Universities studies in both the US and the UK have found that only 2 − 3% of allegations are false.
This is not a fair description. The way people get such statistics is by assuming all accusations are true unless there is strong evidence against them, but there is a large number with no strong evidence either way, and researchers should not just assume they are all true.
A good first place to start is the Wikipedia Article on the subject, which features a wide range of estimates, almost all of which are higher than the 2-3% you say, some of which being dramatically higher.
Your own website lists a slightly higher range, 2-4%
[redacted]
If we look at the source you supply for the 2% we see a different story:
Despite reforms intended to increase the number of rape investigations that proceed to prosecution, the study found that suspects were charged in only 15 percent of the 850 reported rapes. Rape complaints were subsequently withdrawn in 15.1 percent of the cases, and 46.4 percent of the complaints resulted in “no further police action.” For the cases in which complaints were withdrawn, no statistically reliable profile of case characteristics could be determined; however, when complaints were withdrawn, suspects were more likely to be current or former partners of the complainant. Cases that resulted in “no further police action” were more likely to involve younger victims, victims who were acquainted with or had a cursory relationship with the suspect, and victims who had consumed alcohol or other drugs near the time of the offense. Of the 850 cases, 21.3 percent were “still ongoing” or their status could not be determined from the case records. Only a small percentage of cases (2.1 percent) were designated by police as false. Twenty-six percent of the cases involved victims with a psychiatric disability or mental health issue. Charges were more likely to be brought when the rape victim was male (even though the overwhelming majority of complainants were female); had been physically injured; had been medically examined; was not under the influence of alcohol or other drugs at the time of the offense; was subjected to other offenses in the course of the rape; and when the suspect was known to police for previous sexual offenses.
Only 2.1% were designated by the cops as false, but that was after the majority of accusations had been withdrawn or determined ‘no further action’. Many of those withdrawn or no-further-action accusations would also have been designated as false if they had been investigated.
Only 2.1% were designated by the cops as false, but that was after the majority of accusations had been withdrawn or determined ‘no further action’. Many of those withdrawn or no-further-action accusations would also have been designated as false if they had been investigated.
This sounds like you are suggesting (correct me if I’m wrong) that many or most withdrawn and ‘no-further-action’ accusations are actually false, which is not a fair conclusion to draw from the information you presented. It also seems to claim that cases designated as withdrawn or ‘no further action’ are not investigated, which is also not accurate. Specifically, the paragraph you shared mentions many of the reasons victims will withdraw or not seek further action:
Withdraw:
“when complaints were withdrawn, suspects were more likely to be current or former partners of the complainant” Victims of domestic violence are often in immediate physical danger when attempting to leave or report, or are extremely financially limited due to their situation in an abusive relationship. Following through on an accusation could pose more danger to the victim than withdrawing.
Investigating sexual assault cases can be extremely invasive, and often re-traumatize the victim. Cases can then take years to prosecute and involve cross-examining the victim. Some victims withdraw to avoid this process.
No-Further-Action: This is a police designation indicating they decided not to investigate for any number of reasons, including lack of forensic evidence, uncooperative witnesses, lack of resources, or a belief that they will not be able to successfully prosecute. None of these directly imply a false claim.
At least in the US, the reports often initially come through Patrol officers who typically don’t receive the specialized training in responding to sexual assault victims (e.g., trauma-informed training) that investigators working in domestic violence or sex crimes units do, and may be less likely to take reports seriously.
“Cases that resulted in “no further police action” were more likely to involve younger victims” Sexual assault is rarely the only issue in cases with young victims, and gathering evidence is made more complicated by this.
“victims who had consumed alcohol or other drugs near the time of the offense” In addition to the potential for police prejudice against these victims, there’s also the consideration of whether or not a jury is likely to find a victim’s story compelling. This indicates a judgement on the potential for a successful prosecution, not on whether a crime was actually committed.
I agree that this is a “different story,” just not one indicating a secret cache of false claims hidden by statistics. Instead, it’s a depiction of the extremely messy process of reporting, investigating, and prosecuting these cases.
I can’t access the full article from which you pulled that abstract, but I used a number of the resources below in my response. I hope you find them informative:
This sounds like you are suggesting (correct me if I’m wrong) that many or most withdrawn and ‘no-further-action’ accusations are actually false, which is not a fair conclusion to draw from the information you presented.
