Back to the EvMedEd Navigator

The Obstetrical Dilemma


Out of all primates, humans experience some of the most dangerous births. Without access to modern medical care, 0.5 - 3% of births result in the death of the mother, and up to 15% of births result in the death of the infant. Chimpanzee birth appears to be much less risky, especially for the mother, though precise mortality data is not available. One explanation for this difference is that the chimpanzee neonate’s head is much smaller than the mother’s pelvis, allowing it to pass through the birth canal with little chance of becoming stuck. Meanwhile the human neonate's head is approximately as large as the mother’s pelvis, which makes for a tight fit during birth. The risk doesn’t end there. After a successful birth, even healthy human newborns are fragile and have only a fraction of the motor function that newborns of our closest primate relatives possess. Historically, around 27% of human infants die during their first year of life. 

These aspects of human reproduction might seem self-evident. However, the differences between humans and other apes are somewhat puzzling with regard to natural selection. Chimpanzees demonstrate that it is possible to have relatively safe births and precocial infants, both of which seem like they should be advantageous in terms of survival and reproduction. Yet humans have dangerous births and secondarily altricial infants. What ultimate explanations can account for this counterintuitive evolutionary outcome?

Phenotypic Trade-offs

According to one popular hypothesis, both risky childbirth and secondary altriciality result from a phenotypic trade-off acting on the breadth of the female pelvis. This hypothesis is called the obstetrical dilemma (OD), and it has been a leading explanation for these human traits since it was first proposed in 1960 (Washburn 1960). The obstetrical dilemma proposes that a trade-off exists because the female pelvis has two competing roles: birthing large-brained infants and bipedal walking. Intuition suggests that the need to safely birth large-brain infants should favor a wide pelvis, while energetically efficient bipedal locomotion should favor a narrow pelvis. According to the OD, selective pressures for energetically efficient locomotion have prevented the evolution of a female pelvis wide enough to safely birth infants with very large brains. This is used to explain why the neonate’s head can barely fit through the mother’s pelvis during birth, and why humans are born with only around 30% of their adult brain size, instead of 40% like chimpanzees (Dunsworth 2018). 

Sex-specific evolution

The pelvis is one of the most sexually dimorphic features of the human skeleton, and the OD has a number of implications for the causes of this dimorphism. The most apparent one is that sexual dimorphism in pelvic breath exists because the male pelvis can be optimized for locomotor efficiency, while the female pelvis is the product of a trade-off between two opposing functions. This implication leads to a testable prediction; we should expect males to display greater energetic efficiency (on average) than females while walking or running. But what if pelvic breath is not closely tied to locomotor efficiency? 

Hypothesis Testing

In recent years, a number of studies have examined the underlying assumptions of the OD hypothesis. As it turns out, evidence in support of the OD has been sparse. One study tested whether a trade-off exists between pelvic width and locomotor efficiency and found no evidence to indicate that female pelvises are less efficient during locomotion than male pelvises (Warrener et al. 2015). This finding challenges the OD hypothesis, but it also brings us back to the drawing board. If bipedal locomotion doesn’t create selection for a narrower pelvis, then why do sex differences in this trait exist? Why don’t females and males alike possess a pelvis wide enough to birth an infant with 40% of their adult brain size?

One study is generally not enough to completely refute a hypothesis, but other assumptions of the OD have also been tested. For example, the OD suggests that humans are born with smaller-than-expected brains and experience uniquely dangerous births as compared to other primates. However, when we compare neonatal brain size to adult brain size across Old World Primates, then the 30% ratio found in humans is what would be expected given the size of our adult brains. Chimpanzees follow the same trend as humans when it comes to prenatal brain development; they have the second-largest adult brains among primates, and complete the second-lowest percent of brain growth before birth (Dunsworth 2018). Furthermore, humans are not unique in experiencing  a “tight fit” between the neonate’s head and the mother’s pelvis; several other primate species, including gibbons and macaques, face a similar obstetric challenge despite being quadrupedal (Rosenberg and Trevathan 1995).

