Food History: A Roman soldier’s diet

The diet of a Roman soldier consisted of wheat, rations of smoked bacon or fresh meat (when available, usually pork), vegetables, legumes, cheese, vinegar, olive oil, and wine. These rations were issued several times per week so that every soldier would have carried food for around three days. The cost of the ration, around 60 Denarii per year, was deducted from each soldier’s pay.

Food

Just like for Roman civilians the bulk of the diet of a Roman soldier consisted of wheat that was usually eaten as either bread or as a porridge known as puls made from emmer wheat. Of the other grains available, oats were seen as fodder and would only be eaten in times of extreme hunger, millet was only grown in small amounts, rye was only grown in areas too cold for wheat, and rations of barley are recorded being issued to soldiers as a punishment for minor offences.

In the second century BC, Roman soldiers received a ration of about 66 pounds of wheat per month. The grain was issued whole so had to be milled and processed by the soldier himself. The men of each Conturbernium [1] were responsible for preparing their meals, either doing so together or with individual soldiers taking turns. Neither temporary military camps nor long-term Roman forts had any sort of mess hall or big kitchens in which meals could be prepared for the entire unit.

As Romans preferred pork (from Latin: porcus ‘pig’) over beef (Latin: bubula) [2] it is safe to say that the bacon was an important part of the Roman soldier’s diet as it provided him with both fat and a lot of calories. When smoked the bacon would be much easier to transport than, for example, olive oil, another source of fat, which had to be transported in either barrels or clay amphorae, the latter being obviously more fragile.

In addition to the wheat and bacon the ration included legumes like lentils and beans that were high in protein. Cheese (Latin: caseus) was also a staple of the soldier’s diet. Since Romans did not really appreciate beef or, by extension, cow’s milk, cheese was primarily made from either sheep’s or goat’s milk. Cheese in the ancient world seems to be mainly of the soft varieties such as cottage cheese and feta. As is the modern practice, the latter was formed into large blocks and aged in brine (being washed before use to reduce its saltiness). Smoked cheeses were also made to preserve them for longer so may well have been favoured by soldiers when available.

Archaeologists have found additional foodstuffs at several Roman army camps suggesting that soldiers, especially officers, could also buy finer imported ingredients like coriander, oysters or spices such as pepper imported from India.

Drink

Water was essential as every man might consume between 2 and 8 litres (½ to 2 gallons) per day depending on the climate and the physical demands he had to face. One major logistical problem was the supply of water fresh which could prove unreliable at times, especially on a campaign. Consequently, water was often transported in barrels, but this might easily become tainted. Wine and vinegar, therefore, were mixed with the less than fresh water to counteract any aftertaste. By adding vinegar to water the Romans created Posca, an everyday drink popular in the army and also the urban poor. It has a refreshingly sour taste, but one that would have also disguised the smell and taste of stale water. The Roman’s development of posca was also convenient economically since the faulty storage of wine tends to result in vinegar.

According to Andrew Dalby, the initial unfamiliarity with posca in the Greek-speaking East meant writers such as Plutarch used the Greek term oxos ‘vinegar’. The same word was used in the crucifixion narratives which was rendered acetum in the Vulgate Latin translation of the Bible and ‘vinegar’ or its equivalent in many modern versions (Dalby, 2003, 270). Thus, the drink handed to the crucified Jesus was most likely posca and, contrary to the popular belief, may not have been intended as a torment but a sign of compassion. The Roman soldier gave Jesus his preferred drink, and probably from his own ration.

Alongside water, posca, and beer in the northern provinces, wine (Latin: vinum) was a popular drink. In general, most Romans thought the habit of drinking pure wine was barbaric, although extremely good quality wine was drunk pure albeit only in small quantities. ‘Moderation was promoted by the almost universal Greek and Roman habit of drinking wines mixed with water’ (Dalby. 2003, 350). For soldiers on campaign that had the bonus of limiting their opportunity for drunkenness and any resulting disciplinary issues.

