Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

username2 comments on I Want To Live In A Baugruppe - Less Wrong Discussion

44 Post author: Alicorn 17 March 2017 01:36AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Comments (189)

You are viewing a single comment's thread.

Comment author: username2 17 March 2017 08:49:43PM 10 points [-]

I feel a little bit morally obligated to point out the following.

The FBI estimates that each child has almost a 25% chance of being molested, that 4% of adults are sexually attracted to children, and that 70% of children were molested by people they knew and trusted. These number seems to at least roughly comport with my personal understanding of the world and my knowledge of the lives of people close to me.

The horrifying ubiquity of sexual predation of children must at least be mentioned under "Obstacles".

The unfortunate reality is that invitations to group living situations select for predators. No, your radar is not tuned to keep them out. No, you cannot sufficiently vet them after a few hours of interaction and observation of their children. If you think I'm being paranoid, I would argue that no, if 25% of children are likely to be molested, you're probably not being paranoid enough.

I would love it if this weren't true, but this is the world we live in.

I'm sure there are measures that can be taken to ameliorate this issue, but just ignoring it is not one of them.

Comment author: Viliam 18 March 2017 06:20:45PM *  12 points [-]

I want to urge people to not dismiss this without a thought. And it's not just about children.

There are already a few sexual predators hanging around with the rationalist community. I can't say names, because it is typically a "they said, they said" situation, and these types usually have a lot of practice at threatening legal consequences for "slander". (But if you know someone who used to be around and suddenly lost all interest at coming to your meetups, it might make sense to ask them discreetly whether they had a bad experience with someone specifically.)

I personally often don't care much about the statistics for general population, because we are obviously not average. Problem is, "not average" doesn't in itself show the direction. For general intelligence, we are obviously smarter, and that generally correlates with lower (detected?) crime. On the other hand, we also seem to score quite high for unusual sexual behavior in general.

As long as each family has a door they can close (and everything necessary to survive the day is inside), living in a community doesn't seem worse than simply living with neighbors. But there are good reasons why neurotypical people require long time before they start trusting someone; so "they are a member of the same community" should never be used as a replacement for "I have a lot of personal experience from gradually deepening interaction with this specific person". In other words, just because someone says "hi, I also like Less Wrong", doesn't mean I would invite them to my home and leave them alone with my child. Some nerdy people may need to be explicitly reminded of this.

Comment author: ChristianKl 26 March 2017 09:52:25AM 2 points [-]

But there are good reasons why neurotypical people require long time before they start trusting someone; so "they are a member of the same community" should never be used as a replacement for "I have a lot of personal experience from gradually deepening interaction with this specific person".

Given that most abuse happens from people who aren't strangers and successfully passed the filters of neurotypical people that are required to build trust, I don't think trusting a person because you spent a lot of time with them is generally a good heuristic.

Comment author: Viliam 27 March 2017 02:00:05PM *  2 points [-]

trigger warning: discussing rape, in near mode

Now sure how much can I generalize from the few data points of women who trusted me personally enough to tell me about their bad experiences, but within that set, it was neither the archetypal "stranger hiding in a bush", nor the archetypal "lecherous uncle". I remember these three patterns:

a) Girl's mother has a new boyfriend. In mother's absence, the boyfriend starts making sexually suggestive remarks to the girl. Girl complains to her mother. Mother confronts the boyfriend, he dismisses it with laughing, telling the mother that her daughter is simply jealous of him, wanting to keep the mother only for herself. Mother gets angry, scolds the daughter for "lying", and categorically refuses to listen to her arguments anymore. Girl stops reporting to her mother, and their relationship goes from already quite bad to completely ignoring each other. The boyfriend keeps pushing further. (Luckily, in the cases I heard, the predictable bad end didn't actually happened, because something totally unrelated disrupted the setting.)

b) A girl is at a party with her friends. There is also a guy, stranger to her, but friendly with her friends. The party either happens at the guy's place, or at a large place with many rooms. Girl remains talking with the guy, while other people gradually leave. When they are left alone in the room, the guy suddenly becomes physical and rapes her. (In one case, when the girl afterwards starts inconspicuously asking their friends what is exactly is their relationship with the guy, she is surprised to hear almost all of them telling her "actually, I don't like that guy, he seems like an asshole, but he is a friend of my friends, so I just try to ignore him when he comes to a party" or "I noticed him, but didn't pay any attention".)

c) A girl's boyfriend constantly refuses to take "no" for an answer; starting with relatively small things, gradually increasing the requests, until one day he rapes her. The girl keeps dating him, until later something else ends their relationship.

