Recently, an extended discussion has taken place over the fact that a portion of comments here were found to be offensive by some members of this community, while others denied their offensive nature or professed to be puzzled by why they are considered offensive. Several possible explanations for why the comments are offensive have been advanced, and solutions offered based on them:

Each of these explanations seems to have an element of truth, and each solution seems to have a chance of ameliorating the problem. But even though the discussion has mostly died down, we appear far from reaching an agreement, and I think one reason may be the lack of a general theory of the phenomenon of "offense", in the sense of giving and taking offense, that we can use to explain what has happened, so all of the proposed explanations and solutions feel somewhat arbitrary and unfair.

(I think this article has it mostly right, but I’ll give a much shorter account since I can skip the background evo psych info, and I’m not being paid by the word. :)

Let’s consider what other behavior are often considered offensive and see if we can find a pattern:

  • use of vulgar language (where it's not customarily used)
  • failing to address someone by their honorary titles
  • not affording someone their customary privileges
  • to impugn someone’s beauty, intelligence, talent, morality, honor, ancestry, etc.
  • making a joke at someone’s expense

What do all these have in common? Hint: the answer is quite ironic, given the comment that first triggered this whole fracas.

most people here don't value social status enough and (especially the men) don't value having sex with extremely attractive women that money and status would get them

As you may have guessed by now, I think the answer is status. Specifically, to give offense is to imply that a person or group has or should have low status. Taking offense then becomes easy to explain: it’s to defend someone’s status from such an implication, out of a sense of either fairness or self-interest. Let’s go back to the three hypotheses I collected and see if this theory can cover them as special cases.

to be thought of, talked about as, or treated like a non-person” Well, to be like a non-person is clearly to have low status.

analysis of behavior that puts the reader in the group being analyzed, and the speaker outside it” A typical situation in which one group analyzes the behavior of another is a scientific study. In such a study, the researchers usually have higher status than the subjects being studied. But even to offer a casual analysis of someone else’s behavior is to presume more intelligence, insight, or wisdom than that person.

exclusion from the intended audience” To be excluded from the intended audience is to be labeled an outsider by implication, and outsiders typically have lower status than insiders.

But to fully understand why this particular comment is especially offensive, I think we have to consider that it (as well as many PUA discussions) specifically advocates (or appears to advocate) treating women as sex objects instead of potential romantic partners. Now think of the status difference between a sex object and a romantic partner...

Ethical Implications

Usually, one avoids giving offense by minding one’s audience and taking care not to use any language that might cause offense to any audience member. This is very easy to do one-on-one, pretty easy in a small group, hard in front of a large audience (case in point: Larry Summers’s infamous speech), and almost impossible on an Internet forum with a large, diverse, and invisible audience, unless one simply avoids talking about everything that might possibly have anything to do with anyone’s status.

Still, that doesn’t mean that we shouldn’t try to avoid giving offense when we can do so without affecting the point that we’re making, or consider skipping a minor point if it necessarily gives offense.  After all, to lower someone’s social status is to cause a real harm. On the other side of this interaction, we should consider the possibility that our offensiveness sense may be tuned too sensitively, perhaps for an ancestral environment where mass media didn’t exist and any offense might reasonably be considered both personal and intentional. So perhaps we should also try to be less sensitive and avoid taking offense when discussing ideas that are both important and inextricably linked with status.

P.S. It's curious that there hasn't been more research into the evolutionary psychology and ethics of offense. If such research does exist and I simply failed to find them, please let me know.

New Comment
179 comments, sorted by Click to highlight new comments since: Today at 7:05 PM
Some comments are truncated due to high volume. (⌘F to expand all)Change truncation settings

There's another piece to offendedness-- it's not just a an attempt to protect one's status, it's an attempt (sometimes successful) to assert status, and it's possible to have a little too much fun with that. Your theory is good, but it leaves out the way many people seek out things they find offensive.

Over at slactivist, there's an effort to distinguish between being offended and righteous anger-- imho, that distinction hasn't been defined yet, but it might be worth discovering.

Excellent and important point. I think that's a part of my uneasiness that I hadn't been able to verbalize abstractly up until now.

In order to have a comment policy in place, the act of pointing out language-to-avoid should have a standard and formal form which does not decrease the status of the pointed-out speaker, and perhaps even more importantly, should not increase the status of the pointer-outer.

In order to have a comment policy in place, the act of pointing out language-to-avoid should have a standard and formal form which does not decrease the status of the pointed-out speaker, and perhaps even more importantly, should not increase the status of the pointer-outer.

I don't know if this is possible, but that may be because I've been spending time in an environment which is particularly poisonous that way.

I will say that shaming people for ignorance has to be treated as intolerable.... without shaming people who don't know better than to invoke that sort of shame.

Could you clarify: in order to encourage people to admit ignorance and take correction, you want to remove the perception of low status in taking it? That sounds great, but I'm not sure how you get there. Out of people I believe to be mistaken or ignorant, I do respect more those who admit a mistake, than those who defensively generate huge threads full of denial and obfuscation, but in the case where I'm only skimming, I can't often be sure if they're wrong and it's not worth my time to understand and judge. I suppose you're talking about social correctness and ignorance of expected behavior, and I was thinking of general correctness and ignorance. I think there's more status lost here in being incorrect than rude. So maybe discouraging offense-giving is a simpler problem.
To not raise the status of the pointer-outer, we could require that the pointing out is done anonymously. To make it formal, it might be useful to make it part of the UI: rather than posting a reply in free text, you would click a "bad language" button and mark a checkbox for the policy you consider violated. Reducing the bandwidth of the channel in this way might make it harder to communicate elaborate moves in a social-status game using it. Not lowering the status of the pointed-out speaker seems harder: one could make the whole exchange not publicly visible, but would a private message be sufficient to enforce the guidelines, especially since there might be disagreement about what constitutes a violation?
It seems that if you buy this policy, no UI change would be necessary. Just a policy of sending PMs rather than making it public would suffice.
I would generalize, as follows: Allow anonymous posting, with a checkbox. Not really anonymous, you have to be logged in, but non-mods don't see ID. Anonymous posts don't display until a mod explicitly permits them. Special rules for anonymous posts: be very polite, be very on-topic or validly critical of another poster's delivery. Posts without extra effort to politeness beyond the norm will be summarily junked. Result: this gives us criticism with no status modification, it also gives us posts with on-topic views which a poster might have good reason to disown in public, but still consider true.
The only way I can see that working is to avoid doing it publicly, either by telling someone in a message or by some other function, like a 'Private Reply' button that leaves a comment only the target commenter can read.

I'm reminded of the medieval "point of honor" that nobles were entitled to and commoners weren't. Perceiving an offense and successfully thwarting it is a karma mint, so seeking out offense makes sense, especially if other people will likely support you. Not without its harms, of course, because social status is much like a zero-sum game - especially in conflict situations.

Not convinced at all that "offense" is meaningfully different from "righteous anger". It's pretty clear that one of those concepts is meant to be pigeonholed as good and the other as bad, and people are trying to invent a reasonable-sounding definition that would facilitate such pigeonholing; but that's already reason enough for me to discount the whole discussion.

"Karma mint" is a great phrase. I believe that anger is what fuels territorial defense, territorial defense is enough work that it needs some fuel, and that people need some territory, so there might be a useful distinction between the emotions which drive legitimate defense of self and others as compared to looking for conflict for the fun of it or to gain unfair advantages. Maybe some of the discussion at Slactivist can be summarized as checking to see whether the angry person is looking for positive sum or negative-to-zero sum solutions. The situation is made more complicated because a person can be genuinely abused and be looking for negative-to-zero sum solutions.
0Wei Dai15y
What is the medieval "point of honor"? Wikipedia, Google, and Bing are all failing me...

