The Tale of Alice Almost: Strategies for Dealing With Pretty Good People

Sup­pose you value some virtue V and you want to en­courage peo­ple to be bet­ter at it. Sup­pose also you are some­thing of a “thought leader” or “pub­lic in­tel­lec­tual” — you have some abil­ity to in­fluence the cul­ture around you through speech or writ­ing.

Sup­pose Alice Al­most is much more V-vir­tu­ous than the av­er­age per­son — say, she’s in the top one per­cent of the pop­u­la­tion at the prac­tice of V. But she’s still ex­hibited some clear-cut failures of V. She’s al­most V-vir­tu­ous, but not quite.

How should you en­gage with Alice in dis­course, and how should you talk about Alice, if your goal is to get peo­ple to be more V-vir­tu­ous?

Well, it de­pends on what your spe­cific goal is.

Rais­ing the Global Median

If your goal is to raise the gen­eral pop­u­la­tion’s me­dian Vlevel, for in­stance, if V is “un­der­stand­ing of how vac­cines work” and your goal is to in­crease the pro­por­tion of peo­ple who vac­ci­nate their chil­dren, you want to sup­port Alice straight­for­wardly.

Alice is way above the me­dian V level. It would be great if peo­ple be­came more like Alice. If Alice is a pop­u­lar com­mu­ni­ca­tor, sig­nal-boost­ing Alice will be more likely to help rather than harm your cause.

For in­stance, sup­pose Alice makes a post tel­ling par­ents to vac­ci­nate their kids, but she gets a minor fact wrong along the way. It’s still OK to quote or ex­cerpt the true part of her post ap­prov­ingly, or to praise her for com­ing out in fa­vor of vac­cines.

Even spread­ing the post with the in­cor­rect state­ment in­cluded, while it’s definitely sub­op­ti­mal for the cause of in­creas­ing the av­er­age per­son’s un­der­stand­ing of vac­cines, is prob­a­bly net pos­i­tive, rather than net nega­tive.

Rais­ing the Me­dian Among The Virtuous

What if, in­stead, you’re try­ing to pro­mote V among a small sub-com­mu­nity who ex­cel at it? Say, the top 1% of the pop­u­la­tion in terms of V-virtue?

You might do this if your goal only re­quires a small num­ber of peo­ple to prac­tice ex­cep­tional virtue. For in­stance, to have an effec­tive vol­un­teer mil­i­tary doesn’t re­quire all Amer­i­cans to ex­hibit the virtues of a good sol­dier, just the ones who sign up for mil­i­tary ser­vice.

Now, within the com­mu­nity you’re try­ing to in­fluence, Alice Al­most isn’t way above av­er­age any more. Alice is av­er­age.

That means, you want to push peo­ple, in­clud­ing Alice, to be bet­ter than Alice is to­day. Sure, Alice is already pretty V-vir­tu­ous com­pared to the gen­eral pop­u­la­tion, but by the com­mu­nity’s stan­dards, the gen­eral pop­u­la­tion is pa­thetic.

In this sce­nario, it makes sense to crit­i­cize Alice pri­vately if you have a per­sonal re­la­tion­ship with her. It also makes sense to, at least some­times, pub­li­cly point out how the Alice Al­mosts of the com­mu­nity are fal­ling short of the ideal of V. (Prob­a­bly with­out nam­ing names, un­less Alice is already a fa­mous pub­lic figure.)

Ad­di­tion­ally, it makes sense to al­low Alice to bear the usual nega­tive con­se­quences of her ac­tions, and to pub­li­cly ar­gue against any­one try­ing to shield her from nor­mal con­se­quences. For in­stance, if peo­ple who ex­hibit Alice-like failures of V are rou­tinely fired from their jobs in your com­mu­nity, then if Alice gets fired, and her sup­port­ers get out­raged about it, then it makes sense for you to ar­gue that Alice de­served to be fired.

It does not make sense here to ex­press out­rage at Alice’s be­hav­ior, or to “pun­ish” her as though she had com­mit­ted a com­mu­nity norm vi­o­la­tion. Alice is nor­mal — that means that be­hav­ior like Alice’s hap­pens all the time, and that the com­mu­nity does not cur­rently have effec­tive, re­li­ably en­forced norms against be­hav­ior like hers.

