ErikM comments on How to deal with someone in a LessWrong meeting being creepy - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (769)
Hard problem.
"Change your behavior if a significant fraction complains" fails to protect isolated victims, who are likely to be the most common targets of bad behavior and also the ones in most need of support. "Change your behavior if one person complains" is grossly abusable, and the first-order fix to complain about frivolous complaints spirals off into meta. Appealing to common sense, good judgment etc. seems to me like passing the buck back to the situation that created a need for this discussion in the first place.
As a secondary consideration, there's the spectrum between an ex-Muslim requesting that all women present cover up for a few meetings while acclimatising, and a nudist showing up to a meeting and being requested by others to wear clothing while present. At what point does one's apparel start to constitute "behavior" that other people may complain about as creepy?
On thinking about this (five minutes by the clock!) I start to suspect that trying to write rules about creeping is too high-level and abstract, and it would be better to codify rules on what specific behaviors are tolerated or not, and this ruleset could vary by group. Such as:
Edit: oops, list syntax
This rule is always safe to follow, but is suboptimal in that it rules out some contact that both parties would enjoy.
This is mostly an anti-innuendo rule. Just as threats of violence are morally equivalent to acts of violence, entitlement to entering personal space is equivalent to entering personal space.
Threats of violence are bad. Threats of violence are bad because acts of violence are bad. Some of the moral badness of acts of violence flows into threats of violence and makes them bad too. Threats of violence should not be tolerated.
Threats of violence are not morally equivalent to acts of violence. The fact that we're talking about practical real-world morality is no excuse to lose our ability to think quantitatively.
"threats of violence are morally equivalent to acts of violence"
Um, what?
If everyone follows this rule nobody will ever initiate physical contact.
For a better-phrased example of this rule, see the code of conduct from the OpenSF polyamory conference:
The OpenSF code of conduct seems pretty good in general.
It does! Want to clone it for the Singularity Summit?
I don't think a commonsense reading of this rule would prohibit holding one's arms up and saying "Hugs?"
Jandila's response here illustrates the vital point that common sense is not a safe way to read advice in this area. If you need advice, what you consider common sense will often be deeply wrong.
Or possibly just "Hugs okay?", sans the arms outstretched (it can create pressure; the person has signalled very loudly in social terms, so the other person's denial can lead to face loss; people who've been socialized to be sensitive to that, whether for cultural or other reasons, might find the outstretched arms add pressure. Fine for folks who've no issue asserting their boundaries loudly and clearly without concern for face, but that's not even enough of everybody to be a really good rule, I think.)
Is being vague with who it's directed at and counting on something like the bystander effect a good hack for that?
Hm. I've not tried that myself, but as someone who had a lot of past awkwardness and cluelessness in social situations, and now does alright, it strikes me as not a good move. My sense is that it'll just look like a different flavor of awkward-confused, albeit one that puts less direct pressure on the person.
So, I know this funny little trick where you can verbalize a desire and seek explicit permission to act it out while taking care to make sure nothing about the situation seems especially likely to make the other party feel coerced or intimidated into giving an answer out of synch with their preferences. It basically involves paying attention, modelling the other person as an agent, deciding on that basis whether the request is appropriate (while noting the distinction between "appropriate" and "acceptable to the other person") and then asking politely. You do have to take care not to assume that the answer is or "should be" yes, though -- the difference that makes in your approach usually comes off as a bit creepy.
If it happens that you don't know how to perform all of these magical tricks, using your words is a good first approximation. The likelihood of a good outcome is often improved if you ask e.g. "Can I hug you?" as opposed to just bounding up and hugging the person, and your blameworthiness is significantly lowered.
Note please that physically imposing folks who appear to be men and are not charismatic (like social status, but interpreted on an individual basis - the individual one considers highest-charisma is likely to be thought of as creepy by a lot of other people, because it's about walking the fine line between creepiness and friendliness) are most likely to benefit from this. This does stand to be noticed. Cute perky energetic young women can get away with hugging practically anyone without asking. This does not necessarily mean that they should.
Challenging, but certainly possible.
Which bit do you find challenging?
I mean, I was kinda being snarky (I don't think what I suggested is all that hard or unusual at all, though it obviously varies. I've noticed a few reasons for that:
-The person is failing to model the other as an agent, as a center of perspective. Their model of the person starts and stops at their own feelings and reactions; hence, if they find the person attractive, "X is attractive to me" becomes way more salient than it would otherwise be, in determining how they'll attempt interaction. Men do this to women a lot, in general, but there are plenty of other dynamics or situations which can lead to it. Autism or similar psychological variance is massively overstated as an explanation for it; it's way too prevalent a behavior in the general population for that.