In general, look up what constituted a valid reason for dueling in the Middle Ages. For example, here:

To illustrate the judicial duel I will give a synopsis of the account of the combat at Moulins, France on February 17, 1538. During this duel Lion de Barbencois (Sieur de Sarzay) did combat François de Saint-Julien (Sieur de Veniers). A quarrel had been in progress for many years between Sarzay and a gentleman by the name of Sieur de La Tour-Landry. Sarzay had sworn that La Tour-Landry had fled like a coward during the battle of Pravia in 1525 (where Sarzay was not even present). La Tour-Landry's honor was placed in jeopardy by this and demanded that Sarzay give the source of this information. Sarzay related that he had gotten the information from Veniers who vehemently denied the accusation of supplying the information; thus giving Sarzay the Lie. Now, the quarrel changed from Sarzay and La Tour-Landry to Sarzay and Veniers. Thus the duel was not held to clear La Tour-Landry of the charge of cowardice on the field but to dispel the suspicion of a lie. Gossip true, but a lie. In this duel, Sarzay was the injured party since he was accused of lying. After the challenge was given the King was petitioned for a field.

The Wikipedia page on Duel also gives a vivid flavor of the time.

Of course, that leads to offenders trying to assert status by accusing offendees of status-assertion (because we regard status-protection as more worthy than status-assertion), and round and round we go.

But "status", itself, still seems like a black box.

I think "offense" is one emotion caused by a human-universal ability to recognize states of mind in other people that can motivate those people to take actions disadvantageous to oneself or one's allies, and to predict that tolerating an action associated with such a state of mind will set a disadvantageous precedent.

The precedent would be as if the other person had negotiated a right to take those disadvantageous actions, and as if they might later negotiate a right to take actions even more disadvantageous.

The state of mind "I am thinking of someone who has low status" is just one possible such state of mind. Other possibilities are "objectification", "depersonalization", "violent anger", "unwillingness to imitate sanity", "intent to theive", "intent to deceive", "sexual interest", "intent to slack", "unwillingness to obey a shared lord", "intent to obey the letter of the law and not the spirit", "intent to reduce a people to slavery"... But some of these cause "offense" and some ca... (read more)

I think your examples don't necessitate amending Wei Dai's analysis. Objectification and "non-personhood" are obviously about status and the dominance hierarchy. If your peer dares to disobey a shared lord, that implies you have lower status than him: he thinks he can disobey and you can't.
1Wei Dai15y
I agree that "offense is all about status" is probably too simple and that a more complex and refined theory can have greater explanatory/predictive value. On the other hand, the simplicity does have a benefit in that it's easier to apply when you're addressing an audience. It's probably easier to think "will what I write/say cause someone to lose social status?" (with a broad view of what constitutes status) than to try to keep more detailed models of the audience's minds (ETA: except in situations where your social brain works well and does the latter for you automatically). If you disagree, can you try to distill your theory into some practical advice for writers?
The context here is a human dealing with a human. Thus it can be considered a useful heuristic to think "will what I write/say cause someone to lose social status?" and depending on the reply that your brain returns, judge whether it could be considered offensive (since this might prove to be a more accurate means of judging offense than trying to do so directly). Naturally, if you were actually trying to develop an artificial intelligence that needed to refrain from offending people, it probably wouldn't be as easy as just 'calculating the objective status change' and basing the response on that.

I think the article makes a good point. But I also want to point out that practically all human communication involves status games and status transactions. By status transactions I mean things like a person trying to raise or lower their own status or raise or lower the other person's status (or some other group's status). Of course, communication typically conveys other information too, but it almost always is accompanied with some sort of a status transaction.

The form in which these status messages are presented is so subtle and natural to majority of humans [1] (the exceptions being say autistic persons [2]) that it takes a little bit of work to get familiar with. Instead of looking at direct and explicit status messages, you need to look at some normal conversation from this perspective and observe the subtle ways these status transactions are conveyed.

It's much easier to observe status games in face-to-face conversations (when one is not one of the participants in the conversation), where you see the way people make eye contacts, gestures and use space, for example. Keeping the head completely still, for example, sends a very strong high-status signal. You can actually try ... (read more)

I don't think status transactions are completely oblivious to autistic people

You may not notice the status transaction, but the status transaction notices you.

This agrees with my personal experience--that most or all human interaction seems to be permeated by status signaling and other social games. I imagine this is true in most social mammals based on the animal behavior stuff I've read. I find this to be a great aid in arguing effectively. I want to write a longer comment tying this to personal observations on effective communication, persuasion, etc. but I'm at work. Also many readers probably already have a lot of similar (and possibly more cogent) thoughts on the topic.
I don't know if I'm autistic, but I agree with this hypothetical autistic person's assessment.
The keeping head still trick sounds interesting. Do you have a citation? Or other similar tricks?

Several comments have remarked on the obvious-in-retrospect nature of this post. Indeed it seems so obvious now that I find it almost unreal to think of a time (i.e., hours ago) when offense wasn't obviously about status.

Why are we so bad at recognizing some obvious truths, I wonder? I'm pretty sure I had read or at least skimmed that Slate article when it was first published, but it still took me days to make the connection between "offense" and "status" after the controversy arose here. And this was after being primed to think about status from frequently reading LessWrong and OvercomingBias. And the original offending comment itself, which has been quoted multiple times, mentioned status. WTH?

I think maybe status is the wrong approach. Status sounds neutral enough and easily changed. But if we swap it out for the word 'power', then it works better, and now we can handle arguments about status, why people won't change their linguistic ways, etc. I have an old essay on this at I didn't think it appropriate for LW because it seemed a bit too fluffy and based on personal experience, but maybe this post shows I was wrong in that.
Fixed link.
0Wei Dai15y
I thought it would be clear from the context that I meant social status, which is decidedly not neutral or easily changed. I haven't read your essay closely, but your notion of "power" is probably a closely related, if not identical, concept. I'll add a link to my post to prevent future confusion.
Johnstone, mentioned by jajvirta below, posits that his version of status is an eufemism for dominance hierarchy, which is one instance of power.
Funnily enough, one person reading my disrespect essay last week commented that it reminded him a lot of Johnstone, and recommended exactly that book. (I haven't gotten around to reading it yet.)
I read your essay, you should definitely read Impro. When people talk about social status, they are talking about some stable trait-like thing that you can achieve by for example buying a nice car. In your essay, you are talking more about the pecking order or dominance hierarchy, which we constantly maintain and test by our actions. That's why many actions of humans can also be viewed as "status transactions". I also recommend books by Desmond Morris, especially Human Zoo. Many Johnstone's ideas are based on Morris.
Wow. Yeah.
Whenever we talk about status it's always other people. I think maybe this is because keeping track of the conversation about status, and doing your regular subconscious monitoring of your own status, and consciously observing the subconscious monitoring of status by yourself and those around you all at once is just too much. Also, making meta references went out of style a long time ago, and a person who wants to talk about status in general while talking about what status movements are occurring currently is going to have to use a lot (e.g. "I'll bet you wouldn't have made that comment about status if your status was lower.") Reminds me of this C&H strip.
Not true. Signaling is often (always?) done to raise your own status.

Together with A Tale Of Two Tradeoffs and Missing the Trees for the Forest, this post sheds some light on the nature of mind-killing properties of adversarial debate. If adversarial debate is about status, and status-arguing is done by the signaling subsystem of the mind, in the far mode, this explains why the debate is often so myopic. The signaling system of the mind is optimized for deceiving a small group of people to one's advantage, as opposed to making actually good decisions and making accurate estimates where it's possible. It's just bad at seeing the truth, and so the truth gets ignored, while the fight goes on.