It is logical that people would be more likely to withdraw, and the police less likely to investigate, accusations that seem less compelling. If any of these cases are false, the 2.1% is an under-estimate; if they are false at a higher rate than cases accuser and police followed through with, the 2.1% is a significant underestimate.
‘Any’ is different than ‘many,’ as you originally claimed. I think if the report you seem most focused on was estimating that 2.1% of all rape claims are false accusations, your concern would be more understandable. But this is a paper on rapes reported in a specific geographical area over a 3-year period and 2.1 is the percentage of reports designated false by the police, making this an odd choice of information to focus in on as telling a “different story.” Most groups that compile these data in order to make estimates do place those estimates in ranges, anywhere from 2-3 to 2-10%.
if they are false at a higher rate than cases accuser and police followed through with, the 2.1% is a significant underestimate.
This still seems to suggest that cases resulting in withdrawals or no further action are more likely to be false, and you don’t have enough information for that assumption to be well-founded. Further, this assumption feeds into harmful myths about the underlying causes for withdrawal and case attrition in sexual assault reports, many of which are addressed in the sources I’ve linked above.
This thread, and these kinds of discussions are very revealing and worth pursuing to the end.
Let’s go with Scott Alexander’s estimate of a 3% lifetime chance of a man being falsely accused of rape and a 15% chance of a woman being raped. Let’s assume EA is 70% men vs 30% women.
How much “weight” should the community give to guarding against false accusations vs to guarding against the prevalence of sexual abuse (since not believing accusations lets the perpetrator repeat their actions)? This includes a value judgement on the harm from sexual assault vs the harm from reputational damage, etc.
Since this seems to be a difficult tradeoff and the community health team/EA organization leaders are making these tradeoffs that include value judgements (and EAs don’t usually have the same value judgements as the rest of society), the current community members and people in charge should be transparent about these value judgements and overall weights (maybe through a survey?).
It would help current and future EAs decide if they want to be a part of the community based on how they value their own welfare.
I think one question would be the extent to which the types of actions Community Health can take would reduce a perpetrator’s ability to commit future assaults. I don’t know the answer to that but would be interested in what others thought.
I understand the desire to for quantification and transparency, but my guess is this particular quantity is relatively far down the list of parameters to publicly estimate, especially given the methodological difficulty. If you thought it was particularly pressing maybe you could have a go at it? Or perhaps there is some existing economic literature on the subject. I suppose the level burden of proof applied in court cases is a starting point.
I’m more than a little creeped out that you have my website too.
I’m sorry you deleted the original text of your response, as I would have liked to engage with it. Unfortunately I did not save a copy before you deleted and replaced it.
I did however get a copy of this comment you made prior to changing your forum username, where you linked to us all to your website.
Strangely you seem to have deleted this comment now and completely forgotten about having written it. It is very ironic that, in the context of a discussion about the frequency of false accusations, that you would attempt to alter the digital record and then falsely imply I am a creep.
Perhaps this was an accident. It is easy to forget what exactly we have said, especially if we regret it later. If so I hope you take this lesson to heart about how easy it is for even people with the best of intentions to accidentally make a false accusation.
- - -
Screenshot of the above comment in case of further edits:
I’m not sure I understand the rationale for removing information that was supplied by the very same person who now says they want it removed, especially when this information was supplied merely a week ago, on the EA Forum, and in this same thread. The policy that this decision seems to exemplify appears to effectively give anyone the right to censor any information about themselves in posts or comments made by others, regardless of how that information was obtained or how public it is.
Note that the case for disclosing the information in this particular instance was pretty strong: J_J implied that temp_ was a creep for knowing J_J’s website, but it turns out that J_J had included a prominent link to their website in a comment posted just one day earlier. I do not want the Forum to be a place where people can make unfair accusations about others and retain a right to suppress evidence establishing the unfairness of those accusations.
The policy that this decision seems to exemplify appears to effectively give anyone the right to censor any information about themselves in posts or comments made by others, regardless of how that information was obtained or how public it is.
As mentioned in the policy, we do think that there are cases when some personal information is important to share, and we don’t think everyone should have the right to censor any information about themselves. We do consider how public the information is and “when information is easily accessible elsewhere, we will err on the side of keeping it”, but we also strongly consider how relevant the information is to EA. In this case, I felt that the name and website of the user are not relevant enough for effective altruism to justify keeping the information in the comment against the user’s wishes.