Again, not sure how typical these stories are, but... assuming they are relatively frequent, then the dichotomy between a stranger and a non-stranger doesn't properly fit the territory.

Technically, all three cases are "non-strangers". I believe that in many surveys, "mom's boyfriend" would even be classified as a family; the friend-of-friends is a part of the social circle; and the boyfriend is obviously not a stranger, if they already spent some weeks or months dating.

Yet, in the first two cases, the rapist was a stranger to the girl, which makes him quite a non-central example of a "non-stranger". In the first case, she was unable to avoid him; he didn't really "pass her filters". In the second case, that was the kind of error in judgement that I warn against -- believing that the other person was already filtered by someone else, when actually the other people similarly believed that someone else did the filtering, or were just being polite towards a person that didn't pass their personal filter but didn't seem bad enough to initiate a conflict. The third case, yeah, that was a direct failure at filtering.

So, I object connotationally against the suggestion that it is useless to use "spending a lot of time with someone, without seeing something bad" as a heuristics against abuse, because most abusers pass the filters of the neurotypical people anyway. First, we don't know how many abusers didn't pass the filters; maybe without them, there would be much more abuse. Second, in these three examples, the filter (a) couldn't be used by the victim, (b) was used improperly, and (c) wasn't used at all. I am not saying the filters are flawless, just that not using them at all is a fallacy of grey.

For a rationalist community living together, I suspect the first two scenarios could be relevant. A parent, for whom living with other rationalists has high value, might turn a blind eye to the red flags reported by their children or a spouse, and rationalize them away. A person no one actually personally vouches for could be invited, simply because they participated at a LW meetup, said hello to many people, and friended them on facebook.

I think it would be reasonable for the wannabe neighbors to spend some time together before buying the new house. For example, spend a vacation together, preferably at a place where you are expected to cook for yourselves. And maybe, collect some feedback on personal feelings towards each other, in a way that would prevent transitive reporting of "I feel X, but I guess most people are going to say Y, therefore I am saying Y too". Be honest; not having an opinion either way is a valid option. Generally, have an intermediate step between "met each other at a meetup" and "living in a baugruppe".

Comment author: ChristianKl 27 March 2017 08:00:00PM 0 points [-]

I think ⓐ is an example where trust is given because they mother knows the guy and has a relationship with a guy but the trust isn't warranted.

In situations like this our system I is trained to trust and it takes hard system II thinking to acknowledge the problem and respond well to the incident.

The problem is further exacerbated because people treat their stereotypical idea's of how an unsafe person looks as if it would be real knowledge.

This overall conversation is a good example. The guidelines around risk of abuse suggest that having a good support network reduces risk. At the same time you have a person who is afraid of strangers and who thinks minimizing the amount of trusted adult relationships helps to reduce the risk of abuse and they argue their opinion.

An intelligent psychopath doesn't give up the kind of red flags that result in most neurotypical people distrusting them.

I am not saying the filters are flawless, just that not using them at all is a fallacy of grey.

Having a way to filter people is useful for many reasons but at least in our Berlin community we don't lack processes to do that. Both our weekly Dojo and our new biweekly Circling event isn't simply open to everybody and participating at one of the open meetups doesn't automatically qualify a person.

Alicorn also wrote in the OP about having resident- and guest-vetting plans.

More centrally I don't think you should plan in a way that assumes that your filtering process actually keeps out every problematic person. Open sharing of information is important. The way the girl in ⓒ would have been helped is when she shared her issues with friends who talked her through it.

Comment author: Viliam 28 March 2017 11:07:32AM 0 points [-]

I agree that evaluation of other people needs to be an ongoing process. Sometimes people change. Or some people behave differently to different kinds of people, so it's possible that the original evaluator just happened to be one of those towards whom this person feels no hostility.

