Where the Puzzle Began: Identifying a Source
There are several works from the Roman era that give us a glimpse of Roman Cuisine. Cato’s De Agri Cultura (Cato & Dalby 1998), Naturalis Historia by Pliny the Elder (Pliny the Elder & Holland 2012), and de Re Rustica by Columella (Columella & Ash 2007) serve as primary textual evidence of the crops and livestock grown and available for the Roman table as well as recipes for a few of the dishes that were being cooked during this time period, but the largest collection of Roman recipes we have is from the De Re Coquinaria. This collection of recipes is largely attributed to the famous 1st Century CE epicurean Marcus Gavius Apicius. (Apicius & Vehling 1977) While the nearly five hundred recipes contained within the volume are derived from several sources compiled sometime in the late 4th or early 5th Century, approximately three-fifths of them are believed to be Apicius’ own. (Horgan 2017)
Because of the large quantity of material available for the De Re Coquinaria, it was to this recipe collection that I turned when I began to experiment with Roman cooking.
Choosing a Puzzle: the Dish
With so many dishes to choose from, picking a single dish to experiment with was a difficult choice. I wanted to try something that relied on some of the most quintessential elements of Roman cuisine. I settled on a dish that combined the elements of sweet, sour, and salty, as it seemed to me from my reading that these elements formed the basis of Roman taste.
The dish I selected for my experiment was Aliter Pisam Sive Fabam (Apicius V, iii, 7).
Beginning the Puzzle: The Recipe
Aliter pisam sive fabam: despumatam subtrito lasare Parthico, liquamen et caroeno condies. Oleum modice superfundis et inferes.
Beans another way. When skimmed flavor them with crushed Parthian laser, some liquamen, and some caroenum. Pour a little olive oil over these, then serve.
Searching for the Pieces: The Ingredients
Period cooking has always felt like a classic example of experimental archaeology to me. Your given the fragments of recipe, and you have to find the missing pieces in order to construct a reasonably plausible facsimile of what a dish was like in period. The translation formed my jumping off point for redacting the recipe, but before I could begin, I first had to identify all of the ingredients. I turned back to the agricultural texts of the time to see what the authors of the era discussed. I also looked to recent archaeological studies to further inform my understanding of what was available and how certain ingredients were produced.
But, What Kind of Beans?
The first question that I had to ask was what kind of ‘beans’ were available in the Roman era because ‘beans’ are a pretty vague description for a broad category of pulses. I went back to the agricultural treatises from the era. What I found was that Romans had access to a number of pulses, among them lentils, peas, lupins, chickpeas, and fava beans (also referred to broad beans in some texts). (Pliny the Elder & Holland 2012)
From this list, I chose to recreate the dish with chickpeas. I know from my own experience that chickpeas are a very versatile legume which can be cooked in a variety of ways, and while they are flavorful on their own, they are also able to absorb the flavors of the other ingredients that they are prepared with. I did test the eventual final recipe with fava beans, lentils, and peas. The fava beans would be my second choice as they did absorb some of the flavors from the other ingredients. The lentils did not take on much of the flavors at all, and the peas, while flavorful, ended up as more of a pea mash.
Something Fishy: Liquamen
In looking at modern translations, there is some disagreement among translators on the meaning of the word ‘liquamen’. However, by looking to the medical and veterinary texts of the period it is clear that the term ‘liquamen’ is used to describe a liquid made by fermenting small fish with salt. (Grainger 2018) Additionally, archaeological remains from Pompeii attest to the production of this ingredient in the form of amphora with tituli picti (commercial labels) bearing the appellation of liquamen and containing the remains of a fermented fish sauce. (Grainger 2006)
In effort to be authentic, I attempted to produce my own liquamen by following a recipe from a 10th Century Byzantine text entitled the Geoponika. (Dalby et al 2011) Although this text was written much later than Apicius, it contains “material from much earlier in the Roman period and is therefore a valuable source”. (Grainger 2018) The recipe that I chose to follow from this text says to pack small fish in layers of salt in a clay pot which is then left in the sun to ferment. Once the fish has properly fermented, the resulting liquid is strained off for use. (Dalby et al 2011) I found that the process took about three months to produce the desired liquamen. I also found the process to be a rather smelly one, particularly on the days when I would remove the cover of the pot to stir its contents. This had an unfortunate consequence of not making me very popular with my neighbors for a few months. They asked me to kindly never repeat the experiment again.