Note that the case for disclosing the information in this particular instance was pretty strong: J_J implied that temp_ was a creep for knowing J_J’s website, but it turns out that J_J had included a prominent link to their website in a comment posted just one day earlier. I do not want the Forum to be a place where people can make unfair accusations about others and retain a right to suppress evidence establishing the unfairness of those accusations.
I agree, but I think that in the edited comment it’s still clear that J_J had included a prominent link to their website in a comment posted just one day earlier. If that’s not the case I should have edited it differently (possibly writing [user’s website] in the black box). Do you think it should be clarified?
Thanks for the reply. I think the crux of our disagreement may be that I don’t regard “being relevant to EA” as a necessary condition for declining a request to remove personal information, unless that phrase is given a very broad interpretation that includes things like “keeping the EA Forum a place where people can’t make unfair accusations about others”.[1] Separately, if a user voluntarily discloses a piece of personal information, I think this should be beyond the scope of mod action, unless something happened in the intervening period that clearly justifies removing or encoding the information. People can still ask others not to share this info, but I think it should be up to each person to honor those requests, rather than being something enforceable by the admin team.
In this case, as you note, it was possible to remove the personal information while preserving the relevant evidence publicly, although I think the removal made it somewhat more difficult to appreciate what was really going on. But one can imagine other situations in which this cannot be done.
What she said is that she felt creeped out. The implication that the person who’s actions she was creeped out by was a creep is an implication you are adding to her statement, and not one I did personally. It is very much possible to feel creeped out by someone who isn’t a creep.
Reading it as an underhanded way to make an unfair allegation seems very uncharitable to me in light of all her comments.
I don’t understand what rule you think I broke. This was a link to a public website that she herself shared on the thread; it does not fall into any of the categories in the linked website. This is someone who is making serious allegations about EA, and looking to be paid for it—she should not be able to demand others users delete any record of what she has done.
The closest reference I can see in the rules is this:
If information was accessible at the time of posting but isn’t anymore, we might encode it
But then you should have encoded it, not deleted it and edited my screenshots.
I’ve read Slate Star Codex on rape a while ago, and I apologize, but I can’t take his opinions seriously.
Maybe some context would be helpful. There was a previous case in EA with a woman named Kathy Forth who made a bunch of accusations. The consensus seems to be that those accusations were false and/or overblown. It seems many people knew this and shared it through a whisper network, and Scott Alexander was one of a fairly small number of people willing to state it publicly. So, hopefully this explains why dismissing him might be a bit of a red flag with me and some other people here.
BTW, I’m wondering if a good heuristic is: If someone makes a lot of accusations, they’re likely to be a liar. If someone receives a lot of accusations, they’re likely to be guilty. The idea being that genuine victimhood (from a crime/false accusation) happens to people at a fairly even background rate, but people who are bad actors tend misbehave more than once.
In my personal experience talking to victims since, I’ve noticed that this is one of the most scary heuristics of all. Every woman I’ve spoken to in this community who comes forth about SA once feels like they are now more than twice as vulnerable for being targeted by predators.
Very personally, several of my house co-founder’s friends had an intentionally loud conversation outside my door about unfortunate things that might happen to “girls who cry wolf” since nobody will believe them on subsequent claims.
I don’t mean this as a comment on the particular case reported in the TIME article, though I’d reject using naive base rate calculations as a last word on someone’s probabilistic guilt. but the “only 2-3% of allegations are false” stuck out to me because I read a better estimate is probably more like 2-10%. https://slatestarcodex.com/2014/02/17/lies-damned-lies-and-social-media-part-5-of-∞/ there’s a lot of ambiguity here—issues like not every “report” is an “allegation” because sometimes reports don’t name a perpetrator. I have no idea what the correct figure is, but it seems to me the 2-3% figure gets probably bandied around a lot probably with a sense of precision and finality that isn’t warranted by the evidence. Happy to see new evidence or information to the contrary, and whether the rate is 2% or 10% it can certainly be described as “low”.
The article is a nearly a decade old, and for all I know there might be newer research. But I hope when people are thinking about best practices for the future, they do so on the basis of the best evidence available.
Thank you for your explanation. I appreciate you taking the time to explain your reasoning on that point and find it useful for being confident in the rest of what you have to say here.
I think the main reason that EA/rationalist spaces and activist spaces have historically often not gotten along well is that in EA/rat spaces there is a strong norm against a certain kinds of strongly worded moral injunctions.
