Almost 140 years after the publication of the theory of evolution by natural selection1, arguably one of the greatest achievements of science, we persist in failing to appreciate its proper application to the human species. We have a natural tendency to see ourselves as the unchanging pinnacle of creation, the supreme creation of nature, as if evolution were a ladder with Homo sapiens at the top. Instead, we should see evolution as a tree, which proceeds by branching and not (usually) by wholesale transformation and replacement2. Each currently extant twig survives, not because it is ideally suited to its environment, but because it is better suited than the competition. Most twigs eventually become extinct, and major events, such as comet impacts, may wipe out up to 95% of existing species, so that evolution has to recommence from what is left3. Each twig has its own environmental niche in which it is successful. We could not live in a stagnant pond any more than an amoeba could build a house. If we measure success by numbers of individuals, or weight of biomass, then many animals have a better claim to success than humans. For example, beetles make up more than one-quarter of all animal species (over 250,000 types of beetle are known)4. In the Amazon basin the estimated biomass (amount of living matter) of ants is four times the biomass of all the vertebrate animals combined4. It is better to see ourselves as in transition from what we were to what we must become if we are not to follow many previous species into extinction.
It might have been easier to see ourselves as a twig rather than a pinnacle, if species closer to us than chimpanzees and gorillas had still existed during recorded history. They almost did. The earliest bipedal forms appearing in the fossil record were found in East Africa and can be dated to 3.5 million years ago. These hominid species, known as Australopithecines had an almost human-shaped pelvis and walked erect efficiently, but still had a brain capacity equivalent to that of a chimpanzee—about 400 cc5. Various theories have been advanced for the selectional advantage of the upright posturean expanded horizon of vision from standing upright on the veld, less body area exposed to the direct rays of the noon-day sun, ability to go fishing in shallow water, and most importantly freeing of the hands for the purposeful manipulation of tools. One theory suggests that the pelvis, adapted to the upright posture (narrow, to allow the legs to be close together under the spine, which facilitates transfer of forces from femur to spine when running), probably severely limited the possible cranial size of the infant at delivery, until the mechanism of rotational birth evolved during the late Middle Pleistocene era6. Between 2 million and 0.7 (700,000) years ago, the hominid lineage appears to have undergone relatively rapid evolution, with the earliest fossils ascribed to the genus of Homo (Homo habilis) appearing 2 million years ago. The earliest recognisable crania of Homo erectus are from Lake Turkana deposits and date from 1.5–1.6 million years ago, with one complete skull of a female having a cranial capacity of 900 cc7. Homo erectus spread from Africa into Asia, and like most hominid fossils, were marked by great morphological variation-so much so that some researchers believe that some fossils ascribed to Homo erectus actually represent separate species such as Homo lieidelbergensis. By 300,000 to 400,000 years ago morphological distinctions between fossils ascribed to Homo erectus and fossils that can possibly be attributed to early Homo sapiens can be made. By 200,000 years ago Homo neanderthalensis appeared as a further species, with some forms (the ‘classic’ neanderthal) having special morphological adaptations enabling them to cope with the Pleistocene ice ages then spreading and retreating across Europe. Forms of Homo sapiens co-existed with forms of Homo neanderthalensis until as recently as 30,000 years ago7. Evidence indicates that forms of Homo sapiens and Homo Neanderthalensis were probably always separate species and did not interbreed8,9, but the ‘disappearance’ of Neanderthals is still a matter of anthropological controversy. Both had developed large brains but the Neanderthals probably lacked both language and aspects of sophisticated culture—their primitive tools show no evidence of development over 60,000 years and they left no art objects comparable with those of early Homo sapiens10.
The development of the capacity for language as an aspect unique to Homo sapiens11 was probably the crucial development that allowed the development of culture. Reliance on inter-personal communication means that in many respects our main evolutionary rivals are no longer other species, but other members of our own species12. This may be the reason for the massive development of the frontal lobes, essential for dealing with the complexities of human behaviour (famously described by Niccolo Machiavelli in his 1532 publication, The Prince, in which he set out methods for the acquisition and maintenance of political power). The end result has been a brain which is often over 2000 cc in size-and which in the fetus presents such a challenge to the birth process13. Despite Homo sapiens having the lowest ratio of newborn to adult brain size14, the term fetus head takes up most of the space in its mother's pelvis, compared with less than half in the chimpanzee—truly a competition between the need to think and the need to run!
Thus the scene is set for a competition between the fetus and its mother-the fetus survives best if it is large at birth16 but is easier for its mother to deliver if it is small. Indeed, large babies can wreak havoc to the mother's pelvis, causing urinary16 and fecal17 incontinence, and even vesico-vaginal fistula18. Neglected prolonged obstructed labour can in addition lead to renal failure, pelvic inflammatory disease, osteitis pubis, foot drop, and secondary infertility18. The idea that competition is mainly between species (whose individuals support one another) and a hostile environment, is one that Darwin acknowledged as an alternative to competition between individuals19 but which received a setback with Dawkin's concept of the selfish gene20. This proposes that the individual organism is merely a vehicle for its genes; individuals die but its genes are (relatively) immortal, provided the individual reproduces before it dies. The function of an individual is to reproduce its genes; once it has done this it is expendable. Natural selection produces an increase in frequency of successful genes. Haig21,22 has pointed out that arguments for harmony of interest between mother and fetus are flawed because they do not carry identical genes. Genetic actions that enhance the transmission of fetal genes do not always enhance the future transmission of more maternal genes; babies survive occasionally even if the mother dies, and their survival could be enhanced if the mother becomes infertile. Maternal genes have an equal stake in current and future children; the genes of the current child have a greater stake in themselves than in any future child because they may not be present in subsequent offspring, especially if the mother changes her reproductive partner.
