« More Getting Froze | Main | Kid's Rights »

March 12, 2009


Interesting. I've posted this before, but I'd like to see you take stabs at creating a more quantitative language of status. Also, I like that you're taking more comprehensive views. It occurs to me that one element of markets is that they can be decentralized means of producing or altering status (as measured by wealth) that can be more efficient than more qualitative elite assessments. Also, in looking consequentially at how to employ status allocation, I can think of at least two concerns: (1) in hierarchies, we want the most effective resource administrators (including human resources, (2) we want status awards to maximize achievement incentives.
There's a trickier area I've been thinking about how beautiful and entertaining (or other less politically correct primate bais conforming/satiating) people may improve societal productivity --so although they may not be the narrowly technocratic choice, they may be the consequentially best choice. One could call that something like the sticky bias veto. So a very attractive or entertaining President who is fairly competent could be a consequentially better choice than an extremely competent but ugly, boring President --given that the position produces not just a lot of policy, but a lot of leisure and entertainment media.

A good post. You should have linked to your previous ones on socially approved status markers as honest fitness signals and preferences for democracy based on the feeling of increased status for voters. I'll link to them when I find them.

Refusing to acknowledge inequality is not the same as reducing inequality.
Yes, but without any units we can only measure status ordinally and in such case it can NEVER be reduced.

HA, given your obsession with maximizing your persistence odds, I find it surprising that you're taking a somewhat utilitarian approach on leisure/entertainment here.

Personally, I don't feel comfortable conversing about sex or status without a layer of abstraction, which OB helpfully provides.

Perhaps one advantage of disallowing status contests is that people with low status aren't unambiguously shown that. Thus most of the population can believe they are above the median and be more productive for it. Quote G. H. Hardy:

Good work is not done by ‘humble’ men. It is one of the first duties of a professor, for example, in any subject, to exaggerate a little both the importance of his subject and his own importance in it. A man who is always asking ‘Is what I do worthwhile?’ and ‘Am I the right person to do it?’ will always be ineffective himself and a discouragement to others.

Relative to societies where most people have a say in ranking folks around then, status-prudish societies tend to delegate this ranking task to elites.

I'd rank this as a truly fundamental insight!



I think you misunderstand my comment.
I think a certain amount of entertainment probably optimizes productivity given that it's primates whose productivity we're trying to optimize. Also I'm not stating a firm conclusion, I'm highlighting an area that I intuit could be fruitful to research. An alternate hypothesis is that when entertainers usurp high status administrative positions from the domain competent, they're engaging in rent-seeking behavior. Relatedly, I think it's difficult against the gradient of our primate bias to make positions that manage greater resources lower status than positions that manage fewer resources.

Hopefully, I'm skeptical that the high status are the most effective resource administrators.

TGGP, I added one of the links you requested. With more effort we could measure cardinal status.

John, it would be good to study the costs/benefits of status ambiguity.

Eliezer, I was proud of it. :)

Your method of evaluating intellectual ability may be more biased than checking out the speaker's degrees.
If someone has a fancy education or job, he can at least show that he was able to accomplish something minimally substantial over time, which provides a decent sample. If you judge him based on a single conversation, well- What if he was nervous? Sick that day? Expresses himself better through writing than through speech?


Relative to societies where most people have a say in ranking folks around then, status-prudish societies tend to delegate this ranking task to elites. This may in fact produce a better society, but it seems odd to call this more "egalitarian"; people still end up being ranked, and the power to set those rankings is concentrated more into elite hands. Refusing to acknowledge inequality is not the same as reducing inequality.

Two thoughts:

Refusing to acknowledge inequality may lead to a reduction in the sort of inequality people care about. People seem to enjoy feeling better than others, but those others don't necessarily need to be their peers or friends. Likewise, people dislike feeling inferior to others, but they don't care so much about feeling inferior to far-off elites. They mostly dislike feeling inferior to friends and colleagues. It seems to me this leads to an equilibrium where people tend to ignore status with closer friends, while placing greater importance on it when dealing with strangers.

Elites likely better measure the worth of a person than direct status competitions. I'm much more impressed by someone who is promoted because of his talent at his job than a man who wins a pistol duel. I'd rather reward the later behavior than the former, as this obviously transforms a negative sum game into a positive one.

On academia, I'm no academic but I'd guess written arguments are more polite because they level the playing field. Some people aren't good at debate or discussion, but that doesn't make their arguments less correct.

