated on the idea that e.g. a rape threat predicts a rape. I'm saying, "sure, maybe you're right", and rape threats really don't predict rapes. But they probably do predict other threats: career damage, reputation damage from forged posts, online breakin attempts, &c.
- tzs 11y agoAh...yeah, I was focusing on things directed directly to the victim in this part, not on things about the victim directed to third parties or the general public. I would suggest education to combat that--as part of teaching kids early about nasty trolls and how to handle them, teach them that they have to be skeptical of negative personal information about people they get from strangers on the internet. However, I fear that would probably not work. Easily believing rumors without good (or even barely plausible...) evidence seems to be a general tendency of people. Hell, people will believe that bike lanes are part of a UN plot to take away the guns of American civilians [1], so I suppose I should not be surprised that they would readily believe a rumor that someone slept their way into a job or distributed child pornography of their own kids or stole someone else's work. I currently have no ideas on how to effectively fix this. [1] I wish I were making that up...