Wishing to remain on good terms with my neighbors, I began exploring other methods to attain the same flavor without subjecting anyone to the aroma that accompanied the production process. From reading the works of Sally Grainger, I learned that she had been experimenting with a number of Asian fish sauces to compare them to her own experiments with liquamen. I set out to do my own taste comparisons with the liquamen I had created. I purchased several varieties and tasted them. From my taste tests, I learned that Thai fish sauces made with sardines and/or anchovies most closely approximated the flavor of the liquamen produced from the period recipe although the Thai fish sauce is just slightly sweeter in flavor.
Something Sweet: Caroenum
A text from the late 4th Century described caroenum as ‘must’ which has been boiled down so that one-third of its volume is lost. (Palladius 1879) Being familiar with the wine making process, I understood this to mean the juice that is produced from the first crushing of the grapes before fermentation has begun to take place. Some food historians theorize that caroenum was made from the juice of white grapes due to the existence of a sweet wine from Provence which bears a similar name, but with no definite answer to what type of grapes were used to produce the caroenum, I was eager to test the varieties for myself.
I created caroenum from the juice of both white and red grapes. I found that both reduced well and that while both were about the same level of sweetness, the red grapes added a more complex flavor to the dish. For this reason, I chose to use caroenum produced from red grapes for this particular dish.
Parthian Laser: Flavoring your dish with plant resin?
From reading Apicius, I knew that the terms ‘laser’ and ‘Parthian laser’ appeared in several recipes. At first glance, these would seem to be interchangeable terms for the same spice, but they are, in fact, two distinct but related spices. The term ‘laser’ referred to the powder that was created for from “the resin from the stalk, the dried and ground root and the leaves” of the fennel-like Silphium plant that grew wild in Africa and was harvested to extinction sometime around the 1st Century CE. (Grimm 2007)
‘Parthian laser’, on the other hand, is created from the Hing plant, a plant belonging to the fennel family from Asia. It was available contemporaneously with laser although it was considered to be slightly inferior to it. The plant still exists to this day, and the resulting spice created from it is known as Asafoetida. Like silphium, the parts of the plant from which asafoetida is derived are reduced to a resin which is later ground into a powder for use in cooking.
The resin is very aromatic in nature, and the aroma is by no means diminished once it is reduced to powder for use in cooking. The resulting spice has a unique savory umami quality to it that combines the flavors of garlic and onion with an underlying note of sourness to it. Food historians believe that it has much the same flavor as the laser created from the now extinct silphium plant. (Grimm 2007) It is possible that the extinction of the silphium may be what led Apicius to start specifically identifying the ingredient as ‘Parthian laser’ rather than just ‘laser’.
Collecting the Pieces: The Redaction
Step One: Collecting the Hardware
I wanted to prepare my dish in as period fashion as possible. To that end, I began looking at archaeological finds from the Roman era for period cooking utensils. I found what I was looking for in some recent finds from the cities of Pompei, Herculaneum, and Sanisera.
So I set about finding cookware that would closely approximate these items. The most important of these items for me was the cooking pot. I talked with a friend who makes pottery. After looking at pictures and other documentation, she was able to replicate a pot that closely resembled an olla.
I could not find an adequate reproduction of the Roman colander. In its place, I used the closest thing I could find which was a pierced metal colander.
Step Two: Preparing the Beans
With my ingredients identified, I went about my experiment of recreating the dish. I chose to create the dish as an accompaniment to a small family meal, and I based my determination of how much of my main ingredient I would need on this proportion. I began with just cooking the chickpeas. Then, as now, it seemed likely that chickpeas would need to be dried for long term storage.
