Excerpts from a larger discussion about simulacra

Link post

I’ve been dis­cours­ing more pri­vately about the cor­rup­tion of dis­course lately, for rea­sons that I hope are ob­vi­ous at least in the ab­stract, but there’s one thing I did think was share­able. The con­text is an­other friend’s then-forth­com­ing blog post about the poli­ti­ciza­tion of cat­e­gory bound­aries.

In pri­vate com­mu­ni­ca­tion, quoted with per­mis­sion, Jes­sica Tay­lor wrote:

In a world where half the em­ploy­ees with bad jobs get good ti­tles, aren’t their ti­tles pre­dic­tively im­por­tant in that they pre­dict how likely they are to be hired by out­side com­pa­nies? Their like­li­hood of get­ting hired is, un­der these as­sump­tions, go­ing to be the same as that of as peo­ple with good jobs and good ti­tles, and higher than that of peo­ple with bad jobs and bad ti­tles. So, in terms of things like abil­ity to exit (and there­fore ne­go­ti­at­ing abil­ity), there are nat­u­ral clusters of “peo­ple with good ti­tles” and “peo­ple with bad ti­tles”. (Ti­tle is go­ing to have less effect on like­li­hood of get­ting a job than it did be­fore the bul­lshit ti­tles, but it still has a sig­nifi­cant effect)

The Bobs of the world, hav­ing ob­served that em­ploy­ees and Job-seek­ers pre­fer Jobs with high-pres­tige ti­tles, thought they were be­ing benev­olent by mak­ing more Jobs have the de­sired ti­tles.

Some­what re­lated to the pre­vi­ous point, bul­lshit ti­tles ac­tu­ally might end up be­ing in the in­ter­est of the peo­ple with bad jobs, in the sense that they might want oth­ers not to know their job is bad, and de­stroy­ing the lan­guage here makes ac­tual job qual­ity harder to in­fer from ti­tle. Peo­ple do of­ten have a sense that cov­er­ing up em­bar­rass­ing in­for­ma­tion about peo­ple is benev­olent to them. It doesn’t seem like the ar­gu­ment you have pre­sented di­rectly challenges this sense.

Re­lated, we have words for “rich” vs “poor”, which sim­ply name some­one’s po­si­tion in the so­cial sys­tem of money (similar to ti­tle), and we don’t have con­cise ways of talk­ing about the ex­tent to which their wealth re­flects ma­te­rial value that they have cre­ated, which money is at least par­tially about track­ing (but, is also about crony­ism, class in­ter­ests, and theft at the same time). But, “rich” and “poor” are un­de­ni­ably pre­dic­tively use­ful, even though track­ing value cre­ation is also im­por­tant.

There’s some dan­ger that un­crit­i­cally us­ing the lan­guage cor­re­spond­ing to the Schel­ling points cho­sen by an un­just equil­ibrium con­tributes to main­tain­ing that equil­ibrium, by mak­ing these Schel­ling points more nar­ra­tively salient; I think that’s more clear in the jobs ex­am­ple than the money ex­am­ple, but it ap­plies to both.

The bul­lshit ti­tle ex­am­ple quite re­minds me of Si­mu­lacra and Si­mu­la­tion, which I haven’t read yet. From Wikipe­dia:

Si­mu­lacra and Si­mu­la­tion delineates the sign-or­der into four stages:
1 The first stage is a faith­ful image/​copy, where we be­lieve, and it may even be cor­rect, that a sign is a “re­flec­tion of a profound re­al­ity” (pg 6), this is a good ap­pear­ance, in what Bau­drillard called “the sacra­men­tal or­der”.
2 The sec­ond stage is per­ver­sion of re­al­ity, this is where we come to be­lieve the sign to be an un­faith­ful copy, which “masks and de­na­tures” re­al­ity as an “evil ap­pear­ance—it is of the or­der of malefi­cence”. Here, signs and images do not faith­fully re­veal re­al­ity to us, but can hint at the ex­is­tence of an ob­scure re­al­ity which the sign it­self is in­ca­pable of en­cap­su­lat­ing.
3 The third stage masks the ab­sence of a profound re­al­ity, where the sign pre­tends to be a faith­ful copy, but it is a copy with no origi­nal. Signs and images claim to rep­re­sent some­thing real, but no rep­re­sen­ta­tion is tak­ing place and ar­bi­trary images are merely sug­gested as things which they have no re­la­tion­ship to. Bau­drillard calls this the “or­der of sor­cery”, a regime of se­man­tic alge­bra where all hu­man mean­ing is con­jured ar­tifi­cially to ap­pear as a refer­ence to the (in­creas­ingly) her­metic truth.
4 The fourth stage is pure simu­lacrum, in which the simu­lacrum has no re­la­tion­ship to any re­al­ity what­so­ever. Here, signs merely re­flect other signs and any claim to re­al­ity on the part of images or signs is only of the or­der of other such claims. This is a regime of to­tal equiv­alency, where cul­tural prod­ucts need no longer even pre­tend to be real in a naïve sense, be­cause the ex­pe­riences of con­sumers’ lives are so pre­dom­i­nantly ar­tifi­cial that even claims to re­al­ity are ex­pected to be phrased in ar­tifi­cial, “hy­per­real” terms. Any naïve pre­ten­sion to re­al­ity as such is per­ceived as bereft of crit­i­cal self-aware­ness, and thus as over­sen­ti­men­tal.

A pos­si­ble in­ter­pre­ta­tion here is that sig­nifiers origi­nally achieve mean­ing in so­cial sys­tems by cor­re­spond­ing with re­al­ity (stage 1), but once they’re used in a so­cial sys­tem, if the sys­tem doesn’t pro­tect it­self, lies will out­com­pete truth (stage 2); since so­cial sys­tems in­volve Schel­ling games, the lies can be im­por­tant pieces on the play­ing field even when no one ex­pects them to cor­re­spond with re­al­ity (stage 3), and even­tu­ally peo­ple just start treat­ing the state­ments as pieces on the game­board, not even as lies (stage 4). Thus, lan­guage is de­stroyed.

Stage 1 is hon­esty, stage 2 is lies, stage 3 is bul­lshit, stage 4 is pure power games.

I replied:

Let’s ap­ply this to the spe­cific ex­am­ple.

In world 1, com­pa­nies need su­per­vi­sors to co­or­di­nate pro­jects, pro­mote peo­ple who seem gen­er­ally good at things to those roles be­cause it’s im­por­tant for prof­ita­bil­ity to have smart con­scien­tious peo­ple in charge, and have differ­ent ti­tles for su­per­vi­sory and man­age­rial roles vs di­rect la­bor roles in or­der to keep track of who’s do­ing what. As a side effect, com­pa­nies hop­ing to hire some­one for a higher-pay­ing su­per­vi­sory role will fa­vor ap­pli­cants whose ti­tle re­flects that they’ve already (a) been se­lected for such a role by some­one with skin in the game, and (b) done some learn­ing on the job so they already know how to man­age. As an­other side effect, job ti­tle is used for ex­ter­nal so­cial sort­ing, since peo­ple on similar life tra­jec­to­ries have more in com­mon, peo­ple who want to ex­tract money will want to pay more at­ten­tion to peo­ple with higher wages and ex­pected life­time in­come, etc.

In world 2, com­pa­nies have started offer­ing man­age­rial ti­tles to em­ploy­ees as a perk so that they can benefit from the de­sir­able side effects, less­en­ing the ti­tle’s use­ful­ness for track­ing who’s do­ing what work, but pos­si­bly in­creas­ing its cor­re­la­tion with some of the side effects, since the good (i.e., effec­tive at pro­duc­ing the de­sired side effects) ti­tles go to the peo­ple who are most skil­led at play­ing the game. It’s com­mon ad­vice that one of the things you should ne­go­ti­ate if you’re an ear­ly­ish hire at a startup is job ti­tle, since a suffi­ciently im­pres­sive ti­tle will cre­ate path-de­pen­dency mak­ing it awk­ward not to make you a ma­jor ex­ec­u­tive if and when the startup suc­cess­fully grows.

