Open a medieval cookery book for the first time and you may discover something rather startling: the recipe appears to have forgotten to be a recipe.
There may be no ingredient quantities. No oven temperature. No cooking time. Ingredients may have unfamiliar names, familiar words may mean something quite different, and the instructions may consist of a short paragraph that seems to assume you already know exactly what you are doing.
In many ways, that is because the text does assume knowledge on the part of its reader.
Reconstructing a medieval recipe therefore involves considerably more than replacing Middle English words with modern ones. It means reading the culinary instructions in context, identifying what the text actually tells us, recognizing what it leaves unstated, researching unfamiliar ingredients and techniques, comparing related evidence, and finally making a series of informed decisions in the kitchen.
Those decisions are where interpretation becomes reconstruction.
This article explains the general process of reading and reconstructing historical recipes. For the method I developed while working on Give It Forth, including how I research, test, compare, and revise my interpretations, continue with Interpreting Medieval Recipes: My Reconstruction Process.
In This Article
- What Is a Medieval Recipe?
- 1. Start as Close to the Source as You Can
- 2. Read the Entire Recipe Before Reconstructing It
- 3. Mark What You Do Not Know
- 4. Work Through the Culinary Language
- 5. Read the Recipes Around It
- 6. Research the Ingredients Historically
- 7. Notice What the Recipe Does Not Tell You
- 8. Separate Transcription, Translation, Interpretation, and Reconstruction
- 9. Build a Kitchen-Ready Reconstruction
- 10. Test the Recipe
- 11. Compare Other Reconstructions
- 12. Record Confidence and Uncertainty
- See the Method in Practice
- Research Resources
- A Simple Reconstruction Checklist
- Frequently Asked Questions
What Is a Medieval Recipe?
Modern recipes are usually written so that a cook who has never prepared the dish before can reproduce it. We expect an ingredient list, measured quantities, temperatures, cooking times, equipment, and instructions presented in a convenient order.
Medieval culinary texts often work differently.
A recipe might tell the cook to take an ingredient, seethe it, grind it, temper it with broth, cast in spices, let it boil, and serve it forth. The writer may never explain how much broth is required, how finely an ingredient should be ground, how vigorously the pot should boil, or how the cook knows that the dish is finished.
This does not necessarily mean that the recipe was considered deficient by its original users.
The written instructions existed within a larger culinary culture. Skills, equipment, familiar preparations, sensory cues, and working knowledge did not have to be explained every time they were used.
That presents the modern historical cook with a challenge.
We cannot simply fill every silence with whatever seems convenient and call the result medieval.
Instead, we try to determine what can reasonably be recovered from the surviving evidence, where evidence becomes inference, and what choices are necessary to turn the historical text into a dish that can be cooked today.
1. Start as Close to the Source as You Can
Whenever possible, begin with the earliest and most direct version of the recipe available to you.
Depending upon the survival and accessibility of the text, that might mean:
- a digital image or facsimile of the manuscript;
- a reliable transcription;
- a scholarly edition of the culinary collection;
- a contemporary or near-contemporary version of the same recipe;
- or, when more direct evidence is unavailable, a careful translation or secondary study.
These sources can all be useful, but they do different jobs.
A manuscript image lets us see what was physically written, although reading historical handwriting may require specialized skills.
A transcription turns that handwriting into readable type, but a transcriber has already made decisions about letters, abbreviations, damaged words, punctuation, and sometimes word divisions.
A translation adds another layer because someone must decide how historical words and phrases should be expressed in modern language.
A modern reconstruction adds still more decisions.
This is why I prefer to move backward toward the historical source whenever possible rather than beginning with somebody else's finished recipe.
For much of the fifteenth-century English material on Give It Forth, Thomas Austin's edition of Two Fifteenth-Century Cookery-Books, containing Harleian MS. 279 and Harleian MS. 4016 among other material, has been an important working resource.
For other recipes, the best starting point may be a different manuscript, printed early modern book, household text, or surviving ancient culinary collection.
The date a work was originally composed and the date of a surviving manuscript copy are not necessarily the same thing. A culinary work might have been copied, rearranged, expanded, or transmitted through several manuscripts over many years. When possible, note both the historical work you are studying and the particular witness or edition you are using.
2. Read the Entire Recipe Before Reconstructing It
One of the easiest traps in historical cooking is deciding too quickly what a recipe is supposed to become.
A familiar ingredient can pull the modern imagination toward a familiar dish.
Eggs become scrambled eggs.
Something enclosed in dough becomes a baked pie.
A thickened liquid becomes gravy.
Fruit and sugar become dessert.