But I'd still say that new people are a much higher risk, simply because when crappy people are expelled from one community, they are looking for another one, so they are statistically overrepresented among the newcomers. (A similar effect to how software companies, when doing job interviews, mostly find crappy programmers. Because the good ones already have a good job somewhere, but the crappy ones remain endlessly in circulation. If there are 10 competent programmes in the city and 1 crappy one, and 10 software companies, it's possible that each of those companies will interview the crappy guy, and reject him, and then one of the competent guys, and keep him; so even if the crappy guy is only 9% of the population, for each software company he makes 50% of the interviewees.)

Comment author: ChristianKl 28 March 2017 01:35:35PM 0 points [-]

I agree that evaluation of other people needs to be an ongoing process. Sometimes people change. Or some people behave differently to different kinds of people, so it's possible that the original evaluator just happened to be one of those towards whom this person feels no hostility.

Quite a lot of psychopaths do manage to make a good first impression and have charisma they aren't simply crappy people. Still they might misbehave when they believe that it doesn't have negative consequences for them.

Comment author: Rubix 18 March 2017 07:18:55PM 1 point [-]

Endorsed.

Comment author: Alicorn 17 March 2017 11:29:30PM 8 points [-]

I'm pretty sure the solution to this problem is not "trust no one, be a hermit".

Comment author: username2 18 March 2017 01:16:17AM 1 point [-]

I think the word "trust" probably wouldn't exist in a rationally designed language. If I know the base rate of pedophiles is 4% then I will expect that each person I meet has a 1 in 25 chance of being one. If they demonstrate certain qualities to me I will gradually update downward until I reach a point where they become an acceptable risk. There's absolutely no reason to trust anybody on this front without such an analysis.

Comment author: Rubix 18 March 2017 01:49:05AM *  13 points [-]

Source on those statistics, please? I find the claims dubious: in particular, the 25% figure seems to come from this "information packet", which is unsourced and uncited, suggesting that it may not exist. The two Jensens, Cory Jewell and Steve, seem to build a career around inflating the numbers associated with child sexual assault. I can't find sources for either of the other figures.

My stake in the game: I strongly distrust statistics given about child sexual assault unless they are highly specific about what is being discussed, for two reasons.

One is that the definition is incredibly vague: some sources mean "an adult engaging in intercourse with a minor under 13", others mean "touch intended to be sexually gratifying, of a minor under 18, by another party of any age", and definitions run the gamut. Another example: under this website's definition of child sexual abuse, "any sexual activity between adults and minors or between two minors when one forces it on the other (...) like exhibitionism, exposure to pornography", I was sexually abused at 11 when a chatroom troll sent me a link that turned out to be Two Girls, One Cup.

My second reason for reservation around these statistics is that they rarely take into consideration the preferences of the minor. When I was a minor, I had healthy and fulfilling sexual relationships; under many existing definitions, I was sexually assaulted by my loving sixteen-year-old boyfriend when I was sixteen, and under many more I was sexually assaulted by him when he turned eighteen and I was still seventeen. This seems ridiculous and objectionable to me.

A last note: I agree that it is impossible to tell from a few hours of interaction whether someone will abuse your child. Many people can't tell even after years of loving marriage whether their spouse will abuse their children, so "demonstrating acceptable qualities" is not a very good intervention. The absolute best defense against one's children having unwanted/traumatic interactions is to tell them how to set boundaries, tell them to yell if they're touched in a way they don't want, tell them that their body is their own and that nobody gets to touch it without their permission. This has the virtue of defending against all manner of abuse and mistreatment, at the hands of parents, extended family, family friends and acquaintances alike.

Comment author: username2 18 March 2017 04:51:34PM 4 points [-]

I couldn't find the original page I was getting those numbers from, but here's another that gives a bit more granularity. It does seem like that 25% number interprets "sexual abuse" very broadly, but the more detailed numbers are still horrifying and still cause for alertness.

A last note: I agree that it is impossible to tell from a few hours of interaction whether someone will abuse your child. Many people can't tell even after years of loving marriage whether their spouse will abuse their children, so "demonstrating acceptable qualities" is not a very good intervention. The absolute best defense against one's children having unwanted/traumatic interactions is to tell them how to set boundaries, tell them to yell if they're touched in a way they don't want, tell them that their body is their own and that nobody gets to touch it without their permission. This has the virtue of defending against all manner of abuse and mistreatment, at the hands of parents, extended family, family friends and acquaintances alike.

