Charles Darwin’s mother, unwell throughout his childhood, died from an agonising stomach ailment, probably peritonitis, at the age of 52. Charles was then eight years old. As an adult he was obsessively concerned with his own ill-health, particularly the recurrent stomach complaints that recalled his mother’s fatal illness.

Sophioe Kern's illustration for Adam Kuper's article Kissing Cousins

Was his affliction hereditary? Darwin’s mother was a Wedgwood. and the Wedgwoods were notorious for their ill-health. Whenever one of his children fell ill, Charles was inclined to see the same symptoms in himself, and to worry that it exposed a family propensity.

Or were the frequent illnesses of his children perhaps the consequence of inbreeding? He had married his cousin, Emma Wedgwood, his mother’s brother’s daughter. Is that why his children suffered, and his beloved daughter Anne had died in childhood? The poor health of the whole Wedgwood clan might be explained by a long-standing preference for marrying cousins.

These were not simply personal worries. The risks of cousin marriage became a subject of scientific discussion in the 1860s. And in Darwin’s case, his scientific project and his private concerns fed on each other.

Between 1868 and 1877 Darwin published three monographs on cross-fertilisation in animals and plants. In the first of these books, The Variation of Animals and Plants under Domestication, he proposed that “the existence of a great law of nature is almost proved; namely, that the crossing of animals and plants which are not closely related to each other is highly beneficial or even necessary, and that interbreeding [i.e., inbreeding] prolonged during many generations is highly injurious.”

Was this also true of human beings? Darwin thought this probable. He was accordingly fascinated by a study of cousin marriage in Scotland published in 1865 by Arthur Mitchell, Deputy Commissioner in Lunacy for Scotland.

It was widely believed that marriage between close relatives was rampant in remote Scottish regions, particularly the Highlands and Islands. However, Mitchell noted that popular opinion in Scotland condemned “blood-alliances” as “productive of evil”. And indeed national statistics showed that nearly 14 per cent of “idiots” in Scotland were children of kin. In 44 per cent of families with more than one mentally handicapped child the parents were blood relatives. Six per cent of the parents of deaf mutes were close relatives.

Yet Mitchell was not convinced that this was the whole story. Fewer than two per cent of marriages in Scotland were between first or second cousins. The rate was indeed higher in some isolated regions, but the evidence for bad effects was inconclusive. In one small town on the north-east coast of Scotland, nine per cent of marriages were with first cousins and 13 per cent with second cousins. Mitchell acknowledged that the children of these cousin marriages were often unprepossessing, but then many fishing families in the region were “below par in intellect”. A more telling case was Berneray-Lewis (now Great Bernera, off the Isle of Lewis). Here 11 per cent of marriages were with first and second cousins, yet Mitchell remarked that “instead of finding the island peopled with idiots, madmen, cripples, and mutes, not one such person is said to exist in it”.

Perhaps environmental factors – “occupation, social habits, etc.” – influenced the outcome. One “shrewd old woman” remarked to him: “But I’ll tell ye what, Doctor, bairns that’s hungert i’ their youth aye gang wrang. That’s far waur nor sib marriages.” Mitchell concluded that close-kin marriage tended to reinforce “evil influences”.

In 1870 Darwin’s ally John Lubbock proposed an official investigation of the effects of cousin marriage, but could not persuade parliament. Darwin then decided to commission a private study of the topic. He entrusted it to his eldest son, George, who was an accomplished mathematician.

Darwin laid out the research design himself. George was first to establish the incidence of close-kin marriage in the general population. He was then to enquire how many patients in insane asylums were the children of close relatives. If it turned out that marriages between close relatives produced a disproportionate number of “diseased” children, he instructed George, this would “settle the question as to the injuriousness of such marriages”.

The first step was to find out how common it was in England for first cousins to marry. Apparently nobody knew the answer. George Darwin was given estimates that ranged from 10 per cent to one in a thousand. “Every observer”, he concluded, “is biased by the frequency or rarity of such marriages amongst his immediate surroundings.” He would have to discover the facts for himself. Expert in the new statistical techniques that were being developed by his own cousin Francis Galton and by William Farr, George decided to attempt a scientific survey. It was to be one of the very first statistical studies of a social problem.

Using marriage records, press announcements of marriages, and questionnaires, George Darwin discovered that about 4.5 per cent of marriages in the aristocracy were with first cousins; 3.5 per cent in the landed gentry and the upper-middle classes; about 2.25 per cent in the rural population; and among all classes in London, about 1.15 per cent.

