There is a great deal of myth surrounding this strange fish sauce so popular in the ancient world. It was said that it smelt disgusting but this is not quite true. It's production generated such a foul smell that people were often outlawed from making it in their own homes but once made and kept in sealed amphora or bottles it had quite a pleasant aroma. If garum smelt bad in its container it was old and leaking!

Its said that garum was a fish paste not unlike anchovy but this is patently wrong. In the many recipes that survive for the different types of fish sauce, they all require that the finished sauce be strained of any fish pieces. When people attempt to re-create ancient recipes with anchovy paste the resulting food is quite disgusting and to be avoided!!! The natural historian Pliny the Elder, who is writing in the 1st century AD, tells us that it actually looked like aged honey wine and was often mixed with wine to drink. He says that fish, fish intestines and salt are mixed together and allowed to ferment in sealed containers. The poet Martial, writing in the same century, however, tells use that Garum was made from the blood of a still breathing mackerel !! There is some poetic license here, the fish was dead but we can say that the sauce was made from freshly killed fish and not fish that has gone off. Does he mean that the sauce was made only from the blood or the whole fish? Pliny seems to think not. We shall see.

There is also much confusion about the various types of fish sauce. Fish sauce could be called garum, liquamen or muria and there was a further product called allec which was a paste. These sauces are all by-products of the basic process of preserving fish with salt. If the mixture is left for long enough, the salt draws out the water from the fish, fermentation takes place and the solids break down into liquids. This is not bacterial action that results in decay or putrefaction but a process involving enzymes and the more active enzymes are in the gut and blood of the fish. There is sufficient salt for the resulting sauce to be safe and even sterile. Pliny also tells us that the allec, the paste that forms at the bottom of the barrel, was used to heal burns but he also says that it only works if the patient does not know what is being used!!! Fish sauce could be made from one type of fish only like mackerel. This was considered by Martial to be the best kind of garum. He also tells us that tuna fish sauce, which he calls muria, was of second quality. It could also be made from any old rubbish at the bottom of the net and this is clearly the third quality.

Different ratios of salt to fish flesh made a difference to the out come of the sauce and the time of fermentation often varied. Some recipes also call for herbs to be added at the beginning. It is not possible to match the different types of fish sauce described with the names mentioned above with any confidence or agreement amongst food historians. However I believe that, for most people in the Greek/Roman world using fish sauce was exactly like adding pure salt to food today, it was a very basic ordinary product that everyone used. The distinction between the cheapest fish sauce and one made with a particular type of fish was, I believe, largely in the mind of the purchaser. The product was a salty pungent brine that enhanced the flavour of the other ingredients in a dish- no more no less. The degree of pungency is the distinction that matters in defining the various sauces and that was determined by the amount of blood and intestines that were added as well as the fish used.

Martial has given us a hint that a sauce may have been made entirely of fish blood and intestines. There is further evidence that this may be possible from a 1st century poem called Astronomica by Manilius1 The meaning of such a poem as this is often ambiguous and difficult to decipher. He is describing a beach scene when the fishermen have landed their catch. He is obviously talking about fish sauce but he does not name the various types. Fish are brought on to the beach whole and cut up The precious fluid, literally 'flower of the gore' is saved and mixed with salt. He does not say that fermentation takes place but is surely must, before the finished product 'soothes the sense of taste' There is no mention of fish meat or flesh being added at this time and he further indicates that blood alone is meant when he says 'From the one body are put two different purposes....one kind is better with its juices drained and another with them kept in'.

He then says that a separate mass of dead fish, presumably mixed with salt and fermented, sinks to the bottom and dissolves to provide a 'soft complement to foods' He appears to be talking about the fish paste called allec with was hardly ever added to food during cooking but served separately. There would have been a liquid brine at the top of this paste which was itself a fish sauce of some kind.

We then here about another kind of process where vast numbers of small fish are dredged up in a net and turned into a large wine jar called a dolia. We have to assume that salt is added and fermentation takes place before the costly fluid slowly flows out of the fish and they are broken down into liquid gore. This is referring back to the process where the sauce is better if the juices are kept in and they are allowed to dissolve slowly into liquid form. This process appears to have no extra blood added, though just because the poet does not say this doesn't mean it doesn't happen. Can we take the poet literally? This is the problem, can we believe that this is the way that one of the sauces was made. It may be so in the 1st century AD but we do not find detailed recipes until very much later. In fact the most detailed recipe dates from the10th century AD and uses fish flesh and extra blood and calls the sauce liquamen.