Thus it is inappropriate to see human labour as a harmonious process which, because it is ‘natural’, should always go well unless there is disturbance by some external agency. Mother and fetus do have a mutual stake in a successful outcome. but sometimes one will win an advantage to the detriment of the other. Labour should instead be seen as an imperfect solution to a complex problem, and one which is still evolving.
One technological solution to the conflict between the need to think and the need to run is caesarean section (much as tools were a solution to problems with food supply). Increasing use of caesarean section has been associated with the birth of fewer babies with low five minute Apgar scores. Although it is associated with a greater need for blood transfusion, prophylactic antibiotics have meant a fall, rather than a rise, in maternal infections. Subcutaneous heparin can limit the risk of venous thrombosis and embolism. Advances in surgical technique have kept pace with the rise in the caesarean section rate, so that in Massachusetts from 1954 to 1985 there was a quadrupling of the caesarean section rate, but the number of caesarean related deaths per 100,000 live births did not change23. Between 1976 and 1984, there were 649,375 births and 12 1′2 17 caesarean sections, with seven deaths, a rate of 5.8 per 100,000 caesarean sections. In contrast, the rate for vaginal deliveries was 10.8 per 100,000. Although elective caesarean section may increase the mortality risk for the mother by 50% compared with a vaginal delivery in a healthy woman24, numerically this is still a very small risk. The overall mortality/morbidity difference between elective caesarean section and attempted vaginal delivery may well disappear in mothers at increased risk of emergency caesarean, such as women aged over 40 years or with short stature or a large baby, because the risk of an emergency section is probably at least double that of an elective procedure24.
Because the factors involved in human birth are so complex and still poorly quantified, questions such as ‘what is the optimal caesarean section rate?’ are very difficult to answer, and anyway, the answer keeps changing according to both the circumstances and the outcome desired. If we could predict more accurately the women most likely to have a long and difficult labour, elective caesarean section would probably be safer and more acceptable for them than a trial of labour. Some women may prefer an elective caesarean with a small risk of mortality and serious morbidity as a way of avoiding the disabling complications of incontinence (which are at least 20 times as common as common as mortality, so that a woman is far more likely to have heard of someone with stress incontinence following vaginal birth than of someone having died after caesarean section), not to mention avoiding the discomforts of labour. Many see the element of predictability and control of an elective procedure as important advantages25; about 50% of women having a caesarean section would request another one in a future pregnancy26. Provided the caesarean section is not performed before 39 weeks the risk to the fetus is likely to be lower for elective caesarean birth than vaginal birth, especially if there is any abnormality such as growth restriction or abnormal presentation27. Financial data are lacking, but one to one midwifery care of 24 hour labours is not cheap. In any case, most of us are prepared to pay more for convenience, which is why we use our cars rather than public transport.
If we see medicine not as an attempt to defeat natural selection (which it cannot be as the development of the skills of medicine are an integral part of natural selection), but as a way of relieving suffering (whether it be disease or pain), then the ‘correct’ caesarean section rate is the one which produces the maximum maternal satisfaction for the least risk; the balance between these two is a matter of judgment and will vary according to the circumstances. It is conceivable that caesarean section will become so safe that for most women the unpredictable risks of labour will no longer be justified. If caesarean section becomes the norm, average birthweight, being no longer restricted by the constraints of maternal pelvic size, might then increase so that eventually caesarean birth becomes necessary for the majority and is no longer just a choice. Such a situation is already developing in some animals bred selectively for large head size, such as bulldogs, over 60% of whom have to be delivered by caesarean section (R Ballard, personal communication). Whether any additional cost of operative delivery is acceptable depends upon the value assigned to maternal autonomy in relation to convenience, avoidance of pain and damage to pelvic structures, and her desire to protect her baby. Recent papers have emphasised the increasing role of women's choice in the decision to perform a caesarean section, and make it clear that taking heed of women's views is fundamental to achieving a satisfied customer26. Some women ask for a caesarean section because they fear damage to their vagina and to their fetus without there being any medical indication25. In a study of 3150 elective singleton caesarean sections in Scotland, 19.8% were cases where a trial of vaginal delivery was considered medically appropriate (eg, a breech presentation or previous caesarean section), but the mother requested operative delivery28. Rather than indulging in reflex pleas to ‘return to the simplicity of nature’ (which is often ‘red in tooth and claw’), we should be concentrating on making caesarean section even safer, researching ways to predict labours that will have an adverse outcome29–31, and listening to what (properly informed) women want.