I think to make your point you also need to establish that people observing the debate also disregard truth.
I don't think it's true to an unusual extent here, apart from standard inability of people to see the actual problem and not the words in which it's described (which is what the debaters will be doing).
I think you should expand this into a post.

The basic point here seems correct, and obviously so. People take offense at things that lower their own or their ally's status.

"The basic point here seems correct, and obviously so. People take offense at things that lower their own or their ally's status." That some people take some offense at some things that lower their status is trivial. (Introducing allies may be otiose, since the reduction affects the person allying.) Do all people take offense at all things and only those that lower their status? This is the claim, at least as first approximation. Certain commonly known experiments in cognition show that reasoners tend to ignore examining whether scenarios exist such that a status loss causes no offense or where offense occurs without the supposed precondition in status loss. Here, in-depth discussion of counterexamples proposed by posters is avoided repeatedly. How is this different from a discussion occurring anywhere else? What does this state of affairs say about this community's approach to becoming more rational?
Such claims are almost always about tendencies, almost never extreme claims that all or none of something is some way.

I'm having difficulty putting my finger on it, but this concept and definition don't seem to square with my understanding of offense in practice. Your instances do not include politically incorrect statements (racist, sexist, or various other -ists, depending on who exactly is listening), whether factually incorrect or otherwise, which seem to be one of if not the major sources of serious offense. There seems to be a strong bent towards maintaining the existing social order, as opposed to being concerned specifically with the status of the speaker. I'm trying to arrive at insight cogent enough to post, but, since I'm not there yet, I'd just hand-wavingly say that offense has more to do with the preservation of an existing social order than it does with status specifically; if I can back that up rigorously I'll comment or post on it.

As one example, taking offense to vulgar language or imagery does not seem to fit into this mold. If I ran into a church picnic and started yelling obscenities, people would get offended, even though I'm not threatening their "high" status so much as advertising my "low" status. This doesn't seem to be suggesting that a person or grou... (read more)

9Wei Dai15y
A racist statement is usually one that, if accepted by the listener, will tend to lower the status of the targeted race. Same for other -ists. I'm not seeing how it doesn't fit with the status theory. Of course you're threatening their high status. You're implying that vulgar language is appropriate in their social circle, and the only way it could be appropriate is if they have low status. Edit: On second thought, I think what's going on here is that once you're a close enough friend with someone, there is no longer a significant chance that you'd want to intentionally lower their status, so an otherwise offensive comment (especially in private) becomes a signal for close friendship. You're signaling that you believe your friendship is so close that your friend won't think you're intending harm, and by not taking offense, your friend then signals the same thing. This probably takes a mathematical model to make completely clear, but maybe you get the gist.
9Scott Alexander15y
Racist statements don't seem to automatically imply lower status for the offended group. For example, many people found this "joke" offensive, even though the only claim seems to be that black people eat lots of watermelon. Similarly, a statement like "Jews control the financial system" could easily offend Jews, even though if anything it assigns them high status.
9Wei Dai15y
One way to look at this is that racist stereotypes promote viewing members of the target race as an undifferentiated mass with little or no individuality. This lowers the status of that group since individuality is important for status. The watermelon joke invokes such a stereotype of blacks. BTW, I'm afraid that having espoused the idea that offense can be explained in terms of status, and having probably increased my own status in this community as a result, I'm likely quite biased on this issue now. I bet it's much easier now for me to find arguments for this idea than counterarguments. So, reader beware. :)
If a statement concurrently attacks and asserts status of people in different ways, it can still be offensive. "Jews control the financial system" places Jews in out-group, which lowers their status, even though at the same time the statement seems to assert their status.
Stereotypes imply lack of individuality, which is usually low-status. As does grouping them as a single entity, especially if that grouping is made with a hint of sinisterity as would often be the case when talking about financial system -controlling jews.
There may be some "treating as a non-person" involved here: people are individuals with different tastes, goals, etc., and stereotypes like the ones you mention ignore this.
In Impro, Johnstone (mentioned by jajvirta below) actual defines friendship as relationship where we can play with status more freely without retaliation, for example by joking. He compares this to when dogs play catch, always switching pursuer and pursued roles in the middle of the play. Of course dogs can both play catch, and really pursue prey.
i have to agree with psychohistorian with respect to the church-picnic example; it seems a bit of a stretch to say that their status is lowered by other people's obscenities. I think that offense is dependent on the relative positions of the offender and the offended. Who would take offense of what a drunk on the street says?
I would. Especially if in female company. Alone I might well ignore it, with a male group I might engage in 'witty' banter. With women present I might demand an apology or take a swing.
Can you say more about how the presence of women factors into your decision? E.g., say we're both in a mixed-gender group, I am drunk, and I start muttering obscenities (just to be specific, say I start repeating "fuck" over and over). If I've understood you, you're more likely to demand an apology from me (and back up that demand with the threat of violence) than if we're in an all-male group. Do you have any theories as to why? Vaguely relatedly, does your decision change if I'm female?
He could, but I wouldn't expect the set of people who are self aware and forthright about their signalling motives in such cases to overlap to a large degree with the set of people who start unprovoked street fights to prove their dominance through faux-altruism.
It's possible we're talking at cross purposes. I replied "I would" to "Who would take offense of what a drunk on the street says?". I was imagining someone shouting insults at me. When you say "unprovoked" I wonder if you're imagining someone swearing randomly to himself. I'd take no offence at that.
You could be right there. Primed by djcb's comment I was more considering "offense at the degree or nature of other people's obscenities" than "offense at deliberate attempt to insult". In the latter case I would replace "unprovoked" with "ill-advised". It potentially also removes the "faux-altruism", depending on whether it was you or your party member's who the insults were directed at.
Nor would I as a general rule, but people occasionally surprise me, especially around here. I figure it's worth the price of a question to find out.
I'm somewhat curious what the response will be myself, particularly now that I spent the price of an answer in changing his incentives. Now the most obvious (and most boring) 'purely righteous, pro-social' spin comes pre-emptively loaded with connotations of naivety and bullshit. I have no idea what the optimal response is now. It may even be the accurate one!
Indeed! Now torn between 'desire to provide accurate data', 'desire to project correct image for future google searches', 'desire to look clever' and 'worry about advisability of confessing to anti-social personality traits in a public forum using my real name'. I'm slightly reassured by my belief that most men would feel the same way. On the other hand if that belief's wrong, it needs changing.
Ironically any signalling of actual anti-social traits would have already taken place in the earlier declaration of potential violent tendencies. But given that you are signalling approximately normal human behavior---and behavior that tends to be respected in practice anyhow---the only negative signal you could actually give now is a weak signal that you are unable to signal smooth-hypocrisy. Almost universally, at least the 'relative' part. That is nearly all males would be more inclined to take offense and make violent dominance displays with female observers present. There is more to gain by making the move and more to lose by failing to. I am not sure whether the absolute part "I would [take offense at what a drunk stranger says]" applies to most males or not. Possibly. I know it doesn't to myself---in general I don't get offended by direct insults, particularly those with obscenities. Rather, I take offense at passive aggressive insults that superficially conform to polite norms. And as with djcb if the speaker is a random drunk stranger almost nothing they could say will offend me.
Incidentally, do you know of any evidence indicating that a given male's differential likelihood of making violent dominance displays in the presence of female observers correlates, or fails to correlate, with his belief (1) that an arbitrarily selected female observer will be impressed by such displays, and/or with his desire to impress female observers? I mean, "X is more likely to do Y in the presence of Z if and only if X believes Z will be impressed by Y and wishes to impress Z" is kind of an unnecessarily complicated way of expressing something much more general. (1) Naturally, it would be an error to take people's self-reports at face value about such beliefs, which makes testing such a theory tricky, but there are other ways to approach the question.
An actual literal far mode belief interfering with status-and-mating related behaviors in real time? I suppose that could happen. I would confidently predict that the violent dominance displays are more likely soon after exposure to an image of a female they are attracted to than after exposure to a non-attractive visual stimulus. Regarding knowing any citeable evidence: There are plenty of studies of "what males do after exposed to pretty girl pictures". Whether the behaviors actually studied correspond sufficiently to the behaviors we are talking about here will depend on how your model of this kind of social behavior relates to your models human behaviors in things like ultimatum games (on either side), and altruism. ie. The studies constitute evidence for me, given that I already consider human behavior in this situation to be similar to one or both of those other contrived scenarios. Regarding the correlation with desire to impress the selected female observers---I certainly hope there is. Observations of increased aggression with other males when a particular female is present is one of the strongest indications I use when noticing attractions between my peers. I even take it into account when organizing social events. With a couple of my close male friends in particular I know that the presence of attractive females makes it far more likely that we'll have conflict. That being the case I'll preferentially organize and attend social events with such combinations when I'm feeling particularly confident, patient and alert. That way I can absorb and diffuse the aggression rather than engage in tit-for-tat escalation like I may do if on "autopilot".
Dodging your question, I doubt I'd react badly to the situation you describe. I'm British and one can hardly speak British English without saying "fuck". Also we spend quite a lot of time drunk. If a friend got drunk and started to say "fuck" over and over again I'd probably be worried that something was wrong. I have trouble imagining why I'd be offended. Above I meant that I'd take offence in response to a direct insult from a stranger, and my memory informs me that I'm likely to take more offence if there are others present, and even more so if there are women. If the insulter was a woman then I imagine my reaction would be completely different. Mainly confusion and a desire to calm the situation down and get away from the crazy lady as soon as possible. I'm trying to introspect on why and it's surprisingly hard. What I'm remembering and imagining is mostly emotional (and moral in the sense of Pinker's crazy angel), but I'm sure there is a calculation/bluffing/stake raising/status game going on as part of that. Why it seems more important not to back down when there are women present is completely beyond me, but it certainly does. I can feel an essay coming on.
(nods) The reactions you describe aren't uncommon... and, as you intuit, often have a lot to do with status management. In my experience, learning to recognize the forces at play as they arise in real time is extremely useful.
He believes females are more likely to be offended by that than males?
No, definitely not. It would be about me and how I am seen and see myself.
That's certainly one possibility.
Person A of race X takes offense when person B of race X makes an unduly strongly positive comment about race X. Assume no other friends of A know B, and there are no other people are around. Perhaps they're the only two people at a bar. There are two explanations that have been screened off; A can't be worried about what other people will think, because there aren't any. Similarly, A can't be thinking that admonishing B will raise people's perception of his social status, since only B is there to hear, and it won't endear him to B. He could admittedly tell people a story about how he admonished B loudly after he said such an offensive statement, but such a prospect need not touch his mind for him to take offense. So it boils down to Person A getting offended by a comment that raises person A's status. This appears contradictory to your model. I would love to hear how it fails to be, especially if its a simple, intuitive explanation that seems to resemble how people actually think, and not a complex justification. Your model would predict that people of race X would be supportive of such statements (even if they implicitly denigrate other races). This is, unfortunately, sometimes the case, but it is not generally the case (nor should it be) so status does not appear to be the primary operator here. If people can be offended by comments that raise their status or by comments that lower it, "ingroup status lowering / outgroup status raising" do not strike me as being accurate causes. "Threatens social norms" explains all of these instances. It explains why some people get offended by comments about other races/genders/what-have-yous and why others don't - they place different values on maintaining that specific part of the social order. It explains why such comments between friends are acceptable - they know they don't mean it, so social norms aren't threatened. I don't think "status" covers the variety of reactions as well, and it does not seem as close to how it
I think agree with you that status doesn't quite seem to cover everything. But "threatens social standards" seems like too much of a black box to me to be a very satisfying explanation in itself. I guess if it suggests anything, it's that offense, like social standards, have too many distinct, and not always sensible causes to be traced back to a single root.
While Wei_Dai makes a very important contribution, I think there are couple of technical points that are probably more complex. Out-group does not necessarily have lower status. There are groups within the out-group, such as moviestars, we regard as having high status. Threatening somebody with out-group is probably other deeply ingrained mechanism at play rather than status. For gregarious animals, being forced out of group maybe even worse than death, if it includes your offspring. It is not directly about status but survival. Politically incorrect statements (racism etc.) however fit the description of Wei_Dai because groups also have status, and the statements lower the status of the corresponding group.

