Outline: Background

‡ Evolved Mating Psychologies ± Female and male psychologies are different in ways that till
us (humans) toward polygynous mating systems

± Greater absolute wealth differences among males, the
greater the polygyny.

± Substantial polygyny in societies with large absolute
differences in male wealth. ‡ Cultural Evolution and social norms ± Self-re-enforcing stable equilibria ± Internalized motivations ± Competition among societies can favour group-beneficial norms

± Mating vs. marriage systems
12/10/10 1

‡ Anthropological review of marriage
systems ± Consistent with evolutionary expectations? ± Frequencies of spousal # variations ± Should we expect polyandry or group marriage
to also spread?

‡ Historical origins of modern monogamy

± Unusual monogamy ± Cultural practices evolved gradually


‡ Polygynous vs. monogamous marriage comparison crucial ± Societal impacts increases the size of pool of low status males
with little prospect for finding a mate

‡ Increases crime and substance abuse ± Families drive down paternal investment and internal conflict ‡ Young children ± Poorer health and greater mortality ± Results from less paternal investment and internal

‡ Women and girls ± Age of first marriage age gap, and equality ± Psycho-social stress
12/10/10 33

Evolved mating psychology
‡ Why males and female are different ± Eggs are expensive, limited, max 400 in a lifetime ± Sperm are cheap and disposable.. can make 12 million
in an hour.

± Creates an asymmetry in the initial investment in

± Asymmetry in investment magnified through ‡ Gestation weakness, pregnancy sickness ‡ Lactation calorically intensive, suppresses fertility
Henrich Aff #1, pages 27-28 12/10/10 44

Female mating strategies
‡ Women are limited in number of children they can have in their
lifetime (max 10 in ancestral environments), and must invest in all of them.

± The key to their fitness is choosing wisely among potential

± Find the highest quality male who is willing to invest in the
long-run with her offspring. Good genes, and paternal investment (rich, capable, and willing).

± But, can improve this situation by getting high quality genetic
material on the side (extra-pair copulation).

± Long-term and short-term strategies
Henrich Aff #1, pages 27-28 12/10/10 55

Male mating strategies
‡ Men can have literally thousands of offspring, but are also much
more likely to get entirely shut out of the mating game.

± High variance (0 to 1000) polygyny increases variance. ± Most successful strategy is one that has sex with as many
women as possible, and never invests in rearing offspring.

± Many males, however, won t be able to pursue this strategy
because not very many women will have sex with them. ± Alternative (most common) strategy: commit to investing with a woman to form a long-term pair-bond. ‡ Male commits assistance and resources while female commits some degree of paternity certainty. ‡ Forming additional long-term pair-bonds increases fitness ‡ Male also still increases his fitness is each extra-pair copulation
Henrich Aff #1, pages 28-29 12/10/10 66

Mating systems are outcome of the two psychologies
‡ A particular pair-bonding pattern results ± Males will be inclined to make additional pair-bonds, if
they can, and seek extra-pair copulations.

± Females don t benefit much from additional pair-bonds.

‡ She can only get pregnant sequentially ‡ Paternity certainty more investment ‡ Male jealousy conflict ± Lower rate of extra-pair copulations by females
12/10/10 77

A cultural species
‡ Unlike other animals, humans are heavily reliant on cultural
learning, on learning from other group members.

‡ Evolved capacities for cultural learning are a ‡ Humans adaptively learn all manner of social behavior from
others in their social milieu

± How others should be judged, and what affects a person s
reputation or standing in the community.

± Internalize local norms as intrinsic motivation ‡ Social Norms ± Self-re-enforcing social rules for behavior ± When pair-bonds are regulated by social norms marriage
Henrich, Aff #1, page 26 12/10/10 88

Marriage systems
‡ Sets of cultural evolved rules called institutions that
regulate pair-bonding ± Transfer at marriage: brideprice or dowry ± Division of labor, responsibilities to offspring and spouse ± Rules about sex ‡ 90%+ of societies emphasize norms regulating female sexuality ± Residence who lives with whom: patrilocal or matrilocal ± Inheritance
Henrich, Aff #1, page 26 12/10/10 99

Marriage systems
‡ Sets of cultural evolved rules called institutions that
regulate pair-bonding

± Combinations of husbands and wives ‡ Polygyny norms that permit or favor one husband
with 1+ wives.

‡ Polyandry norms that permit or favor a wife with
1+ husbands

‡ ALL marriage systems culturally reinforce the pairbonding aspect of human psychology.

‡ Not all human societies have marriage systems. The Na of
Western China.
Henrich, Aff #1, page 31 12/10/10 1010

Anthropological Terminology
‡ A marriage is long-term pair-bond between two people that
is recognized and sanctioned by the couple s community.

‡ Being married comes with economic, social, and sexual
expectations, prescriptions, and prohibitions (norms) for both parties.

‡ Marriage may or may not be sanctioned by formal laws, and
marriage existed long before formal laws or even writing.