-The person has no sense of whether something is appropriate or not, even though they've modelled the other party accurately ("is agent, has preferences"). This is very common among people who, for whatever reason, have had socialization issues. They usually know there's a bewildering array of possible rules or at least broad patterns that might theoretically bear on the answer, but it's not obvious which ones apply, or that they haven't even thought of. To be honest, even socially-successful people sometimes have trouble navigating that, as soon as they're in circumstances that are unfamiliar to them -- another culture's norms, or when dealing with a known charged dynamic and they're concerned about signalling and how they come off. The trick is that there's usually not any one right answer; it can be as specific as the nonverbal communication between two parties. Is asking for a hug creepy or unnecessary? Sometimes, if you can't read the cues, you really can't know short of asking. This means there's always some subjective sense of risk; the problem is they don't know how to calibrate that to the situation, don't have a model of likely prior probabilities. All they really have is a sense of the variance on the options, which is incredibly wide.
-They're failing to not-assume-yes. This is related to the first problem; the person is failing to be aware of, or consider, the pressure their request creates, or is equivocating the risk of being told "no" or declared "creepy" to be symmetrical with the worst-case scenario on the other person's chart. For one reason or another, it just seems to them that if there's no obvious reason not to, no compelling objection in particular, then obviously the thing they want should happen. "No" isn't heard as a good answer in and of itself, not a sufficient report of the other party's preference; it's felt as somehow keeping them at arm's length, denying them the information they need to know how to get what they want. This sort of thing is very obvious from outside, because it leads to different behaviors and responses, and body language tells, when confronted with a "no."
The hard part is forming an accurate model of the other person and situation.
Your last paragraph is excellent. (Others also good, last excellent.)
There's a bit of confounding between
The usual way is to convey requests and refusals by cues too subtle for status fights. The nerdy way is to always interpret answers as preference reports, not status fights. Bad things happen in the intersection.
That can come across to some women as insecure. (Though I'd expect most of those are in the left half of the bell curve and hence unlikely to be found in LW meetups.)
Some women? And you're Irish? This behaviour is practically tattooing "I have poor social skills or severe confidence issues" on your forehead in any guess culture. Odd is about as positive a description as it's going to get outside of people who've not read a good deal of woman's studies stuff.
It's almost as though some people consider your status hit as something of extremely low importance!
Understood - but essentially no humans consider their own status hits as of extremely low importance. this is so strong that directing other people to lower their status - even if it's in their best long-term interest - is only rarely practical advice.
Oh absolutely. To be clear, I am asserting that people making this recommendation are basically following the FDA playbook. Given a tradeoff between bad things happening and costly safety measures...radically optimize for an expensive six sigmas of certainty that no bad event ever happens, with massive costs to everyone else.
Now, this strategy can make sense, if either:
Arguably, the few people in this thread that are advocating extremely socially costly "safety measures" believe a combination of both.
This is sometimes a fair characterization, but remember that (like this thread has been discussing) the social cost depends a lot on your environment. Better to say that categorically recommending behaviors without understanding the perspectives of those that those behaviors would harm is a problem (obviously somewhat inevitable due to ignorance). (I think we need the term "typical social group fallacy".)
Certainly! As such, we should figure out how to turn geekdoms into ask cultures, when they aren't already. Putting even marginally socially-awkward people in situations where they have to guess other people's intentions, when everyone is intentionally avoiding making their intentions common knowledge, well, that's sort of cruel.
So, this become a problem we can actually try to solve. In a relatively small environment, like a group of a dozen or so, what can one do to induce "ask culture", instead of "guess culture"?
(This should probably be a discussion post of its own... hm.)
My own approach: if I can afford the status-hit, I ask about stuff in a guess culture, and I explicitly answer questions there. In some cases I volunteer explicit explanations even when questions weren't asked, although I'm careful about this, because it can cause a status-hit for the person I'm talking to as well.
Some additional notes:
I was raised in two different guess cultures simultaneously, then transferred to an ask culture in my adolescence, and I'm fairly socially adept. This caused me to think explicitly about this stuff rather a lot, even before I had words for it. That said, I strongly suspect that there's much clearer understandings of this stuff available in research literature, and a good scholar would be invaluable if you were serious about this as a project.
Talking about "affording the status-hit" is oversimplifying to the point of being misleading, since I live in the intersection of multiple cultures and being seen in culture A as deliberately making a status-lowering move in culture B can be a status-raising move in A. Depending on how much I value A-status and B-status, "taking a hit" in B might not be a sacrifice at all. (Of course, being seen that way in A without actually making such a move in B... for example, pretending to my A friends that I am seen as a rebel in B while in fact being no such thing... is potentially a more valuable move, albeit a risky one. As well as a dishonest one, to the extent that that matters.)
The terms "ask culture" and "guess culture" are misleading as well; it's more precise to think in terms of topics for which a given culture takes an "ask" stance, and topics for which it takes a "guess" stance. It's even clearer to think in terms of preferred levels of directness and indirectness when trying to find something out, since successful people don't actually guess about topics for which their culture takes a "guess" stance, they investigate indirectly. But, having said all that, I'm willing to keep talking about "ask" and "guess" culture for convenience as long as we understand the limits of the labels.
