Bryan Caplan  

The Identity of Shame

Henderson on Minimum Wage for ... Supply and demand-side stagnat...
Every large, unselective group includes some villains.  Say whatever you like about the average moral caliber of Christians, atheists, Democrats, Republicans, plumbers, comic book fans, or Albanians.  The fact remains that each of these groups contains some awful people.  While this isn't logically necessary, it is an iron statistical law.  If X has more than a few dozen members, and you can join group X by (a) being born into X, or (b) saying "I'm an X," then X will have some unsavory characters.

When you identify with a large, unselective group, you expose yourself to two dangers.  First, some of the villains in your group may take villainous actions that make you look bad.  Yet on reflection, that's a minor concern: Yes, the bad members of your group make you look bad, but the good members of your group make you look good. 

The second danger is more severe.  Once you identify with any large, unselective group, you will be regularly tempted to commit the villainous act of standing up for your groups' villains.  When they do wrong - as they inevitably will - your impulse will be to ignore, minimize, or justify their misdeeds.  To quote the underrated 8mm, "If you dance with the devil, the devil don't change. The devil changes you."

Look at any large, unselective group you don't identify with.  You see them clearly, do you not?  Some of its members are plainly bad people.  But getting the regular members to unequivocally condemn their bad members is almost impossible.  Evolution has honed their myside bias for millions of years, and that's not about to change.

Fortunately, there are relatively easy ways to avoid these temptations in the first place - to save yourself from all shameful identities.  Namely: Never identify with large, unselective groups!  Instead, restrict your identity to groups that are small, selective, or both.

The nuclear family is the classic small, unselective group; no other group is more deeply founded in human nature.  Fortunately, the moral risk of being part of such a family is usually small.  Even if you have five kids, there a good chance that none of them will be horrendous people.  Circles of friends are the classic small, selective group.  Pick your friends carefully, and you probably won't end up an apologist for evil.

Large, selective groups are riskier.  In principle, they can evade statistical villainy by carefully vetting and excommunicating questionable members.  Unfortunately, myside bias tends to gut the excommunication process.  Fringe movements like Jehovah's Witnesses expel members all the time, but not the Catholic Church. 

The best way to guard against this laxity is to define your large, selective groups in purely intellectual terms.  Identify with liberalism or conservatism, not liberals or conservatives.  This is the kernel of truth behind the "No True Scotsman" fallacy.  Once you insist that "No true libertarian believes in immigration restrictions," you'll feel little temptation to ignore, minimize, or justify libertarians who believe in immigration restrictions.  And this is precisely how you should feel. 

Can anyone really live up to my puritanical advice?  Yes.  Take me.  I identify with my nuclear family, with my friends, and with a bunch of ideas.  I neither need nor want any broader identity.  I was born in America to a Democratic Catholic mother and a Republican Jewish father, but none of these facts define me.  When Americans, Democrats, Republicans, Catholics, and Jews commit misdeeds - as they regularly do - I feel no shame and offer no excuses.  Why?  Because I'm not with them.

COMMENTS (14 to date)
DangerMan writes:

Yes, but the disadvantage you suffer is... they are also not with you.

Joe Marier writes:

I think you're conflating identity and withness.

Doug writes:

This comes down to the halo effect. Humans are much too quick to judge a person's entire character based on a single observed property. If I subscribe to Catholic principles than it certainly makes sense to slightly up-bias my assessment upon learning that someone else is a fellow Catholic. What doesn't make sense, but our monkey brains inevitably succumb to, is to give that fellow a positive evaluation along all dimensions simply due to one positive attribute.

It's like seeing the pictures of Hitler playing with his dog or entertaining children. Every time I see one my mind temporarily freezes up. People who console crying toddlers are good, Hitler's evil, this makes no sense. One even feels twinges of offense at the person linking the photos, as if a picture of Hitler playing with his dog and wearing bunny ears could be a counter-weight to murdering millions of human beings in any rational framework.

But monkey brain doesn't work that way. It only thinks in terms of totally good people and totally bad people. It doesn't recognize that people vary along many attributes, and even the worst scum will likely have many admirable qualities. Monsters almost always compartmentalize the things they do, but at the end of the day they're still monsters.

Hugh writes:
This is the kernel of truth behind the "No True Scotsman" fallacy.

Huh? It is a fallacy because there is no kernel of truth behind it.

I understand that being able to wave away inconvenient occurrences (a libertarian who doesn't believe in Open Borders(?)) allows you to avoid confronting the inconvenience, but that's all.

Lee Kelly writes:


There is a kernel of truth to the slippery slope fallacy--namely, slippery slopes.

Jameson writes:

Every once in a while, Caplan writes a really poorly thought out post.

If "myside bias" is something honed by millions of years of evolution, what on earth makes you think that the desire to belong to a group isn't also ingrained in us? You might call your advice puritanical. Others will just call it weird. So that's one thing.

Another thing you're missing is that each time someone in your group does really horrible things, that can be a time either for humility or for true heroism. Christians, for example, haven't always stood up to defend other Christians, but often to apologize for them or to actively oppose them, in the name of that very faith they hold in common.