Bon appétit!

Reference:

Dalby, A. (2003), ‘Food in the Ancient World from A to Z’, London: Routledge.

Endnotes:

1. As defined in the Oxford Classical Dictionary, contubernium meant a ‘dwelling together’, as of soldiers or animals, but referred especially to a quasi-marital union between slave and slave or slave and free. Within a century, a contubernium consisted of eight soldiers so ten contubernia formed a century of 80 fighting men commanded by a centurio (centurion).

2. For the most part ancient Roman authors on farming practices assume cattle will be kept as working animals rather than for milk or meat.

 

Dispelling Some Myths: Was yellow the colour of prostitution?

Over the years we have periodically encountered claims that prostitutes in the Roman Iron Age and then later in the Mediæval period wore yellow garments or other yellow markings of some sort as signs of their profession. It seems this stems from the idea that people in the past wished to distinguish sex workers from respectable citizens. It is not entirely clear, however, whether a single colour or symbol was ever used to differentiate social class or social roles throughout history. So, in this article we explore what evidence there may be and in the process (hopefully) dispel some myths.

When in Rome

Most scholarship on Roman prostitution implies there was a social hierarchy within the profession. Meretrix (‘woman who earns, paid woman’), for example, seems to refer to a freeborn, higher-class registered prostitute, whereas scrotum (possibly from ‘hides, leather’) denotes an impoverished low-class streetwalker, and amica a purely euphemistic ‘lady-friend’ (Adams, 1983, 321 - 358). Given these distinctions, it is likely that Roman prostitutes dressed differently from respectable citizens. Some modern scholars assert that meretrices wore the toga when in public, either by compulsion or choice, and that the garment may have been imposed on adulteresses as a public signal of their disgrace. Edwards asserts that the toga, when worn by a meretrix set her apart from respectable women, while also suggesting her sexual availability (Edwards, 1997, 81–82). It is worth remembering that the toga was the formal attire of male citizens and not worn by respectable adult freeborn women. The latter, known in Latin as matronae (sing. matrona; ‘matron’), wore stolae (sing. stola) on formal occasions or while out in public. The stola was a long, body-concealing garment typically worn over a long-sleeved tunic, both of which having hemlines at ankle height. Raddicke states that, to indicate their disgraced social position, prostitutes and adulteresses were forbidden to wear stolae (Radicke, 2022, 299–354 & 680–688). Yet, opinions seem divided on the matter of prostitutes wore togas. Some scholars take it literally, while others see toga wearing as a euphemism for a self-assertive, ‘masculine’ woman.

Rather than a toga, references are made to expensive courtesans wearing gaudy, transparent silk garments (Edwards, 1997, 81). Similarly, the wearing of bright colours - ‘Colores meretricii’ - and jewelled anklets also marked them out from respectable women (Balsdon, 1963, 224, 252–4 and p. 327n). Yellow may well have been worn, but the reference to ‘colours’ plural strongly suggests it was not the only colour. We can speculate that what Roman prostitutes wore rather depended on whether they were free or enslaved. As Radicke points out, free prostitutes and adulteresses could wear what they wished, and neither laws nor custom dictated otherwise (Radicke, 2022, 365–374 & 578–581). By contrast, enslaved prostitutes could be made to wear anything their owners or pimps determined, perhaps including a so-called ‘women's toga’. Radicke speculates that this, if it was a ‘thing’, was the toga exigua, a term borrowed from the Roman poet Horace for a short toga worn in the 1st century BC. The assumption being that having less material the short toga would be less costly, more convenient and easier to remove. Moreover, such a garment would have exposed the lower leg and parts of the torso, something no respectable woman would have countenanced. Yet there is little concrete evidence for prostitutes and adulteresses wearing togas. Blanshard draws attention to some Roman authors seemingly indicating that prostitutes displayed themselves naked. In terms of advertising, this does make sense but in the Roman mind nudity and exposure to the public gaze were associated with slavery (Blanshard, 2010, 24). Evidence from wall paintings in the brothels of Pompeii and Herculaneum clearly show men and women naked while actively engaged in sex acts, but they also show some women, presumed to be prostitutes, wearing the Roman equivalent of a bra.

