Winning is for Losers

Link post

This post origi­nally ap­peared on Rib­bon­farm. It was writ­ten as part of the Rib­bon­farm long-form writ­ing course and ed­ited by Joseph Kelly. I owe Joseph and the Rib­bon­farm ed­i­tors (Venkatesh Rao and Sarah Perry) huge thanks for spend­ing the time to make me a bet­ter writer.

Our world is filled with com­pe­ti­tion, fren­z­ied am­bi­tion in ev­ery do­main. In Western na­tions, and above all in the United States, it an­i­mates not only eco­nomic and fi­nan­cial life, but sci­en­tific re­search and in­tel­lec­tual life as well. De­spite the ten­sion and the un­rest it brings, these na­tions are in­clined on the whole to con­grat­u­late them­selves for hav­ing em­braced the spirit of com­pe­ti­tion, for its pos­i­tive effects are con­sid­er­able.

— Rene Girard, The One by Whom Scan­dal Comes

I. Eat­ing Dogs

Hu­man life is all about com­pe­ti­tion, from the micro level to the macro.

We are built by genes that out­com­peted their ri­vals over aeons of nat­u­ral se­lec­tion.

Chil­dren co­op­er­ate less and com­pete more as they grow older, even when com­pe­ti­tion is ir­ra­tional. By the time boys and girls hit pu­berty they start mer­cilessly fight­ing for sta­tus, in ad­di­tion to com­pet­ing for re­sources and at­ten­tion. As peo­ple en­ter the world of dat­ing and find­ing mates, the com­pe­ti­tion for sta­tus only in­ten­sifies. With dat­ing hav­ing moved on­line, ev­ery­one com­petes for the at­ten­tion of their be­loved against thou­sands of other Tin­der matches. And some­times also with the 5 other peo­ple they set up a date with in the same bar. The win­ner takes it all, and nice guys finish last.

We like ex­er­cise, mu­sic, and cook­ing. We like pro­fes­sional sports, Amer­i­can Idol, and cook­ing com­pe­ti­tions even more.

Poli­tics is war. The poli­ti­cal right sees a war be­tween bar­barous for­eign­ers and a civ­i­lized Amer­ica. The left sees a war be­tween eco­nomic classes, or among a mul­ti­tude of iden­tity groups fight­ing to op­press each other. The Liber­tar­ian Party is the only one that doesn’t look at poli­tics as be­ing pri­mar­ily about fight­ing some­one and they con­sis­tently gain less than 1% of the vote, the losers.

Our ed­u­ca­tion sys­tem em­pha­sizes com­pet­i­tive ad­mis­sions, ex­ams, and grad­ing on a curve. This is done to pre­pare stu­dents to com­pete in the job mar­ket and the econ­omy.

Our econ­omy is based on com­pa­nies com­pet­ing with each other in the mar­ket­place. But if you think that em­ploy­ees in the same com­pany will co­op­er­ate for the good of the or­ga­ni­za­tion then you haven’t been pay­ing at­ten­tion Rib­bon­farm: or­ga­ni­za­tions merely set the stage for a Dar­wi­nian con­test in which so­ciopaths pos­sess­ing the will to win op­press the clue­less and ex­ploit the losers.

If you don’t spend your time think­ing of ways to ex­ploit peo­ple you’re prob­a­bly a loser too. You should wake up to the re­al­ity of life as a com­pe­ti­tion and fol­low the ex­am­ple from the Dobu Is­lan­ders of Pa­pua New Guinea, a so­ciety that em­braced this idea com­pletely and with­out reser­va­tions. Ruth Bene­dict and Sam Har­ris de­scribe their cul­ture, the epi­tome of tak­ing this philos­o­phy to the ex­treme:

Life in Dobu fosters ex­treme forms of an­i­mos­ity and ma­lig­nancy which most so­cieties have min­i­mized by their in­sti­tu­tions. All ex­is­tence ap­pears to [the Dobuan] as a cut-throat strug­gle in which deadly an­tag­o­nists are pit­ted against one an­other in con­test for each one of the goods of life. Sus­pi­cion and cru­elty are his trusted weapons in the strife and he gives no mercy, asks for none.

The Dobu ap­pear to have been as blind to the pos­si­bil­ity of true co­op­er­a­tion as they were to the truths of mod­ern sci­ence. Every Dobuan’s pri­mary in­ter­est was to cast spells on other mem­bers of the tribe in an effort to sicken or kill them and in the hopes of mag­i­cally ap­pro­pri­at­ing their crops.