‡ Public rituals usually mark the commencement of a

‡ The key to understanding marriage is recognizing the role of
a community in defining, sanctioning, and enforcing it.
Henrich Aff #1, page 26, footnote 3 12/10/10 1111

Key summary points
‡ Within the context of marriage systems, male mating psychology
strongly favours long-term pair-bonds with multiple partners. Their ability to attract long-term partners will depend on their resources and social status. ‡ Male psychology is generally highly averse to sexually sharing a longterm mate with another male. ± The evolutionary reasons for this involve paternity uncertainty.

± Female ovulation is largely concealed from males ‡ Males should respond to a shortage of females by more fiercely
guarding their mates.

± Sustain paternity certainty ‡ Females are not advantaged by pair-bonding with multiple males.


Henrich, Aff #1, page 32 12

Marriages systems reflect our evolved mating psychology
Table 2. Marriage Systems in Crosscultural Perspective Ethnographic Atlas Type N = 1231 societies (Gray 1998) Monogamous 15% (186) Occasional polygyny 37% (453) Frequent polygyny 48% (588) Polyandry 0.3% (4)

The whole database

Henrich, Aff #1, page 33 12/10/10


Standard Cross Cultural Sample
Table 3. Marriage Systems in the Count (%) Standard Cross Cultural Sample N = 176
Historically more independent societies

Polyandry Monogamy Monogamy with occasional polygyny Polygyny preferred (but <20% of male engage) Polygyny preferred (> 20% of males engage)

2 (1.1%) 27 (15%) 33 (19%) 54 (31%) 60 (34%)

Henrich, Aff #1, page 33 12/10/10


Normative prescriptions
Table 4. Cultural Norms Monogamy prescribed (offspring of nonwives do not inherit) Monogamy preferred but some polygyny Polygyny for exceptional males (leadership, skills) Polygyny for men of wealth, nobility, etc. Polygyny preferred for most men. Most older men should have 2+ wives.
Henrich, Aff #1, page 34 12/10/10

Count (%) N = 183 27 (15%) 32 (17%) 45 (25%) 33 (18%) 46 (25%)


Monogamous marriage may be even more unusual than it appears
Table 2. Marriage Systems in Crosscultural Perspective Ethnographic Atlas Type N = 1231 societies (Gray 1998) Monogamous 15% (186) Occasional polygyny 37% (453) Frequent polygyny 48% (588) Polyandry 0.3% (4) 


Ecological monogamy Historically connected societies (Irish, Rome) Class-based monogamy (Egypt, Babylon)

Henrich, Aff #1, page 33 12/10/10


Types of polygyny
‡ General polygyny ± Wives need not be, and general are not, sisters or
classificatory sisters (cousins)

± Separate the wives ‡ Sororal polygyny ± Wives should be sisters or classificatory sisters. ± Having related wives reduces intra-household co-wife

± Much more closely related households.
Henrich, Aff #1, page 35 12/10/10 1717

Polygyny, status, and wealth
‡ Small-scale hunter-gatherers are polygynous,

‡ In all polygynous societies, a man s social status,
prestige, hunting skill, nobility, and wealth lead to more.

‡ More absolute wealth differences, more variations
in wife numbers (if permitted).

Henrich, Aff #1, page 35 12/10/10


Polygyny, status, and wealth
‡ Betzig (1982; 1993) puts a fine point on this observation by
analyzing what the autocratic leaders of chiefdoms, empires, and early states did regarding wives and concubines.

‡ She reveals a strong pattern that, given the wherewithal to
do so (no internalized social norms or laws to impede them), powerful men consistently assemble immense harems with 100 or more women.

‡ This ranges from High Chiefs in Tonga and Fiji to emperors in
China and the Andes. Harems get bigger and bigger as the societies get larger and more complex.
Henrich, Aff #1, page 35 12/10/10 1919

Polyandrous marriages are quite rare
Polyandry is 1) usually fraternal polyandry, meaning brothers marry the same woman

2) typically found intermixed with other marriage types in the
same society, including both monogamy and polygyny

3) considered to be somewhat unstable with the youngest
husbands leaving the marriage, or taking additional wives themselves (giving rise to polygynandry)

4) principally confined to the Himalayan and, to a lesser degree,
Indian regions of Eurasia, though it has been observed elsewhere, including in the Americas (Levine and Silk 1997). Many researchers have argued that polyandry emerges when sustaining a household requires the input of multiple males (Levine and Silk 1997).



Group marriage?
‡ Scattered reports. ‡ Many of these reports are of dubious quality. ‡ Or, non-anthropological observers have confused marriage
with the custom of wife sharing or loaning, which was common in both aboriginal North America and Australia.

‡ In these societies, which were numerous (and usually
polygynous), husbands controlled sexual access to their wives, and it is considered polite and honourable for them to give those services to close friends or honoured guests for a night or period of time.

