Recreating Roman cheese from ewe and goat milk

Rebuilding a Roman-style cheese from ewe and goat milk is an exercise in careful reconstruction, not the recovery of a single lost recipe. Roman writers mention these milks and different forms of caseus, but no surviving text provides a modern formula with quantities, temperatures and timings. Archaeology adds clues through ceramic vessels interpreted as moulds and strainers, while the literary record shows how the food was used.

This guide develops two experimental methods: a soft, fresh cheese drained in cloth and a pressed wheel suitable for salting and short maturation. They are not claims of archaeological certainty. Each method translates Roman-era ingredients and household techniques into a safe Australian kitchen while keeping the evidence, assumptions and changes visible.

What Roman evidence can tell us

Columella discusses turning milk into cheese and managing the curds, Pliny compares the milks of sheep, goats and cows, and Apicius uses cheese in savoury dishes. The Latin caseus covered a broad family of products rather than one standardized cheese. Fresh curds, pressed forms, salted cheese and aged stores could all sit within that word, which is why assigning every modern cheese a Roman name can be misleading.

No complete Roman household cheesemaking recipe has survived, and “Roman-style” must therefore be stated as an interpretation. Evidence from texts, excavated equipment and modern experiments can support a method, but it cannot establish an exact ancient recipe. Sheep and goat milk are plausible starting points because both were familiar in ancient Mediterranean foodways; goat milk was valued for its distinct flavour, while sheep milk contributed richness and a dense curd.

The broader Archeoricette research follows this evidence-led approach across recipes, event reports and published work. Its experimental reconstructions are valuable precisely because they show the reasoning behind each adjustment, allowing a plausible ancient flavour to emerge without pretending that one reconstructed wheel recreates every Roman cheese.

Choosing milk in an Australian kitchen

Coles and Woolworths usually place bulk cow’s milk in the main dairy aisle, while sheep and goat milk are more often found in a specialist dairy or refrigerated section. In Melbourne, Sydney and Adelaide, independent cheesemakers, food retailers and online dairies often provide a wider choice. If ewe’s milk is scarce, goat’s milk can lead the batch, although the result will be lighter and more sharply flavoured.

The word “ewe” is common in Australian farming and cooking, while “sheep’s milk” appears on many retail labels. Fresh ewe’s milk may follow southern spring and summer lactation, so availability can be seasonal; goat’s milk is often steadier through the year. Check the use-by date, keep milk below 5°C, and avoid cartons that have swollen or leaked. Sale of unpasteurised milk for drinking is prohibited or tightly controlled across most states and territories, so pasteurised milk is the sensible basis for home trials.

Australian recipes use metric measures, making grams and millilitres straightforward for scaling a batch. Australian cooks often say to “give the milk a gentle cook”, a phrase worth preserving here: prolonged boiling can make an acid curd grainy and toughen the final texture. Use a thermometer rather than judging heat from the pan or the weather.

Making a fresh ewe and goat cheese

For the simplest experiment, combine 700 ml pasteurised whole ewe’s milk with 300 ml pasteurised whole goat’s milk. This ratio produces a creamy curd while retaining a goat-like aroma. A reverse ratio will taste more strongly of goat, and a cow’s-milk blend will be milder and easier to texture. Record the ratio because it will shape both the yield and the finished flavour.

Warm the milk slowly to 85–90°C, then stir in live yoghurt or a culture sold for fresh cheese, following the supplier’s rate for one litre. Keep the mixture warm until the whey separates cleanly, usually 30–60 minutes. Ladle the curds into clean muslin suspended over a bowl, let them drain for two to four hours, and avoid squeezing hard. Turn the bundle out, sprinkle lightly with fine salt and refrigerate. Eat within a few days.

The culture supplies the acid that makes this fresh, ricotta-like form, whereas rennet supplies a firmer set and becomes central to the pressed method below. Neither detail should be presented as uniquely Roman. For serving, crumble the cheese over beans, add fruit and herbs, or place it beside bread and olives. The Greek symposion appetisers guide shows how a classical drinking party revolved around savoury foods, giving this fresh cheese a credible social setting without pretending Roman and Greek customs were identical.

