by Hugo Mercier
rather not inherit from one’s elders, from the belief in witchcraft
to the practice of foot binding, these harmful customs are the
exception. On the whole, most culturally acquired beliefs are
sensible enough. Every day, we engage in culturally influenced
be hav iors too numerous to count: being able to speak, for a
start, but also brushing our teeth, getting dressed, cooking, shop-
ping, and so on.
10 ch ap t er 1
Archaeological and anthropological evidence also suggests
that cultural skil s have been crucial to human survival for a very
long time. Members of small- scale socie ties rely on traditional
knowledge and know- how for foraging, hunting, pro cessing
food, making clothing, and producing the variety of tools indis-
pensable to their survival.31
If the simplicity of this “fax model” of cultural transmission
highlights the many benefits of learning from one’s surrounding
culture, its limits are also obvious. For one thing, it vastly under-
estimates the degree of cultural variation pre sent even in the
smallest, most self- contained socie ties. If some be hav iors might
be performed by all group members in a very similar fashion—
some ritual, say— most activities exhibit significant variation.
Not every hunter draws the same lessons from a set of tracks. Not
every forager has the same techniques for finding berries. Not
every artist creates equally appealing songs or sculptures or draw-
ings. So even an individual bent on blindly copying the previous
generation must make decisions: Who to copy from?
One of the most advanced frameworks for answering this
question has been created by an anthropologist, Robert Boyd,
and a biologist, Peter Richerson.32 Known as gene- culture co-
evolution, this theory suggests that genes and cultures have
influenced each other in the course of human evolution. In par-
tic u lar, Boyd and Richerson claim that culture has shaped our
biological evolution. If choosing which bits of one’s culture to
copy is so impor tant, then we should have evolved, through natu-
ral se lection, mechanisms that help solve this prob lem as ef-
fectively as pos si ble. We already have evolved dispositions that
tackle a variety of issues our ancestors faced: forming a broadly
accurate repre sen ta tion of our surroundings, picking edible food,
avoiding predators, attracting mates, forming friendships, and so
forth.33 It would make sense that we had also evolved mecha-
t he c a se f o r g ul l ib il i t y 11
nisms to help us acquire the culture of our peers and our
elders.
To solve the prob lem of who to learn from, we can start by
looking at who performs wel . Alex is an excellent cook; Renée
is great at maintaining good social relationships; it makes sense
to learn from them. But even when we have narrowed down the
prob lem in this way, we’re left with many potential actions to imi-
tate. How do we work out exactly how and why Alex was able
to cook such a great dish? Our intuitions help us rule out some
factors—it prob ably wasn’t his hairdo— but there remain many
possibilities, ranging from the most obvious, such as the ingre-
dients or the cooking time, to the least, such as the specific type
of onions used or how the rice was stirred. As we find out when
we try replicating a cook’s recipe, the determinants of success can
sometimes be quite opaque.34
To help us learn better from others, Boyd, Richerson, and their
colleagues— such as anthropologist Joe Henrich or biologist
Kevin Laland— suggest that humans are endowed with a series
of rough heuristics to guide their cultural learning.35 One of these
rules of thumb extends our ability to learn from the most suc-
cessful. Because it can be difficult to tell which of a successful
individual’s actions are responsible for their success— why Alex
was able to produce a given dish wel , say—it might be safer to
copy indiscriminately every thing successful people do and think,
down to their appearance or hairdo. We can call this a success bias.
Another heuristic consists in copying what ever the majority
does— the conformity bias.36 This bias makes sense under the rea-
sonable assumption that, if each individual has some in de pen-
dent ability to acquire valuable information, then any idea or
be hav ior that is widely accepted is likely to be worth adopting.
It is pos si ble to imagine many other such heuristics. For
instance, Henrich and his colleague Francisco Gil- White have
12 ch ap t er 1
suggested using a variation of the conformity bias to improve
on the success bias.37 They point out that even figuring out who
is successful can be difficult. For instance, in small- scale socie-
ties, which hunter brings in the most game varies widely from
one day to the next.38 In the midst of this statistical noise, how
can we decide which hunter to imitate? We can turn to others.
If many people look up to a given individual—if that indi-
vidual has prestige— then imitating them might be worthwhile.
For Henrich and Gil- White, such a prestige bias is highly
adaptive.