‡ Since these other men are also often married, it might appear
as if some kind of complicated marital arrangement exists. ‡ Polygynandry starts as polyandry, one of low status brothers marries polygynously.
Henrich, Aff #1, page 38 12/10/10 2121

Theoretical Interlude
‡ Why look at history of modern monogamy? ‡ Is this history consistent with the theory? ± Is it rare? ± Is it associated with successfully spreading societies. ‡ Ancient Greece ‡ Rome ‡ Europe ‡ It could have been that monogamous marriage moved with
plow agriculture, or that it s a peculiarly Judeo-Christian idea.

± Not the case



History of Western Monogamy
‡ Traces back through Rome to the Greek city states ‡ Adopted recently in much of the world
± 1880 in Japan, 1926 in Turkey, 1956 in Tunisia, 1953 in China, 1955 in
India (except Muslims), and 1963 in Nepal.

‡ Greek city states first legally instituted monogamy as part of
many different reforms, including elements of democratic governance, which were meant to build egalitarian social solidarity among their citizenries.

‡ Prior to this, all accounts suggest polygynous marriage was
common, at least among the nobility, and monogamy was a strange Greek idea

‡ Instituted legally in the early sixth century BCE in Athens.
Henrich, Aff #1, page 36-37 12/10/10 2323

History of Western Monogamy
‡ While Greek monogamy limited each male citizen to a
single wife, it was considered acceptable to import sex slaves, which wealthy men did. ‡ Interesting approach because it addresses one of the fundamental social dilemmas posed by marriage systems: ± by keeping local women available to low status men for marriage this avoiding the problems created by poor unmarried males, ± while at the same time allowing rich men broad sexual access to imported women. ± This reduces opposition from wealthy and nobility
Henrich, Aff #1, page 36-37 12/10/10 2424

History of Western Monogamy
‡ Romans likely inherited and further developed the
monogamy of the Greeks, though Etruscan marriage norms and relative sexual equality likely had some influence.

‡ Rome outlawed polygamy and regulated this with laws
about sexual behaviour, birth legitimacy, and inheritance (Herlihy 1995; Macdonald 1995).

Henrich, Aff #1, page 37 12/10/10


History of Western Monogamy
‡ Augustus felt Roman morality was declining and
weakening his empire, so he instituted a series of reforms in an effort to get every man from age 25 to 60 to be married. ‡ A series of Roman emperors after Augustus continued to reinforce these legal principles and adapt the law. ‡ Historians: the evolution of this aspect of the Roman legal system is intimately intertwined with the emergence of greater sexual equality under the law (Macdonald 1990; Herlihy 1995; Scheidel 2009).

Henrich, Aff #1, page 37 12/10/10


Into Europe
‡ Early Christian ideas about monogamy and sexual purity
are a combination of the evolving Roman ideals and notions drawn from Greek stoicism. ‡ Christian ideals solidified and eventually spread throughout Europe, which was highly polygynous in the pre-Christian era and during the early days of Christianity. ‡ European aristocracies were highly polygynous in the 5th century. ‡ All sought alliances with the Catholic Church, which worked vigorously to impose monogamous marriage on the aristocracy.
Henrich, Aff #1, page 37 12/10/10


Into Europe
‡ As European kings gradually converted to Christianity the Church
increasingly controlled their marriages, and thus their legitimate heirs (that is, they controlled who had rights to political power). ‡ Since the lower strata of these societies, who were rapidly adopting Christianity, were economically limited to monogamous marriage anyway, the main line of resistance came from the nobility. ‡ Once the nobility began to accept monogamous marriage, general monogamy and associated laws followed (Macdonald 1995). ‡ Historians argue that monogamous marriage in Europe meant that for the first time kings were subject to the same rules as peasants.
Henrich, Aff #1, page 38 12/10/10 2828

History of Western Monogamy
‡ These ideas do not come from Judaism (which permitted
polygynous marriage until at least the 11th century), or the Christian Gospels.

‡ In the Old Testament, the prophets and kings are all

‡ At best, the New Testament offers some vague
recommendations for monogamy among church leaders in the Pastoral Letters (Scheidel 2009).

Henrich, Aff #1, page 37 12/10/10


Summary points so far
‡ Polygynous marriage has been widespread, and emerges
in an immense variety of social contexts.

‡ Wealthy, high status men tend to marry polygynously, if
they are permitted.

‡ Polyandry is quite rare, in a manner consistent with
evolutionary theory.

‡ Imposed universal monogamous marriage is uncommon
too, and seems to have evolved culturally to increase solidarity, foster an egalitarian spirit, and stabilize society.

Henrich, Aff #1, page 38 12/10/10


Marriage on Crime
‡ Married men are much less likely to commit crimes and abuse
substances compared to unmarried men.

± Substantial correlational evidence. Is it causal? ± Longitude studies tracking males from age 17 to 70 show that
the same men are less likely to commit crimes during the married periods of their lives (Sampson et. al. 2006). Massachusetts reform school. ‡ For all crimes, it reduce a man s chances of committing a crime by 35%.