To make mat­ters worse, the Dobu imag­ined that good for­tune con­formed to a rigid law of ther­mo­dy­nam­ics: if one man suc­ceeded in grow­ing more yams than his neigh­bor, his sur­plus crop must have been pilfered through sor­cery. […] The power of sor­cery was be­lieved to grow in pro­por­tion to one’s in­ti­macy with the in­tended vic­tim. This be­lief gave ev­ery Dobuan an in­can­des­cent mis­trust of all oth­ers, which burned bright­est on those clos­est. There­fore, if a man fell se­ri­ously ill or died, his mis­for­tune was im­me­di­ately blamed on his wife, and vice versa. The pic­ture is of a so­ciety com­pletely in thrall to an­ti­so­cial delu­sions.

— Sam Har­ris, The Mo­ral Landscape

Chief Ga­ganamole of the Dobu, a real win­ner, and his wife. Image Credit: Ge­orge Brown.

Huh, this doesn’t ac­tu­ally sound so great.

The prob­lem with liv­ing in a dog-eat-dog world is that dogs just aren’t very tasty. But is it avoid­able? Can you do well in life with­out try­ing to com­pete, dom­i­nate, and win any­thing? Can you even get a date?

I think so. In­stead of eat­ing dogs, we can try to bake pies in­stead.

II. Bak­ing Pies

Life is a game, play to win.

This guy, or this guy, or maybe this guy.

I dis­agree with all three guys, but only with the sec­ond part of the state­ment. Life is a game, but there’s more to play­ing games than try­ing to beat some­one. To un­der­stand this game bet­ter we re­quire some gen­eral the­ory of games. I sug­gest game the­ory.

Game the­ory dis­t­in­guishes be­tween zero-sum games which are purely ad­ver­sar­ial and pos­i­tive-sum games which al­low for co­op­er­a­tion. “Zero-sum” means that any gain for one player means a loss for the other play­ers. In a zero-sum game there are no win-win pos­si­bil­ities and thus no point in try­ing to co­op­er­ate.

Imag­ine some­one emp­ty­ing a bucket full of coins (for my tech savvy read­ers: an ICO of cryp­to­coins) over a busy street. Every per­son in the area now finds them­selves en­gaged in the hilar­i­ous game of look­ing for quar­ters. All the play­ers end up with a pos­i­tive out­come in mon­e­tary terms (if we ig­nore dig­nity), but the game is purely zero-sum be­cause each coin picked up by Mr. Black is one less coin available for Ms. White to find.

If we de­sire to live less like the Dobu we should learn to rec­og­nize zero-sum games and avoid them. The coin game gives us two heuris­tics for do­ing that. The first is that zero-sum games usu­ally take the form of di­vid­ing a fixed pie. In our ex­am­ple, the “pie” was the bucket of coins dumped on the street. The play­ers have no way to get more coins thrown at them, they can only com­pete for the coins that are already there. The sec­ond heuris­tic is that each player is un­happy when more and bet­ter play­ers join the game. As more tal­ented coin scav­engers join, fewer coins are left for you.

In con­trast, a pos­i­tive-sum game in­volves a col­lab­o­ra­tive effort to which many play­ers can con­tribute. Play­ers bake a big­ger pie by co­op­er­at­ing. In pos­i­tive-sum games, the en­trance of new par­ti­ci­pants is ei­ther bad or good for the in­cum­bents, de­pend­ing on the situ­a­tion.

Let’s look at a more com­plex ex­am­ple from an arena that at first glance ap­pears purely com­pet­i­tive – pro­fes­sional sports. Speci­fi­cally, is the NBA a zero-sum or pos­i­tive-sum game for LeBron James?

LeBron James com­pet­ing. Image credit: Ezra Shaw /​ Getty Images.

A sin­gle game of bas­ket­ball is a rel­a­tively zero-sum af­fair, but ath­letes don’t join the NBA for the pur­suit of bas­ket­ball wins in a vac­uum. They get many re­wards for par­ti­ci­pat­ing: money, fame, groupies, and the satis­fac­tion of a bas­ket­ball game played at the high­est level. All of those make up the pie that NBA play­ers bake to­gether.

The ti­tle of “NBA Cham­pion” is a yearly zero-sum game, but it’s an ar­tifi­cial for­mat in­vented by the league. If the league could sell more tick­ets by hav­ing mul­ti­ple con­cur­rent cham­pi­ons or by award­ing style points in­stead of ti­tles, it would.

LeBron wel­comes bet­ter play­ers join­ing the league be­cause that would in­crease the NBA’s pres­tige, pop­u­lar­ity, and prof­its, of which he gets a share. In fact, in 2017 LeBron cost him­self money by beat­ing other teams too quickly – this led to fewer play­off games, which in turn de­creased league rev­enues, to­tal salaries paid to play­ers, and sub­se­quently the value of LeBron’s own con­tract. LeBron wants the league to be as good as pos­si­ble, and the other play­ers are col­lab­o­ra­tors rather than com­peti­tors in the big­ger pic­ture game of the NBA.

Of course, the NBA looks much more zero-sum to a marginal player. Un­like LeBron, a bench­warmer is not happy when more tal­ent joins the league, they may end up tak­ing his job. This points to an­other im­por­tant prin­ci­ple of games: strong play­ers have more room to co­op­er­ate, while weaker play­ers are forced to com­pete with each other.