I have no idea what I'm wading into here, but a few things occured to me reading this:

Taking offense to something relies on status and perhaps more significantly on interpellation. Interpellation and its inherent insistence on dignity create barriers to what I'll call effective communication and introduce a rhetoric of respect. If we wish to be rationalists, really and truly, it seems like we must have a discourse that avoids insisting on respect for anyone or anything. We must all get thick skins, be willing to hear ourselves treated as objects of outside analysis and be willing to be ignored when we have bad ideas. Unwise, "offensive" comments like the one that seemed to kick off this discussion can be assayed because they are examples of poor thinking rather than because they are causes of emotional distress. Here, when it gets down to serious business, we should each have no more merit or status than our own arguments give us.

However, I have no idea how to sum this up in a maxim or otherwise implement this. What I offer is not a solution but an objective. I hope others can flesh it out.


You've invented Crocker's Rules.

If I remember correctly, Eliezer's old site had some references to this. Was this not common invocation on SL4?
As kpreid already said, that's pretty much Crocker's Rules, but few people can manage them, so assuming them or expecting people to declare them is a bad idea.
I think it would help. I declare Crocker's Rules.
I've also declared them, and even though it's been three years, I think it's important enough to warrant posting here. Crocker's Rules are hard - you have to really sit down and think about it, run the simulations, and try to figure out the worst, most offensive things to you personally, and why they truly bother you. It helps to have a small identity. But one of the worst things about Crocker's Rules is using them in polite society. You'd think it would be an advantage; in fact, I've found that it creeps people out. When nothing that is said bothers a person, others subconciously seem to think that there's something wrong. I've started emulating low levels of emotional response in these situations because it seems to make people more comfortable.
I believe you and am curious. Can you give a couple of examples? The meaning of not signalling offense differs depending on what you are not bothered by so I'm not sure which of the possible social scenarios you are referring to. (I'd also reverse the claim adding that one of the worst things about polite society is how you are obliged to do things like be offended at socially prescribed times or face consequences.)
You've put your finger on part of it: not being offended at the 'politically correct' times can be an issue. But I think a greater issue is general lack of emotional response; a lot of people, drama queens in particular, are used to seeing responses and emotional swings in people. If nothing seems to upset you, you start to look abnormal. I've found that an effective way to counter it is to be much more happy and energetic than average. Lack of emotional response to prodding creeps people out, but if you just look like a naturally happy and energetic person, it gets written off as "he's (currently) too happy to be affected".
I've seen people "declare Crocker's Rules" from time to time and here and there, but I've never noticed it make any difference to the resulting conversation. Is it any more than a signal of tribal affiliation?
To the extent that it's a signalling move, I think it's more intended to signal virtue than tribal affiliation. That said, I'm generally willing to take it at face value, as either an attempt to minimize the depth of indirection in the conversation, or to encourage people to speak who would otherwise stay silent because what they have to say might be offensive. I haven't noticed it make much difference either, but it doesn't follow that the person declaring it didn't intend it to make a difference... Crocker's Rules being a constraint-definer on others rather than oneself makes it necessarily a cooperative endeavor.