A downside of asking for things in a guess culture is that people have to give you the things. (Unless you're demanding so much they'd rather refuse and lose you as an ally.) Imposing this cost on people hurts them, as well as lowers your status.
Note that I wrote "asking about", not "asking for".
I agree that turning down requests in a guess culture has social costs, which is one reason the distinction between appropriate and inappropriate requests is considered so important.
Imposing costs on others by making demands of them doesn't necessarily lower my status.
It's probably more accurate to refer to hint cultures rather than guess cultures.
I wish lojban had worked out better-- it would be very handy to have a concise way of indicating whether you're talking about how a culture feels from the inside or the outside.
Probably depends who's talking.
(I'm Italian.)
Forgive me, my memory is poor, I took your references to Ireland to mean you were Irish.
(I studied in Ireland from September 2010 to May 2011.) EDIT: Why were this and the grandparent upvoted?
I wasn't the one who upvoted it, but volunteering extra information that reduces confusions certainly seems worth upvoting to me.
Because we want to see more comments like this (i.e. clearing up confusion), and because in a thread this large it only takes a small percentage of people deciding that a comment is high-quality for it to get upvoted.
It's almost like there is no one magic rule set for interacting with us or something! ;p
On the one hand, emphatically yes - when talking about How To Interact with people of X gender, people tend to make a lot of generalizations.
On the other, feminist scripts seem to be against didactically learning social rules to an extreme extent - instead of pointing out "Hey, this thing works on maybe three out of four women, referring to that subset as 'women' makes you believe less in the other one-quarter," they go the entirely opposite direction and say that learning any rule, ever, is wrong and misleading and Evil. I dislike this, and while your comment is clearly not being this, it can easily be read as it by someone with experience interacting with those scripts.
I often find that what is not creepy for internet feminists can be for women who use other social conventions, and vice versa. Makes it hard when one doesn't know the convention being used. Also makes other-optimising a problem here.
(Edited for clarification)
Heck, I suspect that in a lot of cases what a feminist claims is creepy on the internet, and what the same feminist would find creepy in real life are different things.
That extends to more than feminists, and more than creepiness; people's verbal descriptions of grammatical or moral rules often don't match the judgement they will give to specific cases. More generally, people can't see how their brain works, and when they try to describe it they will get a lot wrong.
Creepiness is partially context-dependent. If you try to list all details, there will be too many details to remember. On the other hand, if you try to find some general rules (such as: "don't make people feel uncomfortable"), some people will have problem translating them to specific situations.
This could be possibly solved by making a "beginners" handbook, which would contain the general rules and their specific instances in the most typical situations (at school, at job, on street, in shop), and later some specific advice for less typical situations (at disco, at funeral, etc.).
But still, even the internet version would probably need different sections for instant messengers, facebook, e-mail... even for e-mail to different groups of people... Eh. Anyway, it could also start with most frequent situations, and progress to the more rare ones.
I suspect one of those negatives still has to go, no?
I think I was really meaning to say "not not creepy" at the time :S
But do you mean to say that the creepiness standards of internet feminists are the same as that for "women who go other social convention"? I was expecting you to mean that they were different.
I like how the guides go about detailing how to do this, rather than simply telling people more things they're doing wrong.
Wait...
You have to follow some extra links to reach the "do" advice., but it's there.
A problem with teasing is that it sets up an environment where it can feel very risky to say "No, I don't like being teased". Will the request be respected, or will it be met with more teasing?
I like the sentence "I am done being teased now". It seems to work pretty well.
Thanks-- I'll keep it in mind. The advantage might be that it has no flavor of "please stop teasing me".
I think it manages to avoid the usual unpleasantness associated with saying, "hey, this is serious now", but then, I prefer bluntness anyway.
It doesn't say please at all. It says "we were doing this thing. Now we aren't anymore."
Since many issues of this type stem not from polite-but-overreaching people but rather the legitimately impolite, this method may not always be hugely effective. Legitimately impolite people would hear something like that and reply "Are you?" with a smirk. Also, if you get angry or seriously assertive, they are likely to assume the problem is on your end and tell people about how "crazy" you are.
The fact that many people reward such behavior is of course a major contributor to this issue.
Yeah, I solve that problem on the meta-level by not hanging out with impolite people after discovering this fact about them.
I like that approach. Unfortunately, for some of the most socially-adept (in other respects), any request not to tease is itself regarded as an invitation for more teasing -- or at least, the "I really need to stop" sensor is way too insensitive to negatives. Even worse, some end up liking the person because of this (which obviously has horrid incentive effects).
The only explanation for this is that it is acceptable for women to initiate physical contact without prior contact by the other party. This is an unconscious double standard.
In many social groups touching initiated from women is often received just as bad as from men, and fairly so. I am sure there are lots of groups with this specific double standard, but it is not universal, not by a large margin.
Also, "only explanation": Really?