We can't seem to get enough of cheese—mooning over mozzarella, pining over parmesan, and ballyhooing over brie. It's unclear exactly when this cheesy obsession began. Ancient humans seemed to start collecting milk shortly after cows, goats, and sheep were domesticated just over 10,000 years ago—and cheese may have soon followed.
Now, a new study in the journal PLOS ONE claims to have found the earliest Mediterranean cheese production, identifying traces of the funky fermented dairy product on pottery shards that are a whopping 7,200 years old. But some scientists are unsure of the find, suggesting that, like a block of Swiss, the work may have some holes.
Ancient Cheesy Love
Researchers have extensively documented milk production throughout the Mediterranean, extending back almost 6,000 years ago. In modern-day northwest Turkey, ancient people were milking their livestock as far back as 8,500 years ago. But genetic analysis suggests that beyond childhood, the ancient people who lived in the region couldn't digest milk, says study author Sarah McClure, an environmental archaeologist at Pennsylvania State University.
“Why are they milking something if they can't consume it without severe gastrointestinal distress?”
Perhaps, her team argues, they partially did it to create cheese. Fermentation reduces the lactose in milk, so if ancient humans could have figured out the process, it might have opened up a delicious and nutritious new source of food. (See where lactose intolerance is highest around the world today.)
Some previous studies document just that. In 2012, researchers identified cheese-making processes in Poland that date back at least 7,000 years. In the Mediterranean, however, scientists haven't found solid evidence for cheesy creations until thousands of years later, starting in the Bronze Age.
Searching for Feta
The researchers of the latest study didn't set out to find cheese, but were interested in Mediterranean food storage. They turned to two Neolithic sites near the coast of Croatia, collecting pottery shards that date to between 6,000 and 5,000 B.C. They extracted traces of fats from the surface of the pottery and analyzed them based on mass and their carbon isotope fingerprints to figure out whether they came from meat, liquid milk, or cheese.
Overall, the lipids seemed to cluster in a few different groups depending on the type of pottery, says McClure. Some pottery types seemed to have been used to store milk, some seemed designated for meats, and a few might have been used for cheese.
Milk seemed to be confined to a relatively scarce bright orange pottery known as figulina, which is made from fine-grained clays and baked with lots of air circulation. “Perhaps part of the fact that it's in this special pottery is that the milk was somehow special or used in a distinctive kind of way,” says McClure.
The researchers also identified traces of what they interpret as milk or fermented milk products on three pieces of pottery rife with holes. Past work on similar sieves found in Poland suggest that such tools were used in the final steps of cheese making, to separate the curds from the whey. This new study, funded in part by the National Geographic Society, marks the first analysis of fat traces on similar pottery in the Mediterranean.
Sniffing Out the Facts
Yet some scientists are concerned about jumping to cheesy conclusions. As Mélanie Roffet-Salque, a postdoctoral researcher at the University of Bristol, writes succinctly in an email: “No evidence for cheese in the rhytons,” the type of solid pottery on which the new study claims to have identified the fermented milk product.
These samples have isotopic fingerprints that fall outside the range common for milk from cows of this period, according to a study Roffet-Salque and colleagues published last year in the journal Archaeological and Anthropological Sciences. That's because many modern animals are often fed what's known as silage—a pre-chopped mix that can include corn, grasses, and legumes that changes the isotopic fingerprint in their fats.
Roffet-Salque instead suggests that the fat traces from the Croatian pottery may actually come from a mix of meats.
Study coauthor Clayton Magill of Heriot-Watt University acknowledges the concern: “Corn mega-messes up all of these kinds of isotopic interpretations,” he says. But he defends their work, saying that the team was most interested in the bacterial processes that create cheese, which some studies suggest also meddle with milk's isotopic fingerprint.
“I would argue that if we go to the microbiological literature, there's a very strong reason to believe that milk fats that were fermented would fall in that range,” he says.
Paul Kindstedt, a professor of food science who specializes in cheese at the University of Vermont, was excited to see the new study. Though he agrees that the differentiation between milk, cheese, and even butter based on these ancient fats is an iffy prospect, the work does represent the first lipid analysis of sieves from this region, which he calls a significant new piece of information.
For these, too, Roffet-Salque adds a note of caution. Of the three sieves analyzed, she contends that only one might be milk, but as with the pots, the isotopic composition differs slightly from what would be expected and needs more investigation.
“The other sieves have an isotopic composition that is too borderline between ruminant adipose and dairy fats to be diagnostic,” she says via email. And just a single sample can't tell you much. For comparison, the ancient cheese identified in Poland, work which Roffet-Salque led, analyzed lipids on 50 fragments of sieves from 34 vessels, in addition to numerous shards from pots.
Magill takes the criticism in stride, noting that more foundational work in both ancient and modern cheeses is needed to strengthen future work in the field.
“It's not the sexiest science, it's not the easiest, and it's quite hard to get funding for it,” he says. “But it's nonetheless essential to the validity and strength of our interpretations, regardless of what you believe about them.”