Pressing curds into a firmer wheel

For a firmer trial, begin with the same one-litre milk blend. Add a small amount of culture according to its instructions, then add animal or microbial rennet at the maker’s stated rate. Coagulate at 34–36°C, rest for roughly 30 minutes, cut the curd into 1 cm pieces and heat slowly to 50–55°C while stirring gently. The lower heat compared with many cooked-curd cheeses should preserve a supple texture.

Drain the curds in a lined food-grade mould and press under gentle pressure for two to six hours. A weight of about 1 kg for every 500 g of drained curd gives enough pressure to expel whey without crushing the grains. Salt the surface at roughly 1.5–2 per cent of the curd’s weight, or use a light brine. For a four-to-eight-week trial, wrap the cheese and keep it in a cool, clean space below about 13°C, turning it regularly and watching for slime, deep cracks or an unpleasant smell.

This is a modern pressed-cheese method informed by Roman use of curds, pressing and storage, not an excavated Roman formula. Summer heat in Australia makes cellar conditions harder to maintain, and ageing requirements differ among states and territories. Use pasteurised milk, clean equipment and food-grade rennet, and do not sell the result without meeting local food-business rules. Anyone at higher risk from foodborne illness should avoid experimental aged cheese. For a modern Italian perspective on seasonal produce and garden-led cooking, Italian garden produce offers a useful parallel.

Pairing cheese with Roman flavours

The fresh cheese needs little interference: ripe figs or dates, a drizzle of olive oil, oregano and a little honey echo the sweet-savoury combinations familiar in ancient Mediterranean cooking. The pressed wheel can be sliced with hard cheese, crumbled into a peppery sauce or served with herbs and preserved vegetables. Black pepper was costly and prestigious in Roman contexts, so a restrained pinch suits the historical setting better than a heavy coating.

Serve the two textures together at a Roman convivium because contrast is useful: soft, tangy curds against a firm, salty rind. Olives, radishes, cucumbers and small portions of bread create a meal resembling opson, the accompaniment to wine described in classical sources. Modern seasonal produce can be substituted without pretending it was standard in every province. Such adjustments are part of the experiment, and a note beside the plate makes the distinction clear.

Reliable choices for an Australian cheese trial

A controlled first batch is more useful than an ambitious wheel. Keep the following in mind before setting up the kitchen:

  • Use pasteurised whole ewe’s or goat’s milk and check the use-by date and refrigeration instructions.
  • Equip yourself with a thermometer, clean tools, muslin and a food-grade mould rather than relying on improvised equipment.
  • Make one-litre batches, recording the milk ratio, culture, rennet, coagulation temperature and drainage time.
  • Compare fresh and pressed methods before extending the ageing period.
  • Buy from specialist dairy counters or growers’ markets such as Melbourne’s Queen Victoria Market, Adelaide Central Market and Sydney Growers’ Market.
  • Keep young cheese below 5°C and follow state and territory requirements before selling or serving it publicly.

At a market stall, ask the producer about pasteurisation, breed and milking date; direct questions often reveal whether a milk is seasonal or pasteurised. Some Australian sellers use “sheep” rather than “ewe” on labels, and both terms refer to the same milk. The project is complete when the cheese tastes coherent and the notes allow another cook to repeat it, not when it matches an idealised ancient image.

Bringing the experiment to the table

A useful result does not have to imitate a named Roman cheese. It can answer smaller questions: does goat milk improve the set, how much whey remains after pressing, or which salt level suits Australian bread? Share photographs of the curd, wheel and tasting plate, record what changed between batches and invite diners to describe aroma, texture and acidity without prompting them toward a “correct” answer.

Researchers and cooks can publish findings or send tasting notes through contact Archeoricette. Make one fresh batch and one pressed wheel this month, document the temperatures and timings, and send the results with a photograph of the finished cheese to keep the reconstruction moving.