Indeed, I didn't say "this is a horrible idea, Alicorn." I was just trying to mention this consideration, which I was a bit surprised not to see mentioned in the original post. If the children are all well-educated about how to respond to attempted abuse, and the adults all know this, a strongly abuse-deterring environment is created.

Comment author: Lumifer 18 March 2017 03:23:46AM 0 points [-]

Y'know, there is a medical diagnosis for this...

Comment author: username2 18 March 2017 09:43:50AM *  0 points [-]

I'm also pretty sure the solution is to NOT put your children in a shared living situation with a dozen other possibly-predacious adults. There is a middle ground of having a secure, private environment for your family with walls and clear separation. Such as most conventional living situations.

Comment author: ChristianKl 26 March 2017 09:16:19AM 0 points [-]

There is a middle ground of having a secure, private environment for your family with walls and clear separation. Such as most conventional living situations.

Most conventional living situations lead to the abuse rate of 25%. I don't think you have provided good evidence that the conventional layout is better.

Comment author: tenshiko 18 March 2017 12:16:33AM 6 points [-]

If these statistics are likewise correct, about half of child molestations involve a direct family relationship. "Stop adding children to your family" seems like a pretty unrealistic method of preventing child molestations from occurring. Then again, a pretty substantial chunk of child molestors are trusted non-relatives, so I see how the baugruppe would disproportionately enable that demographic.

"Do not let parents be alone with their own children" likewise seems pretty unrealistic. Would you want to suggest that a non-parent should be limited in their time alone with a child?

Furthermore, there will only be one baugruppe. Perhaps two. Aren't the participants in such an enterprise disproportionately likely to be economically advantaged with consistently present parents, and therefore less appealing targets for child molestors?

Comment author: username2 18 March 2017 09:53:24AM 2 points [-]

The whole point of the objection. as I (different persona) interpret it, is that a shared living situation is effectively adding more adults to each family. The whole point is to have one big cohesive group living together, which means a lot more people in the same proximity and familiarity as family members in more conventional arrangements.

Put differently, the underlying causes of why most predation seems to come from family has nothing to do with sharing genetic material, but rather things like availability of opportunity and trust. Both of which are features of this shared living environment. So we should expect the individual risk to be higher than 1 in 25 and in fact closer to the rates from family members.

Comment author: Jonathan_Lee 18 March 2017 07:29:23PM 4 points [-]

Even if it's the case that the statistics are as suggested, it would seem that a highly effective strategy is to ensure that there are multiple adults around all the time. I'll accept your numbers ad arguendo (though I think they're relevantly wrong).

If there's a 4% chance that one adult is an abuser, there's a 1/625 chance that two independent ones are, and one might reasonably assume that the other 96% of adults are unlikely to let abuse slide if they see any evidence of it. The failure modes are then things like abusers being able to greenbeard well enough that multiple abusers identify each other and then proceed to be all the adults in a given situation. Which is pretty conjunctive as failures go, and especially in a world where you insist that you know all the adults personally from before you started a baugruppe rather that letting Bob (and his 5 friends who are new to you) all join.

You also mention "selection for predators", but that seems to run against the (admittedly folk) wisdom that children at risk of abuse are those that are isolated and vulnerable. Daycare centres are not the central tendency of abuse; quiet attics are.

Comment author: drethelin 18 March 2017 03:19:28AM 3 points [-]

This is paranoid, but even if it wasn't: The more people living in one house the LESS likely someone is to get away with molesting someone else unnoticed.

Comment author: username2 18 March 2017 04:44:04PM *  3 points [-]

As the other anonymous said, this doesn't follow at all. A group living situation creates a larger field of "trusted adults" per child. Unless all the adults are mindful of these risks, a situation arises where any adult may at any time be put in charge of watching any child or children. This is frankly the textbook definition of what not to do.

If the adults are mindful of the risk, then they can be open about it, and ensure that two or more adults are always tasked with watching children, so that the adults can watch each other. And even this may eventually cease to be necessary.