Constraints on Natural Selection: The EGG hypothesis

While these findings do not completely refute the obstetrical dilemma, they do indicate that understanding the unique qualities of human birth may require considering additional hypotheses. One hypothesis that has gained popularity among researchers is the Energetics of Gestation and Fetal Growth (EGG) hypothesis (Dunsworth et al. 2012). The driving idea behind the EGG hypothesis is that the evolution of gestation length in humans is constrained by the energetics between mother and offspring. According to this hypothesis, birth occurs when the growing energetic requirements of the fetus approach the limit of what the mother can provide through the placenta. The EGG proposes that it is largely a matter of coincidence that the fetal energy requirements reach this limit around the time when the fetus is barely able to fit through the mother’s pelvis. The EGG hypothesis does not deny that sex differences in pelvic shape were likely influenced by selective pressures related to childbirth. It instead challenges the idea that humans would be born with a chimp-like 40% of their adult brain size if not for a selective pressure that favors a narrow pelvis. It argues that human gestation length is governed by energetic constraints that are present across mammals, and requires no unique explanation.

Principles this example illustrates:

Phenotypic Trade-offs

Though it is losing its status as a leading hypothesis, the obstetrical dilemma is nevertheless useful as an exercise in considering how competing evolutionary pressures might shape a single trait (pelvic breadth), and what other consequences such a trade-off might have for the organism. Given its relatively intuitive nature, the obstetrical dilemma presents a good opportunity for students to practice developing a hypothesis using pieces of starting information and the principle of phenotypic trade-offs. The instructor could, for example, provide students with a list of traits/outcomes related to pelvic breath and the question of how natural selection might maximize fitness with regard to maternal & infant survival as well as locomotor efficiency. The students’ task would then be to identify which outcomes (1) are likely to impact fitness, (2) are likely to be a fitness benefit associated with a narrow vs. wide pelvis, and (3) are likely to be a fitness cost associated with a narrow vs. wide pelvis. 

Sexual Selection: Causes of Sex Differences

Both the development and critiques of the obstetrical dilemma present an avenue for exploring evolutionary explanations for anatomical sex differnces. One exercise for doing so can involve students going through the hypothesis-testing process seen in (Warrener et al. 2015). In such a lesson, the instructor can encourage students to make predictions about what relationships we might expect to see between pelvic breath and locomotor efficiency, use these predictions to interpret the data from Warrener et al. 2015, and then make conclusions about about what these result mean for the obstetrical dilemma. For a further challenge, the instructor can ask students to identify ways to strengthen the study, as well as what questions remain to be answered if there is indeed no correlation between pelvic breadth and locomotor efficiency.

Additional resources:




Journal articles

    1. Dunsworth, Holly M., Anna G. Warrener, Terrence Deacon, Peter T. Ellison, and Herman Pontzer. 2012. Metabolic Hypothesis for Human Altriciality. Proceedings of the National Academy of Sciences 109(38): 15212–15216.

    2. Dunsworth, Holly, and Leah Eccleston. 2015. The Evolution of Difficult Childbirth and Helpless Hominin Infants. Annual Review of Anthropology 44(1): 55–69.

    3. Dunsworth, Holly M. 2018. There Is No “Obstetrical Dilemma”: Towards a Braver Medicine with Fewer Childbirth Interventions. Perspectives in Biology and Medicine 61(2): 249–263.

    4. Grabowski, Mark W. 2013. Hominin Obstetrics and the Evolution of Constraints. Evolutionary Biology 40(1): 57–75.

    5. Rosenberg, Karen, and Wenda Trevathan. 1995. Bipedalism and Human Birth: The Obstetrical Dilemma Revisited. Evolutionary Anthropology: Issues, News, and Reviews 4(5): 161–168.

    6. Warrener, Anna G., Kristi L. Lewton, Herman Pontzer, and Daniel E. Lieberman. 2015. A Wider Pelvis Does Not Increase Locomotor Cost in Humans, with Implications for the Evolution of Childbirth. PLOS ONE 10(3): e0118903.

    7. Wells, Jonathan C. K., Jeremy M. DeSilva, and Jay T. Stock. 2012. The Obstetric Dilemma: An Ancient Game of Russian Roulette, or a Variable Dilemma Sensitive to Ecology? American Journal of Physical Anthropology 149(S55): 40–71.

    The International Society for Evolution, Medicine & Public Health is a 501(3)(c) non-profit organization. Copyright (c) 2018. Contact:
    Powered by Wild Apricot Membership Software