The only suggestion that yellow may have been the colour of Rome’s prostitutes stems from one element of the salacious tales woven about Emperor Claudius’ third wife, Valeria Messalina, who enters history with a reputation for being ruthless, predatory, and sexually insatiable. It should be noted that those recording the stories about her were writing some 70 years after the events in an environment hostile to the imperial line to which Messalina had belonged. Two authors were especially instrumental in promoting Messalina's notoriety, their gossiping being later incorporated into official versions of Roman history. One such story is Pliny the Elder’s account of her all-night sex competition with a prostitute that reputedly lasted ‘night and day’ with Messalina winning with a score of 25 partners (Pliny the Elder, The Natural History, Book 10, Chapter 83). More importantly a link to the colour yellow occurs in the poet Juvenal’s sixth satire. The poem contains the infamous description of how the Empress used to work clandestinely all night in a brothel under the name of the She-Wolf (Juvenal, The Satires, VI) [1]: 

‘She would go about with no more than a maid for escort. The Empress dared, at night, to wear the hood of a whore, and she preferred a mat to her bed in the Palatine Palace. Dressed in that way, with a blonde wig hiding her natural hair, she’d enter a brothel that stank of old soiled sheets, and make an empty cubicle, her own; then sell herself, her nipples gilded, naked, taking She-Wolf for a name…’

In his poem, Juvenal not only coined the phrase meretrix augusta (the imperial whore) which was frequently applied to Messalina thereafter, but he also mentions she disguised herself with a blonde wig. Is this then the source of the notion Roman prostitutes wore yellow? Possibly. While prostitution in ancient Rome was legal and licensed, contrary to Maddenholm, there is no reason to believe that Roman prostitutes were distinguished by wearing yellow or that they ‘were required by law to dye their hair blond or wear blond wigs to set themselves apart’ (Maddenholm, 2022, Haaretz Archaeology online). Indeed, Maddenholm’s statement, or a variation thereof, is frequently reproduced in online accounts of Roman prostitution but to date the author cannot identify any specific law requiring prostitute to dye their hair or wear a blonde wig. Where this notion originated is also unclear so for the time being it ought to be considered an urban myth.

Source?

Here is one example of how frustrating it can be to obtain a definitive answer. To a question on the US website Reddit ‘Why was the colour red associated with prostitution?’, an unidentified respondent, replying as ‘postmodernpenguin’, stated:

‘In several cultures, prostitutes were forced to wear different colours than most of the populace to distinguish themselves from the 'respectable' women. Various places used different colours; in ancient Athens saffron dyed cloth was used, in Rome prostitutes were required to wear wigs or dye their hair yellow, but the most prevalent colour used was red, probably owing to the story of 'Rahab the harlot' from the book of Joshua in the Bible, a story in which Rahab the harlot was forced to identify her house with a length of scarlet rope in her window.’

As seems normal with most of these forums, no sources or references are given for the claims made. In other words, apart from the easily checked bible reference, ‘postmodernpenguin’s’ answer is pretty much worthless as all the other statements cannot be independently verified or substantiated.