Boyd, Richerson, Henrich, and others have built sophisti-
cated models showing how reliance on rough heuristics allows
individuals to make the best of their surrounding culture. An-
other advantage of these heuristics is that they are cognitively
cheap, with no need for complex cost- benefit calculations: fig-
ure out what most people believe and adopt the same beliefs,
or figure out who does something best and imitate every thing
they do.39
But what happens when the majority is wrong, or when the
most successful or prestigious individual was just lucky? If these
rough heuristics provide a good bang for the buck— decent re-
sults at a cheap cost— they also lead to systematic mistakes.
Boyd, Richerson, and Henrich are ready to bite the bullet. The
self- sacrifice of the Japa nese kamikaze is accounted for through
a type of conformity bias, which allows cultural ele ments that
are beneficial for the group, but detrimental to the individual, to
spread.40 The prestige bias would explain why people appear
more likely to kil themselves after a celebrity has committed sui-
cide.41 Less dramatically, success bias predicts that people will
buy underwear advertised by basketball star Michael Jordan,
even though his athletic prowess is likely unrelated to his taste
in undergarments.42
t he c a se f o r g ul l ib il i t y 13
Not only do gene- culture coevolution theorists bite the bul-
let, but they do so gleefully. They accept that “to get the benefits
of social learning, humans have to be credulous, for the most part
accepting the ways that they observe in their society as sensible
and proper.”43 Indeed, the fact that reliance on rough heuristics
predicts the spread of absurd beliefs and maladaptive be hav ior,
as well as useful ones, is an “in ter es
t ing evolutionary feature of
these rules.”44 The novelty of this idea— maladaptive culture
spreads because we are adapted for culture— makes it all the
more attractive.
The Case against Gullibility
Many theories in the social sciences can be roughly recast in the
terms of this gene- culture coevolution framework. “The ideas of
the ruling class are in every epoch the ruling ideas,” as Marx
and Engels suggested: success bias.45 People blindly follow the
majority: conformity bias. Charismatic leaders go from being
worshipped by their faction to control ing the masses: prestige
bias. An incredible array of intel ectual traditions— centuries- old
po liti cal philosophy, experimental psy chol ogy, biologically in-
spired modeling— converge on the notion that humans are, by
and large, credulous, overly deferential toward authority, and
excessively conformist.
Could this be all wrong?
Throughout this book, I will chip away at the support for the
idea that the masses are gullible. Here’s the argument in a
nutshel .
Once we take strategic considerations into account, it be-
comes clear that gullibility can be too easily taken advantage of,
and thus isn’t adaptive. Far from being gullible, humans are en-
dowed with dedicated cognitive mechanisms that allow them to
14 ch ap t er 1
carefully evaluate communicated information. Instead of blindly
following prestigious individuals or the majority, we weigh many
cues to decide what to believe, who knows best, who to trust, and
what to feel.
The multiple mass persuasion attempts witnessed since the
dawn of history— from demagogues to advertisers— are no
proof of human gullibility. On the contrary, the recurrent fail-
ures of these attempts attest to the difficulties of influencing
people en masse.
Fi nally, the cultural success of some misconceptions, from
wild rumors to super natural beliefs, isn’t well explained by a ten-
dency to be credulous. By and large, misconceptions do not
spread because they are pushed by prestigious or charismatic
individuals— the supply side. Instead, they owe their success to
demand, as people look for beliefs that fit with their preexisting
views and serve some of their goals. Reassuringly, most popu lar
misconceptions remain largely cut off from the rest of our minds
and have few practical consequences, explaining why we can be
relatively lax when accepting them.
2
VIGILANCE IN COMMUNICATION
the best argument in favor of credulity is that it enables
us to acquire the knowledge of our peers and forebears. Being
disposed to copy what others do or think, and relying on simple
heuristics to figure out who to copy— what most people, or pres-
tigious leaders, do and think— would give us easy access to a
wealth of accumulated wisdom.
This argument, however, fails to take into account the strate-
gic ele ment pre sent in all interactions. It assumes that the indi-
viduals being copied are doing their best to engage in adaptive
be hav iors and to form accurate beliefs. It does not consider that
these individuals might wish to influence those who copy them.
But why wouldn’t they? Being able to influence others is a great
power. And, from an evolutionary point of view, with great power
comes great opportunities.
To understand what happens when individuals evolve to in-
fluence others as well as to be influenced by them, the most suit-
able framework is that of the evolution of communication. The
counterintuitive predictions that follow from this theory are
best illustrated with some puzzling animal be hav iors, which I
now describe, their explanations unfolding throughout the
chapter.
