This story was originally published in our May/June 2023 issue as “The Archaeology of Flavor.” Click here to subscribe to read more stories like this one.

When it comes to ancient foods, archaeologists can easily glean main ingredients based on butchered bones and plant remains found in cooking areas. But every chef — and diner — knows dietary staples don’t make meals delectable or distinctive. It’s flavor, rendered through spices, herbs and culinary craft, that defines a dish. Certain flavors have come to distinguish the cuisines of different cultures, like umami in Japan or herbes de Provence in southern France. In the 1970s, cookbook author Elizabeth Rozin named this phenomenon the Flavor Principle.

“Chefs talk about this all the time, but archaeologists do not,” says Christine Hastorf, an archaeologist and professor at the University of California, Berkeley, who urges her colleagues to think more about flavor. Communities “inherit not just dishes and not just meals, but they inherit flavors.”

Presumably, past cultures also upheld the Flavor Principle. Uncovering those long-gone flavors is tricky, though. Seasonings and prep methods seldom survive the way bones and botanicals do. But thanks to scholars like Hastorf, archaeology has begun to tap into ancient flavors. This new avenue of research is revealing culinary fads and cultural ties that otherwise would be forgotten — and allowing people today to taste the past.

Ancient eating habits

About a decade ago, Hastorf was planning a Bronze Age feast for archaeologists digging in the Italian countryside. Because excavations had recovered cow bones and grains, her menu featured beef and oats. But Hastorf also sought ingredients from a villager who was familiar with the plants nibbled by locals for generations. In the woman’s garden, they picked wild purslane, an herby succulent, which had sprung up between orderly rows of vegetables. Added to the Bronze Age dishes, the weed delivered a tangy crunch and dose of vitamin C.

Wild plants like purslane help create the unique “flavorscapes” of cuisines, says Hastorf. And she believes purslane has figured into the Mediterranean flavorscape for millennia. “I would assume that people in the past would have tested everything that grows in their area,” she explains. The tart, healthy weed likely passed this test and entered the regional cuisine.

Besides heightening that particular feast, the experience later inspired Hastorf to rethink research being done halfway around the globe in Bolivia. There, on a vast plateau two miles above sea level, past people began settling on the shores of Lake Titicaca roughly 4,000 years ago. The ancient highlanders raised llamas, alpacas and guinea pigs, and farmed quinoa and a rainbow of root vegetables. Preserved bits of these crops surfaced in ancient structures and trash heaps excavated by Hastorf and colleague Maria Bruno of Dickinson College. In studies published between 2011 and 2016, the researchers also identified remains from wild species, which they concluded entered the mix, not for culinary reasons, but as uninvited weeds or fodder for grazing animals.

But at the same time, they observed the same “weeds” in meals prepared by the region’s Indigenous Aymara residents. The species also came up when Bruno formally interviewed locals about traditional uses of plants, as part of her earlier dissertation research. And Hastorf remembered the purslane that flavored her Bronze Age meal. In a 2020 Journal of Anthropological Archaeology paper, the archaeologists paired Aymara knowledge with their archaeological evidence to reinterpret ancient eating habits. They realized their original conclusion — assuming wild species were unwanted weeds or fodder — was perhaps misguided. Those species could have been key ingredients that transformed bland eats into cuisine.

For example, today, locals punch up the flavor of stews with k’hoa, a minty shrub that sprouts from rocky soils at even higher elevations than Lake Titicaca. Probable k’hoa remains were found in some excavation pits, suggesting past lakeshore residents also hiked upslope to collect the pungent herb. The study showed flavor preferences have likely endured around Lake Titicaca for over 3,000 years — despite the region’s later rule by the Inca Empire and European colonizers.

Building blocks

While Hastorf and Bruno uncovered past flavors with low-tech tools, other scientists have deployed cutting-edge molecular analyses, like proteomics. This method can output a list of proteins in a substance and read the molecular building blocks of those proteins, which differ by species like DNA code. Plus, proteomics can detect chemical modifications due to reactions the proteins underwent. Thanks to these three layers of information, a scientist could zap a food scrap and discover it contained, say, collagen (a muscle protein) from Ovis aries (the species we know as sheep), which was barbecued (reactions that modified the protein).

Anna Shevchenko, a cell biologist at the Max Planck Institute in Dresden, Germany, was, with archaeological scientist Yimin Yang of the University of Chinese Academy of Sciences in Beijing, among the first to realize proteomics’ potential in archaeological gastronomy. In the early 2000s, she was researching venomous caterpillars native to Brazil when she received a query from Yang: “Instead of caterpillars from Brazil, to try something which is 4,000 years old,” Shevchenko recalls. “We didn’t say no. We like this kind of adventure.”