I wasn't looking forward to reading this, given that the topic had been done to death, but wound up interested in, enjoying, and agreeing with this article. Well done, Wei Dai.

Status seems too general. If a person takes offense to a derogatory comment about that person's own race, they are offended by their status being lowered. If they take offense to a derogatory comment about another race, they are offended by their status being raised (before you object, think of the least charitable hypothetical and see if there's a realistic way in which this comment lowers their status, and that this perception actually inspires the offense; I don't think this is the case). It is possible to describe offense in terms of status, but the on... (read more)

One possible explanation for why I take offense at somebody making a racist comment to me (about other people/groups) is that I do so because I believe racist people are despicable (deserving to be among the lowest of the low in terms of status according to my ordering), and when the person implicitly assumes that I will take the racist remark positively, I equate that with them believing I am the sort of person I would despise and regard as low status. Under this interpretation, there is no raising of status. It is is still always related to some hint of being low status, either in terms of what other people think about my status or in terms of the implications on my self-judged status as a result of other people's remarks. ETA: abbreviated version: we take offense at racist comments because they implicitly assume we are what we might regard as "ignoramus racists", and we regard ignoramus racists as having very low status.
This seems like a clearer idea than the "maintaining social order" one; in particular, it gives a better explanation for third parties taking offense. In general, one could be offended by a remark if silence would imply that you are a member of a group you consider lower-status... and that easily handles the case of say, abolitionists being offended by remarks in favor of slavery, even as a non-abolitionist could've been offended by remarks in favor of abolition, without resorting to elaborate justifications of how some sort of "order" is being threatened or maintained. IOW, the key to the remark is not whether it raises or lowers the status of the remark's referent group, but whether a lack of response would lower the listener's perceived status by the listener's standards. That's a really nice formulation from a PCT perspective as well: PCT posits that remarks which disturb our self-perception of status would motivate behavior intended to restore our reference level of that perception. But PCT or no, this perspective also makes it abundantly clear how the offense occurring really has nothing to do with the remark, the subject of the remark, or the person making it.
Expressing offense and feeling offense are correlated but by no means identical. One can very easily do one without the other. I'm uncertain which of these interpretations this fact comes down in favor of. However, it seems like feeling (and to a lesser degree, expressing) offense (in most cases) is inspired by, "It's just wrong for you to say that and you must be chastised!" rather than by, "You made an implicit assumption about me that I find offensive, and you must be corrected!" Offense feels a lot more like "They broke the rules!" than "They lowered my social status implicitly, kind of!" I see no problem, for either model, coming from the abolitionist/fire-eater example. If you're an abolitionist, you believe "Slavery is very bad." If you're a fire-eater, you believe "Slavery is very good." Believing the opposite would lower your status in your peer group and it threatens the social order you desire (even if you haven't established that order as a standard for society, it is the standard you believe society should operate by). I don't see the latter as being much more "elaborate" than the former. Althought, when two people of opposite political viewpoints argue a contested issue, I think they experience anger much more than they experience offense. They may incidentally offend each other along the way, but I think anger is the dominant feeling, which makes sense, because it's a lot like an actual fight, mentally.
Of the two concepts, the former is easier to implement as an automatic, unconscious response, because it doesn't involve multiple levels of abstraction. That's why I'd tend to favor it over the second version. Even simpler, though, might be to just model certain beliefs as "bad" or "good" and treat those that oppose the "good" as "bad". (However, the controlled variable in that case is still "perceived status", since we learn what beliefs are bad or good through applied status consequences.)
Can I just ask which other groups according to your value system deserve to be the lowest of the low? note: I'm not disparaging or praising your value system in any way I'm just curious :)

Makes sense. PUA comments imply that women should have low status, anti-PUA comments imply that people with PUA attitudes should be shunned by everyone, and so it goes on.

Sometimes, that's in the eye of the beholder, based on a stereotype. As I've pointed out repeatedly, there are entire schools (in the sense of "approaches" as opposed to "teaching organizations") which do not in the least imply that women should have low status, and some individual teachers or organizations that emphasize the need to liberate women from societally-imposed constraints on their sexuality... and even the least enlightened of PUA schools emphasize the need to ensure that a woman doesn't lose face or status within her social group! That being said, there are certainly also individuals (including, unfortunately, some commenting here) who do appear to be impugning the status of women, and I very much disagree with that.

A sharp and obvious-in-retrospect conclusion that somehow failed to be mentioned (or my cursory reading of the discussion failed to notice it mentioned).

P.S. Could you capitalize the title of the post? It hurts my programmer's style sense. Thanks.

1Wei Dai15y
Thanks, done.

Many people are apparently offended when someone uses forbidden language, even when not directed against any person. If I make a mistake and exclaim "Shit!" someone may be offended, but what does it have to do with status? Posters haven't adequately tested this theory againt apparent counterexamples.

Casual use of profanity and obscenity is associated with lower social classes, e.g. "swearing like a sailor." Someone who doesn't want to be considered part of a lower-class social group by anyone who overheard your exclamation, and isn't willing or able to visibly dissociate themselves from you, would be well served by making sure you don't repeat it.
That you feel free to exclaim in front of others may be taken to mean that you aren't sufficiently respectful of their status to restrain yourself. I wonder whether more people who are offended by such outbursts are guessers.

Now that was a reduction.

Short, simple, evopsych, sociological, widely-applicable, insightful.

Standing ovation.

This was roughly my reaction.

Wow, it’s a really cool insight!

I guess the natural question to ask would be: Do people ever get (genuinely) offended by anything that does not threaten their status?

Going further, I don’t know of people directing offense at animals or inanimate objects. Does the offender need to be perceived as intelligent? In that case, are people less frequently offended at those they consider stupid?

Well, I have encountered people being (or claiming to be) offended by what in all rights would be an assault on someone else's status. This could be a form of empathy, or in many cases an attempt to gain status themselves through a show of sympathy. This does seem like a potential occurrence of legitimate offense not caused by a perceived direct or indirect threat to the status of the person being offended, iff the offense is genuine- something which I cannot personally attest to, never having experienced this myself.
Yes. They get offended by things that they think threatens their status but in actual fact doesn't. ;) We call them 'insecure'. (The number one thing I avoid when considering business partners or employers.)
Whoops, didn't make myself clear. Is it the case that normal-functioning humans are (almost) never offended by something they themselves don't perceive as a threat to status? Since the article makes a statement, I'm trying to take it to its logical conclusion; in particular to see what outcomes it prohibits, as per . And non-status-based offenses do seem like an obvious thing it prohibits.
Not at all, I was just playing with the 'status - offence' concept, teasing out another naunce that technically answers your question while also informing on a topic that fascinates me. ;) I of course agree with your analysis, below: Good point. Offence does seem to be a social thing and I cannot offhand think of any instances that are exceptions but such instances would definitely make a lie of the statement. Well, at least make a lie of any claim it could make to being a fully general description.

I have a small question, and this is an abstract question not specifically about any particular controversy on LW: -Suppose there is a statement that happens to be true, but which will also lower someone's or a group's status resulting in offence. Will you chose not to offend and keep the statement to yourself, or will you say it?