Sometimes those conclusions are entirely reasonable.
Sometimes the rest of the recipe quietly destroys them.
I therefore read the complete recipe several times before trying to turn it into a modern formula.
I pay particular attention to the verbs.
- What is boiled or simmered?
- What is baked?
- What is fried?
- What is ground?
- What is strained?
- What is chopped?
- What is mixed together?
- What is added only at the end?
- Is a liquid discarded, retained, thickened, or poured over the finished food?
- Does the recipe tell us to serve immediately after a particular step?
The sequence of actions can be more revealing than the ingredient list.
If the manuscript says to boil something, we should have a very good reason before deciding that what it really meant was bake it.
3. Mark What You Do Not Know
Uncertainty is not a failure of research.
It is information.
If you encounter a word you do not understand, resist the temptation to replace it silently with whatever seems plausible.
Mark it and investigate it.
When I originally worked through Tartlettes from The Forme of Cury, for example, I encountered terms such as y-sode, ayren, raisons of coraunce, powdour fort, foyle of dowh, fars, and sew.
Several of those terms determine something important about the finished dish.
If we misunderstand the dough, the tartlette itself changes.
If we misunderstand sew, the liquid served with it changes.
If we treat powdour fort as though it were a single standardized spice mixture known everywhere in exactly the same proportions, we may create certainty the historical text does not actually provide.
A few apparently small guesses can redirect an entire recipe.
4. Work Through the Culinary Language
Middle English spelling can look intimidating because spelling was far less standardized than it is today.
Once familiar patterns emerge, however, many recipes become surprisingly readable.
Some useful examples include:
| Historical Term | General Meaning |
|---|---|
| ysode / sode / seethe | Boiled or simmered |
| ayren / eyroun | Eggs |
| raisons of coraunce | Currants |
| fars | A stuffing or mixture used to fill something |
| sew | A broth, sauce, gravy, juice, or other liquid preparation depending upon context |
But a dictionary definition gives us possibilities, not always an answer.
The meaning of a culinary term depends upon how it is being used in that particular recipe, manuscript, time, and culinary tradition.
This is why historical dictionaries and culinary glossaries are most useful when combined with contextual reading.
5. Read the Recipes Around It
A medieval recipe should not automatically be treated as an isolated scrap of instructions.
Look at the culinary collection around it.
What recipes come before and after it?
Does the same ingredient appear elsewhere?
Does the writer use the same cooking verb in another recipe where its meaning is clearer?
Is there another version of the dish in the same manuscript or a related collection?
Does the section heading tell us anything about how the compiler understood the recipe?
A preparation may be explained fully in one place and referred to much more briefly later because the writer expected the reader to recognize it.
This sort of contextual reading has become increasingly important in my own work.
For example, when I first reconstructed Tartlettes, the phrase foyle of dowh raised questions about the nature of the dough.
Another recipe in The Forme of Cury, Loseyns, calls for flour and water to be made into paste, rolled into thin sheets, dried, and boiled.
That does not prove that every medieval foyle of dowh was made identically.
It does, however, provide useful evidence that a thin flour-and-water dough intended for boiling existed within the same culinary tradition.
That is the difference between evidence and proof.
6. Research the Ingredients Historically
Recognizing an ingredient's name is only the beginning.
Ask what that ingredient meant to the historical cook.
A modern product sold under a familiar name may not be identical to its historical counterpart. Varieties change. Processing changes. Availability changes. Trade changes. Preservation changes. Even the expected flavor of an ingredient can change.
Useful questions include:
- Was this variety available in the period and place being studied?
- Was the ingredient usually fresh, dried, salted, smoked, fermented, or otherwise preserved?
- Was it a common household food or a luxury import?
- Was it seasonal?
- Was the historical product botanically or zoologically the same as the modern one?
- Does another contemporary source describe how it was prepared?
- Would processing have changed its flavor, texture, or strength?
Spice mixtures deserve particular caution.
Terms such as powdour douce and powdour fort identify broad seasoning ideas, but the surviving culinary evidence does not justify treating them as one universally standardized formula used in every kitchen.
We can create historically plausible working mixtures from comparative evidence.
We should call them reconstructions.
7. Notice What the Recipe Does Not Tell You
Historical recipes frequently omit information modern cooks expect.
Among the missing details may be:
- ingredient quantities;
- ratios;
- cooking times;
- temperatures;
- pan or vessel sizes;
- the amount of cooking liquid;
- the exact texture of the finished dish;
- the yield;
- or the number of diners the preparation was intended to serve.
Silence is not permission to invent freely.