Sorry for the double post, but John Maxwell brings up an interesting point. Since status is a subjective thing, why not have everyone believe they are better than everyone else? This is easier to do with far-off abstractions, e.g. "I'm better than Obama because he's a socialist", or "I'm better than my boss because I'm not rich and greedy", and less easy to do with friends who might share the same values and hobbies.

Friends and peers can implicitly agree to stop any status competitions, and each individual can go on thinking he's better than the others without any conflict. You don't need to do this with far-off individuals, because they probably don't care what you think of them (and the implicit bargaining is more difficult in any case).

The general problem with status is that it's, unlike wealth, zero-sum, because only relative status counts. Someone else gaining status hurts you directly, and destroying the status of someone else (a rival) helps you directly. There aren't too many truly zero-sum goods in the world like status, and status is so directly related with reproduction (women tending to choose high-status men) that I suspect most of the violence and agression that ever happened on the world was due to status.

There are two ways to escape status battles.

One is to be status-egalitarian, which isn't the same as economic-egalitarian. America is more status-egalitarian than Europe, despite Europe being more economic-egalitarian. (Here in Europe, despite the relatively larger income equality, middle classes take much more pains to f.e. speak clearly differently than working classes etc.)

The other is fixed status: aristocracy, fixed hierarchy etc.

“Refusing to acknowledge inequality is not the same as reducing inequality.”

This may be not totally accurate. All people are different. However, to turn “difference” according to some criterion into “inequality”, you need people to acknowledge this criterion to be important. For instance, having a different skin color might be of no big deal in some cultures, but of paramount importance in others. Higher intelligence may be envied in college, but lead to ostracism in high school.

It is probably true though that you can not make inequality go away completely. Still, you might change the relevant criteria for comparisons.

Grant, it find it unlikely that people really only dislike hearing about their low status, rather than being low status, but I'm open to data on the subject. And why can't we say for any preference people have that they really only want to believe that their preference has been satisfied? That is, how does this argument apply with any extra force to status?

Miklos, how can status be both a zero-sum rank and allow some societies to be more status-egalitarian?

Leonid, you lost me.

Robin: In large part I agree with many of the points you make about status and enjoy your posts on the subject. I also agree with Hopefully Anonymous in that your arguments would be more compelling if there was some way to quantify status and its consequences. I respectfully suggest that until you are able to measure status in some sort of index or quantitative measure, it is going to be tough to prove any of the points you are making, even though they intuitively seem likely.

Addressing your larger point about cultures having differing tolerances for allowing “status jousting”. Maybe it comes down to the costs involved. In animal societies, the cost of challenging for status are high, it usually involves a physical confrontation where the loser is either killed, maimed, or driven off and their survival potential diminishes. The net result is that these types of challenges occur rarely. So the higher the costs of status challenges the less often you would expect to see them. Consider a work example, you would expect to see more status challenges between people of approximately the same status because the costs are lower, versus examples where an employee openly challenges the boss, which may result in them getting fired.

I expect to see similar effects regarding signaling sexual interest. The higher the costs of the signaling the less you are likely to see of it. Consider the example of a secretary, she is much more likely to flirt with a random stranger in a bar, that has little downside, than with a superior at work. If her advances at work fail, her reputation may be damaged, and the social costs much higher.

So both with status signaling and with sexual signaling, I would argue its prevalence is related to the costs of failure, and the higher the costs, the less it occurs.

We have all been at conferences,where the person asking the question of the speaker, is not interested in the answer. So more often than not they already know the answer, the questioning is merely a mechanism to signal their status and perhaps challenge the status of the speaker. Less direct challenging of the speaker is likely to occur if the costs are higher. For example, one is less likely to aggressively challenge a Nobel Laurette publicly, for the risk that he shoots you down, and embarrasses you (costs you social capital) versus a middle of the road person, where your status failure challenge is less obvious and less costly. I see this in the presentations at Stanford. When a real big wig comes to town the questions are much more deferential because people are reluctant to take on the big wig and be publicly embarrassed. With lessor known speakers, there is much more of tendency to be more confrontational... to basically say you are wrong, and here is why ...

Russ, animal status isn't that simple. A lot of status confrontations are non-violent (birds singing) or there's a possibility of submission rather than major injury.

Status and reproduction is complicated, too. The alpha males can spend so much attention on each other that the females choose betas. Not betas exclusively, but enough that being an alpha is not a simple best strategy.

I think this collapses too many dimensions into one and then suggests a tradeoff between the endpoints of the collapsed line. Maybe that tradeoff is there in practice, but it would take more examples to convince me.