You haven't given enough information. There are a lot of offensive things that "happen to be true" that you don't say all the time. It being offensive is a good reason not to say it, but presumably you have in mind some reason to say it. One would need to evaluate that against the 'giving offense' to see which one wins.
Quick heavy-handed illustration: You meet someone who was badly disfigured in an accident, let's say this guy, and after taking a look at him say "Holy shit, you're ugly!" This is a completely true statement (being scalped by industrial machinery will do that), also rude and offensive, and has very little reason to need saying.
What about "everything that can be destroyed by the truth should be"? There might be an inconsistency between saying maximally true things and not offending people. What is the priority on LW? On a somewhat related note, I can see it already. You spend years carefully programming your AI, calculating it's friendliness, making sure it is perfectly bayesian and perfectly honest. You are finally done. You turn it on and the first line it prints: Oh dear, you are quite ugly.
That statement obviously only applies when there is falsehood to be destroyed. I'm sure that guy knows he is not especially pretty. Telling him he's ugly may be truthful but it's also kind of like yelling "You're really hot" at the sun.
If I were badly disfigured I would much prefer people saying that to my face than babbling off a lie. I would even prefer it to silence. If I'm talking to someone and they say nothing of my appearance I'm sure I could tell that they were at least thinking about it. I would much prefer that they think about it out loud than "behind my back" in their head. Plus, my experience has always been that mutual insults are a great way for people to become friends. Maybe it's the "civilized" version of play fighting that many animals (and human kids) engage in.
Beware of hypotheticals. Would you also prefer it to talking about something relevant, like whatever you happened to be meeting about? 1. "Hi, I'm Amy. Nice to meet you. Now to get right down to business, let's look at this chart..." 2. "Hi, I'm Amy. Nice to meet you. Man, you're ugly... In most cases, there's nothing to be gained by #2.
Under which circumstances would 'saying true things' win and under which other circumstances 'not saying anything' would win? I would also add, under which circumstances would you 'say something you believe to be false' or 'agree with something you believe to be false' in order to avoid offense?
Um. That's a very complicated question about life, the universe, and everything. There are many circumstances during which saying particular things are beneficial. Maybe some examples would help? You and your friend Anna (to pick a name) are having ice cream. She's talking about how she felt when her mom died. You point out in response that the atomic weight of molybdenum is 95.94. This is very unhelpful and probably should not have been said, and she's offended that you don't care about her feelings as much as she thought you did. Now you and your friend Anna are being held at gunpoint. You are hooked up to a lie detector and asked if you think Anna looks fat in those jeans. If you lie, they shoot you both. Anna would be offended if you think she looks fat. (ETA: as it happens, you do think so.) In this case, it is probably best not to lie. And there are various other sorts of cases as well.
I think this is a life skill which you've developed already but aren't thinking of as the same thing. The stereotypical example: your mother looks terrible in a dress but really loves it and she doesn't need to impress anyone, so you might say she looks good anyway. That's just one point in an extremely wide spectrum the issue encloses, in which you probably have feelings already on what's best when. I think the decisions are personal, dependent on the situation, and often hotly debated.
How do these questions relate to your first comment? Are you asking if I would lie to not give offense?
Statements that are true, important to the development and practice of rationality, and lower a group's status should be said. Persons stumbling upon information of this nature would benefit from doing their best to present the information in a way that will sting the least for this group's status (post would be less controversial, would have to hear less distress calls). The more altruistic might consider how the community benefits from conscientiously proffered information. A person who may have been offended despite all this will probably feel compelled to comment. I think it is important not to just be irritated at someone reacting this way, but to consider whether they have any valid point. Any time someone is offended by what we say, we may have an opportunity to learn how to prevent offense in the future (while still conveying the information accurately). In some cases, we might even learn we missed something and were wrong.
I have another question: Would statements of the type made by Lawernce Summers* be considered too offensive for LW or is discussion allowed? *
The only way something can be offensive enough to justify banning is if its author's intended goal was to offend, rather than to seek truth. I don't think that Lawrence Summers was trolling, and his remarks have a non-negligible chance of being true.
False dichotomy. People can be offensive accidentally and/or ignorantly in the course of making all kinds of statements; it doesn't have to be either deliberate or a side feature of an attempt at seeking truth.
Indeed, people can be offensive accidentally and/or ignorantly. And yet, people can also be offensive while seeking truth. The offensive thing might not even be true, but to reach a conclusion either way, the offensive thing must be entertained as a hypothesis. What if there was a hypothetical situation where the goal of seeking truth and the goal of not offending were at odds with each other such that one had to chose one or the other? Which goal is more important, truth seeking or non offense? People have hinted that it depends on the situation. Pretend I'm an alien and and I know nothing of our society. How would you explain this to me? What algorithm is used to decide when it's ok and when it's not ok to offend? And on a related note, why is it ok to offend certain groups but not ok to offend other groups? Eg. It's ok to offend religious people but not ok to offend ...
Do you want the real answer? Humans don't use algorithms for communication. At the very least, they don't use explicit ones.
You may offend people who do or believe foolish things, unless they meet the criteria for mental illness/retardation or you can avoid it without changing the substance of your claim. Of course, there are some situations where the need not to offend an off-limits person or group must take the backseat to some greater need.
"You may offend people who do or believe foolish things" How does one determine which acts and beliefs are foolish? "Of course, there are some situations where the need not to offend an off-limits person or group must take the backseat to some greater need." What are these situations? Who are the off limits people? What is the greater need? I'm not asking to annoy, I really want to know. I don't get it. Please explain.
I don't have an airtight definition handy, but it seems to me that (at least in the modern day in the developed world), religion is foolish, and (for instance) being a particular gender / race / etc. is not.
Hmmm. I can see a difference between religion and gender / race etc. So is the rule: "It's ok to offend a group if individuals have the option of belonging or not belonging to that group, but if individuals don't have such a choice, then it's not ok to offend them."?
That seems like an acceptable gloss of the distinction, although there are probably fine-grained intuitions it won't cover.
I don't like this rule. I don't like rules that restrict truth seeking. I think this amounts to modern day heresy.
I don't know what you mean by this. What's 'heresy' outside of a religious context, and why should we care?
I don't think that's very helpful. It doesn't seem to me that he as a very good grasp of what 'heresy' means, and you didn't explain what you meant by it in context. Did you mean: * What you're doing is the modern-day equivalent to heresy (which I'd need explained) * What you're suggesting is something that could serve the same purpose of accusations of 'heresy' in past days * something else Note that for literal readings, the modern-day equivalent of 'heresy' is 'heresy'. It still means the same thing and is used the same way by the Church.
By heresy I mean preemptively denouncing an idea because it doesn't adhere to some doctrine with no regard to whether the idea is true or false.
That's definitely not a way that I've heard 'heresy' used. From Wikipedia: In the Catholic tradition, heretics (people committing heresy) were condemned for leading people to believe that Catholicism is about something that it is not, and therefore putting their souls in danger. Perhaps in the definition you gave above, you were referring to the attitude the Church had towards heresy, rather than heresy itself? Also, your definition doesn't fit what Alicorn suggested above. It might if you replace 'idea' with 'utterance' and understand 'denounce' to not mean 'reject as false'.
It should definitely be allowed, but the poster should also expect many to react in disagreement. Myself, I think anyone calling Summers sexist for his comments is incorrect. But I also think Summers is incorrect in his conclusion, that the biological difference between the sexes is not great enough to be a cause of the low enrollment discussed. Maybe we should all play a game of rationalist taboo with the words "sexist"... and another game with the word "insensitive"...? It seems to me we all have different ideas about these and related concepts. Maybe this could be a helpful response to offenders and offendees in the future-- whenever and whyever it pops up.
Depends on the reason the true statement needs to be said.