Yang sent about a dozen yellowish, penny-sized lumps, which had been pried from the necks and chests of mummies discovered in the dune-swept Taklamakan Desert of northwest China. Between 1980 and 1450 B.C., the so-called Xiaohe culture managed to reside, herd, and farm on that desolate terrain. The people also buried their dead in boat like coffins. The arid desert naturally mummified the bodies and preserved other perishables stored within the graves, including the musty lumps Yang sent to Shevchenko.

As hoped, the analysis revealed the substance’s protein makeup and clues about its recipe: It seems the lumps started as cow’s milk, which must have been strained because the proteins more closely matched skim versus raw dairy. Then, ancient people likely curdled the milk by microbial fermentation. Had this step been done another way — like the rennet method used for cottage cheese — a protein called kappa-casein would have shown telltale breaks. (It didn’t.) The recipe resulted in a kefir cheese similar to labneh common in Middle Eastern countries today, the scientists reported in a 2014 Journal of Archaeological Science paper.

The researchers next analyzed milky residues found in grave baskets at Gumugou, a slightly older cemetery also in the Taklamakan Desert. Despite the sites’ closeness in time and space, “dairy production at Gumugou and Xiaohe was totally different,” says Yang. The Gumugou baskets contained strained sour milk, whereas Xiaohe people made solid cheese, according to the team’s study published in a 2016 issue of Quaternary International.

However, knowing these recipes doesn’t answer why the groups interred distinct dairy. “We don’t completely restore the ideas of ancient people,” says Yang. He proposes that the recipes differed because a colder climate during Xiaohe’s time caused the people to be more mobile, with a need for longer lasting, solid cheese. As for why the people laid dairy in graves: “Maybe they bury it for eating after death … but it’s difficult to confirm,” he says.

Since these landmark studies, the scientists have applied proteomics to diverse cuisines. In a 2022 Heritage Science paper, Yang analyzed remnants of baked grain cakes and mutton kebab, discovered at a 2,000-year-old site along the Silk Road.

Shevchenko recently detected proteins probably left by caviar prepared by Mesolithic people in Germany 6,000 years ago.

And the kefir cheese, which began this research for Shevchenko, even entered her kitchen. As part of the analysis, she prepared modern comparisons and kept the bacterial starter alive for several years — making cheese about twice a week for her family. Unfortunately, the microbes died while she was on vacation. “I was not crying, but I was sad,” she says.

In the jar

Molecular methods less complex than proteomics have revealed surprising tastes and even trade connections. In a 2022 PLoS ONE paper, archaeologists examined 13 clay jars that were unearthed from Jerusalem buildings that had been destroyed when Babylonians sacked the city in 586 B.C. Some of the vessels bore rosette stamps signaling they belonged to the kingdom’s royal economy. Scientists crushed and dissolved fingernail-sized pieces of the pottery and ran the extracts through an instrument that identifies compounds based on their weight and chemical charge. The analysis detected acids, alcohols, sugars and other molecules, which indicate the vessels stored olive oil and wine. But several wine jars also harbored key molecules in vanilla.

Study co-author Yuval Gadot balks at the thought of vanilla-flavored wine: “I don’t think I can sell that now in a restaurant,” says the Tel Aviv University archaeologist. However, the foreign plant, which must have been grown in India or East Africa, shows Jerusalem’s then-rulers accessed prestigious trade routes run by Assyrians, Egyptians and Babylonians. “Sometimes the taste is completely cultural. If you say, ‘This is exotic and only [the] ruling elite has the ability to have that,’ then it’s tasty. Even if it’s not,” Gadot explains.

The elite’s hankering for exotic flavors likely started centuries earlier in the region. A 2013 study showed that 11th to ninth century flasks, discovered in treasuries and temples, contained a main molecule of cinnamon, which only grew in southern Asia at the time. The finding suggests this spice made it to the Mediterranean long before ancient texts document its presence in the sixth century B.C.

Beyond identifying ancient flavors, Gadot is also part of research that has resurrected some of them. He and other archaeologists shared 2,000- to 5,000-year-old alcohol jugs, uncovered at sites in Israel, with microbiologists from Hebrew University of Jerusalem. The scientists extracted yeast cells from pores in the pottery, which turned out to be alive. Genomic analysis revealed the cells belonged to species that can ferment. The researchers used the awakened yeast to brew beer, wine, and mead, following recipes written in ancient Egyptian texts. Given a sampling, members of the Beer Judge Certification Program proclaimed the beer tasted like English ale.

How the yeast survived over two millennia remains a mystery. It’s possible the fungi went into a dormant state, but the team thinks it’s more likely that the organisms continued to reproduce, slowly. In that case, the yeast cells that entered today’s brew were the descendants of the ones that fermented the ancient drink.

Either way, sipping this grog approaches culinary time travel. Perhaps the yeasty tang of fermented fruits and grains transcends the Flavor Principle, pleasing tongues across eras and cultures.