Just as many people may claim their favorite cheese — Gouda, Cheddar, Brie — stands above the rest, Pliny the Elder thought a variety from Gaul was superior. Here, a look into curdled milk and a recipe for this prince among cheeses.
The ball of cheese I made on a plate, garnished with a wreath of thyme.
Today, there are, really, uncountable varieties of cheese. Cheese can be made from cow milk, or sheep milk or buffalo milk or camel milk. Cheese can be fresh and soft or aged and hard. Cheese can be smoked, or purposefully inoculated with mold (blue cheese finds a fan in me). Or cheese can be flavored with chives, or with jalapeños, or cranberries, or dill.
The options are infinite. But how did we end up with this food that can so artfully be worked into pasta, potatoes, eggs, and sandwiches?
Some 7,000 years ago, members of the Linear Pottery culture inhabited a swath of land from the Netherlands to Romania. Named for the defining designs on their earthenware, archaeologists have uncovered a handful of important sites mostly clustered in along the Elbe, Rhine, and Danube rivers. One of these prehistoric finds, located in Kujawy in north-central Poland, bore fragments of clay sieves among other potsherds.
Chemical analysis of these chips of pottery revealed traces of lipids on their surfaces. Based on this discovery and their resemblance to modern strainers, it’s believed the Linear Pottery culture was busy producing cheese well before anybody assumed. This cheese was simple, probably similar to cottage cheese today, and there wouldn’t likely have been multiple types like there are now.
As cheesemaking evolved, varieties arose. Later Egyptian murals and Sumerian writings mention cheese. We can’t say what even those cheeses would have been like, but it’s evident that by the time of the Roman empire there were established cheese preferences, formed from a wide range of products made around the Mediterranean. Pliny the Elder says the following about the best cheeses by Roman standards:
The kinds of cheese that are most esteemed at Rome, where the various good things of all nations are to be judged of by comparison, are those which come from the provinces of Nemausus, and more especially the villages there of Lesura and Gabalis; but its excellence is only very short-lived, and it must be eaten while it is fresh.
So ancient gourmets decided some of the best cheeses came from France — not much has changed! The “provinces of Nemausus” would have been Gallia Narbonensis or southern Gallia Celtica, in south-central France, and Nemausus the city of Nîmes. Initially settled by a tribe of Gauls, Nemausus became important to the Romans as a city along the Via Domitia, the road connecting Italy with the province of Hispania.
Even many centuries later, modern Nîmes is like Rome away from Rome. The city boasts an excellent amphitheater that hosts mock gladiator fights each spring, has one of the best-preserved Roman temples anywhere, and was the destination of the Pont du Gard, an aqueduct used to supply much-needed water to its citizens.
The cheese produced around Nemausus in the grassland of southern Gaul was tasty, as we can tell from Pliny’s description. We also know it was fresh. In the original Latin text, this is caseus musteus — though it looks like “musty,” Pliny adds this to note the Gallic cheese was new, not aged.
These things we can tell, but what was it made from? What comes after the passage above may clue us in:
Goats also produce a cheese which has been of late held in the highest esteem, its flavor being heightened by smoking it. The cheese of this kind which is made at Rome is considered preferable to any other; for that which is made in Gaul has a strong taste, like that of medicine.
We can assume the “cheese of this kind” is goat cheese, maybe specifically smoked. If the cheese from Nemausus was deemed to be the best but Pliny said the Gaulish goat cheese had a strong medicinal taste, it doesn’t seem like chèvre would’ve been the cheese in question.
While we work to narrow our search, archaeological evidence can be an especially useful tool. As French archaeologists sifted through the remains of Entremont, an oppidum (walled town) near Aix-en-Provence and not too far from Nîmes, they excavated a number of animal skeletons.
Just under half belonged to pigs, but I would reckon these were mostly for meat — Varro writes that the Gauls exported ham to Rome (De re rustica II). Pig milk is gamy and watery, and sows don’t take kindly to people pinching them.
The next most common find was sheep, then cattle. Among the livestock remains at Entremont were cheese baskets that would’ve been used to shape each wheel. Cheese production, as we can see, wasn’t a foreign practice to the Gauls in southern France.
Both sheep and cattle are milked today to produce southern French cheeses: ewe’s milk makes Roquefort, cow’s milk makes Cantal. So these two types of milk are possibilities, but archaeological evidence doesn’t conclusively rule in favor of one or the other.
Entremont is located at the edge of a plateau, where sheep are often raised today. I figure there is geographic continuity: the conditions that make tending sheep in that area common today would have been similar for the Gauls living at the ancient site of Entremont.
The name of one of the cheesemaking villages Pliny mentions — Gabalis — comes from a tribe of Gauls, the Gabali, who lived in the area just north of Nemausus. Their capital, though actually named Anderitum, was given the name civitas Gabalum, “city of the Gabali,” in Latin. This town eventually became Javols, and sits on the eastern end of the Plateau de l’Aubrac, a volcanic highland covered with meadows.