What about a simpler standard:

Comments should be relevant and at least aspire to be rational.

If a comment is rational and relevant, I can't really see why it should be avoided -- even if someone might not like it. This means that you can (and should!) point out the use of stereotypes or prejudice and other biases. But the mere fact that someone could take offence is not the deciding factor.

While it's generally good to avoid offending people, it's cannot always be maintained -- for example, many of the atheism-themed posts would be offending to some the... (read more)

Relevant to what standard? A comment's relevance reflects its usefulness to a section of the community/audience. If the comment is offending a portion of that audience, does it hinder its own relevance? I think the whole gender hullabaloo has provided a decent enough answer to this question. A lot of the petty, potentially offensive things I notice are particularly not-relevant snippets smashed into otherwise decent posts.
I think the 'relevance' is usually rather easy to discern -- it simply means that a comment attempts to contribute something to the subject under discussion. I added 'relevance' only as criterion for completeness' sake, because it could make an otherwise rational, intelligent comment still be out of place. In practice, irrelevant comments are not a problem on LW. Offensiveness is orthogonal to relevance, I would think. But my question stands -- is there any comment that is both relevant and rational, but should yet be considered offensive or inappropriate? The 'petty, potentially offensive' things can be countered with normal rational reasoning, pointing out e.g. stereotyping, without the much more ambivalent rule of "don't be offensive". Relevant truths should be spoken, even if someone finds its offensive.

I think I agree with most of this. I did like the comments pointing out that we can be offended by having our status raised, but actually I think in that situation we're really offended on behalf of someone else when their status is lowered. Someone who puts down jewish people in front of me offends me because I object to the lowering of jewish people's status, rather than the raising of my own relative to them.

However, people also use the term "offended" when they're angered by something they think is morally wrong. So for example "I am off... (read more)

7Wei Dai15y
If someone were to speak favorably of Nazi Germany in your presence and you appeared to assent, that would impugn your morality and lower your status in your social circle, so you have to take offense and make clear to everyone around that you are absolutely not in favor of Nazi Germany. I think this is basically the same example as being offended by vulgar language. You want to avoid lowering your status by association with anything vulgar.

I think that if we are going to have a comment policy that takes offense into consideration at all, the burden of proof needs to be on the offendee, not the alleged offender.

Please everyone, try to ignore your monkey-brain-bits a tiny bit more.

I thought the burden of proof was on the offendee already, but it's not a bad idea to point this out in the comment policy. "If you see something offensive which detracts from a post, please politely describe what is offensive about it so that the author can correct it." Or words to that effect.
What could be used for proof? This is about emotions. We can't completely get away from them, but they're something of a rubber ruler.

I think this is sufficiently straightforward and comprehensible to be included in the Comment Policy - probably as a line saying, "Don't be offensive", with a sentence summary and a link back here.

And yes, I'm aware that we don't seem to have a Comment Policy. Given the past few days, I think it would be good to make one.

I'd emphatically disagree. This dispute has lead to some rather valuable and insightful posts, imo. It did not seem to get particularly heated, and people discussed it, for the most part, quite civilly. If future problems arise, a simple comment with a few links to these posts that have sprung up in the last few days should quickly correct the offender. An overt policy will, first of all, probably be read by two people, one of whom wrote it, and second of all may stifle useful discussion should a similar but not similar enough problem recur. The depth of discussion about this seems to be proportional to the complexity and intrigue of the topic matter, not the actual offense caused by the comment that started this all.
Well, it did get about as heated as things get around here. It's not like someone was trying to say that the airplane moves, or that you shouldn't switch envelopes, or something.
The point of having an explicit Comment Policy is that we are presently (or, at least, were very recently) driving away half our possible audience with the way many of us were talking. At least one person commented explicitly to that effect on "Sayeth The Girl". By having a Comment Policy that we regulars can agree upon and that we can point the new readers to, we can make fairly sure that there's a widespread knowledge of the kind of norms we want here, in a way which won't encourage flamewars every six months.
To be fair, the female audience isn't here mostly for other reasons. Also, the pick-up talk seems to be a fairly recent thing which appeared along with the self-help (akrasia) and rationality = winning talk.
The rationality = winning talk has been around for rather longer than the rest of it - indeed, longer than this site.
This is true, but OB was mostly focused on epistemic rationality. Interestingly enough, I think posts on instrumental rationality will probably attract more women, since more women are interested in practical stuff rather than in the theoretical and philosophical.
And you know this how?
Okay, I don't want to argue from anecdotal evidence and stereotypes, so I will admit I don't really know it. But gender and sex differences are possibilities which we can't just ignore - Of Gender and Rationality points (4), (5), (6), (7).
"Gender differences", where "gender" means the social construct of gender identity, are almost certainly relevant; I don't think anyone would claim that different genders aren't socialized differently starting from a young age. Genetic sex differences have a much higher burden of proof, because there has to be some evolutionary benefit to such differences developing. That said, it would still be nice if you had direct evidence for your particular assertion ("women prefer practical stuff"), because I've seen this assertion before but don't find it persuasive. It doesn't match my own anecdotal experience, and it sounds a lot like other hypotheses on systematic differences between superficially obvious groups that haven't panned out (cf. this post and related posts on Language Log about gender speech differences).
This is known to be false. Many differences are purely random.
It is known that many differences between related species are due to genetic drift rather than selection. Is it known that differences between males and females within a species are due to genetic drift?
There is as much genetic drift on the X and Y chromosomes as on any of the other chromosomes, so at least some of the differences between sexes must be caused by it.
The Y chromosome is physically much smaller than all the other chromosomes. In particular, it's not an equal partner to the X chromosome. From Wikipedia, I learned that the human Y chromosome contains 86 genes, which code for only 23 distinct proteins.
One example of a sex-linked trait that's wasn't due to selection is colorblindness, which is common only in males because the relevant genes are recessive and located on the X chromosome.
At some point, a population of organisms developed some sex determination genes. Over time, this system developed into the XY sex determination system: sexual dimorphism due to the segregation of sex-determining genetic information onto a single highly degenerate chromosome. I think it's very implausible that this segregation is without evolutionary benefit. The occurence of sex-linked monogenic traits like the example you give here are a side effect whose evolutionary cost was outweighed by the benefit of the development of the XY system. So I don't think this is a good example for the claim you are trying to defend.
Random differences still need to be roughly fitness-neutral, else they will be selected against. Given the stunted nature of the Y chromosome, as Cyan notes, and the fact that all other genetic material is shared, this still means the burden of proof is stacked against the idea of non-obviously essential sexual dimorphism, and strongly so in the case of proposed differences that are large compared to intra-sex individual variation.
I don't think simply counting genes tells us much of anything about the amount of sexual dimorphism in a species, one way or the other. The vast majority of genetic material is shared between sexes in any species, and some species don't even use genes to determine sex. If the fact that most genetic material is shared between sexes really did stack the deck against large sex differences, then we would never see large sex differences anywhere. Futhermore, I'm not sure what you mean by "non-obviously essential." Obvious to whom? It's not at all obvious to me why (say) it would be adaptive for human males to have so much facial hair compared to human females, and yet human males really do have a lot of facial hair.
The discussion here was about behavior, not physical differences. My apologies if I was unclear about that. Furthermore, you seem to be reading "stack the deck against" as referring to likelihood of differences arising, I meant it more in the sense of "here is an observed behavioral difference between human sexes, is it due to 1) statistical noise 2) learned social behaviors 3) intrinsic genetic differences, &c." It seems reasonable to have a fairly low prior for #3 vs. #2. Also, "obviously essential" in the the sense of "the whole point of sexual dimorphism". Suckling an infant is pretty clearly essential behavior. "Women are more practical", not so much.
I agree that in the case of behavioral differences, we have a prominent "learned social behavior" hypothesis that we do not have in the case of physiological differences, but it's not because of the number of genes shared between sexes; it's because of the common-sense intuition that culture influences behavior in a dramatic way that it doesn't influence physiology. I agree here. (In particular, "Women are more practical" is vague to the point of not-even-wrong-ness.) However, it does seem worth noting that if there are non-obviously-essential physiological differences (such as male facial hair), then it's at least not implausible that there might also be non-obviously-essential brain development differences that manifest as behavioral differences.
Well, in a counterfactual world where somehow genetic differences between the sexes were much larger, comparable to the genetic differences between humans and other primates, genetic reasons for behavior differences would be a lot more plausible, just as genetic differences explain behavioral differences between us and other primates now. That this is not the case in reality is why "learned behavior" is a stronger, common-sense hypothesis. Of course. In fact, there are probably quite a few. But for any given observed behavioral difference, it's sensible to assume it's a learned behavior lacking strong evidence otherwise (such as consistent observation of the same difference in multiple unrelated cultures). I think we're arguing 95% terminology and 5% substance here.
I suspect there is a substantive disagreement lurking here. Specifically, as much as it hurts my liberal feminist heart to say it (or it did hurt, before I got jaded), I'm going to have to deny this: Maybe we're tripping over this word genetic? When I say that the number of shared genes doesn't matter, what I'm getting at is that while SRY may "just" be "one gene," it triggers this entire masculinizing developmental process, and while I haven't studied the details (yet), it doesn't look trivial. Obviously culture exists, but the capacity to generate and transmit culture is a specific ability of human brains that happens in a specific manner, and if there are innate sex differences in human brains, then we are not justified in assuming that a given behavioral sex difference is a strictly cultural artifact that could have just as easily gone the other way. Culture is---I don't have the word for it---informed, constrained?---by human nature. We have to reason these things out on a case-by-case basis. Suppose---suppose American males score better than females on a test of mental rotation by eight-tenths of a standard deviation, and suppose we don't have any cross-cultural data, however dearly we might wish for it. I can't bring myself to presume a social explanation.
Another data point for contextualism in epistemology! ETA: (to self) confirmation bias much?