So, I began by soaking the beans overnight to soften them. I then drained them and covered them with fresh water before starting to cook them. I brought the liquid up to a boil and boiled the chickpeas until they were tender enough to eat, but not so soft that they turned to mush when being stirred. During the cooking process, I also used a spoon to skim off any foam that rose to the top of the cooking vessel.Then I drained off the cooking liquid. This seemed to me consistent with Apicius’ direction to “skim” the beans. Once I was sure of my cooking times, I began to experiment with the flavors.
Step Three: Building the Flavor
My first question was this meant to be a sauce or a dressing. As there is no mention of the ingredients being cooked together, I determined that it was meant to be a dressing to the dish. This led me to the idea that smaller increments of the ingredients were used as it would not require a great quantity to dress the chickpeas. Using a notebook to record my variations, I began to test this portion of the recipe.
Some iterations proved entirely too salty when liquamen was the largest component. Too much asafoetida produced a dish that was both overly sour and a little bit stinky. Caroenum, on the other hand, could produce a dish that was entirely too sweet when it was used too liberally. After several iterations, and many taste tests by my friends and family, I could create a recipe that I believe reflects a balance of the three competing flavors. The resulting dressing was very aromatic in nature. While the dish was quite light, the flavor was a subtle mixture of the sweet, salt, savory, and sour.
Completing the Puzzle
Putting Together the Pieces
I tested the recipe many times to make sure that the recipe was reproducible, and I asked a friend to test it as well to make sure that my results could be recreated by someone other than myself. I also tested the recipe using entirely period cooking methods as well as modern ones. I achieved the same results using both methods, though I will note that the period method took slightly longer than the modern method. The recipe below represents my best approximation of how I believe the dish was prepared in period.
- 1 cup dried chickpeas
- ½ tsp asafetida
- 2 tbsp liquamen
- 1 1/3 tbsp caroenum
- Olive oil
Soak chickpeas overnight in cold water. Drain and place in pan. Cover with twice the amount of water and bring to a boil. Cover and simmer for 1 hour or until tender. Drain. In separate bowl mix liquamen, caroenum, and asafoetida. Pour mixture over chickpeas in serving bowl and mix well. Drizzle with olive oil and mix well. Serve.
The Big Picture
How did it turn out?
I found the dish to be really very flavorful. There was a richness to the complex flavor of the dressing, and yet it was light enough that it did not overwhelm the flavor of the chickpeas themselves. I did find that there was a fine line when cooking the chickpeas between tender enough and cooking them to the point of mushiness. I found that overall, I much preferred the texture of the dish when the chickpeas were tender but still firm enough to retain their shape until eaten. I did not enjoy the texture of the dish when the chickpeas were cooked to the point that they became a mush when mixing the dressing into them. I will also say that I prefer this dish when it is made with liquamen that I made as opposed to when it is made with commercially available fish sauce. The extra sweetness that is present in the commercially available fish sauce somewhat upsets the balance of the flavors. I plan to make more of my own liquamen thanks to the auspices of a friend who has offered to let me make it on a remote portion their property where neighbors will not be impacted by its production.
What did other people think?
Ok, but what did other people think of the dish? I tested this dish on people both inside and outside the SCA. Friends outside of the SCA were initially leery about trying “strange medieval food”, but they agreed to try it nonetheless. Of the crowd of non-SCA friends who tasted it, most said it was not a flavor combination that they would have thought to put together, but for the most part, they really enjoyed the dish. The one person who didn’t like it said that the flavor was good, but they just don’t care for chickpeas no matter how they were prepared. It went over so well with my colleagues that I now get frequent requests to bring this dish to work potlucks.
I tested this first to my closest SCA friends, many of whom are already medieval foodies but who have little to no experience with Roman cuisine. One of the most common comments that I got was that people were surprised with how light the dish was, and what bright notes that got from the flavor combination. Another comment was that it had a very nice balance of savory, acidic, and salty that is unusual in medieval dishes. Emboldened by my friends’ praise, I decided to try it in the larger context of served dish at a feast. The dish seemed to go over well as I had many requests for additional helpings of the dish. In talking to feast attendees, one of the things that they mentioned being concerned about was that the dish would be too salty because of the liquamen. They were pleasantly surprised at the balance of the overall flavors, and several people asked for copies of the recipe so they could make it home. At the end of the feast, there was only a very small amount, perhaps a couple servings, of the dish leftover.
