Liars and Outliers (14 page)

Most punishments are even less extreme. We may still hang around with some friends, but not rely on them as much or not tell them our intimate secrets. We may still invite
those
relatives to the family's holiday party, but not talk to them much.

Shame is a common
reputational punishment, and—as a result—an important social emotion. Much of this is the informal kind of shaming we've all experienced amongst our friends and colleagues. More formal examples include police blotter reports, IRS quarterly listings of Americans who renounce their citizenship, public disclosure of
excessive CEO pay
, televised arrests, deadbeat dads in the media, and TV shows like
America's Most Wanted
. Of course, these all have a technological component, and some might be more properly put into the category of institutional pressure.

Informal punishments are so common we can miss them if we're not paying attention. An employee who has frequent conflicts with his colleagues may find himself shuffled into a dead-end position, or assigned the graveyard shift. A husband spied flirting with the housekeeper may be told by his friends and family that such behavior is unacceptable. An entertainer who espouses unpopular political positions may encounter a dip in his popularity at the box office as moviegoers boycott his films.

Informal punishments
are common in our society. They're certainly prevalent through childhood, from early play amongst small children through social ostracism in school. Some groups define themselves through the exclusion of others.

Remember the Bad Apple Effect from the previous chapter? As you might expect, the effects of those
bad apples diminishes
if punishment is threatened.

There is an old idea that punishment can be transferred from one person to another. In many traditions, God punishes a person's relatives in addition to punishing the transgressor.
21
In some societies,
if Alice kills Bob
, one of Bob's relatives is allowed to kill one of Alice's relatives in retribution. The
Nazis instituted this
as government policy; it was called “
sippenhaft
.” This practice is a form of societal pressure. If Alice is considering defecting from some group norm—killing another person, committing adultery, whatever—she not only has to worry about human or divine retribution against her personally, but also retribution against members of her family. And threatening Alice's family should she defect both raises her perceived cost of defection and enlists her family members in persuading her not to defect in the first place. The Israeli government's current practice of bulldozing the homes of suicide bombers' families is an example. Of course, sometimes this goes very badly. Think of “honor killings” of rape victims, or blood feuds in
various cultures
throughout history, like the
Hatfields versus the McCoys
.

There's a variant of the Hawk-Dove game that demonstrates how
reputation can solve
a societal dilemma. It's designed so doves are more likely to interact with doves. When this happens, hawks can be isolated and their numbers reduced.

Compared with the basic Hawk-Dove game, cooperation turns out to be an even better strategy. Stable populations have even fewer hawks because doves, by preferring to interact with other doves, can effectively isolate them. Left to fight amongst themselves, hawks tend to kill each other off. Returning to human society, we are at our most cooperative when we seek out other cooperative people and avoid those who would take advantage of us. We learned this in Chapter 3 when we looked at the evolution of cooperation: cooperators do better when they can recognize each other. Reputation not only encourages cooperation, but also marginalizes defectors to the point where there ends up being fewer of them to deal with.

This is important. We've been talking about societal dilemmas as if they're always decisions to either cooperate or defect. In the real world, we often have a choice of people with whom to interact. We don't walk into stores randomly, wondering if the merchant will cheat us. We only walk into stores where we believe the merchant will not cheat us. Instead of defecting and cheating the merchant as punishment, we prefer to shop elsewhere.

In Chapter 3, we learned that two things are required for cooperation: reciprocal altruism and a calculating intelligence. Morals and reputation, the two things I've been calling our primitive toolbox of social pressures, provide that reciprocal altruism. Even so, reputational societal pressure can fail in many ways.

Defectors take steps to hide facts that can harm their reputation, or manipulate facts to help their reputation
. Recall that in the mid-1970s,
John Wayne Gacy
managed to rape and kill 33 young men. All the while, his Chicago neighbors and colleagues on civic and charitable committees never suspected “Pogo the Clown” Gacy was involved in any work more diabolical than entertaining children for good causes. In the UK,
Dr. Harold Shipman
had a similar story. Described as “a pillar of the community” by his neighbors, he killed at least 250 people, mostly elderly widows, before he was caught. Most examples are less extreme. A politician might go to church and publicly pray, to encourage people to think he's honest: a whited sepulchre. An American trekking through Europe might sew a Canadian flag on his backpack.

Confidence tricksters spend a lot of time manipulating reputation signals. They employ all sorts of props, façades, and other actors—shills—to convince their victims that they have a good reputation by appearing authentic, building confidence, and encouraging trust. Corporations and political candidates both do similar things; they use paid supporters to deliberately spread artificial reputational information about them. This is becoming even more prevalent and effective on the Internet. Hired hands write fake blog posts, blog comments, tweets, Facebook comments, and so on. Scammers on eBay create fake feedback, giving themselves a better reputation. There are even companies that will give you fake Facebook friends, making you seem more popular with attractive people than you actually are.