Let’s con­sider ed­u­ca­tion, speci­fi­cally go­ing to a pres­ti­gious uni­ver­sity. If you’re a bor­der­line can­di­date for a uni­ver­sity, strong ap­pli­cants re­duce your chance of ad­mis­sions. Once you’re in, they make your grad­ing curve steeper and com­pete for on-cam­pus lead­er­ship po­si­tions and ul­ti­mately for jobs. Com­pet­ing against stronger stu­dents can have de­mor­al­iz­ing effects that per­sist long af­ter school is over.

This isn’t the case for the stu­dent who is much smarter than her peers. She wel­comes stronger class­mates. They im­prove her learn­ing op­por­tu­ni­ties and in­crease the over­all pres­tige of the uni­ver­sity, with­out be­ing a threat.

There are two ways to be­come a stronger player and “rise above the com­pe­ti­tion,” as it were. You can try to out­work ev­ery­one else, or you can look to be a big­ger fish in a smaller pond. Both op­tions can work for a col­lege ap­pli­cant, al­though prob­a­bly not as much for a bas­ket­ball player. The NBA is the only game in town, and NBA play­ers are pre­sum­ably already work­ing as hard as they can.

How­ever, there’s an­other way to avoid the grind of com­pe­ti­tion: in­stead of be­ing the strongest player, be the strangest.

If you pos­sess a unique skill, it com­ple­ments the skills of other play­ers in­stead of com­pet­ing with them. NBA play­ers have built lu­cra­tive ca­reers as “the guy who just blocks shots and has a sweet ’fro” or “the white guy who just stands in the cor­ner and makes threes.” It’s enough to do only one thing well if that thing is rare.

But for avoid­ing com­pe­ti­tion, hav­ing unique skills isn’t half as im­por­tant as hav­ing unique de­sires. The philoso­pher René Girard de­scribed the mimetic con­ta­gion of de­sire: peo­ple in­stinc­tively imi­tate the de­sires of those around them, which leads to ev­ery­one chas­ing the same prizes. Th­ese prizes of­ten have no in­her­ent value other than be­ing the ob­jects of shared pur­suit. When those prizes are in limited sup­ply, this pur­suit cre­ates zero-sum com­pe­ti­tion and leads to bit­ter ri­valries.

The two ways of be­ing similar re­in­force each other. When peo­ple go af­ter the same prizes, they will de­velop similar skills in the pur­suit. When peo­ple’s skills don’t set them apart, the will try to stand out by com­pet­ing ever more des­per­ately for the com­mon prizes.

We talked be­fore about how pres­ti­gious uni­ver­si­ties set the scene for end­less com­pe­ti­tion among the stu­dents at ev­ery stage of their ed­u­ca­tion. Dan Wang ties this to Girard’s idea of com­pe­ti­tion stem­ming from similar­ity and mimet­ics:

The closer we are to other peo­ple—Girard means this in mul­ti­ple di­men­sions—the more in­tensely that mimetic con­ta­gion will spread. Alter­na­tively, com­pe­ti­tion is fiercer the more that com­peti­tors re­sem­ble each other. When we’re not so differ­ent from peo­ple around us, it’s ir­re­sistible to be­come ob­sessed about beat­ing oth­ers. […]

It’s hard to con­struct a more perfect in­cu­ba­tor for mimetic con­ta­gion than the Amer­i­can col­lege cam­pus. Most 18-year-olds are not su­per differ­en­ti­ated from each other. By con­struc­tion, what­ever dis­tinc­tions any does have are usu­ally earned through bru­tal, zero-sum com­pe­ti­tions. Th­ese tour­na­ment-type dis­tinc­tions in­clude: SAT scores at or near perfec­tion; be­ing a top player on a sports team; gain­ing mas­ter sta­tus from chess matches; play­ing first in­stru­ment in state or­ches­tra; earn­ing high rank­ings in Math Olympiad; and so on, cul­mi­nat­ing in gain­ing ad­mis­sion to a par­tic­u­lar col­lege.

Once peo­ple en­ter col­lege, they get so­cial­ized into group en­vi­ron­ments that usu­ally con­tinue to op­er­ate in zero-sum com­pet­i­tive dy­nam­ics. Th­ese in­clude or­ches­tras and sport teams; fra­ter­ni­ties and soror­i­ties; and many types of clubs. The biggest source of mimetic pres­sures are the classes. Every­one starts out by tak­ing the same in­tro classes; those seek­ing dis­tinc­tion throw them­selves into the hard­est classes, or seek tute­lage from star pro­fes­sors, and try to earn the high­est grades. […]

No one has ever asked me how one should es­cape mimetic con­ta­gion on cam­pus. Still here’s my an­swer: If one must go to col­lege, I ad­vise cul­ti­vat­ing smaller so­cial cir­cles. In­stead of go­ing to class and prepar­ing for ex­ams, to go to the library and just read. Fi­nally, not to join a fra­ter­nity or fi­nance club, but to be part of a knit­ting cir­cle or hik­ing group in­stead.