For example, you could imagine two societies in which status is determined by perception of wealth, one in which wealth is public and a prudish one in which wealth is the subject of gossip. In neither one would I expect the elites to play a role in bestowing status, but in one skill in gossip is relevant.

In this example, something that I would call prudishness has the opposite effect on the relevance of gossip skills.

"Hopefully, I'm skeptical that the high status are the most effective resource administrators."

I share your skepticism that there's a perfect correlation (hence I used the term rent-seeking). My point was that one is probably working against a strong gradient in any human culture to build social consensus that larger resource administrators (effective or not) are lower status than smaller resource administrators.

Nancy great criticism of a good comment by Russ. I encourage both of you to comment/critique on my blog, too.

Nancy & everyone,
Animal combat and contention are very interesting. The key reason why it is so often non-violent (or "slappy" rather than injurious) is much like what Nancy mentioned: when two parties duke it out, the real winner is often the third, fourth, and fifth (etc) competing parties, which benefit (at no cost) from the contenders' weakening or injuring each other.

Thus, if there are two competing ant colonies on a tiny island in a lake, it would probably behoove the larger one to totally destroy the smaller one, even if it suffers major casualties in doing so. On a large island with 80 competing colonies, the optimal level of aggression may be much lower.

Intergroup lethal violence in chimps usually involves three or more allies attacking one foe. At this ratio, the cost-benefit analysis changes radically, because two of the attackers can restrain the victim's limbs. Thus the attacking party faces little risk.

Most species of birds are more or less monogamous (with or without modest polygamy), and do not often engage in serious combat. A lot of the ritualized soft combats seen in various bird species, many of which don't include any bodily contact at all, seem to me like they may primarily be displays of reaction time (and muscle quickness). If I can see that a rival is 150 milliseconds faster than me, the odds in a mortal combat may then be 75/25 in his favor, and I would have a negative expected gain by fighting to the death rather than submitting. If the amount of territory or food I'm willing to cede grows large enough, he also has a negative expected gain by killing me, since he faces some risk of death or injury in the process of doing so. So it goes for birds. When the prize is immense, eg control of an entire harem, severe violence is much more common (elephant seals).

... which is why male elephant seals are about 3 times as large as female elephant seals.

There are two different kind of status, although there is some interdependence between the two.

First one is similar to dominance hierarchy in animals. This type of status is visible only in social interactions and it is dynamic. Almost everything people do (words, body language, voice tonality) can viewed as a status transaction, mean to lower or raise somebody else's or your own status. This kind of status you can't "have" but you always "do". Even person who typically has high status may have down day, and be visibly less confrontative, and have lower position in dominance hierarchy than usual.

Second type is what people commonly mean by status: reputation, position, car, hotness of wife, money, etc. This kind of status you "have". This second type of status usually raises your self-confidence, which "leaks" to the first type of status. Also, this kind of high status etc. may influence other people to perceive you more socially dominant.

Mikko, in your formulation, why do you think the second type isn't similar to dominance hierarchy in animals?

HA, depends on the definition of "similarity". Second type operates on symbols, and animals can't do that. First one is very similar to dominance hierarchy in other primates.

Mikko, the examples you chose (treatment by other social group members, treatment by high status females, attractiveness of habitat) seem to me like they all have popularly known analogues in nonhuman animal species. It's weird that you seem to be ignorant of what I think is common knowledge. You also don't seem to be a domain expert discounting popular misconceptions due to superior information. So your posts on this topic seem a bit weird to me.

Interesting, but I'm unconvinced.

1) Consider: why are people status prudish? Non-prudishness seems more common in low status groups, so perhaps being prudish about status demonstrates that you are high status. This makes sense because it shows you are powerful and knowledgeable about the required things enough to manipulate your position subtly. Then we aren't delegating, but making the costs of status (in effort to know the right info and behavior) higher, as we seem to continually do with the costs for status.

2) In all societies elites appear to have as much say as they want in status; part of having high status is the demonstration that you have power over other people's. If someone elite can be bothered intervening in your status it would probably override what your friends thought anyway, so any delegating we did to elites shouldn't make a difference. Elite influence just looks more proportionally if you aren't keeping up with subtly influencing your own status in the harsher competition.

3) Most elites don't know you - how do they decide who to give status to? Probably the signals they see, which you do control. Am I missing who you mean by elites here? I'd like more examples - in Australia you get into college purely based on a number derived academically - having gone to a good college offers status still, but it looks like people have a lot more say in their own if this matters proportionally more than in status non-prudish groups - if there any reason it is less likely to go this way?

