Cooperation is for Winners
Cross-posted from Putanumonit.
My last post was about learning to compete well from sports. But why compete at all? After all, I’m the one who wrote that winning is for losers, and we should avoid getting sucked into zero-sum contests.
Competing well doesn’t necessarily mean having to compete more, for one thing. Instead, it does allow you to choose where to compete. More importantly, sports-like competitions create prestige hierarchies (as opposed to dominance hierarchies). Climbing those is not intended to make you a fearsome boss but a valuable ally, one that others want to cooperate with. Getting to cooperation often requires winning competitions.
Wei Dai commented on the previous post:
More seriously, these days I think of competition as more of a problem than a solution. Some of the most important x-risks (e.g., advanced AI) are x-risks mainly because of competitive dynamics. If people weren’t competing for the prestige/power/money of being first to create AGI or to make advances in AI in general, we’d be able to solve AI safety problems at leisure.
How does one get to solve AI safety problems at their leisure, in a cooperative environment? Obviously, by working at the Machine Intelligence Research Institute.
But to work at MIRI one must first get a job at MIRI, and since MIRI has few spots and many applicants this means outcompeting other candidates for the job. MIRI itself is funded by donations, so it has to outcompete other non-profits for grants.
Hopefully, both competitions are conducted in a sportsmanlike way by demonstrating research achievements instead of by sabotaging competitors. But it’s still a competition – applying for jobs or grans is stressful and entails a high risk of failure, often due to circumstances outside one’s control.
Organizations have to manage very carefully the process of directing the winners of a competitive selection process to an ultimately cooperative endeavor. Here’s a wonderful post looking at this issue in two almost identical situations: chicken coops and academia.
So in the late 70s and early 80s there were active breeding programs to produce hens who laid more eggs. Take the highest egg layers in each generation and let them be the only ones that breed to produce the next generation. […]
Only one problem. High productivity egg-laying is associated with aggression – indeed the highest egg layers are basically the ones that beat up the other hens in the coop with them and capture the most resources. This led to a choice between two routes. If nothing was done, total egg-laying went down as there were too many injuries and too much mortality. The alternative was to continue the selection process but to supplement it by engaging in what many would regard as barbaric practices – beak trimming, declawing etc to prevent injuries from the aggression.
Then in the 1980s people got the idea to use group selection. Instead of picking individuals that were most productive, they selected entire hen houses that were most productive to produce the next generation. […]
Now, when I tell fellow faculty this story, many immediately respond all on their own “that’s like the selection being imposed for high-performing individual faculty members at the cost of their departments and the larger mission”. Are Dean’s and higher ups committing the same error as hen breeders and relying too much on a one-dimensional individual level selection that results in destroying overall productivity? Are we producing professors who need to be declawed and have their beaks trimmed to maintain productivity (or less figuratively is departmental collegiality suffering as a result)?
Having to compete for the opportunity to cooperate is how dating works too. Your goal is to reach cooperation (for a night or for a lifetime) with the person you’re pursuing. But you’re competing with everyone else: competing to be noticed, to go on a date, to earn their commitment. A lot of people I know miss out on relationships because they’re afraid of losing the competition – getting rejected.
They are ways to sneak around this fear, but they’re not optimal. Guys can use Tinder, which is set up to make “not being swiped right on” feel as little as possible like rejection. But Tinder is superficial, has poor norms around it, and only really works for the 20% of men who are most attractive.
Women can just choose not to ask anyone out at all, whether online or in physical spaces. But that limits a woman’s choices to her OkCupid inbox or the men who hit on her, which selects for men who are aggressive flirters instead of the other attributes that are much more important in a boyfriend. I think that most women can do much better by reaching out themselves to better guys than their inbox provides.
Getting over the fear of rejection is hard, but here again, sports can show the way. Losing a tennis match is less bad but has a lot in common with getting turned down in a romantic setting. In the moment, it feels terrible. But after a while, you reflect that losing the match/date is not a wholesale condemnation of your worth as a person or your social standing. It tells you where your expectations should be set in the future and, most crucially, how you should improve.
There are competitive elements even in things that don’t look like a competition at all, like blogging. I enjoy reading all across the rationalist blogosphere. We build on each other’s ideas, link to each other’s posts, and sustain a shared community. But nobody (with one exception) has time to read all the blogs, so we are also competing for reader attention, links, and space on Scott’s blogroll.
This doesn’t mean that I post with the intention of stealing readers away from John or Sarah. But it implies that I think that my blog is competitive with theirs, that it’s worth someone’s time to read Putanumonit on their restroom break. Blogging also means I’m willing to take an occasional L: writing something stupid or political or just poorly written and being told by people that I wasted their restroom break and they’re not getting it back. Hopefully, these losses make Putanumonit better.
Only 20% of LessWrong participants active enough to fill out a survey have ever written a post. When I ask my friends who are in the remaining 80%, they don’t say it’s because they have literally nothing to contribute. They’re afraid that what they write will not be good, and if it’s good it won’t be original, and if it’s good and original then Scott will write about the same thing better in a few weeks anyway. They don’t like the karma system of upvotes and downvotes. In short: they know it’s a competition, and they don’t want to compete.
No one gets to hear their ideas, enjoy their writing, or correct their mistakes. If cooperation is happening, they’re not a part of it.
Winning is for Losers was about one sort of mistake people can make: they see a hierarchy of points and rankings (e.g., the entire American education system) and dedicating all their resources to winning instead of asking whether the game is even worth it. But I’m seeing the opposite mistake more and more, people seeing competition and flinching away instinctively. They tell themselves that they’re the good guys, cooperating instead of fighting.
But if winning is for losers, cooperation is for the winners.