‡ For property and violent crimes, marriage cuts the
probability of committing a crime by half.

‡ When men divorce or are widowed, their crime rates go

‡ It s not the men, it s the marriage.
12/10/10 Henrich, Aff #1, page 40-41 3131

Marriage on Crime
‡ Horney et. al. (1995) working with Nebraska inmates (recall data). ± Moving in with a wife reduces the probability of a man
committing a crime by roughly half, controlling for employment, drug abuse, etc.

± Effect is strongest for assault and weakest for property crimes. ± Marriage effect is similar to entering school. ± Cohabitating with a girlfriend (as opposed to a wife) either
increases or does not impact individuals crime rates.

± The positive effect on crime of living with a wife is larger than
the negative effect created by heavy drinking.

± Taking drugs had the biggest effect on increasing crime rates. ± Horney et. al. may underestimate the total impact of marriage
because marriage also reduces binge drinking and drug use.

‡ Same story in London, Farrington and West.
12/10/10 Henrich, Aff #1, page 41 3232

Polygyny and Crime
‡ Do these individual-level effects aggregate up to the societal level? ‡ Analytical challenge: polygynous marriage isn t randomly
distributed across different societies. Tends to co-occur with low GDP per capita, low female equality, little democracy, etc.

1) Run cross-national regressions using a polygyny variable to
predict different crimes, controlling for Economic Development, Economic Inequality, Population Density, and Democracy.

2) Replace polygyny variable with % of unmarried males. If our
causal story is corrected, this will be at least as strong.

3) Look within countries and longitudinally using sex ratios as a
proxy for polygyny in different regions.
12/10/10 Henrich, Aff #1, page 42 3333

Cross-national analyses
‡ Very conservative analysis since Economic Development,
Economic Inequality, and Democracy are controlled for.

1) Monogamous marriage may cause economic development,
greater equality, and democracy.

2) Polygyny variable is vary noisy. ‡ Kanazawa and Still coded all of the cultures in the
Encyclopaedia of World Cultures on a four point scale (from 0 = monogamy is the rule and is widespread, to 3 = polygyny is the rule and is widespread), and then developed a country-level value by aggregating all of the cultures within a country.


Henrich, Aff #1, page 42-43


Cross-national analyses
‡ Prong (1) polygyny and crime ± Positive relationship between polygyny and murder,
rape, assault, and robbery. ± Murder and rape are significant at conventional levels, and marginal on robbery. ± Coefficient on assault is large but not well estimated. ± Being fully polygynous increases murders and rapes by 12 per 100,000. ± Being fully polygynous increases robberies by 64 per 100,000.


Henrich, Aff #1, page 43


Cross-national analyses
‡ Prong (2): Replace polygyny with % unmarried men over 15. ‡ Now the % of unmarried men is positively related to murder,
rape, robbery, and assault.

± Murder, rape, and robbery now are all significant at
conventional levels.

‡ An increase of 40 in the % of unmarried men predicts an
increase of

‡ 11 per 100,000 in murders. ‡ 22 per 100,000 in rapes ‡ 180 per 100,000 robberies ± Assault is marginally significant ‡ 108 per 100,000 assaults

Minimum effect sizes, since we are controlling for GDP per capita, equality, and democracy

Henrich, Aff #1, page 43-44 3636


Cross-national analyses
‡ Prong (3): Within country analyses ‡ By themselves, cross-national analyses don t permit any
assessment of causality.

‡ Use sex ratio as a proxy for effects of polygyny could create the
same kinds of effects for crime.

± India son preference create sex ratio favoring males. Varies
across districts in India.

‡ Because it is within one country it deals with many of the
problems of cross-national analysis.

± China son preference plus one child policy enacted at
different times allows us to get closer to causality.

± U.S. could this work in the a Western democracy?
12/10/10 Henrich, Aff #1, page 44 3737

‡ Rising murder rate in India (Dreze and Khera 2000) ‡ 319 districts, 90% of population ‡ Murder avoids reporting biases ‡ Controlling for economic and demographic differences, more males to
females predicts higher murder rates.

‡ Showed that it s not just more males. Higher sex ratios make the males
more murderous.

‡ The effect is large: going from a male to female ratio (in Uttar Pradesh)
of 1.12 to one (in Kerala) of 0.97 cuts the murder rate in half.

‡ Literacy is also an important independent predictor of murder rates
across districts, though poverty and urbanization are not.


Henrich, Aff #1, page 45


‡ Crucial natural experiment ‡ In China, adult sex ratios (males to females) rose markedly from
1.053 to 1.095 between 1988 and 2004, nearly doubling the unmarried or surplus men (Edlund et al. 2007).

‡ At the same time, crime rates nearly doubled 90% of which
were committed by men.