Defectors try to minimize the effects of their bad reputation
. People get new friends, move to another city, or—in extreme cases—change their names, get plastic surgery, or steal someone else's identity. Philip Morris renamed itself Altria, because who would want to buy their Kraft Mac and Cheese from a cigarette company? ValuJet, its brand ruined after Flight 592 crashed in the Everglades in 1996, now operates as AirTran Airways. Blackwater, the defense contractor notorious for numerous Iraq war abuses, became Xe Services and then Academi. The School of the Americas, implicated in training many human rights–abusing military staff in Latin America, rebranded itself as the Western Hemisphere Institute for Security Cooperation.

Corporations work to minimize the effects of negative reputation on their brands through advertising and public relations. Multinational food and consumer product companies like Unilever and Procter & Gamble deliberately downplay their corporate brand and focus attention on their sub-brands. There are a lot of marketing reasons to engage in sub-branding, but the security thinking is that if there is a problem with one of their brands, the negative publicity won't spread across the company's entire product line.

Some people simply don't care about reputation
. Like our individual morals, our individual concern about reputation varies—from person to person as well as from situation to situation. Some of us care a lot; others, not so much. Of course, this is contextual. We all have different reputations in different groups with respect to different personal attributes.

Some people end up with the wrong reputation
. Even if someone does nothing wrong, there's no guarantee that his reputation is accurate. Untrue stories can circulate by mistake. Someone else might lie to give him a bad reputation. We all know people who have reputations they don't deserve, both good and bad.

Defectors band together in subgroups that have different reputational rules
. Gang members thrive in groups. Sure, they have a terrible reputation in the broader community, but they care primarily about their reputation within their gangs. This dynamic is also true for defectors who have a different moral system from the dominant culture: a lone pot smoker in a pot-free community is going to have a lot harder time than one who finds other pot smokers in the vicinity. His friends will help him defect. In effect, he will choose to cooperate with the smaller society of defectors, rather than with the pot-abstaining majority. The same is true for those worshipping in secret out of fear: early Christians in the Roman Empire, pagans afterwards, Jews in post-expulsion Spain, devout Russian Orthodox in the former Soviet Union. We'll talk about this more in Chapters 11 and 12.

The value of defecting might be worth the reputational damage
. Maybe it's a single large transaction, and the merchant is willing to sacrifice her reputation for the money. Or maybe it's a situation where the merchant can outrun his reputation.
22
We've all heard stories of home remodeling contractors that score a big contract, and then either don't do the work or do a quick, shoddy job, and disappear with the money. “Fly by night,” it's called. They've made the risk trade-off and decided their reputation wasn't that valuable. If they're career scammers, a big payoff may even enhance their reputation among their fellow scammers. A restaurant owner in a tourist area could serve lousy food, confident that the reputational damage matters less when there's no repeat clientele. A corporate CEO might decide his company's ability to repair reputational damage allows it to get away with misdeeds he wouldn't have authorized if he didn't have such an effective public relations department.

The most important reason reputational pressure starts to fail is that groups get too
large.
23
Assisted by technology, reputational pressure can scale globally. Think of the reputations of public figures and celebrities, companies and brands, or individuals on the Internet. Think of eBay's reputation system, review sites like Yelp, or how we can make friends on shared-interest websites. Think of the FBI's criminal databases, the information about you kept by credit bureaus, or Google's database of your interests. Think of passports, driver's licenses, or employee badges. These are all reputational systems, and all serve to apply reputational pressure in different risk trade-offs.

But these systems can have all sorts of inaccuracies. What we know about celebrities, corporations, and people in faraway places doesn't always match reality. It's not only the natural errors that creep into any large-scale process, it's that these systems can be manipulated and the technologies used to support them can be attacked. In order for reputation to scale, we need to trust these reputational systems, but sometimes that trust is not well-founded. We'll talk about this more in Chapter 10.

Reputational pressure works best within a group of people who know each other: a group of friends or coworkers in an office, compared to a bunch of strangers on a bus or a city full of people. Neighbors are good at
settling disputes
; people who don't live so close to each other are less good at it.

However, once the group size grows larger and the social ties between people weaken, reputation alone doesn't cut it.

Commenting on Hardin's original Tragedy of the Commons paper, psychologist Julian Edney wrote that “the upper limit for a simple, self-contained, sustaining, well-functioning commons may be as low as 150 people.”
24

Eleven years later,
Dunbar wrote
:

The Hutterites, a group of contemporary North American religious fundamentalists who live and farm communally, regard 150 as the maximum size for their communities. What is interesting is the reason they give for splitting communities at this size. They find that when there are more than about 150 individuals, they cannot control the behaviour of members by peer pressure alone.