Sumptuary laws in Mediæval Europe

In Europe, from the ancient Greeks and Romans onward, nearly all societies have used sumptuary laws (Latin: sūmptuāriae lēgēs) to control and reinforce social hierarchies and morals by placing restrictions on clothing, food, and conspicuous spending. Sumptuary laws attempted to regulate the balance of trade by limiting the market for expensive imported goods to a town, region or country in favour of supporting the local economy. Such laws were intended to make it easy to identify social rank and privilege or prevent commoners imitating the appearance of aristocrats. Nevertheless, they also could be employed to discriminate against disfavoured groups. One obvious example is the infamous ‘Yellow Badge’, also known as yellow patch, Jewish badge or yellow star (German: Judenstern, lit. 'Jew's star'). Marking the wearer as a religious or ethnic outsider, Jews were ordered to wear it by some Muslim caliphates and European states during the Medieval and early modern periods, and by the fascist dictatorships of Italy and Nazi Germany in World War II.

For the present purpose, however, sumptuary laws introduced in the 13th century established special forms of dress for prostitutes and courtesans. At various times and places, prostitutes were not allowed to wear jewels, embroidery, or other finery, as was often the case for other women of low rank, like domestic servants. In Marseille, a striped cloak was to be worn, while in England a striped hood was the distinguishing garment. Prostitutes and courtesans were explicitly banned from wearing fur-lined hoods in England, or fur-lined clothing in 1360’s France.

Over time, these markers tended to be reduced to distinctive bands of fabric attached to the arm or shoulder, or tassels on the arm. As to colour, laws were passed in 13th century Zurich requiring prostitutes to wear red caps. The website ‘Beachcombing’s Bizarre History Blog’ rather usefully has collated various sources to produce a list of symbols required by different European cities and regions to distinguish prostitutes from ordinary citizens:

Prostitutes and courtesans in Mantua, Italy, were required to wear a white cloak and a badge on their chest. What the badge consisted of is unclear, but a record in London reveals that, in 1516 at least, a prostitute was punished by being forced to wear a yellow ‘H’ for harlot.

As can be seen, a yellow cloak, a yellow headband, yellow scarves and even a yellow ‘H’ are all mentioned. Interestingly, Danish teacher Sidsel Pederson notes in an article on her website ‘Postej & Stews’ that ‘we do see that yellow scarfs and trims have been a sign of prostitution in Italy, but in Northern Europe it was not the case’ (Pederson, 2019). It is quite apparent that the sumptuary laws across Europe were inconsistent. Just because yellow scarfs were worn by Italian prostitutes does not imply this was the case elsewhere. As far as Pederson can ascertain, at no point in the Mediæval or Renaissance periods did Danish law require prostitutes to don yellow garments or symbols. In fact, the late 15th century King Hans (1496) ‘wrote into law that Danish prostitutes had to wear a cap, half black, half red and only to wear cheap cloth. Uncovered hair was a sign of virginity and so prostitutes were not allowed to wear their hair uncovered’ (Pederson, 2019). As far as late Mediæval Denmark was concerned, prostitutes wore red/black caps and cheap clothing, although what that meant in practice is anyone’s guess.

The ‘Yellow’ Ticket

One country where the colour yellow was officially ascribed to prostitutes is Russia of the late 19th and early 20th centuries. ‘Yellow ticket’, ‘yellow passport’ or ‘yellow card’ are translations of an informal Russian term (жёлтый билет) for a personal identification document carried by prostitutes in the Russian Empire between 1843 and 1909. The document combined an identity card, a residence permit, a license to practice prostitution, and prostitute's medical check-up card. The document was officially referred to as a ‘medical card’ (медицинский билет) or a ‘replacement card’ (заменительный билет) as well as various other titles. Calling it a ‘replacement card’ reflects that upon registration, the prostitute left her original passport or residence permit (вид на жительство) in the local police office, it being replaced with the ‘yellow card’ as a personal identity document. Up until 1909, when the requirement was dropped, card holders were subject to periodic medical check-ups which were recorded on the card (Pravda, 2002).

It is traditionally thought that the term ‘yellow card’ is derived from the document’s colour (see right). Alternatively, the association between yellow and prostitution has been attributed to Tsar Pavel (‘Paul’) I. It is alleged that the Tsar’s passion for uniforms and decorations led him to compel prostitutes to wear special yellow dresses (Pravda, 2002). Whether they did or not is unclear, but it was enough for the yellow colour to become the symbol of the ‘world’s oldest profession’, at least in Russia [2].