I think the reason for this is that EA tends to attract people driven to do good but with perhaps an OCDish tendency to worry if they’re doing enough and push themselves too hard. If we instead went around saying it was shameful people were still eating meat and also shameful people weren’t giving more money instead of using it to vist friends and also shameful they weren’t thinking enough about how to prevent sexual assault people would pretty quickly burn out.
(There are certainly people in the community who can and do translate a lot of things activists say into EA/rat friendly language)
Of course, if you assume the best of too many people you let in bad actors and I suspect a lot of people might find it stifling. But it does create a safe space for many of us.
fwiw if I were were CEA’s health team I would phrase this as “although they are a small minority we have occasionally come across accusations which were as far as we could tell false” or some such. I’m not sure what can be gained over nitpicking over the exact number—this is going to vary from place to place
I think the “don’t want change” faction is a good deal smaller than the “we would like time to digest and decide what changes seem like a good idea and what typical problems actually look like (given how heterogeneous EA is) and how to make these changes while keeping some aspects of the culture we find valuable” faction
(also, as a trans women, I’m going to have to raise an eyebrow at “men and AMAB”—i think the actual risk factor is hormones fwiw but...
ed: you know what I think these are “rape by envelopement doesn’t count” numbers, which exclude the overwhelming majority of rape women commit. these are regrettably just what a lot of official sources will give you
my reading on this is EA/rationality has quite a few people who are very well informed and thoughful and like to geek out about this stuff, although culturally not quite the same as you and also more than a few people (and I say this while still saying I have much respect for the guy) who should read something on this topic by someone not named Scott Alexander
I think you’ll find varying views on this, some people talk about how trans female upbringings/ability to assimulate to the local “male” culture are in many ways different to baseline cis men and there is a lot of truth in this as well, certainly the median pretransition trans woman is probably several sd less assertative than the median cis man.
I guess I’m nonzero of a biodeterminist of this because of things like sex drive being pretty different, I don’t think this necessarily implies a pessimistic viewpoint on this. People who steal cars are more likely to like cars, but presumably there are other character traits that cause people to steal cars and we don’t throw up our hands and say solving car theft is impossible
ok I guess trans women are more likely to be in very sexually open subcultures, unclear if this helps or harms
(in general I endorse the style guide for a lot of this stuff being ‘cis men are like so, cis women are like so, i have no idea what trans people are like’ instead of guessing)
It sounds like you want to engage constructively to reduce abuse in the community, and I appreciate that. The community will be stronger in the long run if it can be a safer and more welcoming space.
I know we’re a bunch of weirdos with a very specific set of subcultural tics, but I hope everyone appreciates your efforts to help. I think people here really are unusually motivated to do good and there is a lot of goodwill as a result. On the other hand, I think a lot of that is ego driven. And it’s a very nerdy culture, male-dominated and probably many people here have a predictable set of blind spots as a result.
Wish I had more to say, or could do more to help, but I’m not in the bay area, don’t work in tech, and don’t have very much context for the cultural problems you’re encountering.
Scott Alexander said “0.3% chance of false accusation”, I calculated 1⁄4000 ( This number is based on the FBI UCR study: 93,000 reported rapes yearly X (.97 X .03 X .2 chance of falsely naming a man) X 75 years / 160000000 men = 0.025% or 1 in 4000 chance of being falsely accused of rape. FBI: UCR. “2015 Crime in the United States.” FBI. Accessed July 23, 2022. https://ucr.fbi.gov/crime-in-the-u.s/2015/crime-in-the-u.s.-2015/offenses-known-to-law-enforcement/rape)) Either way, still rare.
… Men have a 1 in 4,000 or 0.025% (or 0.3%, if we use Scott Alexander’s number) chance of being falsely accused.
This is a highly misleading summary. 0.3% is not Alexander’s estimate, it is his attempt to get an ultra-conservative impossible-to-be-lower-than-this lower bound. His best guess was literally 10x higher:
So greater than 0.3% of men get falsely accused of rape sometime in their lives, and the most likely number is probably around 3%.
This is over 100x higher than your number. And it’s not even his upper bound, which is significantly higher still.
EA has copped a lot of media criticism lately. Some of it (especially the stuff more directly associated with FTX) is well-deserved. There are some other loud critics who seem to be motivated by personal vendettas and/or seem to fundamentally object with the movement’s core aims and values, but rather than tackling those head-on, seem to be trying to simply through everything that’ll stick, no matter how flimsy.