‡ The increase in sex ratio was created by the gradual
implementation of China s one-child policy. Each province implemented the policy at different times for idiosyncratic reasons (unrelated to crime rates or sex ratios)

‡ Excellent opportunity for statistical analyses of the impacts of the
policy and the alterations in sex ratio it created.
12/10/10 Henrich, Aff #1, page 44 3939

Ideal natural experiment for two reasons:

1) Limiting child number through potent family planning led to
preferences for male children. That means that male children are exceptionally valued.

± They would benefit from heavy parental investment, and one
should expect, if anything, that such children would be less likely to commit crimes than the boys of previous generations.

2) Limiting family size means the population began to shrink, a
demographic shift that opens up opportunities in the labour market and ought to decrease people s likelihood of committing crimes. 

So it is significant that, despite these downward forces, crime
still went up.
12/10/10 Henrich, Aff #1, page 44 4040

‡ Controlling for many other economic and social variables,
a 0.01 increase in sex ratio is associated with a 3% increase in property and violent crimes.

‡ Inequality, unemployment, and urbanization also show
positive effects on crime rates, but the effect of sex ratio is independent of these effects.

‡ These analyses also indicate that the effect arises from
an increase in the size of the pool of low-status unmarried men, and not from an increase in men in general.


Henrich, Aff #1, page 44


‡ Causality ‡ The authors show that the year of implementation of the
one-child policy predicts sex ratio, and this predicts crime rates across provinces.

‡ The sex ratio decades later cannot be causing the
implementation of the one-child policy (in the past).

‡ The one-child policy caused families to strongly favour
males, creating a pool of unmarried low status men.

‡ This pool generates massive increases in crime rates
despite economic growth, and a shrinking population.


Henrich, Aff #1, page 44


Historical sources
‡ Courtwright in his book, Violent Land, indicates that pools of
unmarried men have created similar effects (high crime, violence, drug abuse, etc.) in a variety of circumstances.

‡ The violent character of the American West arose principally
from the large pool of unmarried men who migrated there.

‡ Variation in crime rates, including murders, in America
corresponds to the spatial distribution of sex ratios in the 19th century.

‡ As sex ratios move toward 1 in regions, crime rates drop. ‡ Similar cases in New South Wales in Australia in the late 1700s
and the Argentinean Pampas in the gaucho era.
Henrich, Aff #1, page 45



Polygyny, Age of Marriage, Age Gaps & Gender Equality
‡ Polygynous marriage ignites a competitive process for wives
among men that increases males efforts to control females

± drives down the age of first marriage for females while increasing
the age gap between spouses ± Seeking out the only available females for both first and subsequent wives, men of all ages pursue younger females.

‡ This is necessarily true of men choosing to strictly adhere to
monogamous marriage.

± Competition drives men to use whatever connections, advantages,
and alliances they have in order to obtain wives.

± Once girls become wives, older husbands will strive to protect
their young wives from other males dampen women s freedoms & exacerbate inequality.

‡ This effect is further exacerbated by age gaps.
12/10/10 Henrich, Aff #1, page 49 4444

Three lines of evidence
1) Macro-level data: examine macro level (country level) data,
comparing highly polygynous countries with

± less polygynous countries ± monogamously marrying countries from between 20 degree
south latitude and 20 degrees north latitude (i.e., tropical non-African nations).

2) Micro-level data: examine four case studies and compare the
age of first marriage for women and age gap between married people in polygynous and monogamous marriages.

3) Sex ratio as a proxy for the effects of polygyny on women and
gender equality.
12/10/10 Henrich, Aff #1, page 49 4545

Macro-level Data
Table 6. Comparison Data from Highly Polygynous, Less Polygynous and Comparable Monogamous Countries North Highly Comp. Less America/Wes Polygynous Monogamous Variables tern Europe (>10% married Polygynous Countries men) # Countries 28 20 58 24 Female age at first 19.9 22.7*** 25.0*** 29.6*** marriage Age gap (with first 6.4 3.9*** 2.8*** 2.4*** wife only) Total fertility 6.78 5.97** 4.62*** 1.84*** Child mortality 19.4% 18.3% 11.6%** 1.4%*** rate, 1980 Infant mortality 12.2% 11.5% 6.9%** 1.2%*** rate, 1980 GDP per capita, $975 $1574* $2798*** $11,950*** 1985 12/10/10 Henrich, Aff #1, page 50 4646

Micro-level Data
Table 8. Age at first marriage among Bedouin Arabs Women Variable Avg. age at 1st marriage Avg. # of years younger than husband Monogamous marriage Polygynous marriage 


19.46 3.49


Click to edit the outline 22.95 26.99* text format

Monogamous marriage

Polygynous marriage


Second Outline Further polygynously marrying men are seeking out younger first wives Level
Table 9. Age at first marriage in a municipality in south eastern Turkey  Third Outline Monogam Polygynous- Polygynous- Statistical ous wives senior wives junior wives significance Level Avg. age at 1st marriage % of women married under age 15 12/10/10 17 15 18 

