Why is childbirth so hard for humans – and is it getting even harder?

The female pelvis may become too narrow for vaginal childbirth, meaning that caesareans could become the only option for delivering babies. At least, that’s the claim made by a research team last year. Their calculations suggest that in Australia, Mexico and Poland, the average female pelvis is now 4.2 centimetres narrower than it was in 1926. If the trend continues, they wrote, surgery may be the only way for our species to continue.

Untangling whether this bold and controversial claim has merit means exploring one of human evolution’s most famous – and most eyebrow-raising – hypotheses: the obstetrical dilemma. The idea is that an evolutionary battle has been raging over the female pelvis for millions of years, with the structure being simultaneously under pressure to be narrower to assist with walking on two legs and wider to make childbirth easier.

The new claim is based on an updated version of the dilemma, which argues that modern medical technology allows women to give birth no matter the shape of their pelvis. This, it is suggested, has redrawn the battle lines in a way that allows for the narrowing of hips. In the past, a narrow birth canal or a too-large baby would have probably proved fatal for mother and infant in childbirth. But because caesareans reduce this evolutionary pressure, the genes that code for narrow hips can be passed on.

Things, of course, are rarely so simple. Over the past 15 years, some researchers claim to have found evidence against the hypothesis, while others offer evidence in its favour. Yet more have broadened it out to incorporate surprising factors, from the advent of farming through to changes in our modern diet. The big question is, have their efforts brought us any nearer to explaining why childbirth is often so difficult – and to predicting whether it will become even more so in future?

Evolution of a narrow pelvis

Childbirth is notoriously painful. It is also potentially dangerous, claiming tens of thousands of lives each year. Countless more women are left with life-changing injuries, including pelvic organ prolapse. “It’s astonishing that in 2026, with all our medical technology, maternal and fetal mortality and morbidity rates are still so high – including in wealthier countries,” says Nicole Webb, a palaeoanthropologist at the Senckenberg Research Institute and Natural History Museum in Germany. This seems to make little sense. After all, childbirth is critically important for the survival of our species. “As an evolutionary biologist, it is difficult to reconcile the fact that the very process that enables our reproductive fitness can also hinder it,” says Webb.

For decades, the obstetrical dilemma has been the go-to explanation for why childbirth is such a risky process. Although the roots of the idea can be traced back to the early 20th century, it was first properly described only in 1960, in an article published by the late physical anthropologist Sherwood Washburn. Curiously for such an influential idea, it didn’t appear in an academic journal, but in the pages of a popular science magazine, Scientific American, where it was mentioned only briefly in an article primarily concerned with the role of tools in human evolution.

Here, Washburn suggested that our ancient ancestors evolved a narrower pelvis – and by association a narrower birth canal, the vertical opening running through the pelvis – to allow for more efficient walking on two legs. Then, however, our ancestors also underwent selection for larger brain size. This created a problem – Washburn’s signature obstetrical dilemma – because large-brained adults begin life as large-brained babies, and babies with heads above a certain size can’t fit through a narrow birth canal.

For Washburn, evolution solved this dilemma by delivering babies at an earlier stage of development, when they were somewhat smaller – although not too much so, given that survival rates fall as birthweight drops. This solution had profound knock-on effects, he wrote, because those babies were then far more immature at birth and so required much more maternal care. Consequently, new mothers became helpless themselves, reliant on the males in the social group to hunt and provide for them. This, wrote Washburn, fundamentally shaped how humans organise socially, with a firmly patriarchal slant.

Given this conclusion, the obstetrical dilemma isn’t the most enlightened of hypotheses. In fact, many researchers now question the assumption that ancient women were incapable of hunting or becoming powerful members of society. Still, the basic idea that the female pelvis is shaped by these two competing factors is accepted by lots of people working in the sphere of evolution, despite being frustratingly ambiguous. Even today, for example, there is no consensus on whether “efficient” walking should be measured in terms of the energy needed to move, or simply in terms of how easily – and, in particular, how quickly – someone can get from point A to point B.

A caesarean-only future?

All this means that testing the obstetrical dilemma is difficult. Gathering the necessary data is challenging, as this ideally includes detailed pelvic measurements of a large number of individuals whose behaviour and medical history are then tracked through time. Last year, though, Marianne Brasil, an anthropologist and biologist at Western Washington University in Washington state, and her colleagues explored the dilemma with a sample of more than 30,000 people – both male and female – using data from the UK Biobank, a biorepository that contains medical and lifestyle information from a large cohort of people living in the UK. The researchers found that people with larger birth canals had slower self-reported walking speeds, in line with the obstetrical dilemma.