Summary

Despite repeated assertions that yellow was the distinctive colour worn by prostitutes throughout European history, the evidence is inconclusive at best. Prostitutes in the Middle Ages may well have worn ‘yellow stockings’ or ‘yellow dresses’ or indeed ‘something yellow’, but so did many other people leaving the whole notion of being distinctive rather moot.  Yet, there is some evidence for the wearing of yellow garments or yellow symbols in certain cases as stipulated in sumptuary laws at different times and places. Yet, as we have seen many other colours are known to have been used to distinguish sex workers in Mediæval towns and regions. The evidence for a Roman law stipulating prostitutes had to dye their hair blonde or sport a blonde wig is lacking. Where commentators assert such a law exists, no source or reference is routinely given. Without corroboration, the critical thinker is left with the strong suspicion that this Roman ‘law’ is simply an urban myth.

Until next time, bon appétit!

References:

Adams, J. N., (1983), ‘Words for "prostitute" in Latin’, University of Köln, Available online (accessed June 10th, 2023).

Balsdon, J.P.V.D., (1963), ‘Roman Women. Their History and Habits’, Bodley Head.

Blanshard, A.J.L., (2010), ‘Sex: Vice and Love from Antiquity to Modernity’, Wiley-Blackwell.

Edwards, C., (1997), ‘Unspeakable Professions Public Performance and Prostitution in Ancient Rome’, in Hallett, J.P. & Skinner, M.B. (eds.), ‘Roman Sexualities’, Princeton University Press.

Juvenal (Decimus Junius Juvenalis), The Satires (Saturae), VI (114-135).

Maddenholm, T., (2022), ‘A Brief History of Prostitution in Ancient Greece and Rome’, Haaretz Archaeology, Available online (accessed June 10th, 2023).

Pedersen, S., (2019), ‘The yellow dress – a medieval sign of prostitution?’, Postej & Stews, Available online (accessed June 9th, 2023).

Pliny the Elder (Gaius Plinius Secundus), The Natural History (Naturalis Historia), Book 10, Chapter 83.

Pravda, (2002), ‘Three Centuries of Russian Prostitution’, Available online (accessed June 9th, 2023).

Radicke, J., (2022), ‘4 stola/vestis longa – a dress of Roman matrons’, in ‘Roman Women's Dress: Literary Sources, Terminology, and Historical Development’, (2023), Berlin: De Gruyter.

strangehistory.net, (2016), ‘Prostitutes’ Symbols’, Available online (accessed June 9th, 2023).

Endnotes:

1. The Latin word for wolf is lupus and its derivative lupa also means harlot, prostitute, whore, and she-wolf (Lupus femina), a connection Juvenal clearly intended his readers to make. Similarly, a brothel was known as a lupanar. All of which makes the founding story of Rome rather more interesting. Were the brothers Romulus and Remus raised by a wolf or a prostitute?

2. The epithet ‘world’s oldest profession’ is sometimes equally applied to espionage and spying, yet it is probably more accurate to say the ‘oldest professions’ were hunters, farmers, and shepherds.

A Brief History of Food: Chicken à la Marengo

Chicken à la Marengo is a French dish, similar to chicken à la Provençale, that consists of a chicken sautéed in oil with garlic and tomato, and garnished with fried eggs and crayfish. Over the last two hundred years the ingredients have varied considerably. Many modern recipes add olives but omit the eggs and crayfish (or prawns).

History?

According to a popular myth, the dish was first made after the army of French Emperor Napoleon Bonaparte defeated the Austrian army at the Battle of Marengo, located south of Alessandria in Italy, in June 1800. The story goes that because the supply wagons were too distant, Napoleon’s resourceful chef, Dunant, foraged in the town for ingredients and created the dish from whatever he could gather. Napoleon is said to have enjoyed the dish so much that he had it served to him after every battle. It is also said that later, when he was better supplied, Dunant substituted mushrooms for crayfish and added wine to the recipe, but Napoleon refused to eat it believing that a change would bring him bad luck. 