Instead, ask whether the missing information can be estimated from the recipe itself, related recipes, historical culinary practice, surviving equipment, the physical behavior of the ingredients, or practical kitchen testing.
If the evidence still does not settle the question, say so.
For example:
I used enough broth to produce a spoonable pottage.
That tells the reader exactly what decision was made.
It is more responsible than writing "2 cups broth" in a modern recipe in a way that implies the manuscript gave us two cups.
8. Separate Transcription, Translation, Interpretation, and Reconstruction
This is one of the most useful distinctions a historical cook can make.
Translation explains those words in modern language.
Interpretation asks what those words mean in their historical and culinary context.
Reconstruction turns that interpretation into a dish that can actually be cooked.
Those layers should not quietly collapse into one another.
If a manuscript says:
Take pork y sode and grynde hit smal...
a conservative modern translation might begin:
Take boiled pork and grind it finely...
That is relatively close to the historical instruction.
If I then decide to use eight ounces of pork, simmer it for twenty minutes, mince it in a food processor, and season it with one-eighth teaspoon of a reconstructed spice mixture, I have moved into reconstruction.
Those measurements and procedures are decisions made to make the dish reproducible in a modern kitchen.
They were not hiding invisibly in the manuscript waiting for me to discover them.
This is why Give It Forth generally preserves the historical recipe or transcription alongside the modern kitchen-ready version whenever possible.
Readers should be able to see where the historical evidence ends and my reconstruction begins.
9. Build a Kitchen-Ready Reconstruction
Once the language, ingredients, techniques, and contextual evidence have been considered, the recipe can begin moving toward the kitchen.
This is where practical decisions must be made.
How much should the first test batch make?
What modern vessel best approximates the required cooking environment?
How much liquid is needed for the described consistency?
How much spice produces a detectable but balanced result?
What modern ingredient is the closest practical equivalent when the historical ingredient cannot be obtained?
I generally begin with a small batch.
A small reconstruction wastes fewer ingredients if the interpretation fails spectacularly, and it makes it easier to adjust proportions during testing.
When I substitute an ingredient, I document it.
There is an important difference between:
The recipe contains partridge.
and:
I used chicken because partridge was unavailable to me.
The second statement preserves both the historical evidence and the practical reconstruction.
10. Test the Recipe
Eventually the research has to encounter a saucepan.
Kitchen testing can answer questions that textual research sometimes cannot.
Does the proposed ratio work?
Does the mixture bind?
Does the sauce thicken as expected?
Does the dough survive boiling?
Does the spice mixture overwhelm everything else?
Does the finished texture make sense in light of the historical instructions?
Testing does not prove that a reconstruction is historically correct.
It establishes whether the interpretation is physically and culinarily plausible.
Sometimes the first test is wonderful.
Sometimes everyone at the table becomes extremely interested in the bread instead.
Both results are useful.
Record what happened.
If the first reconstruction is changed, document what changed and why.
11. Compare Other Reconstructions
I prefer to examine other modern reconstructions relatively late in the process.
Looking at somebody else's finished dish too early can create an anchor.
Once you have seen the recipe interpreted as a pie, every ambiguous phrase suddenly begins trying very hard to become a pie.
After I have developed my own interpretation, however, comparison becomes extremely valuable.
If another researcher reached a similar conclusion through independent analysis, that may strengthen the interpretation.
If several careful researchers reached dramatically different conclusions, that may be even more useful.
I ask:
- Where did our interpretations diverge?
- Did we translate a word differently?
- Did one researcher use another manuscript witness?
- Did we understand a technique differently?
- Did one reconstruction rely on an assumption the others did not?
- Does another version reveal evidence I missed?
- Does the disagreement identify something that deserves additional research?
Different reconstructions do not automatically mean somebody cooked the recipe incorrectly.
Sometimes disagreement shows us precisely where the surviving evidence becomes ambiguous.
12. Record Confidence and Uncertainty
One of the most important changes in my own reconstruction work has been becoming more explicit about uncertainty.
Not every element of a reconstructed dish deserves the same confidence.
One ingredient may be directly named in the manuscript.
Another may require a translation.
The cooking method may be explicit.
The final texture may be inferred.
The quantity may come entirely from modern testing.
Treating all of those decisions as equally certain hides the most interesting part of the research.
High confidence: The historical source states the ingredient or procedure clearly and the terminology is well understood.
Moderate confidence: The source is less specific, but related recipes, linguistic evidence, historical context, or practical testing provide good support.
Lower confidence: The reconstruction requires choosing among several plausible possibilities because the surviving evidence does not resolve the question.