15
16 ch ap t er 2
Animals Acting Weird
In the woodlands of eastern Australia, one sometimes stumbles
upon strange construction: tiny house- like structures made of
grass, decorated with berries, eggshel s, bits of metal, and sun-
dry colorful objects. These constructions— called bowers— are
not made by the local human population but by spotted bow-
erbirds. Do the birds use these painstakingly constructed
structures for protection against the weather or against preda-
tors? No. They build more typical nests in trees for this purpose.
Why do bowerbirds bother building bowers?
With a sleek shape, long horns, elegant black streaks on their
flanks, and a bright white rump, Thomson’s gazelles are gorgeous
animals— gorgeous, but maybe a bit daft. Packs of wild dogs
roam the savanna, ready to chase and devour them, yet when a
gazelle spots a pack of dogs, it often fails to flee at full speed. In-
stead, it stots, jumping on the same spot while keeping its legs
straight. It stots high, sometimes as high as six feet.1 It stots in
the absence of any obstacles. It stots even though stotting slows
it down. Why does the stupid gazelle not stop stotting?
Like the spotted bowerbird, the Arabian babbler is a brown-
ish bird, just short of one foot long. As the name suggests, though,
babblers do not build bowers; they babble. Besides its vocal dis-
plays, a striking feature of the Arabian babbler is its cooperative-
ness: groups of a dozen babblers look after their offspring
together, clean each other, and act as sentinels. When they see a
predator approaching, these sentinels engage in a be hav ior that
seems much more sensible than that of our gazelles: they give
alarm calls. When the predator is still quite distant, the sentinels
emit barks (two relatively low calls) or tril s (a higher, longer vi-
brating call). When the predator draws closer, the sentinels
start emitting tzwicks (three short, higher- pitched calls). These
v i g il a n c e in c o mmuni c at i o n 17
calls allow the other group members to hide from some preda-
tors and to mob others. So far so good. Some babblers, however,
have solitary lives: they do not live, let alone cooperate, with
other babblers. Yet when these so- called floaters spot a preda-
tor, they give the same calls as the sentinels.2 Why would lonely
babblers give such wanton warning wails?
In humans, as in other mammals, pregnancy causes many
changes in a mother’s body. Some are obvious— the enlarged
belly— but others are subtler, such as the change in the way in-
sulin is produced. Insulin is the hormone that tel s the body to
turn blood sugar into fat. After a sugar- rich meal, blood sugar
levels increase, insulin is secreted, and the sugar is stored away
in the form of fat. Toward the end of pregnancy, however,
mothers start generating increasingly large quantities of insulin
after their meals. This might seem weird: the growing fetus asks
for a huge amount of energy, which it draws from the sugar in
the mother’s blood. Even weirder: in spite of this huge spike in
insulin, blood sugar levels remain elevated for longer than
/>
usual.3 Why would a mother’s body labor to limit the little
one’s resources? And why would it fail?
For animals with such tiny brains, bees are very sophisticated
foragers. They scout for nectar- rich flowers and keep track of
their location. When they return to the hive, the bees use their
famous waggle dance to tell their hivemates where to find food.
To forage effectively, bees are apt to use both personal experi-
ence (where they have found good flower spots in the past) and
social information (the dances of other bees). In order to test
how much weight bees put on personal versus social informa-
tion, entomologist Margaret Wray and her col eagues performed
a series of ingenious experiments. They put a feeder (an artifi-
cial source of sugar) in the middle of a lake. Some bees flew over
the lake, found the feeder, and returned to the hive with the good
18 ch ap t er 2
news. Now, flowers do not grow in the middle of lakes. The bees
back at the hive, upon seeing a dance pointing at the lake, would
have been quite justified in thinking the returning bees mistaken.
But they did not. The bees dutifully left the hive— indeed, they
left the hive at the same rate as when the feeder had been placed
in a much more plausible location.4 Why would these intel igent
insects ignore their individual intuitions and follow implausible
instructions?
Conflicts and the Evolution of Communication
The key to understanding these strange be hav iors is also the key
to understanding how we evaluate what we are told: the theory
of the evolution of communication.
For genuine communication to exist, there must be dedicated
adaptations on the side of the entity sending the signals, and on
the side of the entity receiving the signals.5 For example, vervet
monkeys have a sophisticated system of alarm calls, allowing
them to warn each other of the presence of ea gles, snakes, leop-
ards, and other predators. Vervets must be equipped with
mechanisms that trigger the correct call when each of these pred-
ators has been spotted, as well as with mechanisms that trigger
the appropriate reaction to each of these calls— climbing trees
is not very helpful when an ea gle draws near.6 These alarm calls
clearly fit the bill for an evolved communication system.