— Dan Wang, Col­lege as an In­cu­ba­tor of Girar­dian Terror

Most of the prizes stu­dents com­pete for aren’t re­ally worth­while even when the temp­ta­tion to com­pete for them is over­whelming. Is the point of at­tend­ing col­lege to be elected fi­nance VP of some fra­ter­nity? Col­lege should be a place to have fun, get laid, make friends, learn some­thing, and figure out which ca­reer suits your in­di­vi­d­ual skills and tastes. Th­ese are mostly co­op­er­a­tive pur­suits, and Girar­dian com­pe­ti­tion stands in the way of achiev­ing them.

The strongest and strangest (e.g. knit­ting cir­cle) stu­dents won’t get sucked into the com­pet­i­tive vor­tex. They’ll spend time in the library study­ing what­ever weird sub­ject they’re ob­sessed with, they’ll make friends with fel­low geeks, and they’ll won­der why most of their class­mates are per­pet­u­ally mis­er­able.

III. Tits and Tats

We have started build­ing a frame­work of com­pet­i­tive and co­op­er­a­tive situ­a­tions. Com­pe­ti­tion stems from zero-sum con­tests over a fixed pie, where ad­di­tional play­ers are never wel­come. Co­op­er­a­tion comes from an op­por­tu­nity to bake a pie col­lab­o­ra­tively, and strong play­ers are wel­come if they con­tribute. End­ing up in the lat­ter situ­a­tion re­quires be­ing more ca­pa­ble than any­one else, or re­ally differ­ent from ev­ery­one else.

This foun­da­tion is enough to sur­vive in Har­vard or the NBA, but it’s in­suffi­cient for a real challenge like OkCupid. For a strat­egy that works in on­line dat­ing we need to dig deeper into game the­ory, and the one par­tic­u­lar game that is most heav­ily the­o­rized about.

The same way that biol­o­gists are sup­posed to study all liv­ing crea­tures but end up mostly fo­cus­ing on mice and fruit flies, so it is with game the­o­rists. They’re os­ten­si­bly study­ing all pos­si­ble games, but a huge chunk of the liter­a­ture is ded­i­cated to a sin­gle one, the pris­oner’s dilemma. There are many analo­gous fram­ings of the dilemma, I pre­fer this sim­ple, pris­oner-free for­mu­la­tion:

You re­ceive a wid­get with two but­tons on it, la­beled “co­op­er­ate” and “defect.”You are in­formed that an­other per­son some­where in the world re­ceived the same wid­get. If you press “defect,” $1,000 will be im­me­di­ately de­posited into your bank ac­count while the other player, whom you’ll never meet, gets noth­ing. If you press “co­op­er­ate,” the other per­son gets $3,000, but you get noth­ing ex­cept for a warm feel­ing. You both make your choices with­out know­ing what the other per­son chose.

That’s it, that’s the game.

The salient fea­ture of the pris­oner’s dilemma is that choos­ing “defect” makes a player $1,000 richer re­gard­less of what the other player is do­ing. In the ab­sence of mechanisms to in­fluence each other, this usu­ally leads to both play­ers defect­ing. Of course, if both play­ers chose to co­op­er­ate they’d each be bet­ter off by $2,000.

The “strong or strange” prin­ci­ple ap­plies here as well. A billion­aire may be happy to let a ran­dom per­son take $3,000, so may the guy who lives out of a van and climbs gi­ant cliffs with­out a rope. The former has enough money and the lat­ter doesn’t even need it to buy a rope. But for peo­ple who are nei­ther very strong or very strange co­op­er­a­tion is difficult and defec­tion is tempt­ing.

This sim­ple setup be­lies a rich uni­verse of hu­man in­ter­ac­tion. In his book, Mo­ral Tribes, psy­chol­o­gist Joshua Greene shows that most of our so­cial in­tu­itions and moral emo­tions evolved as means to co­op­er­ate in pris­oner’s dilem­mas with other peo­ple. Em­pa­thy and com­pas­sion al­low play­ers to co­op­er­ate by mak­ing it in­trin­si­cally re­ward­ing to benefit oth­ers. The ca­pac­ity to feel self-righ­teous or guilty sig­nals a per­sonal com­mit­ment to do­ing the right thing. Emo­tions like trib­al­ism and loy­alty al­low co­op­er­a­tion to be en­forced by a broader col­lec­tive or by an au­thor­ity figure.