Modern testing: Quantities, timing, temperatures, substitutions, or procedures developed in my kitchen to make the historical instructions usable by a modern cook.
A confidence level is not a grade.
Lower confidence does not mean a reconstruction is worthless.
It means that part of the reconstruction rests on less certain evidence.
Making that visible allows readers to judge the reasoning for themselves.
Historical cooking becomes much more interesting when we can say:
Here is what the source tells us. Here is what the evidence suggests. Here is what remains uncertain. And here is what I chose to do.
See the Method in Practice
Different recipes create different research problems. These Give It Forth reconstructions demonstrate several ways the process can unfold.
Arbolettys: Medieval Cheese, Egg & Herb Pottage from Harleian MS 279
Arbolettys is a useful example of reconsidering the expected form of a dish. My original reconstruction called it a cheese soup. Returning to the recipe, its placement among the Potages dyuers, its dairy-first method, and comparison with related herb-and-egg preparations helped refine that interpretation into a soft cheese, egg, and herb pottage.
This is a good example of why a reconstruction can remain broadly sound while the explanation surrounding it becomes more precise.
Autre Vele en Bokenade: Veal in Spiced Almond Pottage
This reconstruction demonstrates the importance of following the manuscript's sequence of operations rather than allowing a familiar modern dish to dictate the method. It also illustrates the difference between the historical ingredient, veal, and a practical modern substitution used during an early kitchen test.
Beren Briw: Reconstructing an Anglo-Saxon Barley Pottage
Beren Briw shows what happens when the evidence is much thinner.
Rather than beginning with a surviving Anglo-Saxon recipe that exactly matches the finished dish, the reconstruction required separating Old English food terminology, early textual evidence, later comparative recipes, and a modern published reconstruction that had previously blended those layers together.
Here, confidence levels are especially important because the historical evidence does not support every part of the finished modern recipe equally.
Brǣdan Fisc: Reconstructing Apicius' Roman Fish Baked with Coriander
This article demonstrates another kind of revision.
The historical recipe itself survives and gives a clear sequence involving fish, salt, coriander seed, enclosed oven cooking, and a sharp vinegar finish. Returning to that source revealed that an older modern version had accumulated ingredients from neighboring Apician recipes.
The reconstruction therefore became more historically conservative by removing things rather than adding them.
The companion article Interpreting Medieval Recipes: My Reconstruction Process follows the research method I developed for working from historical text to tested modern recipe.
Research Resources
Historical recipe reconstruction rarely depends upon a single cookbook.
Depending on the question, I may consult manuscript transcriptions, parallel recipes, historical dictionaries, household accounts, estate records, herbals, medical texts, agricultural manuals, archaeological evidence, and modern scholarship.
Each kind of source answers a different question.
A cookbook can tell us what somebody recorded as a preparation.
A household account may tell us what people actually purchased.
An agricultural text may help establish how an ingredient was grown or managed.
An herbal or dietary work may tell us how that food was understood.
A dictionary may recover a lost culinary meaning.
Modern scholarship may bring several kinds of evidence together.
Rather than reproduce the same long list of research links in every reconstruction, I maintain them in one central Give It Forth library.
Visit Historical Herbal & Culinary Resources for medieval cookbooks, manuscript resources, household accounts, herbals, agricultural texts, sources for diet and humoral theory, stillroom and household manuals, and modern scholarly gateways used throughout this project.
You may also find these Give It Forth research hubs useful:
- Start Here: Historical Recipes for Modern Cooks - an orientation for readers new to historical cooking.
- Historical Cooking Basics - basic techniques, ingredients, measurements, and interpretive tools.
- Harleian MS. 279 Recipe Index - the Give It Forth hub for one of the manuscripts at the heart of this project.
- The Steward's Table - a scaling tool for turning tested Kitchen Copy recipes into quantities suitable for larger meals and feasts.
A Simple Reconstruction Checklist
- Locate the best historical source available to you.
- Identify the particular manuscript, edition, or witness you are using.
- Read the complete recipe several times.
- Mark unfamiliar and ambiguous words.
- Research vocabulary and culinary terminology.
- Pay attention to the sequence of cooking verbs.
- Examine surrounding and parallel recipes.
- Research important ingredients in historical context.
- Separate transcription from translation.
- Separate translation from interpretation.
- Identify information the historical source leaves unstated.
- Build a small, testable modern reconstruction.
- Document substitutions and modern decisions.
- Cook the recipe.
- Record what actually happened.
- Compare the result with other researchers and related recipes.
- Revise when new evidence or testing warrants it.
- State clearly what is certain, probable, possible, and modern.