Now, maybe the com­mu­nity should have stronger norms against her be­hav­ior! But you have to ex­plic­itly make the case for that. If you go around say­ing “Alice should be jailed be­cause she did X”, and X isn’t ille­gal un­der cur­rent law, then you are wrong. You first have to ar­gue that X should be ille­gal.

If Alice’s failures of V-virtue are typ­i­cal, then you do want to com­mu­ni­cate the mes­sage that peo­ple should prac­tice V more than Alice does. But this will be news to your au­di­ence, not com­mon knowl­edge, since many of them are no bet­ter than Alice. To com­mu­ni­cate effec­tively, you’ll have to take a tone of ed­u­cat­ing or shar­ing in­for­ma­tion: “Alice Al­most, a well-known mem­ber of our com­mu­nity, just did X. Many of us do X, in fact. But X is not good enough. We shouldn’t con­sider X okay any more. Here’s why.”

En­forc­ing Com­mu­nity Norms

What if Alice is in­side the com­mu­nity of top-1%-V-virtue you care about, but no­tice­ably worse than av­er­age at V or vi­o­lat­ing com­mu­nity stan­dards for V?

That’s an easy case. En­force the norms! That’s what they’re there for!

Con­tin­u­ing to en­force the usual penalties against failures of V, and mak­ing it com­mon knowl­edge that you do so, and sup­port oth­ers who en­force penalties, keeps the “floor” of V in your com­mu­nity from fal­ling, ei­ther by de­ter­rence or ex­pul­sion or both.

In terms of tone, it now makes sense for you to com­mu­ni­cate in a more “judg­men­tal” way, be­cause it’s com­mon knowl­edge that Alice did wrong. You can say some­thing like “Alice did X. As you know, X is un­ac­cept­able/​for­bid­den/​sub­stan­dard in our com­mu­nity. There­fore, we will be pe­nal­iz­ing her in such-and-such a way, ac­cord­ing to our well-known, es­tab­lished tra­di­tions/​code/​policy.”

Splin­ter­ing off a “Rem­nant”

The pre­vi­ous three cases treated the bound­aries of your com­mu­nity as static. What if we made them dy­namic in­stead?

Sup­pose you’re not happy with the stan­dard of V-virtue of “the top 1% of the pop­u­la­tion.” You want to cre­ate a sub­com­mu­nity with an even higher stan­dard — let’s say, draw­ing from the top 0.1% of the pop­u­la­tion.

You might do this, for in­stance, if V is “de­gree of al­ign­ment/​agree­ment with a policy agenda”, and you’re not mak­ing any progress with dis­course/​col­lab­o­ra­tion be­tween peo­ple who are only mostly al­igned with your agenda, so you want to form a smaller task force com­posed of a core of peo­ple who are hy­per-al­igned.

In that case, Alice Al­most is nor­mal for your cur­rent com­mu­nity, but she’s no­tably in­fe­rior in V-virtue com­pared to the stan­dards of the splin­ter com­mu­nity you want to form.

Here, not only do you want to pub­li­cly crit­i­cize ac­tions like Alice’s, but you even want to spend most of your time talk­ing about how the Alice Al­mosts of the world fall short of the ideal V, as you ad­vo­cate for the ex­is­tence of your splin­ter group. You want to reach out to the peo­ple who are bet­ter at V than Alice, even if they don’t know it them­selves, and ex­plain to them what the differ­ence be­tween top-1% V-virtue and top 0.1% V-virtue looks like, and why that differ­ence mat­ters. You’re, in effect, em­pow­er­ing and en­courag­ing them to no­tice that they’re not Alice’s peers any more, they’ve lev­eled up be­yond her, and they don’t have to make ex­cuses for her any more.

Just like in the case where Alice is a typ­i­cal mem­ber of your com­mu­nity and you want to push your com­mu­nity to do bet­ter, your crit­i­cisms of Alice will be news to much of your au­di­ence, so you have to take an “ed­u­ca­tional/​in­for­ma­tional” tone. Even the peo­ple in the top 0.1% “rem­nant” may not be aware yet that there’s any­thing wrong with Alice’s be­hav­ior.