The next step was to gather statistics from mental asylums. Charles Darwin wrote on George’s behalf to the heads of the leading institutions. Several provided detailed responses. These indicated that only three to four per cent of patients were the offspring of marriages between first cousins. This was not out of line with the incidence of first-cousin marriage in the population at large. George concluded that “as far as insanity and idiocy go, no evil has been shown to accrue from consanguineous marriages.”

Other studies suggested that the offspring of cousin marriages were more likely to suffer from blindness, deafness, or infertility. George accepted that these conditions were highly hereditary, but saw no convincing evidence that they were caused by cousin marriage. In fact, first-cousin marriages were, if anything, more fertile than others. He suggested that a man was more likely to marry a cousin if he had many to choose from. First-cousin marriage would therefore be more common among people who came from large – and so presumably fertile – families.

Only one small piece of evidence gave George pause. Among men who had rowed for Oxford or Cambridge, men who were obviously the fittest of the fit, sons of first-cousin parents appeared slightly less frequently than might have been expected (2.4 per cent as opposed to 3-3.5 per cent among their peers).

To be sure, George Darwin was well aware that his conclusions flew in the face of a common and ancient prejudice. He conceded that marriages between cousins might be quite all right for the rich but bad for the poor. After all, “my father, Mr Charles Darwin, has found that in-bred plants, when allowed enough space and good soil, frequently show little or no deterioration, whilst when placed in competition with another plant, they frequently perish or are much stunted.” In short, cousin marriages caused no harm in the best families, but the inhabitants of slum tenements should probably avoid them.

Charles Darwin endorsed these conclusions. In later editions of Variation he modified his original rule, weakening the claim: “it is a great law of nature, that all organic beings profit from an occasional cross with individuals not closely related to them in blood.” (Emphasis added.) On the other hand, the experience of animal breeders indicated that the advantage of inbreeding “as far as the retention of character is concerned, is indisputable, and often outweighs the evil of a slight loss of constitutional vigour”.

The densely intermarried Wedgwoods liked to joke that any hint of laziness or illness was an infallible sign of familial degeneracy. However, the Darwinian establishment was now convinced that the risks of cousin marriage were slight, at least within prosperous families. Francis Galton wrote enthusiastically to George Darwin that he had “exploded most effectually a popular scare”. He added that his cousin could make a fortune from his discovery.

“Thus: there are, say, 200,000 annual marriages in the kingdom, of which 2,000 and more are between first cousins. You have only to print in proportion, and in various appropriate scales of cheapness or luxury: WORDS of Scientific COMFORT and ENCOURAGEMENT to COUSINS who are LOVERS then each lover and each of the two sets of parents would be sure to buy a copy; i.e. an annual sale of 8,000 copies!! (Cousins who fall in love and don’t marry would also buy copies, as well as those who think that they might fall in love.)”

Galton’s protégé Karl Pearson, a stalwart of the eugenics movement, made a follow-up study in 1908. He was less systematic than George Darwin, relying on information volunteered by readers of the British Medical Journal. These select respondents reported a very high incidence of first-cousin marriages in their own families. A smaller proportion of marriages were with more distant cousins, but Pearson remarked that second and third cousins in these families were often related in more than one line. He lumped them all together and concluded that “consanguineous marriages in the professional classes probably occur in less than eight per cent and more than five per cent of cases.” Yet only 1.3 per cent of patients in the Great Ormond Street Hospital for Children were the children of cousins. Pearson concluded that “the diseases of children are not largely due to any consanguinity between their parents.”

Endorsed by the Darwinian establishment, George Darwin’s conclusions reassured many people whose family trees featured marriages between cousins. Englishmen could also rest more easy when they considered that Queen Victoria was married to a first cousin, and that several of her descendants had also married cousins. And Darwin’s conclusions seemed only common sense to landowners in the House of Lords, who knew that the inbreeding of good stock was sound policy.

American scientists disagreed, however. The Bemiss Report to the American Medical Association in 1858 claimed that marriages between cousins were responsible for a number of birth defects. Despite its slapdash methodology, the Bemiss Report got wide publicity. Politicians and journalists began to demand a ban on cousin marriage. Judges and clergymen weighed in with solemn warnings.

Kansas banned the marriage of first cousins in 1861. Ten of the states that joined the union in the second half of the 19th century passed similar legislation (although not California or Texas). Several of the older states introduced a ban on first-cousin marriage, beginning with New Hampshire in 1869. Others, notably Massachusetts, New York and Pennsylvania, still allowed cousins to marry, but everywhere cousin marriage became less common. (First-cousin marriage has since been banned in Kentucky (1946), Maine (1985) and Texas (2005).)