Also, I find that your definition of paranoid must be different from mine if you look at those statistics and think "nothing risky going on here". I have to assume you have no personal experience with this issue. I can't help but feel like people in this thread are conflating a feeling of "I don't want this to be true and I don't want to have to think about it" with "this is obviously overly paranoid".

Comment author: drethelin 18 March 2017 09:44:28PM *  4 points [-]

I think the statistics you quote are exaggerated in order to terrify. When I tried to look up "4% of adults are sexually attracted to children," for example, I found nothing. Similarly, the news is often full of stranger danger fears because terror is what gets attention and therefore revenue and funding. And as others have said, they also include stuff like 18 year olds having sex with 17 year olds, which some people may find unacceptable but I don't.

Comment author: gjm 19 March 2017 01:38:41AM 8 points [-]

Note also that "4% of adults are sexually attracted to children" is a very different statement from "4% of adults are likely to molest children if left alone with them".

(I suspect rather more than 4% of adults are sexually attracted to Angelina Jolie[1], but that doesn't mean they'd molest her if left alone in a room with her.)

[1] Chosen by putting "famous actress" into Google and picking the first name it gave me. If she isn't your type -- she isn't particularly mine, as it happens -- feel free to imagine I chose a different name.

Comment author: ChristianKl 26 March 2017 09:00:11AM *  0 points [-]

Even if 4% of adults are sexually attracted to children that doesn't mean that they are going to abuse children. There are guy's in this communities who are sexually attracted to women but who never had sex and also wouldn't rape a woman just to have sex.

If it's clear a rationalist that abusing a child will mean that he get's expelled from the community in which he lives and might face legal challenges than I think most of the people in this community wouldn't act on a system I desire to engage in sexual abuse because their system II is strong enough to think through the situation.

Practically that means that it's important to have an environment where open communication happens so the expectation that a child will communicate about situations with whom they are uncomfortable exists. I think a lot of abuse does happen in environments where that open communication is lacking and a child will stay silent about abuse.

Comment author: username2 19 March 2017 02:05:30AM 0 points [-]

Sounds a lot more like rationalization than rationalism.

Comment author: drethelin 19 March 2017 02:52:34AM 0 points [-]

This is why we need downvotes.

Comment author: username2 19 March 2017 05:38:44AM 0 points [-]

I was being quite serious. When given a quantitative argument you responded with a grab bag of abstract objections not backed by data but vaguely supporting your original viewpoint. A natural human response designed to keep one from changing their mind, generally called rationalization. I encourage becoming aware of when this is happening and use that awareness to improve your model of the world.

Comment author: ozymandias 19 March 2017 03:12:06PM 3 points [-]

I think an accurate qualitative argument is better than a sourceless quantitative argument.

Comment author: Lumifer 19 March 2017 04:19:05PM 1 point [-]

When given a quantitative argument

Numbers are not particularly magical and being quantitative doesn't imply the argument is more likely to be correct. After all, "there are lies, damn lies, and statistics".

Comment author: ChristianKl 26 March 2017 08:55:36AM 2 points [-]

As the other anonymous said, this doesn't follow at all. A group living situation creates a larger field of "trusted adults" per child.

Do you have sources that suggests that having a larger circle of trusted adults per child increases the likelihood of getting abused?

Comment author: username2 26 March 2017 05:38:43PM 0 points [-]

Summation of probabilities.

Comment author: ChristianKl 26 March 2017 06:00:25PM *  2 points [-]

You could also subtract properties or multiply or divide them.

More trusted adults might increase the chances that the child isn't isolated and talks about his experiences with someone which makes them less susceptible to be a victim.

The WHO for example says that among the risk factors for abuse there are:

being isolated in the community or lacking a support network

a breakdown of support in child rearing from the extended family.

If your true concern is the children not getting abused it makes sense to look at the actual risk factors that the literature supports.

Children in this project might actually be less at risk because there's a support network. The textbook says "have a strong support network" and not keep the support network small to reduce the number of trusted adults.

Comment author: Viliam 20 March 2017 10:41:50AM 2 points [-]

ensure that two or more adults are always tasked with watching children, so that the adults can watch each other.

This may feel exaggerated, because many people not living in communities are not following this rule consistently either. People often leave their children alone with grandparents or babysitters. Sure, there is a risk involved, but... life sometimes gives you constraints.