How does it fit in the larger puzzle of the Roman diet?
So where does this dish fit in the context of Roman cuisine at large? Apicius was a famous epicurean, and historians have long assumed that the recipes contained in the text that is attributed to him reflect only the cuisine of the elites of Roman society. This would suggest, then, that this was a dish that belonged solely to the elites. However, new archaeological discoveries from the cities of Pompeii and Herculaneum are changing our understanding of the diet of the average Roman citizen. This new evidence is suggesting the average Roman citizen had access to many of the same the ingredients that the elites had, although perhaps in lesser quantities. (Fina MacGrioghair A.S. LIV/2020)
So if I had to guess where this dish fit in the context of overall Roman cuisine, where would I put it? My best guess is that is a dish that crosses socio-economic boundaries. The ingredients were widely available, and the recipe is one which is easy to make. This suggests to me that it could have been eaten both the elites as well as the average citizen. These same considerations suggest that it would also be a dish that could easily be prepared and served in private homes or at tabernas and thermopalia, making it available as part of the “fast food” culture that existed in many Roman urban population centers. I think that is also a dish that works equally well as a snack as it does as an accompanying side dish to a larger meal.
I hope that this exploration of a Roman dish has sparked your interest and your curiosity with regard to Roman cuisine. For myself, it is a topic of intense fascination. What started as curiosity has blossomed into passion. In addition to exploring new recipes, I have incorporated many of the recipes that I have already redacted into my mundane menu. This dish is a particular favorite of both my roommate and myself, and we have found that it is a tasty accompaniment to roast lamb. I encourage you to try the recipe for yourself, and I hope that you enjoy it as much as I do. I have also attached a copy of my complete documentation of the recipe below, as well as a copy of my research paper on the archaeology of the Roman diet if you are interested in learning more.
- Apicius, M.G., Flowers, B., & Rosenbaum, E. (2012). The Roman Cookery Book. Mansfield, CT. Martino Publishing.
- Apicius, M.G., & Vehling, J.D. (1977). Cookery and dining in imperial Rome. Courier Corporation.
- Cato, M., & Dalby, A. (1998). Cato on Farming: De Agriculture. Marion Boyars.
- Columella, L.I.M, & Ash, H.B. (2007). On agriculture. Cambridge, MA. Harvard University Press
- Dalby, A., Bassus, C., & Constantine, V. I. I. (2011). Geoponika: Farm Work: A Modern Translation of the Roman and Byzantine Farming Handbook. Prospect Books.
- Fina MacGrioghair, The Archaeology of the Roman Diet. (A.S. LIV/2020)
- Grainger, S. Garum and Liquamen, What’s in a Name?. J Mari Arch 13, 247–261 (2018). https://doi.org/10.1007/s11457-018-9211-5
- Grainger, S. (2006). Towards an authentic roman sauce. In Authenticity in the Kitchen: Proceedings of the Oxford Symposium on Food and Cookery 2005 (pp. 206-10). Oxford Symposium.
- Grimm, V. (2007). The Good Things That Lay at Hand. Tastes of Ancient Greece and Rome. Paul FREEDMAN, ed., Food, The History of Taste, London, Thames & Hudson Ltd.
- Horgan, J. (2017, November 10). Marcus Gavius Apicius. Ancient History Encyclopedia. Retrieved from https://www.ancient.eu/Marcus_Gavius_Apicius/
- Palladius, R. T. A. (1879). Palladius on husbondrie. Early English Text Society.
- Pliny (the Elder) & Holland, Philemon. (2012). Pliny’s Natural History in Thirty-seven Books. Hardpress Publishing.
14 Replies to “Reconstructing a Roman Dish: Putting Together the Puzzle Pieces of “Beans, Another Way””
This is really great work! I enjoyed the walk-through of your research process and I like the way you break down the ingredients and the cooking method to combine it into a reconstruction. And then follow through with to who and how it was available. I also enjoyed your attached research pdfs.
