Is Religion Adaptive?

Integrating cognition and function

Robin DunbarBritish Academy Centenary Project

Institute of Cognitive & Evolutionary Anthropology

University of Oxford

Two Evolutionary Questions about Religion

Nothing as costly as religion could possibly be a maladaptation or a mere by-product

• What does [did] religion do for us?

• Why did religion [apparently] evolve only in humans?

Four [Traditional] Functions for Religion

• gives coherence to a complex world [Freud]

• psychological wellbeing

• social bonding [Durkheim]

• enforces conformity

[moral codes] [Marx]

Does Religion Benefit You?

Frequent claims that religion has no adaptive benefits at the individual level notwithstanding...

Compared to others, actively religious people:

• live longer

• are more content/happier

• are less stressed

• suffer fewer psychological problems

• recover faster from surgery

[Data from extensive sociological and

epidemiological studies in past decade]

BUT, even if none of these was true….

The World of Multilevel Selection• Multi-level social systems are common in

mammals• When sociality involves an implicit social

contract….Fitness accrues at the level of the individual, but through benefits generated by the group

• It requires a more subtle understanding of fitness – Hamilton‟s original concept [neighbour-modulated fitnesses]

• This is NOTgroup selection

Elephants at Amboseli

Human social networks

scale hierarchically

Old World monkeys

Feral goats on Rum

Back To The Beginning….

• Social Brain Hypothesis

• An explanation for the evolution of large brains in primates

• Evidence: group size [and many aspects of “smart” behaviour] are a function of neocortex volume

Apes

Dunbar 1992, 1998

Neocortex ratio = neocortex vol/rest of brain

[i.e. “thinking” part of brain]

Humans and the Social Brain

• Predicted group size for humans is ~150

[Dunbar‟s Number]

…BUT primate societiesare very intimate

HumanSocial Groups

All these have mean sizes of 100-200

Neolithic villages 6500 BC 150-200 Modern armies (company) 180Hutterite communities 107„Nebraska‟ Amish parishes 113business organisation <200ideal church congregations <200Doomsday Book villages 150C18th English villages 160GoreTex Inc‟s structure 150Research sub-disciplines 100-200

Small world experiments 134Hunter-Gatherer communities 148Xmas card networks 154

Maximum Network Size

350-374

325-349

300-324

275-299

250-274

225-249

200-224

175-199

150-174

125-149

100-124

75-99

50-74

25-49

0-24

Nu

mb

er

of

Ca

se

s

10

9

8

7

6

5

4

3

2

1

0

“Reverse”Small World Experiments

Killworth et al (1984)

1

10

100

1000

10000

0 10 20 30

Hunter-Gatherer Societies

Dunbar (1993)

Xmas Card Networks

Hill & Dunbar (2003)

Individual Tribes

The Freerider Problem

Barrett, Dunbar & Lycett: Fig 9-6

0

2

4

6

8

10

0 2 4 6 8 10 12

Coalition Time

Searc

h T

ime

Freeriders are not

successful

T = 0%

T = 25%

T = 100%

Freeriders

are

successful

Enquist & Leimar (1993)

All primate societies are based on animplicit social contract [cooperation]

All such systems are susceptible to freeriders

Dispersed social systems are especially susceptible

Punishment [stick] has attracted mostattention...

BUT it’s only as effective as thedetection rate

Voluntary commitment [carrot] ismuch more effective

Barrett, Dunbar & Lycett: Fig 9-9

0

20

40

60

80

100

0 5 10 15 20 25 30 35

Number of Generations

Fre

eri

de

rs (

% o

f P

op

ula

tio

n)

50%

30%

10%

Rate of dialectic change per generation

Nettle & Dunbar (1997)

Freeriders very

quickly drive

cooperators to

extinction

Freeriders are more successful

over a wide range of conditions

The Twin Pincers of Primate Sociality

• The intensity of intimacy

• Mentalising [understanding other‟s minds]

1st 2nd 3rd ....order

The Levels of Intentionality

The Intensity of Intimacy

• Social bonding primate-style

• Intimacy influences trust and obligation

• Grooming increases with group size

• Grooming releases endorphins and creates an opiate “high”

Group Size

160140120100806040200

So

cia

l T

ime

(%

)

50

40

30

20

10

0

Predicted for Humans

Mean grooming

time vs mean

group size for

individual primate

species

Grooming Time in Humans?

• Grooming as the bonding agent in primates

• Grooming time is a linear function of group size

Group Size

160140120100806040200

So

cia

l T

ime

(%

)50

40

30

20

10

0

Predicted for Humans

Actual

social

time in

humans

Three Ways to Bridge the Gap?

Millions Years BP

3.53.02.52.01.51.0.50.0-.5

Pre

dic

ted G

room

ing T

ime (

%)

50

40

30

20

10

Laughter

a cross-cultural trait

shared with chimpanzees

Singing and dancing

Religion and its rituals

Australopiths

Modern humans

H. erectus

Archaic humans

The

Bonding

Gap

Laughter… the Best Medicine?

0

0.5

1

1.5

2

2.5

3

3.5

4

Strangers Acquaintances

Cooperation(in GBP)

Neutral

Comedy

Ss were more generous to strangers

(but not “friends”)

after watching a comedy video

In a Public Good Game P=0.559 p<0.001

Procedure:

pain test – video/activity – pain re-test

Musical Endorphins

Procedure:

pain test – activity – pain re-test

Musical performance

facilitates endorphin

release, but listening

to music

does not

Dunbar, Kaskatis, MacDonald &

Barra (submitted)

Singing Drumming Dancing Listening

to music

Sing/No sing

Drum

circle

music

shop

video

Fast

Slow

Dance

group

Music

practice

An Opium for the Masses?

