In many ways, humans are weird mammals. And our relationship with milk is especially weird.
In every mammalian species, females produce milk to feed their young. The nursing babies digest the milk with the help of an enzyme called lactase, which cuts milk sugar into easily absorbed fragments. When the young mammals are weaned, they stop making lactase. After all, why waste energy making an enzyme you no longer need?
But it is common for our species to keep consuming milk into adulthood. What’s more, about one-third of people carry genetic mutations that allow them to produce lactase throughout their lives, making it easier to digest milk.
Scientists have long suspected that dairy consumption and the persistence of lactase rose together in human history. When people started herding cattle and other livestock some 10,000 years ago, the theory went, those with a mutation for lactase persistence gained a new source of calories and protein. People without the mutation, in contrast, became sick when they tried to consume milk and so did not take advantage of the new milk supply.
But a new study of ancient human DNA and milk-drenched pottery shards suggests that the traditional story does not hold up. “Something was not quite right with the received wisdom,” said Richard Evershed, a biogeochemist at the University of Bristol in England, and an author of the study.
Dr. Evershed and his colleagues found that Europeans were consuming milk without lactase for thousands of years, despite the misery from gas and cramping it might have caused. The scientists argue that the lactase mutation only became important to survival when Europeans began enduring epidemics and famines: During those periods, their poor health would have exacerbated gastric distress, leading to life-threatening diarrhea.
The study, published in the journal Nature, emerged out of a collaboration among more than 100 scientists with very different kinds of expertise, including genetics, archaeology and epidemiology. For his part, Dr. Evershed pioneered methods in the 1990s to detect traces of milk fat left behind on ancient pots.
Over the years, he and his colleagues have found milk on thousands of pottery fragments across Europe and neighboring regions. For the new study, the scientists used this database to create a map of milk consumption over the past 9,000 years.
The oldest evidence of milk came from Turkey, which was home to some of the world’s first agrarians. Those farmers then moved across Europe, taking their cattle and other livestock with them. By 6,000 years ago, they had arrived with their milk in England and Ireland.
Dr. Evershed and his colleagues found that some societies took up milk while neighboring ones did not. They also found that milk production went through boom-and-bust cycles over the centuries.
Mark Thomas, a geneticist at University College London, led the team’s analysis of lactase persistence. He and his colleagues analyzed DNA harvested from 1,786 ancient skeletons found across Europe and neighboring regions. They looked for a mutation that kept the lactase gene switched on during adulthood.
The oldest mutation they found dated back about 6,600 years ago. But in their collection of ancient remains, it stayed rare until 4,000 years ago. For those 2,600 years, in other words, Europeans were consuming milk despite almost none of them being able to make lactase as adults.
To see how this mutation affected people today, the researchers joined forces with George Davey Smith, an epidemiologist at the University of Bristol. Dr. Davey Smith has carried out a number of studies on the health of living British people by analyzing a large database called UK Biobank. Hundreds of thousands of volunteers have submitted their DNA to the effort, along with their electronic health records and answers to questionnaires.
Dr. Davey Smith sifted through the UK Biobank for information about milk and lactase, comparing 312,781 volunteers who carried the lactase mutation to 20,250 who did not.
The analysis delivered some surprising results: People without the lactase mutation consume about as much milk as people who carry it. Yet people who cannot make the enzyme do not suffer any significant health problems. They do not die at a higher rate, they do not have weaker bones and they have just as many children as people with the mutation do.
Together, these parallel lines of evidence suggest that early Europeans made milk a part of their diet, even without lactase. It is possible that some of them occasionally suffered some uncomfortable cramps and gas, but it was not enough to affect their health.
Early Europeans may have also lessened the painful effects of milk sugar by fermenting milk into cheese or turning it into butter. (In Ireland, people who harvest peat from bogs have occasionally found massive containers of “bog butter” dating back thousands of years.)
Consuming milk without lactase became riskier later, in times of crisis, Dr. Evershed and his colleagues argued. Starvation has been shown to shift mild symptoms, such as gas and cramps, to more dangerous ones, like diarrhea.
During periodic Bronze Age famines, people may have tried to survive by drinking milk when other kinds of food became scarce. Those without lactase might have been more likely to perish, while the lactase mutation might have helped others survive.
When people began living with large numbers of domesticated animals, their interactions created the opportunity for pathogens to jump the species barrier and cause outbreaks. Many of those pathogens may have triggered more diarrhea-causing diseases. People who could not make lactase might have suffered even more diarrhea during their infections, pushing them closer to death.
“You dehydrate and die,” Dr. Evershed said.
But for those who could tolerate it, milk could have restored fluid levels, making it more likely that they could recover from the infection.
“I think these ideas are really solid,” said Shevan Wilkin, the head of the ancient protein laboratory at the University of Zurich, who was not involved in the study. “I don’t know if they’re exactly right. However, they’re the first people who said, ‘Let’s actually work this out with data and see if we can support our ideas.’”
Dr. Wilkin said that other scientists will likely study whether the same patterns applied to ancient populations outside of Europe. Societies in parts of Africa, the Middle East and South Asia have also evolved ways to keep lactase production through adulthood — in some cases thanks to the same mutation, but in other cases with a different one. It is possible that the same evolutionary forces were at play in those regions, but it is also possible that the story will be different.
“That’s why I think this paper is so cool: It’s really going to open up so many more studies looking at this in a number of different ways,” Dr. Wilkin said.