Not Fourth Outline significantly different




Henrich, Aff #1, page 53

Level  Fifth p=0.01 Outline 4747 Level

Table 11. Ages at first birth and age difference between spouses among Aboriginals in Arnhem Land, Australia Monogamous  Polygynous editStatistical Click to the outline marriages marriages significance text format Avg. age difference between husbands 7 17.1 p=0.0001  Second Outline and wives Avg. mother s age at 1st birth Avg. father s age at 1st birth

19.32 28.71 19.19  36.27 

not statistically Third Outline different

Level Fourth Outline Level  Fifth Outline 4848 Level


Henrich, Aff #1, page 54

Gender Equity?
‡ UNDP s Gender
Empowerment Index
Table 7: Female Equity Comparison Highly Polygynous Comparable Monogamous

± Male/female income


± Female
representation in high status jobs

Gender Empowerment Measure (GEM), 2003 Ratio of adult female to male literacy rates, 2005



± Canada ranked 4th in
2009 (0.83).




Henrich, Aff #1, page 51


Effects of polygyny on gender equality
‡ Use sex ratio as a proxy to determine the effects of
polygyny on women.

± High sex ratio will increase competition for women.  Age of marriage will go down and age gap expands  Men will tighten their control on wives, sisters and
daughters. Higher sex ratio mean less gender equality



Effects of polygyny on marriage age and age gap
‡ An analysis of 117 countries found that in countries with a high
ratio of males to females, females married younger (South and Trent 1988).

‡ The rising sex ratio in China has caused rich families to acquire
infant girls to guarantee their sons have wives (Hudson and den Boer 2004). NOTE the typo ( declining sex ratio , pg. 55).

‡ Similarly, in some regions of India more than half of females in
some regions are married before age 15 (Burns 1998).

‡ In the American frontier where females were in short supply,
brides were reported as young as 12 and 13 (Courtwright 1996).

‡ This converges with our findings above, indicating that
competition for scarce females drives the age of first marriage down.
12/10/10 5151

Cross-national analysis: sex ratio and gender empowerment
‡ South and Trent analyzed data from 117 countries around the
world. The sample included countries from the full spectrum of development, but with a bias towards more developed countries.

‡ The sex ratios based on number of males and females between
the ages of 15 and 49 from any year between 1973 and 1982

‡ Analyses controlled for ± the reliability of the sex ratio data for each country ± the socioeconomic development of each country ‡ an index composed of variables including GDP per capita,
infant mortality, percentage of population living in urban areas, and life expectancy.
12/10/10 5252

Cross-national analysis: sex ratio and gender empowerment
‡ Higher sex ratios (i.e., more males than females) predict ± lower participation of women in the labour force ± lower illegitimacy rates (women more frequently have
babies outside of marriage when they have more power)

± lower divorce rates (women can divorce)


Henrich, Aff #1, page 55-56


Apply to developed countries?
‡ Limit sample to developed countries ONLY ‡ sex ratio had a greater effect on indicators of women s
roles than in less developed countries, with the exception of participation in the labour force.

‡ In more developed countries, higher sex ratios predict for
women a

± (a) lower age at first marriage, ± (b) higher fertility rate, and ± (c) lower literacy
12/10/10 Henrich, Aff #1, page 56 5454

The Role of Sexual Jealousy and Age Disparity
‡ Breitman and Shackelford (2004) show that in Chicago,
controlling for a wide range of other factors, the age gap between husbands and wives is strongly related to homicide rates.

‡ The larger the age gap, the more likely it is that a husband will
kill his wife, and vice-versa (the young wife murders her husband).

‡ The effect is highly non-linear: in moving from spouses being the
same age to husbands being between 13 and 15 years older, the rate increases from 5 per 100,000 to 8 per 100,000. ‡ Then, it spikes to almost five times the same-age rate. 

Domestic violence will be worse in polygynous societies because
of the age gap. 

FLDS age gaps of 16 years are not uncommon
12/10/10 Henrich, Aff #2, page 7-8 5555

North America?
A 15 year long ethnographic study in the FLDS community in Colorado City/Centennial Park in the U.S. reveals similar dynamics. Anthropologist William Jankowiak writes (p.171): There is a shortage of eligible women to marry in every polygynous society, and this is a primary factor responsible for intergenerational conflict in Colorado City/Centennial Park. Senior males are always on the marriage market and thus compete with younger men for mates in a limited pool of eligible women in the 1960s a local policeman, without approval of the religious leadership, would threaten to arrest unmarried males who did not leave the community The competition for mates is acute Young men know that if they do not find a girlfriend before they graduate from high school, they probably never will have one. Without a girlfriend, they will leave the community to find a wife (p. 172-173).