None of that excuses dismissal of the concerning patterns of abuse you’ve raised, but I think it explains some of the defensiveness around here right now.
You’ve mentioned here a few times that you spoke to Julia about a few of the people implicated in the TIME article. Were you happy with what Julia/the community health team did as a response? How much of this can you share? Do you know if the decision of the victims to talk to the journalist was prompted by the community health team’s response?
This is not a fair description. The way people get such statistics is by assuming all accusations are true unless there is strong evidence against them, but there is a large number with no strong evidence either way, and researchers should not just assume they are all true.
A good first place to start is the Wikipedia Article on the subject, which features a wide range of estimates, almost all of which are higher than the 2-3% you say, some of which being dramatically higher.
https://en.wikipedia.org/wiki/False_accusation_of_rape
Alexander also has a good blog post on this:
https://slatestarcodex.com/2014/02/17/lies-damned-lies-and-social-media-part-5-of-%E2%88%9E/
Your own website lists a slightly higher range, 2-4%
[redacted]
If we look at the source you supply for the 2% we see a different story:
Only 2.1% were designated by the cops as false, but that was after the majority of accusations had been withdrawn or determined ‘no further action’. Many of those withdrawn or no-further-action accusations would also have been designated as false if they had been investigated.
https://www.ojp.gov/ncjrs/virtual-library/abstracts/study-reported-rapes-victoria-2000-2003-summary-research-report
This sounds like you are suggesting (correct me if I’m wrong) that many or most withdrawn and ‘no-further-action’ accusations are actually false, which is not a fair conclusion to draw from the information you presented. It also seems to claim that cases designated as withdrawn or ‘no further action’ are not investigated, which is also not accurate. Specifically, the paragraph you shared mentions many of the reasons victims will withdraw or not seek further action:
Withdraw:
“when complaints were withdrawn, suspects were more likely to be current or former partners of the complainant” Victims of domestic violence are often in immediate physical danger when attempting to leave or report, or are extremely financially limited due to their situation in an abusive relationship. Following through on an accusation could pose more danger to the victim than withdrawing.
Investigating sexual assault cases can be extremely invasive, and often re-traumatize the victim. Cases can then take years to prosecute and involve cross-examining the victim. Some victims withdraw to avoid this process.
No-Further-Action: This is a police designation indicating they decided not to investigate for any number of reasons, including lack of forensic evidence, uncooperative witnesses, lack of resources, or a belief that they will not be able to successfully prosecute. None of these directly imply a false claim.
At least in the US, the reports often initially come through Patrol officers who typically don’t receive the specialized training in responding to sexual assault victims (e.g., trauma-informed training) that investigators working in domestic violence or sex crimes units do, and may be less likely to take reports seriously.
“Cases that resulted in “no further police action” were more likely to involve younger victims” Sexual assault is rarely the only issue in cases with young victims, and gathering evidence is made more complicated by this.
“victims who had consumed alcohol or other drugs near the time of the offense” In addition to the potential for police prejudice against these victims, there’s also the consideration of whether or not a jury is likely to find a victim’s story compelling. This indicates a judgement on the potential for a successful prosecution, not on whether a crime was actually committed.
I agree that this is a “different story,” just not one indicating a secret cache of false claims hidden by statistics. Instead, it’s a depiction of the extremely messy process of reporting, investigating, and prosecuting these cases.
I can’t access the full article from which you pulled that abstract, but I used a number of the resources below in my response. I hope you find them informative:
https://www.ojp.gov/pdffiles1/nij/grants/252689.pdf (Particularly the qualitative interviews where police describe lack of resources, lack of training for patrol officers)
https://www.uml.edu/News/stories/2019/Sexual_Assault_Research.aspx (short summary of the above)
https://www.london.gov.uk/sites/default/files/london_rape_review_final_report_31.7.19.pdf (seemingly a similar type of report to the abstract you included, but with context around the findings)
https://www.psychologytoday.com/us/blog/finding-new-home/201810/rape-allegations (on some reasons people don’t report, which seem relevant here)
It is logical that people would be more likely to withdraw, and the police less likely to investigate, accusations that seem less compelling. If any of these cases are false, the 2.1% is an under-estimate; if they are false at a higher rate than cases accuser and police followed through with, the 2.1% is a significant underestimate.