But Anna Warrener at the University of Colorado Denver sees problems with that conclusion. Not only are self-reported walking speeds potentially unreliable, she says, but Brasil and her colleagues had to use the low-resolution X-ray scans contained within the UK Biobank to measure birth canal size. These scans provide only a face-on, two-dimensional picture of the pelvis, whereas the birth canal is – obviously – three-dimensional. Brasil concedes that this was a limitation of the study.

A further issue is that the obstetrical dilemma assumes that the shape of the pelvis is controlled only by walking efficiency and childbirth capacity. Over the past 15 years, that assumption has been challenged.

Some researchers say that pelvic floor health is also important, to take one example. A narrower pelvis might offer better support for the internal organs – and for the fetus during pregnancy. In line with this idea, Brasil and her colleagues found that individuals in the UK Biobank with wider birth canals were more likely to experience pelvic organ prolapse.

Researchers, including Brasil, think this could influence the evolution of the pelvis. The argument is that women who develop pelvic floor problems during their first experience with childbirth may be unwilling or unable to go through the process again, and so contribute fewer genes to the next generation.

On top of these potential additional complexities, there is a broader problem with the obstetrical dilemma: if, as Washburn suggested, it was solved millions of years ago, why is childbirth so hard for many women today? A decade ago, theoretical biologist Philipp Mitteröcker at the University of Vienna in Austria and his colleagues suggested an explanation. They argued that the key to the mystery is to remember that natural selection works at the level of populations. “It doesn’t optimise individual health,” he says.

At the population level, he says, there has been selection to ensure that the average female pelvis is large enough for childbirth. But because pelvic shape varies naturally across a population, some women inevitably develop a pelvis that is simply too narrow for vaginal childbirth. So, while such childbirth is possible – if painful – for most women, it is simply impossible for others, with terrible and even potentially fatal consequences, says Mitteröcker.

In their study, Mitteröcker and his colleagues then threw yet another variable into the mix: our sophisticated human culture. Over the past few centuries, medics have developed obstetrical instruments and procedures that can help deliver even uncomfortably large babies – and, since the mid-20th century, caesareans have become increasingly available for the safe delivery of babies who won’t fit through the birth canal.

Mitteröcker and his colleagues argued that these cultural inventions may have relaxed the selective pressure on women to grow a large pelvis. A mathematical model the team developed suggests that, since the 1950s, there may have been a 20 per cent increase in the number of women with a pelvis too narrow for childbirth. “Everybody thinks in terms of thousands or millions of years when talking about evolution,” says Mitteröcker. “But to increase or decrease certain body dimensions, that can go pretty fast.”

His view is far from accepted, however. Others are sceptical that we would see so much change over such a short period. “I would expect that it would take, very likely, much more than a few hundred years, especially with our relatively long lifespans,” says Brasil. Things become murkier still when you consider that childbirth itself has become a divisive topic in mainstream society, and the obstetrical dilemma – even if unintentionally – could filter into the debate.

One of the issues here, Warrener wrote in a 2023 paper, is that childbirth became increasingly medicalised during the 20th century. To some, it has become over-medicalised. To counter this trend, there are now advocates for “natural” childbirth without medical help – some extreme versions of which have been linked with baby deaths. This general trend, wrote Warrener, can leave pregnant women in a difficult position: expressing doubts about any medical intervention might mean being called “ignorant” by some in society, but if they ultimately need medical assistance during birth, they may be labelled a “failure” by others.

For Webb, though, the obstetrical dilemma hypothesis is valuable here because it can help women contextualise any problems they may experience during childbirth. “I think one should be able to say that there are evolutionary forces at play that can explain birth difficulty,” she says. “We want to empower women to know that if you can’t give birth naturally, it’s not your fault.”

Which brings us back to the study on the remarkable narrowing of pelvises in Australia, Mexico and Poland, and the suggestion that evolutionary forces will eventually leave no woman able to give birth without surgical intervention. Mitteröcker is convinced that the study, conducted by Maciej Henneberg at the University of Adelaide in Australia and his colleagues, has identified a real trend, and that the team is correct to tie it to changes in medical care during childbirth. “To show this in three different populations in three different countries is, I think, convincing,” says Mitteröcker. Webb, too, says that the researchers are on to something with their results.