Comment author: ChristianKl 26 March 2017 08:48:51AM 0 points [-]

A group living situation creates a larger field of "trusted adults" per child. Unless all the adults are mindful of these risks, a situation arises where any adult may at any time be put in charge of watching any child or children. This is frankly the textbook definition of what not to do.

Could you point out a textbook that describes that is isn't what should be done?

Comment author: username2 18 March 2017 09:56:45AM 1 point [-]

That seems entirely unjustified. More people and ore space means more threats and more opportunities. Most child molestation happens from a SINGLE close and trusted family member, acting alone. Yet your same argument could be applied to the single-family household -- with a family living together it is more likely that someone else in the family will notice. But the data doesn't seem to support that.

Comment author: drethelin 18 March 2017 09:49:45PM 1 point [-]

most single family households have a lot fewer than 10-20 people in them.

Comment author: RedMan 20 March 2017 12:29:38PM *  0 points [-]

This problem is an easy one to solve. Implicit association tests are effective at discriminating pedophiles from non-pedophiles.

Have 'age of consent' set as a community norm (keep it legal folks!) and 'must score under community-assessed maximum tolerable score on the pedo-IAT' as a condition of moving in.

You'd be the first neighborhood association to take this sort of strong, invasive(?) measure for preventing child abuse in a close knit community, I don't think they do this in Celebration, Florida (or whatever the Disney World community is called)

https://www.ncbi.nlm.nih.gov/pubmed/23613137

Note, any implementation should also include eye tracking, or another analytic to detect a user looking away and clicking at a constant rate. Looking away and clicking to advance at random or at a constant rate is the only mechanism for defeating an IAT.

How do I apply for a slot in the house? I'm tired of living by the sword and would love to relocate to a tech hub but don't know anyone.

Comment author: philh 22 March 2017 12:22:17PM 3 points [-]

I am skeptical that the IAT can accurately detect pedophiles. The article is paywalled, so I can't say anything specific about it, but I can gesture in the general direction of the reproducability crisis.

Comment author: RedMan 25 March 2017 06:44:02PM *  0 points [-]

The IAT is part of a somewhat widely used assessment of sex offenders, the Able Assessment, and is less invasive than penile plethysmography (ref1). Unfortunately, IATs have been shown to be unhelpful for identifying female child sex offenders, as their cognitive approach to offending is different from that of men ('I was coerced by a man/lonely and horny' vs 'entitled and attracted to the bodies of children') (ref2).

There is likely a false positive rate for an IAT, enough that it is relegated to the realm of polygraphy, and inadmissible in court...but I am not particularly concerned, as it is likely not large enough to render the test worse than random, and for a community like this, given that no additional discrimination will be applied beyond 'please live somewhere else', males in this specific, vulnerable community should be fine with submitting to an IAT. Given the 'male coercion' factor in female sex offenders, denying access to men who 'fail' the IAT would probably reduce the liklihood of female offending as well.

The reproducibility crisis is real, and most psychometric tests are lousy for a number of reasons, but it is possible to extract data that is useful, though not perfect, for making decisions. This is not a fire-and-forget solution to the problem, but in concert with normal behavior intended to reduce harm, it will hopefully help prevent the 'Rationalist Baugruppe' from devolving into a 'Rationalist Pitcairn Island'

Assessment survey: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2993520/ Survey on women: https://beta.bps.org.uk/news-and-policy/developing-assessment-and-treatment-practices-female-sexual-offenders

Comment author: philh 27 March 2017 05:50:45PM 0 points [-]

This isn't something I find sufficiently compelling to spend a lot of time on.

But I note that https://en.wikipedia.org/wiki/Abel_Assessment does not seem particularly reassuring. I note that your first reference does not mention the IAT under that name, and from skimming doesn't appear to talk about it under a different name. I note that the IAT not working on women is consistent with the IAT simply failing to replicate. And I note that

The reproducibility crisis is real, and most psychometric tests are lousy for a number of reasons, but it is possible to extract data that is useful, though not perfect, for making decisions.

Sounds semantically similar to "a lot of things are failing to replicate, but I think this thing works anyway".

So I remain unconvinced.