The eas­iest way to achieve mu­tual co­op­er­a­tion is by re­peated pris­oner’s dilem­mas played with the same part­ner. This al­lows each player to play tit for tat – re­ward a co­op­er­a­tor with co­op­er­a­tion in the next round of play, and defect against a defec­tor. Tit for tat is im­ple­mented in na­ture by ev­ery­one from fish to birds to mon­keys. It’s such a use­ful co­op­er­a­tion strat­egy for Homo sapi­ens that we evolved a whole suite of emo­tions that help us im­ple­ment it: anger, trust, venge­ful­ness, and grat­i­tude.

Tit for tat works when you’re deal­ing with the same few play­ers over long time-frames. The strat­egy doesn’t work if play­ers don’t ex­pect to in­ter­act in the fu­ture. The in­cen­tives of fu­ture co­op­er­a­tion or pun­ish­ment lose their bite when deal­ing with large groups of play­ers, or when those play­ers are only con­cerned with im­me­di­ate out­comes.

Tits on a tat

What kind of game is on­line dat­ing? It can be a short-term and multi-player game in which ev­ery­one screws each other (in the literal and good sense, but also in the figu­ra­tive and bad sense). But if two peo­ple are try­ing to build a real re­la­tion­ship, dat­ing needs to be a co­op­er­a­tive, long-term, two-player game. If that’s the game you’re play­ing, tit for tat is your strat­egy.

Now, it may seem ob­vi­ous that find­ing a ro­man­tic part­ner should be a col­lab­o­ra­tive pur­suit and not a hos­tile con­test, but it’s not the nat­u­ral ap­proach. The way of dat­ing in na­ture is spiky dicks.

IV. Spiky Dicks

“Every­thing in the world is about sex ex­cept sex. Sex is about power.”

— Os­car Wilde

Wilde’s fa­mous quote sum­ma­rizes all the re­search in evolu­tion­ary biol­ogy which shows that through­out the an­i­mal world sex is about com­pe­ti­tion, and com­pe­ti­tion is usu­ally about sex. An­i­mals may co­op­er­ate with each other to ac­quire food and avoid be­ing eaten. But as soon as that’s taken care of, it’s back to vi­cious con­tests over mat­ing.

We nor­mally think of mat­ing-re­lated com­pe­ti­tion as hap­pen­ing among mem­bers of the same sex, par­tic­u­larly males. For ex­am­ple, male elephant seals fight so sav­agely for ac­cess to fe­males that at the end of their mat­ing sea­son 4% of males will have had most of the sex but 90% of males will carry scars and in­juries from fight­ing. From Achilles vs. Paris to Swaggy P vs. D’Lo, many a his­toric beef among men has started over a woman.

But the real ac­tion is in male vs. fe­male con­flicts. Those are no less vi­o­lent, and of­ten a lot more cre­ative.

In sev­eral species of bee­tles, the males have evolved sharp spikes on their gen­i­talia which an­chor the fe­male in place dur­ing cop­u­la­tion. Fe­male bee­tles evolve more soft tis­sue in the cop­u­la­tory duct to pro­tect them­selves from in­jury, which in turn leads males to evolve ever scarier look­ing dick­heads.

Cal­loso­bruchus analis pe­nis (bee­tle dick). Image credit: Wikipe­dia.

Ducks have taken this idea one step fur­ther, and then fifty more steps in re­ally weird di­rec­tions.

In­stead of choco­late and roses, male ducks usu­ally go for the “forced cop­u­la­tion” ap­proach to dat­ing. In re­sponse, fe­male ducks evolved corkscrew vagi­nas, which made male ducks evolve spring-loaded foot-long corkscrew penises (with spikes on them, of course). Fi­nally, fe­male ducks evolved branch­ing labyrin­thine vagi­nas so they can send the sperm of a male they don’t like to­wards a literal and re­pro­duc­tive dead end.


This sort of sex­ual arms race is the norm in the an­i­mal world. So far we hu­mans haven’t sprouted spiky gen­i­talia, our main weapon in in­ter-sex con­flict is ly­ing and de­cep­tion. Mem­bers of both sexes pre­tend to be fit­ter and more faith­ful to their part­ner than they re­ally are. Bet­ter pre­tense leads to bet­ter de­tec­tion of trick­ery, which leads to ever more so­phis­ti­cated ly­ing. Even­tu­ally, peo­ple evolved the abil­ity to con­vinc­ingly lie to them­selves, all the bet­ter to fool oth­ers about their com­mit­ment and at­trac­tion to a po­ten­tial mate.

This is as true to­day as it was on the sa­van­nah. In on­line dat­ing men lie about their height and in­come, women lie about their age and weight, and a quar­ter of pro­files have pho­to­shopped pic­tures. And you thought the news was fake.

Bul­lshit­ting is a use­ful strat­egy for spread­ing your genes widely with a min­i­mal com­mit­ment of re­sources, or for beat­ing your room­mates in a com­pe­ti­tion to sleep with more women. Most lies don’t sur­vive be­yond the first date, but they get a lot of peo­ple to go on that first date and get drunk enough to jump into bed with you. This is a very short-term and multi-player ap­proach to dat­ing, and some peo­ple as­sume that this is the only one.