But neither Mitteröcker nor Webb thinks this will lead to a caesarean-only future. This is partly because any selective pressure for women to develop a narrower pelvis is probably weaker today than it was in the past. For instance, walking efficiently was vital for early humans on the savannah, but it is less important for modern humans, whose lives are far more sedentary. Even Henneberg is softening his initial bold claim: “I understand the scepticism,” he says. He adds that he really meant to imply that “practically all” births could require medical assistance with tools and techniques such as forceps or vacuum extraction, as well as just caesareans.

Diet’s impact on childbirth

As we have seen, some researchers, including Brasil, take this scepticism further, questioning whether new medical practices would prompt evolutionary change on a timescale of decades. But if such changes would take longer to register, how else could we explain the pelvic narrowing that Henneberg and his team reported? Some researchers think it is worth considering one final factor: diet.

Although there is now strong evidence that pelvic proportions are heritable, as implied by the obstetrical dilemma, we also know that genetics isn’t the whole story. Even someone with genes favouring a wide pelvis may develop a narrow one if they experience a vitamin D deficiency during childhood, for example.

Diet more generally also influences pelvic development. Jonathan Wells, an anthropologist and paediatric nutritionist at University College London’s Great Ormond Street Institute of Child Health, says this suggests that the advent of farming roughly 12,000 years ago was a watershed moment in the history of childbirth. Abandoning the hunter-gatherer lifestyle led to a diet richer in carbohydrates and calories. This is generally agreed to have contributed to a worsening of health that restricted growth during childhood, while also, strangely enough, promoting fetal growth during pregnancy. The result, Wells and his colleagues argue, is that farming left women with smaller birth canals, but larger babies to deliver, leading to greater childbirth difficulties.

Wells speculates that the link between diet and childbirth may even help explain one of the greatest mysteries of recent human evolution. About 7000 years ago, some humans developed the ability to digest dairy milk well into adulthood. This “lactase persistence” then spread through many populations over just a few millennia. “It was like wildfire,” says Wells. “People have long been wondering why.” A few years ago, he and his colleagues said childbirth might have been a factor. A dairy-rich diet is more nutritious, promoting skeletal growth – including the bones defining the birth canal. As such, being able to digest dairy milk may have translated to fewer women dying in childbirth, he says, meaning that consuming dairy spread as a practice, because it brought a dramatic boost to communities.

Researchers are still exploring the connection between diet and childbirth, but some societies seem to have long been aware of the link. Last year, Wells and his colleagues published a study exploring dietary food choices during pregnancy in Madagascar. They discovered that cultural taboos on the island discourage pregnant women from eating too much bread or other carbohydrate-rich food precisely because islanders are aware that the foods promote fetal growth and make childbirth harder.

Unfortunately, such local knowledge might be losing its power in today’s connected world. Ultra-processed foods that are high in calories but low in important nutrients are now widely available, contributing to a global rise in obesity. Wells thinks this might help explain why death during childbirth remains a real risk in low-income countries. Because processed foods are cheaper than healthier foods and readily available, women might struggle to eat a nutritious diet, which Wells says has exacerbated childbirth difficulties in these countries, leading to what he has dubbed a new obstetrical dilemma.

Even if he is correct, however, there may still be a positive message here: if it ever proves possible to reverse obesity trends by improving diets, childbirth may become easier for many women in low-income countries. “You’re not going to see changes overnight, but a healthier diet could actually contribute to better childbirth outcomes,” says Wells.

The obstetrical dilemma was introduced almost 66 years ago in a few brief lines buried in a popular science article. Since then, Washburn’s hypothesis has been defended by some and rejected by others. But in the past 15 years, it has also evolved into a much grander idea that incorporates all manner of other variables, including pelvic floor health, medical technology and even junk food to explain why childbirth continues to be difficult and potentially dangerous. We now know that Washburn’s idea wasn’t the final word on the subject, but merely the starting point for an investigation that has expanded in strange and unexpected directions – even if some recent claims, like the eventual end of childbirth without medical intervention, are on the shakier side. Grappling with why delivering human babies is so fraught for so many comes with no neat answers. Exploring it, though, illustrates the profound ways that childbirth has shaped our evolution – and continues to do so.