‘Any’ is different than ‘many,’ as you originally claimed. I think if the report you seem most focused on was estimating that 2.1% of all rape claims are false accusations, your concern would be more understandable. But this is a paper on rapes reported in a specific geographical area over a 3-year period and 2.1 is the percentage of reports designated false by the police, making this an odd choice of information to focus in on as telling a “different story.” Most groups that compile these data in order to make estimates do place those estimates in ranges, anywhere from 2-3 to 2-10%.
This still seems to suggest that cases resulting in withdrawals or no further action are more likely to be false, and you don’t have enough information for that assumption to be well-founded. Further, this assumption feeds into harmful myths about the underlying causes for withdrawal and case attrition in sexual assault reports, many of which are addressed in the sources I’ve linked above.
I’ve removed the website of a user from the above comment after a request to the mod team, in light of our new policies on revealing personal information on the Forum
This thread, and these kinds of discussions are very revealing and worth pursuing to the end.
Let’s go with Scott Alexander’s estimate of a 3% lifetime chance of a man being falsely accused of rape and a 15% chance of a woman being raped. Let’s assume EA is 70% men vs 30% women.
How much “weight” should the community give to guarding against false accusations vs to guarding against the prevalence of sexual abuse (since not believing accusations lets the perpetrator repeat their actions)? This includes a value judgement on the harm from sexual assault vs the harm from reputational damage, etc.
Since this seems to be a difficult tradeoff and the community health team/EA organization leaders are making these tradeoffs that include value judgements (and EAs don’t usually have the same value judgements as the rest of society), the current community members and people in charge should be transparent about these value judgements and overall weights (maybe through a survey?).
It would help current and future EAs decide if they want to be a part of the community based on how they value their own welfare.
I think one question would be the extent to which the types of actions Community Health can take would reduce a perpetrator’s ability to commit future assaults. I don’t know the answer to that but would be interested in what others thought.
I understand the desire to for quantification and transparency, but my guess is this particular quantity is relatively far down the list of parameters to publicly estimate, especially given the methodological difficulty. If you thought it was particularly pressing maybe you could have a go at it? Or perhaps there is some existing economic literature on the subject. I suppose the level burden of proof applied in court cases is a starting point.
This debate seems like an unhelpful tangent given that the original comments about false accusations weren’t even referring to sexual assault, based on Julia’s clarification: https://forum.effectivealtruism.org/posts/JCyX29F77Jak5gbwq/ea-sexual-harassment-and-abuse?commentId=tFxjdj34q8xoaWut3.
I’m sorry you deleted the original text of your response, as I would have liked to engage with it. Unfortunately I did not save a copy before you deleted and replaced it.
I did however get a copy of this comment you made prior to changing your forum username, where you linked to us all to your website.
Strangely you seem to have deleted this comment now and completely forgotten about having written it. It is very ironic that, in the context of a discussion about the frequency of false accusations, that you would attempt to alter the digital record and then falsely imply I am a creep.
Perhaps this was an accident. It is easy to forget what exactly we have said, especially if we regret it later. If so I hope you take this lesson to heart about how easy it is for even people with the best of intentions to accidentally make a false accusation.
- - -
Screenshot of the above comment in case of further edits:
I’ve removed the name and website of a user from the above comment after a request to the mod team, in light of our new policies on revealing personal information on the Forum
I’m not sure I understand the rationale for removing information that was supplied by the very same person who now says they want it removed, especially when this information was supplied merely a week ago, on the EA Forum, and in this same thread. The policy that this decision seems to exemplify appears to effectively give anyone the right to censor any information about themselves in posts or comments made by others, regardless of how that information was obtained or how public it is.
Note that the case for disclosing the information in this particular instance was pretty strong: J_J implied that temp_ was a creep for knowing J_J’s website, but it turns out that J_J had included a prominent link to their website in a comment posted just one day earlier. I do not want the Forum to be a place where people can make unfair accusations about others and retain a right to suppress evidence establishing the unfairness of those accusations.
Thank you for the feedback,
As mentioned in the policy, we do think that there are cases when some personal information is important to share, and we don’t think everyone should have the right to censor any information about themselves.
We do consider how public the information is and “when information is easily accessible elsewhere, we will err on the side of keeping it”, but we also strongly consider how relevant the information is to EA.
In this case, I felt that the name and website of the user are not relevant enough for effective altruism to justify keeping the information in the comment against the user’s wishes.