Religious practices are often well suited

to stimulate endorphins

Stigmata of

Padre Pio

Ecstatic states:

make you relaxed

enhance sense of

communalityMedieval flagellants

Whirling dervish

So, why not get your kicks on your own?

….because creating a sense of

“bondingness” [commitment to

community and prosociality] requires

doing it together [i.e. interacting]!

Synchony Ramps up the Endorphins

Alone Group Alone Group

Change in pain threshold

before and after 45 mins

rowing work-out on

ergometers in the gym:

Alone vs in a virtual boat

Cohen et al (2008)

Why Does Religion Have This Effect?

There are two likely mechanisms:

• direct influence of endorphins [endorphins seem to “tune” the immune system]

• sense of belonging and communality acting directly OR indirectly through support from network members

Mean Relatedness in Plymouth Colony

(winter 1620)

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

Survived Died

… being a member of a large

kinship group reduces

(1) illness rates in children

[Newcastle and Trindad]

(2) death rates in adults

[the Mayflower colonists

in 1620 and the Donner

Party of 1846]

The Transcendental Edge…?

Sosis & Alcorta (2003)

Something about a transcendental dimension

raises commitment to the project

C19th US utopian cults

Religious cults last longer than secular ones

The Ecology of Community Size

• Small communities in the Tropics, larger ones at higher latitudes

• Religion density and collectivism are a function of disease prevalence [aka latitude]

Fincher & Thornhill (2008)

Fincher et al (2008)

Barrett, Dunbar & Lycett: Fig 13-4

-7

-6

-5

-4

-3

-2

-1

0

0 2 4 6 8 10 12

Mean Growing Season

Lo

g R

esid

ual N

um

ber

of

Lan

gu

ag

es

Nettle (1999)

Cults are a Small-Scale Phenomenon

• Foundation size for C19th US utopian cults

• Against expectation (scale-free effect from Zipf‟s Law), N<30 is uncommon, and 150>N>400 is the most common

Dunbar & Sosis (in prep)

Is There an Optimal Cult Size?

• Very small and very large foundations don‟t survive well

• Optimum size ~150

• For secular communes, optimal size may even be smaller (~50)

50

Dunbar & Sosis (in prep)

The Role of Social Cognition

• Intentionality as a reflexively hierarchical sequence of belief states

• …that may be very costly in information processing terms

Level of intensionality

9th8th7th6th5th4th3rd2nd1st

Fre

quency o

f fa

ilure

20

10

0

Level of intensionality

1086420

Clique s

ize

30

20

10

0

Kindermann et al (1998)

Stiller & Dunbar (2007)

Religion at the Limits of Cognition…?

Belief as a personal phenomenon“I believe that God wants us to act

with righteous intent” [3]

Belief as a social phenomenon“I intend that you believe that God

wants us to act with righteous intent” [4]

…. BUT why should you care?

Belief as a communal phenomenon:“I intend that you understand that we believe that

God wants us to act with righteous intent” [5]

Level of intensionality

9th8th7th6th5th4th3rd2nd1st

Fre

quency o

f fa

ilure

20

10

0

0 1000 2000 3000 4000 5000 6000

Frontal Lobe Volume (cc)

1

2

3

4

5

Ac

hie

vab

le In

ten

tio

na

lity

Lev

el

Monkeys

Apes

Humans

Dunbar (2003)

Why is Religion Unique to Humans?

• If achievable level of intentionality really is determined by capacity of frontal lobe…

• …only humans have sufficient capacity for 5th order

Can We Date the Origins of Religion?

• Maybe….

• If we can estimate frontal lobe volume from cranial capacity…

…then we can estimate achievable level of intentionality

0

1

2

3

4

5

6

0 200 400 600

Frontal Lobe Vol. (cc)

Ach

iev

ab

le I

nte

nti

on

ali

ty

0.00 1.00 2.00 3.00

date (millions of years BP)

300.0

600.0

900.0

1200.0

1500.0

1800.0

cc

0 200 400 600 800 1000 1200 1400

Total Brain Volume (cc)

0

200

400

600

800

1000

Fro

nta

l L

ob

e V

olu

me

(c

c)

Cranial volume

Dating the Origins of Religion

• Sufficient frontal lobe volume appears very late

• … not earlier than the appearance of Homo sapiens

[Don‟t get too excited about the Neanderthals] 0

1

2

3

4

5

6

7

0 2 4

Million of Years Ago

Ach

ieva

ble

In

ten

tion

alit

yAustralopiths

H. erectus

Archaic humans

Modern humans

Neanderthals

Conclusions

• Social contract societies risk collapse from freeriding unless mechanisms are in place to enforce social cohesion

• In primates, this involves both cognitive and psychopharmacological mechanisms

• Voluntary commitment [carrot] works better than punishment [stick]

• Religion and ritual seem to function in just these ways

• Religion is: a small scale phenomenon, very susceptible to fragmentationwell designed to reinforce

in-group/out-group effects [Durkheim was basically right]

• Other benefits are by-products

• Religion may be a very recent evolutionary phenomenon