Every­one com­plains that while on­line dat­ing made it eas­ier to get a first date, turn­ing that first date into a re­la­tion­ship be­came a lot harder. Most dat­ing ad­vice an­swers this co­nun­drum with “keep do­ing what got you the first date, just more and bet­ter.” This is dumb and doesn’t work. What gets you first dates is mass-ap­peal and ly­ing. Those are defec­tion strate­gies, they benefit the player while mak­ing dat­ing harder for both the gen­der they pur­sue and the one they com­pete with. On­line dat­ing isn’t defec­tive, it’s the play­ers who keep defect­ing.

Get­ting one per­son to spend a thou­sand nights with you is the ex­act op­po­site of get­ting a thou­sand peo­ple to spend one. It re­quires play­ing the op­po­site kind of game: long-term and fo­cused on a sin­gle per­son (or three, but not fifty). The strat­egy in this sort of game is to play tit for tat to achieve mu­tual co­op­er­a­tion with the per­son you will even­tu­ally end up with. You can play co­op­er­a­tively with that per­son even if you haven’t met them yet. In fact, your first shared goal is to find each other, and then build the foun­da­tion for a re­la­tion­ship that will make both of you happy.

The first step to­wards this is com­plete hon­esty. If the other per­son is so am­biva­lent about meet­ing you that an inch of height or a year of age would tip the bal­ance, you prob­a­bly won’t end up pick­ing baby names to­gether any­way. You shouldn’t lie on your pro­file even if ev­ery­one else does. The nov­elty of see­ing some­one who fulfills ex­actly what their pro­file promised will kick off your first dates on a note of pleas­ant sur­prise in­stead of dis­ap­point­ment.

But just be­ing hon­est is not enough. In ac­cor­dance with “stronger or stranger,” to avoid com­pet­ing with ev­ery­one else for your part­ner’s at­ten­tion, you have to be re­ally ir­re­sistible or re­ally weird. The lat­ter is much eas­ier and works just as well as the former.

OkCupid’s data shows that con­ven­tion­ally at­trac­tive pro­file pic­tures get far fewer mes­sages than pho­tos that elicit strong pos­i­tive and nega­tive re­ac­tions. As long as at least a few peo­ple re­ally dig you, hav­ing a lot of haters is not to your detri­ment. When I started dat­ing on­line, I wasn’t sure if I should use the photo be­low as my main pro­file pic. But when two women wrote me just to say that they would never date a man with a photo like this, I knew this was the right one.

My goal was to make it easy for my as yet un­known part­ner to find me, so I made my pro­file idiosyn­cratic enough to filter out most of the users that weren’t her. In­stead of a self-sum­mary, I started the pro­file with a stupid poem. I men­tioned all my es­o­teric in­ter­ests like Bayesian episte­mol­ogy. I listed sev­eral rea­sons not to date me. As I kept mak­ing my pro­file quirk­ier, the women it at­tracted were a lot more in­ter­est­ing to me.

Fi­nally, I got my re­ward:

The point of tit for tat is to defect against defec­tors (the 99% of women who aren’t re­ally into me) and to co­op­er­ate with co­op­er­a­tors (the few who are). For pro­file de­sign, this means scar­ing away the peo­ple who are at­tracted to you su­perfi­cially and ap­peal­ing to those who like your unique quirks. On first dates, it means cut­ting off those who aren’t ready to risk mak­ing a small com­mit­ment to you, and build­ing some­thing with those that are.

At the start of a re­la­tion­ship, the “defect” move is to go along on a few dates while swiping for other matches in the mean­time. It keeps your own op­tions open but does the op­po­site for the per­son you’re see­ing. The temp­ta­tion to do this ex­ists be­cause the 1,000 po­ten­tial peo­ple you haven’t met yet ap­pear perfect in the fuzzy light of imag­i­na­tion, while the ac­tual per­son in front of you has shown a wart or two. But mu­tual defec­tion has costs: it pre­vents both part­ners from mak­ing the effort to build the re­la­tion­ship on a stronger foun­da­tion than just mu­tual lust. Without that foun­da­tion, the lust hor­mones dis­si­pate af­ter a cou­ple of months and both peo­ple are back where they started, slightly frus­trated and two months older.

Tit-for-tat­ting the first date mostly means go­ing against com­mon ad­vice.

Every­one says to avoid heavy top­ics on the first date. But why would you waste time with some­one with whom you can’t have a se­ri­ous con­ver­sa­tion about the mean­ing of life or the min­i­mum wage? If these top­ics aren’t deal break­ers, you should be able to talk about them with open-mind­ed­ness and hu­mil­ity. If they are, you should use them to filter out in­com­pat­i­ble matches and get back to look­ing for the ones who un­der­stand la­bor eco­nomics.