Katja, social norms like prudishness apply to everyone in a culture; distinguish poor cultures from poor people in a culture. Higher status people quite often do not have all the say they want over the status of others.


"Most elites don't know you - how do they decide who to give status to? Probably the signals they see, which you do control"

This seems obtuse to me. Elites easily work through groups. Outgroups are quickly regulated and controlled through democracy. An individual can control only a few important status indicators. Others are your lucky or unlucky accident of birth, sadly. And the accidental ones tragically are often the most dominant in status issues.

I don't know what hanson or elkins meant about elites. Could you give concrete examples?

Here's a concrete example that doesn't seem to fit what either of you is saying: in the US college admission, especially to elite colleges is not transparent. It has to do with some simple numbers, but it also has to do with essays and interviews. The children of alumni are admitted to elite schools ("legacy admissions"), producing identical high status through a different channel.

This is compatible with KG's "elites don't know you." On the other hand, the importance of interviews means that skills for interviews and knowing what elites want to hear is more important than skills for gossip and knowing what your circle wants to hear.

Most of us can effect some improvement in our status if we put in the effort. When a competent team psychiatrist was replaced by an arrogant and less competent one, the psychologist and I discovered after a few months that we were engaging in more status-enhancing behaviors, both within the team and in the hospital at large. These took both the positive form of volunteering for extra bits of work and the negative form of quietly dissing others. Recognising this change, we were able to discontinue it as unnecessary. We had reacted automatically to a perceived threat. We could both observe what strategies we believed would "work" for us, and that in itself became a year-long discussion.

"3) Most elites don't know you - how do they decide who to give status to?"

By default, perhaps? Elites know elites. If they don't know you, you're not elite.

Robin, agreed, but doesn't alter my point. In many cases a behavior is a status signal, then is taken up by all, but the low status do it less well (which was why it signaled status to begin with). e.g. cars show status as rich have them -> everyone gets them, but poor people still have shoddier ones. Gifts are another e.g. Similarly everyone in a culture must be prudish once it begins, but the high status do it better. Those who do it worse in a given culture aren't less prudish, they are less successful at demonstrating status under those conditions. Gossip skills aren't less useful, they've just got so good more people can't, or have better things to do than, participate. They become lower status, not less influenced by the ranking of those around them, or their own behavior.

Are you sure there are female societies which don't revolve around rumor-mongering? Even the male ones I know of seem to.

Are you arguing that the relative status of people who are close in status are determined by elites more than elsewhere, or that this detail of status competition has disappeared somewhat, leaving differences between larger groups (that elites might bother to differentiate between) to make up more of a person's status (without this kind of activity increasing)?

When you say 'delegate', do you mean in any conscious way or just by default?

Frelkins, if elites control the status of whole groups by any means that seems irrelevant to the issue of who determines status between closer people who share the same groups and would in other societies be participating in ranking one another, which is what I thought we were talking about - I may have misunderstood. Accidental status determiners also seem irrelevant here, as they would apply in all societies.


Grant, it find it unlikely that people really only dislike hearing about their low status, rather than being low status, but I'm open to data on the subject. And why can't we say for any preference people have that they really only want to believe that their preference has been satisfied? That is, how does this argument apply with any extra force to status?

I think it applies with extra force to status because there is no objective measure of status. That is, there is no scoreboard of status ranking everyone can see. Unless there are actual status competitions, people's internal ordinal status rankings (if they even have any; I'm not sure I do for a lot of people) can be totally mis-matched from reality ('reality' being whatever results would come of status competitions).

Status competitions seem to be considered especially rude or childish when they are among friends or peers. People who think they will loose status competitions obviously don't want to, and may leave their social group if they feel their status has been lowered. People who think they may win status competitions may not be so adverse to them, but they may not want to loose friends or colleagues who may stop associating with them if they are humiliated.

Coupled with status competitions being negative-sum for other reasons, I think status prudishness is very rational. In a more tribal environment where people couldn't simply go somewhere else and associate with another group of people, they may have made more since. If that environment a humiliated opponent had no choice but to remain in the tribe.

I don't have any empirical data, only the anecdotes I've observed.

The comments to this entry are closed.

Less Wrong (sister site)

May 2009

Sun Mon Tue Wed Thu Fri Sat
          1 2
3 4 5 6 7 8 9
10 11 12 13 14 15 16
17 18 19 20 21 22 23
24 25 26 27 28 29 30