By this same token, then, as cattle farming continues to be one of the main activities on this plateau, it seems feasible that the native Gabali would have herded not sheep, but kine. As Pliny mentions “kinds of cheese” that came from Gaul, it’s likely both sheep and cow milk were used. With geography in mind, though, the latter is what I’ve decided to pick.
Another reason for my choice: I’d love to make cheese with sheep milk, but it’s hard to find, period (especially now, a little too early in the year), and the milk I could potentially get would be ultra-pasteurized and useless for these purposes (as we’ll see later).
This dish, unlike some of the other ones I’ve recreated, is less faithfully concocted. I suppose that’s the nature of looking beyond the cookbooks of Rome — there are scant written records in Gaulish, let alone information on food or recipes. Take this recipe cum grano salis — with a grain of salt — if you wish, but I think it’s exciting even trying to fathom what life with the neighboring Celts might have been like.
Photos of the Maison Carrée temple, the Arena of Nîmes, and pastureland on the Plateau d’Aubrac (where the Gabali would have herded).
In 2018, archaeologists chanced on jars of 3,200-year-old cheese in the tomb of Ptahmes at Saqqara in Egypt. This is a remarkable discovery — it’s some of oldest cheese found, and was preserved well — but it would’ve had some nasty side effects. Traces of the bug Brucella melitensis were detected in the cheese, and its eaters probably would have contracted brucellosis, causing sweating, myalgia, and bouts of vomiting and diarrhea.
Fortunately for us, heating milk eliminates this fear. It also eliminates useful bacteria in milk, and denatures some proteins necessary for curdling. As I was perusing the dairy aisle at my grocery store, all of the milk I came across was pasteurized (alas, just not the same as a byre in southern Gaul!). I wasn’t able to find raw milk anywhere, but in my fraught searching, I found there’s a fix: vat pasteurization.
Yes, it’s still pasteurized. But vat, or batch, pasteurization heats the milk at a lower temperature for longer, leaving more of the enzymes and beneficial bacteria in the final product. Raw cow juice would be best, but I didn’t want to wind up sick.
To ensure better curd formation, I decided to add kefir, a probiotic-rich drink made from fermented milk. It contains many of the mesophilic bacteria destroyed by pasteurizing milk, necessary for developing the flavor and texture of the cheese. Kefir is also a more accessible alternative to a designated cheese culture packet, and does mostly the same job.
Some of the bacteria that would have landed in the bucket of milk in a Gallic cowshed, I reckon, could be restored after modern heat treatment. But without further ado, here’s a recreated version of a Gallic curd cheese that could have been made in the pastures north of Nemausus:
½ gallon of milk, whole and NOT ultra-pasteurized
½ cup of kefir
5-6 drops of rennet (vegetable or animal, or mineral if you’re a modern Major-General)
1 tbsp of salt
Pour the milk and kefir into a pot and mix together well. The milk I used wasn’t homogenized, so I had to break apart globs of fat with my spoon. Heat on low (100ºF) for around 90 minutes, making sure the temperature stays constant. I found a film collected on the surface of the pot while mine was cooking.
Dilute the rennet in about ¼ cup of warm water. Take the pot off of the heat, and slowly stir in the rennet mixture. Let the milk sit for half an hour or so (I’m imprecise with timing, but I think it depends on what you’re working with) to coagulate.
Bring the milk back up to temperature on low heat. I noticed the milk was wrinkly on the surface at this point. Score the milk in the pot with a knife (this it to break up the curds) so you end up with about one-inch cubes. Gently stir the curds around for about 15 minutes — you may see that they start to clump together.
Line a colander with cheesecloth. Fish the curds out of the pot with a slotted spoon, draining off any excess whey. Salt the curds as you add them to the cheesecloth. A whole tablespoon may seem like a lot, but it helps to draw out moisture and will eventually trickle away.
Pulling the corners together, twist the cheesecloth and wring out what moisture is left. Then tie the cloth off and hang it over a bowl for 24 hours to condense and dry out completely. The following day, unwrap the cheesecloth, slice, and enjoy!
Photos of the ingredients, milk with rennet, milk after heating with rennet (visible separation), curds in cheesecloth, and the final product.
For my first time making ancient-like cheese, this wasn’t bad at all! I did find it was a little rubbery and slightly bland. I skimped on salt and added a dash too much rennet, I think, but these are several things I’ve adjusted with the recipe above.
The options for serving this cheese are endless. Slice it and eat it with some panic bread as its makers might have done, or enjoy it in Roman fashion: a foreign delicacy topped with slathers of epityrum and served alongside figs and apples. Reading about ancient food is one thing, but recreating it is much more gratifying — especially when the end product is grate.