It's not status that's the issue, it's the offendee's conception of reality. Status is just the most common (by far) example of this.

Whenever a person has an "image" (subconscious subtle bias) of how things work, without consciously being aware of it, that person's perception of reality is distorted by the image/bias. Then, whenever some input does not fit with the image; either due to someone else asserting their own conflicting image of reality, or due to someone speaking bluntly about a conflicting observation of reality, the biased person su... (read more)

If the offender wishes to avoid the consequences that come from the offended individual successfully portraying them as offensive or otherwise doing reputation damage then yes, it is the offender's job not to offend people. It is likewise the offender's job to avoid offending people if they happen to intrinsically value other people not being offended by them, where the degree of value is less than the cost of limiting their freedom of expression and extra modelling of expected reactions. Not true. Have the offender publicly executed via gruesome torture. That can make future instances highly unlikely. (Reduce the degree of 'tantrum throwing' as necessary.) You can change the environment to be more like how you want it to be. It is also worth noting that avoiding the emotional experience of being offended is not the point. In the same way avoiding the emotional experience of being angry, sad or afraid is not the point. Those feelings are usually there to indicate that we are best served by taking actions in the actual real world so that outrageous, sad or scary things don't happen to us as much. Removing the feeling without removing the stimulus is not always a good thing.

I'll probably be referring this post to some people who don't actively read LW. For that purpose, it might be good if you could make the references to Alicorn's, orthonormal's and Eliezer's proposed explanations into links to them.

2Wei Dai15y
Sure. BTW, I wonder if anyone can suggest a better link for "status" than what I'm using now, which is
Thanks! (rewarded you with a karma point) Well, there's obviously always the wikipedia article, at least.
1Wei Dai15y
Wikipedia is the first place I looked, of course. :) Unfortunately that article just isn't very good.

"If such research does exist..."

Perhaps tangentially related:

Conservatives are more easily digusted

It will be nice to come up with a more precise definition of 'lowering the status'. For example, if some person treats me like a non-person, all he is doing is expressing his opinion about me being a non-person. This being the opinion of just one person, should not affect my status in the whole society and yet, I feel offended. So the first question is whether this should be called lowering of my status.

Also, let us assume that one person treating me like a non-person does lower my status in some way. Even then, shouting back at him and informing him that ... (read more)

If somebody treats you like a nonperson one of two things is true. Either 1) You are a mere possession, and thus have a very low status or 2) the person treating you that way did something inappropriate. Being offended and making a scene is a way to show that 2 is the case, and thus defend yourself against the decreased status.
Shouting at him may not change his behavior, but it may be good for your status. It certainly tells onlookers something.
Actually, my experience is, that when I protest against the behaviour, which I perceive as somewhat offensive, the other party gets the message and either stops or at least becomes less intensively offensive. I try to protest in a peaceful manner, not shouting, or offending more, etc. The idea is, that sometimes people do not realize they make someone feel threatened for their social status. By protesting, I am giving them a feedback and a chance the war will not be initiated. Funny enough, my boyfriend often believes, that by giving such a feedback, he is showing he feels threatened, makes himself more vulnerable and it is "not a good way of gaining back the lost status". Well, maybe there is a gender difference. My boyfriend also says, that women live in a different reality than men, because they are known to be generally less aggressive. As a consequence, they are perceived as less threatening in the social interactions, and people are just NICER to them. So there is still a possibility, that the peaceful protest, which often works very well for me, will not work for men. However, I suggest to give it a try.

I think this is the most useful way of framing the situation yet. I now realize that the things to which I took the most "offense" were what I interpreted as gambits being made, control being sought, etc. Status grabs.

I'd put it this way: offense is given by someone who violates a social norm the offended rely on to justify or enforce their status claims. This principle excludes the variety of status claims that don't give rise to offense. It explains why obscenity and profanity offend some people — those who assert superiority by obtaining deferential and proper speech from others. It also explains why the reaction is punitive: it's the way social norms generally are enforced. I expand on this position (in a legal-writing context) in a blog entry "Formula and Formality": Stephen R. Diamond
I think I stated the governing principle accurately here. It accounts for all the countexamples mentioned (although my account predates this discussion). Others must disagree; I wonder why?

Another thing to consider: if you are correct, then to speak of one group of people as subjects and one as objects naturally gives the subjects higher status than those that are the objects. In the case of the typical PUA community talk, the male is the subject and the female the object.

This particular problem can be avoided by speaking of both as objects (as in the case of scientific study) and distancing one self from the subject matter.

Compounding this is that when men are subjects and women are objects in a rationalist forum, this draws on some long-standing tropes about men being essentially more rational than women.

"After all, to lower someone’s social status is to cause a real harm."

To lower someone's social status is to cause that person harm. So is taking a person's job, where, also, one's loss is another's gain. Assaults on status do not necessarily lower total utility. Since status is relative, where one loses another gains.


"If such research does exist..."

Perhaps tangentially related:

Conservatives are more easily digusted


After all, to lower someone’s social status is to cause a real harm.

If status is a zero-sum game, raising someone's status also causes real harm, or rather both actions are neutral.

It makes sense to strive for an equitable distribution of status, though.