12/10/10 Henrich, Aff #2, page 9


North America
For those possessing contemporary Western values regarding gender equality it may be difficult to imagine how educated, modern North Americans could ever come to compete for women, or treat them as resources to be divided up. In this setting, fathers often exchanged daughters in order to marry them men wanted to marry off their daughter before they could decide to select from within their age cohort. By the 1990 s Second Ward fathers began to negotiate marital exchanges not for themselves but for a favorite son, or in some cases a grandson (p. 171). Within the First Ward, families also gave/give daughters as a kind of patronage to the prophet. Jankowiak writes: The prophet s age does not restrict families from offering their daughters to him the reason why fathers give their daughter to the prophet (often with a wife s encouragement) are to gain prestige and to obtain material and spiritual benefits (p. 171).
12/10/10 Henrich, Aff #2, page 9 5757

A What If Economic Model of Imposing Monogamy
‡ Tertilt developed an economic decision-making model using
standard modeling tools from economics to examine what gives rise to polygyny, how it affects an economy, and what impact, if any, imposed monogamy has (2005).

‡ Her model assumes both men and women care about both having
children and consuming, but that men can continue to reproduce their entire lives while women are limited to only a portion of their lives

± She also assumes that men tend to prefer younger women. ‡ She shows that her model produces polygynous mating patterns
under a wide range of conditions, and that it can produce results that match real-world patterns related to age-gaps, fertility, and saving rates for polygynous countries.
Henrich, Aff #1, page 32 12/10/10 5858

Imposing Monogamy
‡ She then calibrates the model to the Highly Polygynous data
shown above (and other related data) and then asks the question of what happens when she imposes monogamy on everyone (in the model). ‡ The result is that ± the fertility rate goes down ± the age gap goes down ± saving rates go up, bride prices disappear ± and GDP per capita goes way up. ± This occurs because men (in the model) can t invest in obtaining wives or selling daughters (which they do massively, otherwise), and instead they save and invest in production and consumption.
Henrich, Aff #1, page 33 12/10/10 5959

What about empowering women?
‡ Tertilt (2006) uses the same model to compare the effects of
imposing monogamy legally to the effect of increasing the power of women.

‡ Tertilt alters her model to shift reproductive decision-making from
men to women, but leaves polygyny legal.

‡ In this female-choice model, ± The #of wives per husband declines a little bit, as does fertility. ± GDP per capita also goes up some ± Savings rates go up substantially. ± Overall, however, empowering women does not have nearly the
impact on GDP per capita and fertility as imposing monogamy.

± This underlines the point made above that giving women free
choice does not necessarily yield monogamy, low fertility, or economic growth, though it does yield less intensive polygyny.
12/10/10 Henrich, Aff #1, page 33 6060

Child mortality: North America
‡ Low male parental investment, low relatedness, and
high internal conflict mean influence the health of children.

‡ 19th century Mormon communities in Utah (Heath and
Hadley 1998)

‡ Compare the family composition and child survival data
from 90 households consisting of 45 headed by wealthy men (top 2% of wealth in that community) and 45 headed by poor, but still married, men (from the bottom 16%).


Henrich, Aff #1, page 47


Child Mortality: North America
‡ The first thing these data show is that wealthy males
had on average 3.2 wives compared to 1.4 among the poor.

‡ All but five of the wealthy men had more than one

± One rich man had 11 wives. ‡ Overall, the wealthy men controlled 120 women while
the poor men controlled 63. This means that 90 husbands had 183 wives, which implies roughly 93 missing men had no wives.

‡ Ergo, the pool of unmarried low status men.
12/10/10 6262

Child Mortality: North America
‡ Wealthy men had more total offspring and longer
reproductive careers (33 years for wealthy men compared to 22 years for poor men)

‡ The children of poor men had better survival rates to age

± For poor men, 6.9 of their offspring (per wife) survived
on average to age 15.

± For wealthy men, only 5.5 of their offspring survived. ‡ This is amazing, given that the poor men had less than 10%
of the wealth of the rich men, and the rich men had significantly more total offspring (including those that did not make it to 15).
12/10/10 Henrich, Aff #1, page 47 6363

It s not the dilution effect alone
‡ Compare the poor men with one or two wives with the
rich men with one or two wives.

‡ Among men with one or two wives, ± For these poor men 6.9 children per wife survived ± For these rich men 5.7 children survived per wife ‡ This supports the idea that poor men with limited
resources for another wife tend to invest more in their existing offspring while rich men with the same number of wives invest less in offspring because they are busy seeking additional wives.
12/10/10 Henrich, Aff #1, page 47 6464

Child Health and Mortality
‡ These historical North American patterns are similar to those
observed in recent studies of contemporary polygynous African societies.

‡ Children from polygynous families have an increased risk of
diminished nutritional status, poor health outcomes, and higher mortality.

‡ Quantitative studies in Tanzania and Chad found that children in
polygynous households had poorer nutritional status than their counterparts in monogamous households, as indicated by the children s height and weight measurements (Begin et al. 1999; Sellen 1999; Hadley 2005).