I agree, but I think that in the edited comment it’s still clear that J_J had included a prominent link to their website in a comment posted just one day earlier. If that’s not the case I should have edited it differently (possibly writing [user’s website] in the black box). Do you think it should be clarified?
Thanks for the reply. I think the crux of our disagreement may be that I don’t regard “being relevant to EA” as a necessary condition for declining a request to remove personal information, unless that phrase is given a very broad interpretation that includes things like “keeping the EA Forum a place where people can’t make unfair accusations about others”.[1] Separately, if a user voluntarily discloses a piece of personal information, I think this should be beyond the scope of mod action, unless something happened in the intervening period that clearly justifies removing or encoding the information. People can still ask others not to share this info, but I think it should be up to each person to honor those requests, rather than being something enforceable by the admin team.
In this case, as you note, it was possible to remove the personal information while preserving the relevant evidence publicly, although I think the removal made it somewhat more difficult to appreciate what was really going on. But one can imagine other situations in which this cannot be done.
J_J did not accuse temp_ of being a creep. J_J said she was creeped out. There is a subtle but important difference between these two statements.
As I wrote, “J_J implied that temp_ was a creep”, and implying that someone is a creep is a way of making an accusation, in this case an unfair one.
What she said is that she felt creeped out. The implication that the person who’s actions she was creeped out by was a creep is an implication you are adding to her statement, and not one I did personally. It is very much possible to feel creeped out by someone who isn’t a creep.
Reading it as an underhanded way to make an unfair allegation seems very uncharitable to me in light of all her comments.
I don’t understand what rule you think I broke. This was a link to a public website that she herself shared on the thread; it does not fall into any of the categories in the linked website. This is someone who is making serious allegations about EA, and looking to be paid for it—she should not be able to demand others users delete any record of what she has done.
The closest reference I can see in the rules is this:
But then you should have encoded it, not deleted it and edited my screenshots.
Maybe some context would be helpful. There was a previous case in EA with a woman named Kathy Forth who made a bunch of accusations. The consensus seems to be that those accusations were false and/or overblown. It seems many people knew this and shared it through a whisper network, and Scott Alexander was one of a fairly small number of people willing to state it publicly. So, hopefully this explains why dismissing him might be a bit of a red flag with me and some other people here.
BTW, I’m wondering if a good heuristic is: If someone makes a lot of accusations, they’re likely to be a liar. If someone receives a lot of accusations, they’re likely to be guilty. The idea being that genuine victimhood (from a crime/false accusation) happens to people at a fairly even background rate, but people who are bad actors tend misbehave more than once.
In my personal experience talking to victims since, I’ve noticed that this is one of the most scary heuristics of all. Every woman I’ve spoken to in this community who comes forth about SA once feels like they are now more than twice as vulnerable for being targeted by predators.
Very personally, several of my house co-founder’s friends had an intentionally loud conversation outside my door about unfortunate things that might happen to “girls who cry wolf” since nobody will believe them on subsequent claims.
That’s horrifying, sorry to hear about it
wait, what? how do we know this?
I underwent a palpable shock at this factoid, it was unexpected, while also being sad etc.
I don’t mean this as a comment on the particular case reported in the TIME article, though I’d reject using naive base rate calculations as a last word on someone’s probabilistic guilt. but the “only 2-3% of allegations are false” stuck out to me because I read a better estimate is probably more like 2-10%. https://slatestarcodex.com/2014/02/17/lies-damned-lies-and-social-media-part-5-of-∞/ there’s a lot of ambiguity here—issues like not every “report” is an “allegation” because sometimes reports don’t name a perpetrator. I have no idea what the correct figure is, but it seems to me the 2-3% figure gets probably bandied around a lot probably with a sense of precision and finality that isn’t warranted by the evidence. Happy to see new evidence or information to the contrary, and whether the rate is 2% or 10% it can certainly be described as “low”.
The article is a nearly a decade old, and for all I know there might be newer research. But I hope when people are thinking about best practices for the future, they do so on the basis of the best evidence available.
Thank you for your explanation. I appreciate you taking the time to explain your reasoning on that point and find it useful for being confident in the rest of what you have to say here.
I think the main reason that EA/rationalist spaces and activist spaces have historically often not gotten along well is that in EA/rat spaces there is a strong norm against a certain kinds of strongly worded moral injunctions.