Every­one says to avoid talk­ing about your ex on the first date. Maybe that’s a good idea, but for an en­tire year while I was dat­ing, I shared a one bed­room apart­ment with an ex-girlfriend. In New York City, it takes more than a bro­ken heart to give up pay­ing half-rent for a sweet pad. When I started dat­ing again I didn’t feel com­fortable bring­ing this up, but then I re­al­ized that I should talk about it un­apolo­get­i­cally on the first date. I would ask my dates to trust me that this was a tem­po­rary habita­tion cir­cum­stance, not a per­ma­nent emo­tional one.

This con­fes­sion ac­tu­ally worked to my ad­van­tage, it sent a strong sig­nal that I have noth­ing to hide and there were no other shoes wait­ing to drop. The women who were will­ing to trust me re­cip­ro­cated by tel­ling me some­thing em­bar­rass­ing about them­selves, and we turned an awk­ward situ­a­tion into an op­por­tu­nity to build mu­tual trust.

Every­one says to hold off on tex­ting af­ter the date, so as not to ap­pear des­per­ate. I as­sumed that if I had an hon­est and deep con­ver­sa­tion with some­one on the first date, she would have plenty of in­for­ma­tion about my value as a ro­man­tic part­ner with­out hav­ing to de­duce it from the timing of my text. Wait­ing 3 days to text cre­ates un­cer­tainty, and in the co­op­er­a­tive game of dat­ing un­cer­tainty in­creases the odds of defec­tion.

In­stead of the 3 day rule, I went with the −1 day rule. If I en­joyed the date I would say: “Hey, I re­ally en­joyed this date! I’m go­ing to text you to­mor­row at 8 pm to see if you want to go on an­other one.” This is a tit for tat move. I’m clearly show­ing that I’m play­ing co­op­er­ate, and I set clear ex­pec­ta­tions for re­cip­ro­ca­tion. Be­cause I let the girl know in ad­vance when I’ll reach out, if I don’t get a re­ply rel­a­tively quickly the next day I can safely as­sume that she’s not in­ter­ested, which saves me from chas­ing ghosts.

One piece of com­mon wis­dom that is ac­tu­ally true is that vuln­er­a­bil­ity is the key to build­ing in­ti­macy. And yet, very few peo­ple are will­ing to be vuln­er­a­ble in front of po­ten­tial ro­man­tic part­ners. Vuln­er­a­bil­ity is the ul­ti­mate tit for tat strat­egy: there’s a lot to gain if the other player re­cip­ro­cates, and a lot of pain if they defect.

Of course, it’s pos­si­ble to be too vuln­er­a­ble on the first few dates, just as it’s pos­si­ble to be too weird, too deep, too hon­est or too de­mand­ing. But in my ex­pe­rience, peo­ple are afraid of be­ing too open much more of­ten than they ac­tu­ally are.

The strate­gies above do of­ten fail, in the sense of scar­ing some­one away from a sec­ond date. But if they only “fail” in cases where the sec­ond date wasn’t go­ing to lead to a tenth, that’s a fea­ture. As with star­tups, if you’re go­ing to fail you should fail quickly, and move on to some­one who tits your tats.

And when those strate­gies suc­ceed, they do so mag­nifi­cently. If you lead off the first date with hon­esty, vuln­er­a­bil­ity, and com­mit­ment and the date turns into a re­la­tion­ship, the re­la­tion­ship will also be based on hon­esty, vuln­er­a­bil­ity and com­mit­ment. This is worth a lot when so many re­la­tion­ships are based in­stead on pre­tense and power games.

Tit for tat doesn’t ap­ply to all as­pects of a first date. Where it doesn’t, just co­op­er­ate un­con­di­tion­ally. Take a shower, show up early, com­mute to the other per­son’s neigh­bor­hood, turn off your phone, offer to pay. Don’t be a spiky dick.

I learned those les­sons over a cou­ple of years a cou­ple dozen OkCupid dates. Like ev­ery game, dat­ing is a skill that im­proves with prac­tice. With my tit for tat game and with the help of a spread­sheet, Bayesian episte­mol­ogy girl and I are get­ting mar­ried in the fall hope­fully got mar­ried in the time be­tween me writ­ing this post and pub­lish­ing it. If we didn’t, it’s go­ing to be re­ally awk­ward.

V. Fight­ing Moloch

So far I’ve talked about how co­op­er­at­ing in­stead of try­ing to beat some­one leads to per­son­ally benefi­cial out­comes. But there’s more at stake here than your next first date.