Henrich, Aff #1, page 48


Cross-culturally: Child Health and Mortality
‡ In Hadley s (2005) Tanzanian study, despite these favourable
conditions, the children of polygynously married mothers were more likely to be underweight, and were relatively shorter and gained less weight and height during the study than children of monogamously married mothers.

‡ In Sellen s (1999) Tanzanian study, children of polygynous
mothers had lower weight for age scores (WAZ) and height for age scores (HAZ) than children of monogamous mothers.

‡ The analyses controlled for wealth and child and maternal


Henrich, Aff #1, page 48


Cross-culturally: Child Mortality
‡ Children in polygynous families are also at an elevated risk of
mortality compared to children in monogamous families.

‡ A study using data from 22 sub-Saharan African countries found
that polygyny is a significant risk factor for child mortality (Omariba and Boyle 2007). ± Children in polygynous families were 24.4% more likely to die compared to children in monogamous families.

‡ A study of six West African countries found that infants in
polygynous families had a 60-70% greater risk of dying compared to children in monogamous families (Amey 2002).

‡ Among the Dogon of Mali (Strassmann 1997), children (under
age 10) in polygynous households were 7-11 times more likely to die than their counterparts in monogamous households. ‡ Not due to wealth dilution.
12/10/10 Henrich, Aff #1, page 48 6767

Cross-culturally: Child Mortality
NO Mortality Differences?

‡ Anderson (2000) South Africa: cited on benefits of polygynous

‡ 22 monogamous and 22 polygynous ‡ Surviving children per wife ± Monogamous 4.01 ± All Polygynous 3.87 ± Sororal Polygyny 4.7 ± Non-close relative 3.8 ‡ Claim is 4.01 and 3.8 are actually equal. ‡ No p-value, no statistical information at all. ‡ # of studies found showing better survival rates in polygyny = 0
12/10/10 Henrich, Aff #1, page 48 6868

Women s Psycho-Social Impacts
‡ Internal conflict vs. co-wives cooperation?
Studies showing NEGATIVE impacts of polygyny vs. monogamy

‡ Studies among Arabs in Israel (Al-Krenawi and Graham 2006) and
in Turkey (Ozkan et al. 2006) found significantly higher rates of psychological distress and disorders among polygynously married women compared to their monogamously married counterparts.

‡ Among the disorders/distress experienced at significantly
elevated rates by polygynously married women in the Arabic sample are depression, obsession-compulsion, hostility, anxiety, phobia, psychoticism, and paranoid ideation (Al-Krenawi and Graham 2006).
12/10/10 Henrich, Aff #1, page 58 6969

Women s Psycho-Social Impacts
‡ Women in polygynous marriages also reported significantly more
problems in family functioning and marital relationships and less satisfaction in life than monogamously married women in their societies (Al-Krenawi and Graham 2006).

‡ In Turkey, the increased likelihood of having a psychological
disorder among senior wives compared to monogamous wives was 1.6 times for conversion disorder (blindness, paralysis, fits) and 2.4 times for somatisation disorder (headaches, pain, etc.).

‡ The other disorders were not significantly different in prevalence
between monogamous and polygamous wives.

‡ Wife order mattered, and impact of order varied cross-culturally
Henrich, Aff #1, page 59



Women s Psycho-Social Impacts
‡ Internal conflict vs. co-wives cooperation?
Studies showing NO impacts (polygynous vs. monogamous)

‡ A study among East Africans did not find any difference in
rates of anxiety or depression between women in polygynous versus monogamous marriages (Patil and Hadley 2008).

‡ Could be a wealth effect more wives generate more
economic production. Studies showing POSITIVE impacts of polygyny

‡ Failed to locate any such studies
12/10/10 Henrich, Aff #1, page 59 7171


On the origins monogamous marriage Cultural group selection for social norms/laws that have group
beneficial effects. It s spread because it gives an advantage to a society in competition with other societies. and paternal investment elements of our evolved psychology to stabilize pair-bonds, suppress mating efforts by males, and channel males energies into productive labor and child investment.

‡ Monogamous marriage has evolved to harnesses the pair-bonding

± Reduces crime (benefits commerce, frees revenues, etc.) ± Maintains internal harmony ± Increase solidarity/equality by leveling social differences. More
equal societies experience better social outcomes, longevity, etc. May sow seeds of democracy in process. ± Favoring gender equality by reducing mate competition

± Promotes investment in offspring & saving economic growth.
12/10/10 Henrich, Aff #1, page 59-60 7272

Table 5: Household Arrangements Marriage Arrangements Independent nuclear, monogamous Independent nuclear, polygyny Preferentially sororal, cowives in same dwelling Preferentially sororal, cowives in separate dwellings Non-sororal, cowives in separate dwellings Non-sororal, cowives in same dwellings Independent polyandrous families Missing data Ethnographic Atlas % (N = 1267) 14.6% (186) 35.7% (453) 5.4% (69) 1.4% (18) 27% (344) 12.4% (157) 0.32% (4) 2.8% (36)



Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.