How­ever, you’re now speak­ing pri­mar­ily to the top 0.1%, not the top 1%, so you can now af­ford to be some­what more in­sult­ing to­wards Alice. You’re try­ing to cre­ate norms for a fu­ture com­mu­nity in which Alice’s be­hav­ior will be con­sid­ered un­ac­cept­able/​sub­stan­dard, so you can start to in­tro­duce the frame where Alice-like be­hav­ior is “im­moral”, “in­com­pe­tent”, “out­ra­geous”, or oth­er­wise failing to meet a rea­son­able per­son’s min­i­mum ex­pec­ta­tions.

Ex­pand­ing Com­mu­nity Membership

Let’s say you’re do­ing just the op­po­site. You think your com­mu­nity is too se­lec­tive. You want to ex­pand its bound­aries to, say, a group drawn from the top 10% of the pop­u­la­tion in V-virtue. Your goals may re­quire you to raise the V-lev­els of a wider au­di­ence than you’d been speak­ing to be­fore.

In this case, you’re more or less in the same po­si­tion as in the first case where you’re just try­ing to raise the global me­dian. You should sup­port Alice Al­most (as much as pos­si­ble with­out your­self imi­tat­ing or com­pound­ing her failures), laud her as a role model, and not make a big pub­lic deal about the fact that she falls short of the ideal; most of the peo­ple you’re try­ing to reach fall short even farther.

What if Alice is Dilut­ing Com­mu­nity Values?

Now, what if Alice Al­most is the one try­ing to ex­pand com­mu­nity mem­ber­ship to in­clude peo­ple lower in V-virtue … and you don’t agree with that?

Now, Alice is your op­po­nent.

In all the pre­vi­ous cases, the worst Alice did was drag down the com­mu­nity’s me­dian V level, ei­ther di­rectly or by be­ing a role model for oth­ers. But we had no rea­son to sup­pose she was op­ti­miz­ing for low­er­ing the me­dian V level of the com­mu­nity. Once Alice is try­ing to “pop­u­larize” or “ex­pand” the com­mu­nity, that changes. She’s ac­tively try­ing to lower me­dian V in your com­mu­nity — that is, she’s op­ti­miz­ing for the op­po­site of what you want.

This means that, not only should you crit­i­cize Alice, en­force ex­ist­ing com­mu­nity norms that for­bid her be­hav­ior, and ar­gue that com­mu­nity stan­dards should be­come stric­ter against Alice-like, 1%-level failures of V-virtue, but you should also op­ti­mize against Alice gain­ing more power gen­er­ally.

(But what if Alice suc­ceeds in ex­pand­ing the com­mu­nity size 10x and rais­ing the me­dian V level within the larger com­mu­nity by 10x or more, such that the me­dian V level still in­creases from where it is now? Wouldn’t Alice’s goals be al­igned with your goals then? Yeah, but we can as­sume we’re in a regime where in­creas­ing V lev­els is very hard — a rea­son­able as­sump­tion if you think about the track record of try­ing to teach ethics or in­still virtue in large num­bers of peo­ple — so such a huge per­sua­sive/​rhetor­i­cal win is un­likely.)

Alice, for her part, will see you as op­ti­miz­ing against her goals (she wants to grow the com­mu­nity and you want to pre­vent that) so she’ll have rea­son to op­ti­mize gen­er­ally against you gain­ing more power.

Alice Al­most and you are now in a zero-sum game. You are di­rect op­po­nents, even though both of you are, com­pared to the gen­eral pop­u­la­tion, both very high in V-virtue.

Alice Al­most in this sce­nario is a So­ciopath, in the Chap­man sense — she’s try­ing to ex­pand and dilute the sub­cul­ture. And So­ciopaths are not just a lit­tle bad for the sur­vival of the sub­cul­ture, they are an ex­is­ten­tial threat to it, even though they are only a lit­tle weaker in the defin­ing skills/​virtues of the sub­cul­ture than the Geeks who founded it. In the long run, it’s not about where you are, it’s where you’re aiming, and the So­ciopaths are aiming down.

Of course, get­ting locked into a zero-sum game is bad if you can avoid it. Misi­den­ti­fy­ing Alice as a So­ciopath when she isn’t, or miss­ing an op­por­tu­nity to di­alogue with her and come to agree­ment about how big the com­mu­nity re­ally needs to be, is costly. You don’t want to be hasty or para­noid in read­ing peo­ple as op­po­nents. But there’s a very, very big differ­ence be­tween how you deal with some­one who just hap­pened to do some­thing that blocked your goal, and how you deal with some­one who is per­sis­tently op­ti­miz­ing against your goal.