Mu­tual co­op­er­a­tion gets harder the more play­ers are in­volved. At the ex­treme, a pris­oner’s dilemma played by an en­tire so­ciety of­ten re­sults in ev­ery­one defect­ing against each other. As in the two per­son game, “defect” is any move by a player that nets them a small gain while im­pos­ing a large cost on oth­ers. Here are some ex­am­ples of defec­tions in so­ciety-wide games: send­ing a mar­ket­ing email, us­ing an­tibiotics, burn­ing some coal, call­ing some­one a Nazi on­line. The cor­re­spond­ing out­comes: per­va­sive spam, drug re­sis­tance, global warm­ing, Twit­ter. Th­ese out­comes are com­mon and tragic, so they’re known as tragedies of the com­mons.

There’s a view that failure to co­op­er­ate on mul­ti­player pris­oner’s dilem­mas is the great­est threat to our civ­i­liza­tion, or any civ­i­liza­tion for that mat­ter. This po­si­tion is best ar­tic­u­lated by Scott Alexan­der, who gave it a name: Moloch.

Moloch is why, when food is scarce, the an­i­mals (and hu­mans) that breed and kill most effi­ciently out­com­pete and de­stroy those that don’t. Moloch is why gov­ern­ments race to the bot­tom and provide cor­po­rate welfare. Moloch is the force be­hind arms races, en­vi­ron­men­tal de­struc­tion, and click­bait – com­pe­ti­tions that leave ev­ery sin­gle par­ti­ci­pant worse off.

Hu­man­ity cur­rently en­joys a mo­ment where the re­sources available to us ex­ceed our abil­ity to ex­ploit them. We can af­ford to en­gage in ac­tivi­ties that aren’t part of a ruth­less com­pe­ti­tion for re­sources: art, leisure, blog­ging. But once our ca­pac­ity for ex­ploita­tion in­creases – for ex­am­ple with the ad­vent of smarter-than-hu­man AI – art, leisure and blog­ging will be­come un­af­ford­able lux­u­ries.

Scott offers an es­cape: tran­shu­man­ism. The goal is to cre­ate some­thing or some­one that shares our val­ues, nd is so strong that it doesn’t have to sac­ri­fice those val­ues for the sake of com­pe­ti­tion.

I know, I know, this sounds pretty in­sane. Whether one thinks that this plan is fea­si­ble or not de­pends on many things, like one’s ge­o­graphic dis­tance from the Bay Area. But here’s the fun part – it’s a great way to fight Moloch even if it doesn’t work.

Imag­ine if we were try­ing to de­sign a com­mu­nity of peo­ple de­voted to co­op­er­a­tion, based on ev­ery­thing we learned about com­pet­i­tive and co­op­er­a­tive games. How should we ap­proach this?

We would build a com­mu­nity ded­i­cated to cre­at­ing some­thing new, a freshly baked pie. It would have to be a long-term pro­ject. It would have an im­por­tant and purely col­lec­tive re­ward at stake, like pro­tect­ing against a com­mon tragedy. It would in­volve a bunch of weirdos.

It would be some­thing like tran­shu­man­ism.

Tran­shu­man­ism in­her­ently cre­ates a co­op­er­a­tive cul­ture among those in­volved in it. The pur­suit of an out­landish goal in the far fu­ture, like friendly AI, cry­on­ics, cur­ing ag­ing, or has­ten­ing the sin­gu­lar­ity, is a re­mark­able way to turn nat­u­rally un­co­op­er­a­tive geeks into a col­lec­tive.

Does that make tran­shu­man­ism sound like a re­li­gion? The two main faults of re­li­gions are that they turn their fol­low­ers to vi­o­lence against the out­group, and that they un­tether their fol­low­ers from re­al­ity. En­courag­ing their fol­low­ers to co­op­er­ate and to think long-term is over­all a pos­i­tive as­pect of re­li­gions. Tran­shu­man­ists try to be at­tuned to the phys­i­cal and tech­nolog­i­cal re­al­ity, and the in­group of tran­shu­man­ism is the en­tire hu­man species. As far as re­li­gions go, it gives you most of the good stuff with lit­tle of the bad.

Of course, it’s hard to join a com­mu­nity you don’t be­lieve in just for the benefit of a co­op­er­a­tive cul­ture. There’s an­other way to achieve the same goal: cre­ate that cul­ture your­self. Ul­ti­mately, “cul­ture” is just a set of norms that peo­ple fol­low. You don’t need a com­mu­nity to start liv­ing by those norms your­self, and watch them spread to those around you.

Whichever game you’re play­ing, lead with co­op­er­a­tion and play tit-for-tat. Co­op­er­ate at times even when the other per­son seems to defect, just in case. Be hon­est and rad­i­cally trans­par­ent to re­duce the cost of in­ter­act­ing with you. Pur­sue weird in­ter­ests and goals. Write hon­estly about your weird in­ter­ests and goals, and pub­lish them for free on­line. Don’t be a dick. Deal with ev­ery per­son as if you’re go­ing to be play­ing re­peated games with them for the next 10,000 years.

If the tran­shu­man­ists get their way, it may ac­tu­ally hap­pen.