There was a strong 'snob' factor in relation to food in Roman society just as in today's attitudes to food: In the case of garum it was the type of fish that was used that mattered not its strength. Apicius, for instance, considered the best garum to be made from mullet, he then recommended that another mullet should then be cooked in this sauce of its own companions! At the other end of the scale of quality, we hear that in the new province of Judaea an ordinary fish sauce was available that was kosher i.e. it was made without shell fish which the Jewish community could not eat. The implication here is that most ordinary fish sauces did contain just about anything that was caught in the net.

We will now look at the origins of these sauces. The Romans did not in fact invent Garum but, as with most things of quality, they took the idea from the Greeks. The original motive for its production is probably to do with the preservation of vast quantities of small fish that were too small for consumption but too numerous to be wasted. The Greeks developed a taste for this sauce, which they called garos and used it as a salt substitute to add a distinctive flavour to their food. It may have originally been made from a local fish called garos or from all manner of different fish. We do not know if the Greeks made their original sauce differently from the Romans. All our information date from the 1st century AD and beyond and so we can not know more about its origins.

We are told, through numerous references to garum in literature, that fish sauce was very popular in Roman society but this is not truly brought home until we read the recipes in the 'De re Coquinaria'- the cookery book attributed to Apicius that dates from the mid 4th century AD. It is possible to count on one hand the number of recipes that do not have some form of fish sauce in them. This, of cause, is not surprising as salt is crucial to enhance the flavour of cooked food. The problem is that Garum as a term is not used in Apicius apart from a few compound words donating fish sauce and other ingredients like wine or vinegar mixed together, possibly at the time of production. Liquamen is the term used in Apicius and its also the term used in the recipes and references for fish sauce that post date the book. It is vital for me that the definition of garum and liquamen is settled. Most of the recipes that I deal with come from the Apician recipe book and if there is a major difference between the original garum and liquamen then I need to know. Liquamen appears no where in 1st century AD literature as a term for fish sauce. It arrives in the Apician text and appears afterward more frequently. There are late Latin/early medieval references which actually tell us that garum and liquamen are the same thing. What is not understood is why the name liquamen is unknown in the early period. It is always expected that a paper like this should have an opinion on the issue under discussion and not sit on the fence. I therefor offer the following possible explanation.

At one stage a sauce made entirely from fish guts was called garum/os. In this form it may have been Greek in origin. In order not to waste the vast quantities of very small fish that are always caught, they were added to the mixture and a separate sauce is developed. Both arrive in Rome along with the slave cooks and recipes. When Rome becomes enthusiastic about these sauces the concept of varying the type of fish is introduced, which had not been the issue before. Gradually the sauce made entirely from intestines is abandoned. If a sauce is to be made from one particular fish then it is sensible to use the whole fish- it is the flesh surely that would have any distinctive contribution to make. The name garum is retained for the quality product now made from mackerel and fish blood to distinguish it from one made with tuna called muria. The name liquamen was given to the cheap every day variety of fish sauce, which may have originally only contained small fish, to distinguish it from the other sauces. The absence of literary evidence is explained by the fact that liquamen was a cheap every day sauce and not worthy of note by the food writers of the 1st century AD, who were, without exception, wealthy enough to aspire to the quality products. As the empire expanded fish sauce was standardized and exported to every frontier. The market for a fish sauce with a specific breed of fish fades. A sauce made from both blood and assorted fish types became the standard recipe which retained the name liquamen and the term garum faded into disuse.

Enough about the definition of the stuff, what does it taste like. Well you can find a very similar product in a Chinese supermarket. Its called Nuc Nam or Nam Pla and appears to have a very similar method of production.. However, and its a big however, it does not have any extra fish intestines added, only those present in the fish anyway. You may think this is quite a good thing but of course it does mean that the pungency is greatly reduced. It is our only substitute and will serve. If my theory is correct and I make no claim that it is, then, at one time, a fish sauce just like the Vietnamese Nuc Nam did exist in the Roman world.

Roman food is quite distinctive and very reliant on fish sauce for its success. Wine, honey, vinegar,oil and fish sauce combine to create a balance of sweet-sour-salt that is quite unique and well worth trying. I have conducted experiments with recipes, using salt instead of fish sauce and found the result to be far less appetizing. I hope I haven't put you off the idea of trying Roman food, it really is very good!

Source: SALLY GRAINGER