The reaction against cousin marriage gained ground in Britain after the First World War, and by the 1920s British eugenicists were routinely condemning cousin marriage. Leonard Darwin, another son of Charles Darwin, followed his cousin Francis Galton as President of the Eugenics Education Society and joined the chorus of disapproval, despite the fact that he was himself the son of cousins and had married a first cousin once removed. In the 1930s the liberal geneticists JBS Haldane and Lionel Penrose denounced eugenic racial theories. Nevertheless they agreed that cousin marriages should be discouraged because of their link to recessive disorders, notably congenital deaf-mutism and certain mental defects.

In the English upper and upper-middle classes the prevalence of first-cousin marriage had remained steady at between 4 and 5 per cent for much of the 19th century – that is, one marriage in every twenty to twenty-five. Then quite suddenly, after the First World War, cousin marriage became very unusual. By the 1930s, only one marriage in 6,000 was with a first cousin. A study of a middle-class London population conducted in the 1960s found that just one marriage in 25,000 was between first cousins.

There are many reasons for this shift, not least the decimation of young men in the trenches of Flanders. The changes in scientific opinion were probably not decisive. After all, Darwin had played down the risks to offspring. However, as the practice changed so did the beliefs. Today there is a common feeling that a liaison between cousins is incestuous, and that if it is not forbidden it should be, if only because it carries unacceptable risks of genetic damage to offspring. And yet the consensus amongst experts today is much closer to the conclusions of George Darwin than it is to the popular view.

Any mating is risky to some degree, one danger being that offspring may be born with a defect if parents have a deleterious recessive gene in common. The chances of birth defects and of infant mortality are roughly doubled for the children of first cousins, but in normal circumstances that means that only some additional two per cent of children may be affected. Reviewing a number of modern studies, the geneticists AH Bittles and E Makov concluded that “the risks to the offspring of inbred unions generally are within the limits of acceptability. For first cousin progeny, it also must be admitted that they appear to be in remarkably close agreement with the levels calculated by [George] Darwin in 1875”. In the USA, the National Society of Genetic Counselors recently convened a panel of experts to review the risks of first-cousin marriage. They reported that the small background risk of congenital defects is raised by some 1.7 to 2 per cent in the case of children of first cousins. There is also an additional 4.4 per cent chance of pre-reproductive mortality.

The risks are significantly higher if cousin marriages are repeated over several generations. This may be a factor in the high prevalence of birth defects in the British Pakistani population, amongst whom cousin marriages are common.

In Pakistan, and in the Pakistani diaspora, a preference is widely expressed for marriage within the extended family or biradari, and in fact marriages between close relatives are common in most regions of Pakistan. Perhaps unexpectedly, the rate of cousin marriage is even higher among Pakistani immigrants to Britain than it is in rural Pakistan. And the present generation of British-born Pakistanis has the highest rate of all. Around a third of the marriages of the immigrant generation were with first cousins, but well over half the marriages of the British-born generation are with first cousins.

A preference for cousin marriage is often put down to religious dogma, but the real reasons are in every sense closer to home. The remarkable preference of British Pakistanis for cousin marriage is largely a consequence of immigration, and more specifically of British immigration regulations. It is very difficult for people to enter Britain unless they are married to people already here. In most cousin marriages, one partner is brought to Britain from Pakistan. Alison Shaw found that 90 per cent of the first-cousin marriages in a sample of British Pakistanis in Oxford involved one spouse who came directly from Pakistan. There are often debts to family members back home, who helped to finance the migration. Above all, there is family loyalty. Roger Ballard points out that if a British-based family refuses a marriage offer from relatives in Pakistan, “they are likely to be charged with having become so anglicised that they have forgotten their most fundamental duties towards their kin”.

The odds of genetic misfortune rise sharply if cousin marriages are repeated over several generations, but it is not easy to separate the effects of cousin marriage, heredity and environment. One study of the genetic consequences of cousin marriage in families of Pakistani origin in Norway found that the risks were rather low. Other studies are less reassuring. The authors of a recent prospective survey of Pakistani families in Birmingham suggests that if they “ceased to marry relations, their childhood mortality and morbidity would decrease by 60 per cent”.

The tabloid press – and populist politicians – claim that the preference for cousin marriage overloads the health services and prevents integration, and put the blame on superstitious beliefs and patriarchal attitudes. And yet the family life of British Pakistanis is close to the Victorian ideal. Indeed, the great bourgeois clans of Victorian England – like the Darwin-Wedgwoods – had a great deal in common with a modern Pakistani biradari. The 19th-century English bourgeoisie would have been sympathetic to the motives that incline these new Britons to marry their cousins.

Adam Kuper‘s new book Incest and Influence: The Private Life of Bourgeois England (Harvard University Press) is out in November