I think the reason for this is that EA tends to attract people driven to do good but with perhaps an OCDish tendency to worry if they’re doing enough and push themselves too hard. If we instead went around saying it was shameful people were still eating meat and also shameful people weren’t giving more money instead of using it to vist friends and also shameful they weren’t thinking enough about how to prevent sexual assault people would pretty quickly burn out.
(There are certainly people in the community who can and do translate a lot of things activists say into EA/rat friendly language)
Of course, if you assume the best of too many people you let in bad actors and I suspect a lot of people might find it stifling. But it does create a safe space for many of us.
fwiw if I were were CEA’s health team I would phrase this as “although they are a small minority we have occasionally come across accusations which were as far as we could tell false” or some such.
I’m not sure what can be gained over nitpicking over the exact number—this is going to vary from place to place
I think the “don’t want change” faction is a good deal smaller than the “we would like time to digest and decide what changes seem like a good idea and what typical problems actually look like (given how heterogeneous EA is) and how to make these changes while keeping some aspects of the culture we find valuable” faction
(also, as a trans women, I’m going to have to raise an eyebrow at “men and AMAB”—i think the actual risk factor is hormones fwiw but...
ed: you know what I think these are “rape by envelopement doesn’t count” numbers, which exclude the overwhelming majority of rape women commit.
these are regrettably just what a lot of official sources will give you
ed2: an explanation https://thingofthings.wordpress.com/2015/02/28/346/
this has updated me strongly towards “I would personally not trust you if I were to report a sexual assault”)
my reading on this is EA/rationality has quite a few people who are very well informed and thoughful and like to geek out about this stuff, although culturally not quite the same as you and also more than a few people (and I say this while still saying I have much respect for the guy) who should read something on this topic by someone not named Scott Alexander
I got a lot out of archive binging Ozy Brennan’s blog (probably helpful to search by tags eg sex positivity, feminism)
https://thingofthings.wordpress.com/
https://thingofthings.substack.com/
I think you’ll find varying views on this, some people talk about how trans female upbringings/ability to assimulate to the local “male” culture are in many ways different to baseline cis men and there is a lot of truth in this as well, certainly the median pretransition trans woman is probably several sd less assertative than the median cis man.
I guess I’m nonzero of a biodeterminist of this because of things like sex drive being pretty different, I don’t think this necessarily implies a pessimistic viewpoint on this. People who steal cars are more likely to like cars, but presumably there are other character traits that cause people to steal cars and we don’t throw up our hands and say solving car theft is impossible
ok I guess trans women are more likely to be in very sexually open subcultures, unclear if this helps or harms
(in general I endorse the style guide for a lot of this stuff being ‘cis men are like so, cis women are like so, i have no idea what trans people are like’ instead of guessing)
It sounds like you want to engage constructively to reduce abuse in the community, and I appreciate that. The community will be stronger in the long run if it can be a safer and more welcoming space.
I know we’re a bunch of weirdos with a very specific set of subcultural tics, but I hope everyone appreciates your efforts to help. I think people here really are unusually motivated to do good and there is a lot of goodwill as a result. On the other hand, I think a lot of that is ego driven. And it’s a very nerdy culture, male-dominated and probably many people here have a predictable set of blind spots as a result.
Wish I had more to say, or could do more to help, but I’m not in the bay area, don’t work in tech, and don’t have very much context for the cultural problems you’re encountering.
This is a highly misleading summary. 0.3% is not Alexander’s estimate, it is his attempt to get an ultra-conservative impossible-to-be-lower-than-this lower bound. His best guess was literally 10x higher:
This is over 100x higher than your number. And it’s not even his upper bound, which is significantly higher still.
https://slatestarcodex.com/2014/02/17/lies-damned-lies-and-social-media-part-5-of-%E2%88%9E/
EA has copped a lot of media criticism lately. Some of it (especially the stuff more directly associated with FTX) is well-deserved. There are some other loud critics who seem to be motivated by personal vendettas and/or seem to fundamentally object with the movement’s core aims and values, but rather than tackling those head-on, seem to be trying to simply through everything that’ll stick, no matter how flimsy.
None of that excuses dismissal of the concerning patterns of abuse you’ve raised, but I think it explains some of the defensiveness around here right now.
You’ve mentioned here a few times that you spoke to Julia about a few of the people implicated in the TIME article. Were you happy with what Julia/the community health team did as a response? How much of this can you share? Do you know if the decision of the victims to talk to the journalist was prompted by the community health team’s response?