Nicely done. The explanation of your process was very cogent and clear. I look forward to seeing your next project.
I am so glad you participated, it made me recollect what a pleasure it was to judge your presentations at Kingdom A&S. I am still all pouty that you didn’t save some liquamen to bribe your judges.
A question that I didn’t think at the time, but as I’ve been re-reading yours and reading some other culinary presentations: how do you handle negative reactions to the taste of some Roman dishes? I have given up using my husband as a taste tester, as he pretty emphatically doesn’t like the last of a lot of the dishes I’ve redacted.
Roman flavors have a lot more in common with modern cuisine from India than they do with modern western European or American food. So some palates just don’t find those flavor combinations enjoyable. What have found to be more of an issue is that people find the ingredients weird, and they are unwilling to try the dish for this reason. I often find that if I can get someone to try a dish before knowing all the ingredients, they often enjoy the flavor without being put off by the ingredient that seems odd to them. I only do this, of course, when I know that there is nothing in the dish that they are allergic to or that would violate a dietary restriction that they may have. Case in point, I prepared this dish for some friends whose dietary needs were familiar to me. One friend’s partner thoroughly enjoyed the dish, and even asked for second. The partner later told me that if they had known it contained “fish sauce” they would have assumed it would be too salty and wouldn’t have tried it. They said that they were glad they didn’t know because they would have missed out on something tasty.
In a feast environment, blind tasting just doesn’t work. I am very conscientious about giving full ingredient list for every dish on my menu and avoiding cross contamination of any kind. So everyone knows what’s in the dish, and some may choose to avoid things that sound like they have weird ingredients. That being said, I prepared this dish for a feast recently. I was told by one feast attendee that they were reluctant to try it based on the odd ingredients (asafoetida in this instance), but after seeing how much their table mates were enjoying it, they decided to sample it. They loved it and asked for the recipe. This is an isolated example of course. Not everyone will try it, nor will everyone who tries it like it. My hope is that my enthusiasm and passion will at least encourage someone to step out of their comfort zone and try something that they might otherwise avoid.
Thank you for giving us an up close look at your end to end process, I especially appreciate that once you had made the liquamen you compared against modern product to see if you could find a reasonable facsimile.
This was a fantastic presentation! Well researched and written.
oh and dangerous to the pocketbook, but because of your article, I just found this lovely reproduction:
Wants it we does precious!
Very enjoyable! I’m impressed that you made your own liquamen & that the neighbors didn’t call the Health Department 😀
Did you use a particular variety of grape for your caroenum? I’ve only made saba with Sangiovese grapes, but I have to assume there would be a change in flavor from one grape to another, just as there is in wine….
By “period cooking methods” I assume you mean cooking over a direct fire?
Did you look at the other legume recipes in Apicius for comparison as well? I find that in addition to the generic Pisam (Peas), he specifically calls for Lenticula (Lentils) Fabam (Fava), Faseoli Virides (Green beans*), and Cicer (Chick Peas).
I think it would be fun to gather a group of like minded cooks & compare several of the Roman legume dishes at once to see how the variations taste.
Thank you for your presentation!
*not our modern green beans which are a new world variety, as I’m sure you know…
I enjoyed reading about your process and experiments. For the caroenum, did you purchase the juice itself or make your own from crushed grapes?
It was made from crushed grapes from a friend’s garden.
this is awesome. thank you for all your hard work.
This is awesome 😉 I know I have had a number of discussions about liquamen over the years with various people and I laud your brave soul for attempting it! 😉
I would love to try this dish 😉
I very much enjoy and appreciate your complete process. Looking at the whole, then looking at each broken down part, then pulling it together. Going too far in each direction in order to find balance is beautiful.
What quantities of liqaumen and caroenum did you produce?
So far I’ve only made a few quarts of liquamen as I was working in small batches. I hope to increase my batch sizes for this now that I understand the process and have a place where it can be made with offending anyone’s senses. As for the caroenum, at this point the largest batch I have made resulted in about half a gallon of caroenum but I did learn that it has to be stored carefully because if it isn’t, it can still ferment or be subject to spoilage.