In short, in world 2, the sys­tem is wire­head­ing it­self with re­spect to ti­tles, but in a way that comes with real re­source com­mit­ments, so peo­ple who can track the map and re­al­ity sep­a­rately, and play on both game­boards si­mul­ta­neously, can ex­tract things through ju­di­cious ac­qui­si­tion of ti­tles.

In world 3, the sys­tem starts us­ing ti­tles to wire­head its em­ploy­ees. Ti­tles like “Vice Pres­i­dent of Sort­ing” are use­less and played out in the in­dus­try, in­ter­view­ers know to ask what you ac­tu­ally did (and prob­a­bly just look at your body lan­guage, and maybe call around to get your rep­u­ta­tion, or just check what par­ties you’ve been to), but maybe there’s some con­no­ta­tive im­pres­sive­ness left in the term, and you feel bet­ter get­ting to play the im­prov game as a Vice Pres­i­dent rather than a La­borer. You’re given so­cial per­mis­sion to switch your in­ner class af­fili­a­tion and feel like a mem­ber of the man­age­rial class. Prob­a­bly mom and dad are im­pressed.

In world 4, some of the prac­tices from world 3 are left, and it’s al­most uni­ver­sally un­der­stood emo­tion­ally that they don’t re­fer to any­thing, but there’s noth­ing real to con­trast them with, so if you tell a story about your­self well enough, peo­ple will go along with it even though they know that all the “ev­i­dence” is mean­ingless. E.g. Trump man­ages to play a great busi­ness­man on TV, and this is (plus a start­ing en­dow­ment of money and some ba­sic pri­mate cun­ning) enough to start off his pres­i­den­tial run in the genre “suc­cess­ful busi­ness­man com­ing to clean up Wash­ing­ton.” Eliz­a­beth Holmes was also play­ing in world 4.

Note that as we progress through these wor­lds, the ti­tle be­comes less use­ful to peo­ple like [friend]. I think this needs to be made very ex­plicit for the ar­gu­ment to reg­ister to LessWrongers. The sort of per­son who can hold a bul­lshit job maybe does bet­ter in world 2 than in world 1, but [friend] doesn’t play that game, he wants to do work that mat­ters on the ob­ject level and be justly re­warded for it. (Though he’s cur­rently, un­der­stand­ably too dis­tracted by cul­tural forces threat­en­ing to de­stroy world 1 al­to­gether to fo­cus on his ob­ject-level work.)

If World 3 were to ar­rive uniformly it wouldn’t be very use­ful to any­one, but it doesn’t—it always ar­rives un­evenly, so that in the early stages while cyn­i­cal man­agers are still me­tab­o­liz­ing world 2 into world 3, peo­ple who can most savvily lev­er­age class priv­ilege into bul­lshit jobs know which ti­tles to stay away from, and in the late stages out­right con artists bring about world 4 when enough of the power land­scape has been me­tab­o­lized into world 3.

This is no­tably similar to the stages of a fi­nan­cial spec­u­la­tive bub­ble, though I think there are some differ­ences that would be worth mod­el­ing.

Jes­sica’s re­ply:

This all seems right, thanks for the ad­di­tional ex­pla­na­tion. The naive ver­sion of the Bau­drillard for­mu­la­tion (which is naive since I haven’t read the book) un­for­tu­nately as­sumes that wor­lds are uniform, and which world ev­ery­one is in is mu­tual knowl­edge, when ac­tu­ally some peo­ple are much more savvy than oth­ers (in terms of both know­ing what game is be­ing played and skill at play­ing the game), ex­ploit­ing the la­bor of peo­ple who think they are in world 1 when ac­tual ma­te­rial/​in­for­ma­tional work is nec­es­sary, or when im­prov of such is called for.


Re­lated: There is a war, The Scams are Win­ning, Anatomy of a Bub­ble, On the con­struc­tion of bea­cons, Ac­tors and Scribes, words and deeds, Naive episte­mol­ogy, savvy epistemology