How My Own Process Has Changed
I first wrote publicly about my reconstruction process in 2017 after I had been asked to teach a class on how I interpreted historical recipes.
The foundation of my method has remained remarkably consistent.
Start as close to the source as possible.
Read before deciding what the dish should become.
Research unfamiliar language.
Look for related evidence.
Develop an independent interpretation.
Test it.
Then compare it with the work of others.
What has changed is my caution.
I am now much more deliberate about distinguishing transcription, translation, interpretation, and reconstruction.
I pay more attention to manuscript context and parallel recipes.
I am less willing to allow an assumption to disappear into a polished modern recipe without being identified as an assumption.
And when I return to an older Give It Forth article and find that I no longer agree completely with my younger self, I revise it.
That is not a flaw in historical research.
That is historical research working properly.
New evidence can change an interpretation. Better resources can answer an old question. Another manuscript may reveal a useful comparison. A kitchen test may expose something that seemed perfectly sensible on paper but simply does not work.
A reconstruction should not become a little culinary monument that can never be touched again.
Sometimes the responsible thing to do is cook it again.
Frequently Asked Questions
Do medieval recipes usually give exact measurements?
No. Many surviving medieval culinary recipes provide ingredients and procedures without the quantities, cooking times, temperatures, or yields expected in modern recipes. Those details often have to be reconstructed from context and kitchen testing.
Is translating a medieval recipe the same as reconstructing it?
No. Translation attempts to explain what the historical words say. Interpretation asks what those words mean in their historical culinary context. Reconstruction then makes the practical decisions necessary to turn that interpretation into a dish that can be cooked.
Do I need to be able to read an original medieval manuscript?
No. Manuscript images are extremely valuable, but reliable transcriptions and scholarly editions make historical cookery accessible to people who are not trained paleographers. When several sources are available, comparing them can help identify transcription problems or variant readings.
Why are there no quantities in so many medieval recipes?
The surviving text often assumes practical culinary knowledge that did not need to be explained to its intended users. Quantities may also have varied according to household needs, available ingredients, vessel size, desired consistency, and the experience of the cook.
Should I look at modern reconstructions before making my own?
They can be extremely useful, but I prefer to examine them after working through the historical source independently. This reduces the risk that another cook's finished dish will influence how I interpret an ambiguous instruction.
What should I do if I cannot identify an ingredient?
First determine whether the problem involves spelling, translation, historical terminology, or identification of the plant, animal, spice, or prepared food. Consult dictionaries, culinary glossaries, parallel recipes, contemporary sources, and modern scholarship. If an exact ingredient cannot reasonably be obtained and a substitute is used, document the substitution and explain why it was chosen.
Can I substitute a modern ingredient?
Yes, when necessary for availability, legality, safety, cost, or practicality, but the substitution should be identified. A reconstruction becomes misleading when a modern substitute quietly replaces the historical ingredient without explanation.
How do I know whether my reconstruction is correct?
Absolute certainty is often impossible. Instead, evaluate the strength of the evidence supporting each decision. Some parts of a recipe may be directly supported by the historical source, others strongly suggested by context, and others dependent upon modern testing. A good reconstruction makes those differences visible.
Can medieval recipes have more than one reasonable reconstruction?
Yes. Ambiguous vocabulary, missing quantities, different manuscript versions, and unstated culinary knowledge can lead careful researchers to different conclusions. Comparing those differences can reveal where the surviving evidence is strongest and where interpretation is necessary.
Does kitchen testing prove that a reconstruction is historically accurate?
No. Kitchen testing demonstrates that an interpretation is workable and culinarily plausible. Historical accuracy must still be evaluated through the surviving evidence.
One Final Thought
The goal of historical cooking is not to pretend that we can stand in a medieval kitchen and recover every lost movement of the cook's hand.
We cannot.
What we can do is follow the evidence as carefully as possible.
We can read the words that survive.
We can compare manuscripts, recipes, ingredients, accounts, and techniques.
We can identify what the source tells us and what it does not.
We can test our interpretations in the kitchen.
We can explain where evidence becomes inference.
And when better evidence changes our understanding, we can change our minds.
For me, that is one of the pleasures of reconstruction.
Every historical recipe is a small conversation across centuries.
Sometimes the old cook answers our questions.
Sometimes they leave us staring thoughtfully into the pot.
Either way, there is usually something worth cooking.
No comments:
Post a Comment
Thank you for taking the time to leave a comment on this blog. Please note blatant advertisements will be marked as spam and deleted during the review.
Anonymous posting is discouraged.
Happy Cooking!
Note: Only a member of this blog may post a comment.