Anthropology 174AW

Sexual Practices & Harem Size

This webpage explores the correlation between despotism and harem size with sexual attitudes and practices among the various cultures. My hypothesis is that there seems to be a strong correlation between sex believed to be dangerous and maximum harem size. My assumption is that people who tend to believe that the act of sexual intercourse is dangerous may withdraw from being an active participant. This may in turn, result in their spouse looking to alternatives means to satisfy their sexual needs and desires. One option is have harems. As the needs of the spouse are continually unmet, the harem size may increase depending on the maximum amount of harems allowed. The maximum harem size varies according to the distinct cultures. This webpage will explore the various cultures that support this theory.

What is a Harem?

Harem can be defined as a secondary wife of inferior social status, in certain polygamous societies. The word harem is derived from the Arabic word harim, meaning "that which is prohibited".


Correlations which run from -.5 to Ė1.0 or from +.5 to +.1 show a very close connection between two variables. After running a crosstabulation on the variables V161 Sex Believed Dangerous and V1133 Maximum Harem Size, I found that the two variables had a very strong postive correlation of .541*. This displays the correlation at a significance level of .05 on a two-tailed distribution. Following is a graph of the results of my crosstabulation.


V1133 Maximum Harem Size * V161 Sex Believed Dangerous

The results of the above graph illustrates the relationship between two existing variables. If sex is not believed to be dangerous, then the ratio remains three out of five men who will have a maximum of three conjugal relationships or less. Then if the sample size is increased to eight men, the odds are half, four out of eight men have a maximum of four to ten conjugal relationships. This is with the variable sex believed to be dangerous then the odds are two out of three men will have a maximum of 11 to 100 conjugal relationships.

V200 region Crosstabulation


The above graph includes the variable V200 Region, as well as V161 Sex Believed Dangerous and V1133 Maximum Harem Size. This graph displays the areas of low, or no harems and high or concentrated regions with harems. The regions with low or no harems are East Eurasia, Insular Pacific, North America, and South America. As it turns out, lots of low danger, no harems are in the New World. The regions with high or localized harem are Africa, and Circum-Mediterranean. These two regions have the high danger harem cases. As a result, there is a possibly that these are just large cultural regions that have these characteristics, which may lower the statistical significance level.


I wanted to place two very interesting maps on this webpage, but I was unable to do so. However, I would like you to see the actually regions where harem exists. If you just click on these sites it will take you there. (



If youíre interested in learning more about related topics, here is a list of references. However, Iíll have to warn you, I had some trouble finding these sources. Good Luck!

Betzig, L. L. (1982). Despotism and differential reproduction: A cross-cultural correlation of conflict asymmetry, hierarchy and degree of polygyny. Ethnology and Sociobiology, 3(4), 209-221.

Betzig, L. (1986). Despotism and differential reproduction: A Darwinian view of history. New York: Aldine. Introductions and explanations of the variables by author were published in world cultures, 1988, Volume 4, Number 4.

Betzig L. (1992). Little history of Darwinian history. Ethnology and Sociobiology, 13(5-6), 303-307.

Betzig, L. (1989). Rethinking ethnology: A response to some recent critiques. Ethnology and Sociobiology, 10(5), 315-324.

Betzig, L. (1992). Roman polygyny. Ethnology and Sociobiology, 13(5-6), 309-349.

Broude, G. & Greene, S. J. (1976). Sexual attitudes and practices. Ethnology, 15, 409-429.