Inconsistencies

One critic of the Dunant version, the late food writer Alan Davidson, argued that the first published recipe for the dish omits tomatoes, but also that there would have been no access to them at the time.  Unfortunately, this is not entirely true. Tomatoes are indeed not native to Europe as they originated in the South American Andes. As a food source, they first used in Mexico before being spread throughout the world after the Spanish colonisation of the Americas. Tomatoes were not grown in England until the 1590s, with one of the earliest cultivators being John Gerard, a barber-surgeon. Gerard's Herbal published in 1597, which largely plagiarised continental sources, is one of the earliest discussions of the tomato in England. He knew the fruit was eaten in Spain and Italy, but he nonetheless believed it was poisonous. The plant and raw fruit do have low levels of tomatine but are not generally dangerous. Gerard's views were influential such that the tomato was considered unfit for eating, but perhaps not necessarily thought poisonous, for many years in Britain and its North American colonies. By the mid-18th century, however, tomatoes were widely eaten in Britain, and before the end of that century, the Encyclopædia Britannica stated the tomato was ‘in daily use’ in soups, broths, and as a garnish. They were not part of the average person's diet although by 1820 tomatoes were described as ‘…seen in great abundance in all our vegetable markets’ and to be ‘used by all our best cooks’. Reference was made to their cultivation in gardens albeit ‘for the singularity of their appearance’, while their use in cooking was associated with exotic Italian or Jewish cuisine. In Elizabeth Blackwell's ‘A Curious Herbal’ the tomato is named ‘Love Apple’ (Amoris Pomum) and described as being consumed with oil and vinegar in Italy. So, tomatoes were eaten in Italy at time of the Battle of Marengo. Whether they were available in the area for Dunant to forage is difficult to prove.

In his book ‘Napoleon's Chicken Marengo: Creating the Myth of the Emperor’s Favourite Dish’ author Andrew Uffindell examines in detail the military history of the campaign, the Battle of Marengo and the men involved. He also explores Napoleon’s eating habits and preferences and significantly the food stuffs available in the area at the time. The classic ingredients of the dish were not available around Marengo. Moreover, Dunant was not even in Napoleon’s service at the time of the battle. He was in Russia in June 1800 and did not become Napoleon’s chef until 1801 so he simply could not have created the dish after Marengo. Uffindell therefore concludes that the association of the dish with Napoleon after the battle is a myth which probably began with spin perpetrated by the Emperor himself. In the years following the Battle of Marengo Napoleon rewrote the history of the campaign to place himself and his military abilities in a more favourable light.

The recipe…

It is said that Napoleon particularly enjoyed chicken ala Provençale, often eating the dish for breakfast. It is just possible that chicken ala Provençale was the inspiration for the Chicken à la Marengo recipe that Dunant first recorded in 1809. Since then, however, the recipe’s ingredients of have varied considerably. Chefs have re-interpreted the recipe according to personal preference substituting rabbit and veal for the chicken and adding such things as mushrooms, truffles, black olives, white wine, different herbs and spices, and so on. Even the crayfish and fried eggs, once part of the original recipe, have been left out or substituted over the last two hundred years. The following recipe is an attempt to recreate the original, albeit for a modern audience:

Despite the story’s inconsistencies or whether the recipe’s origin stems from a myth, Chicken à la Marengo is worth a try. We hope you are inspired to do so. Bon appétit!

References:

Uffindell, A., (2011), ‘Creating the Myth of the Emperor’s Favourite Dish’, Frontline Books.

‘Napoleon’s Chicken Marengo’, Food.com, Available online (accessed October 31^st, 2023).