In general, I think we should consider all other human beings a part of our identity. We do no good by dismissing the evil that others do as "not me." So, again, while you might call your advice puritanical, others might call it differently. Self-centered, perhaps?

Richard writes:

To identify mainly with your nuclear family is to live in a world of petty concerns and vendettas. It is the morality of the mafia.

To widen one's identity to include all Christians, all Jews, all Americans, or indeed all humans, is the morality of progress, which culminated in the Enlightenment.

Bryan dresses up his prescription in what sounds like sophisticated analysis, but it is really the formula of primitivism. People who are happy, and who prosper, broaden their circle of identity and regard as widely as they can.

RPLong writes:

Richard, what?

To identify mainly with your nuclear family is to live in a world of petty concerns and vendettas. It is the morality of the mafia.
Would you like to spend Thanksgiving at my house this year?

Regarding group affiliation, it is a mechanism that makes human cooperation possible. I think every group has a "join me" phase and a "No Homer's Club" phase. No one ever faults a group for its "join me" phase, which is after all an invitation to cooperate with other people and coordinate efforts around a single purpose. What's not to like?

It's that "No Homer's Club" phase that makes group affiliation bad, the phase in which "bad" members are ostracized or expelled, the phase in which minor differences are intolerantly stamped out, and "unworthy" non-members are kept out or even attacked.

My own solution to this problem is to only associate with groups that appear to be in a "join me" phase. Once that phase ends, it's time to move on. People cooperating to serve a (good) common goal? I can get behind that. People being jerks and making non-members the brunt of their jerkiness? Not for me.

Pat writes:

If you can resist the temptation to be part of a group, you can resist the temptation of standing up for villians in your group.

It seems silly and lonely to avoid associating with people simply because you worry that someone in that group might be bad and you might want to defend them.

deluks917 writes:

Slippery slope clearly is not a fallacy in general. An obvious and important example is judicial precedent. How a court rules a case has a clear casual effect on the outcome of later cases. In any situation where "natural law" is important slippery slope is nothing like a fallacy.

In debates over policies the governments previous actions feature heavily. For example in Europe a very common talking point for those favor more speech restrictions is "we banned Holocaust denial and nothing bad happened." Removing legal protections often has a casual effect of making it easier to remove even more protections.

The libertarians aren't the only ones who often are right when they make slippery slope arguments, the reactionaries usually are too. As long as gay marriag eis kept illegal it really will be harder for polyamorous marriages to eb legalized. Once a group excepts the logic of "classic" MtF and FtM transexuals it is hard for them to deny people who identify as other genders (genderqueer, etc).

I personally long for a society where everypne's gender expression is not questioned. And I strongly support multiple marriages and especially respect for the openly polyamorous. But I am not goign to pretend the reactionaries are "Crazy" when they say legalizing gay mariages will help push a country further down the "liberal" path. Its just that I think its an actively good thing we go down that path.

ThomasH writes:

I have a different strategy from Caplan's: I identify only with with groups that I know have a large percentage of people I don't agree with.

As a Catholic, I'm "liberal" on doctrine and social matters and "conservative on the "economic" teachings, approximately the opposite of the US bishops. As a Liberal, I am very conservative about economic regulation and taxation of income (believing in progressive taxation of consumption). And as a patriotic American I think most US foreign policy stinks.

I realize this still leaves me more prone to forgive Harry Ried's sins than Mitch McConnell's and Obama's more than Putin's, but at least I know where (some of) my biases lie.

Jeff writes:
The best way to guard against this laxity is to define your large, selective groups in purely intellectual terms. Identify with liberalism or conservatism, not liberals or conservatives.

Identifying too closely with certain ideologies has its own, very well-known problems, some of which have been discussed on this blog before, so I'm not sure this is such a great solution.

The other problem here is that if you want any of your ideas to be implemented, you need political power, and groups of people are much better at attaining and retaining political power than individuals, so if you and everyone else who thinks like you refuses to join a group because of the dangers of group social dynamics, they'll completely surrender the field to those who have no such qualms. This is, in short, not a recipe for furthering your ideas.

BZ writes:

I think Dr. Caplan's point is that he'd rather protect perception of particular ideas from the taint of the people proclaiming them than tie himself up in a movement of people. Not my own choice, but I see his point.

There is also quite a bit of confusion on what the Slippery Slope fallacy is, or perhaps even what a logical fallacy is. An informal logical fallacy is a proposed relationship between a premise and a conclusion that does not always, everywhere, and necessarily hold. Theoretically, there are an infinite number of logical fallacies -- however, we give names to the most common ones. Just because one can think of some incidental causal relationships does not mean that there is a logical necessary relationship. For example: In practice, a broad judicial Precedent almost always affects related case law. However, the argument "There was precedent X, therefore there will be related consequences Y and Z" would indeed commit the Slippery Slope Fallacy, because it is not Necessarily true. It's truth depends on other premises, such as "Judges always and everywhere respect precedent."

MW29 writes:

I'm with you on this one. I have not examined the entire body of work, but I've noticed when you write things that could be construed as "anti-American", of course, you immediately get battered in the comments. This is a classic case of "if you're not with us, you're against us" and that has always sickened me. Keep laying the track out in front of you and moving forward. I enjoy your writings.

Comments for this entry have been closed
Return to top