Tuesday, September 8, 2015

FRENCH BREAD HISTORY: Making medieval/Renaissance bread

Further information on bread history can be found on Facebook in the Bread History Lounge.

A number of sources on the Web claim to provide instructions on how to bake medieval bread. But few cite any period sources at all and those that do sometimes cite recipes which are not actually for simple bread, but things like
rastons (“little rats”) which are in fact a variant of a French pastry. One uses a sponge, a method which is not documented until centuries later. Etc.

This is not surprising; no medieval recipes for bread survive at all nor do many details on how it was made. Still scattered data does exist, at least for the French side, allowing the committed historical baker to narrow the parameters of what they make as “medieval bread”.

The purpose of this post is simply to gather all such items together. Though I try to draw some conclusions here, I am not an experienced, much less an expert, baker and ideally those who are will adopt the source material here with the aid of their wider knowledge and experience. Some items are relatively easy to apply; others will require more effort or equipment than any but the most dedicated recreationists will be able to deploy. However one chooses to use them, these offer documented parameters to narrow one's attempts to reproduce early bread towards something closer to, if not exactly like, it.

Note that the focus here is on French bread, not only because that is my own subject, but because substantially more information exists for French bread than for English for the medieval period and that right after it (approximately the Renaissance, though that term has very different meanings outside of Italy). If your main interest is in English bread, the most important difference to note is in the use of yeast, rather than sourdough, to leaven bread. This may not have always applied, but certainly is well-documented enough to take into account. See “Leavening” for more information.

Key points

What follows is highly detailed; the idea after all is to substantiate each point with as much as data as is available. But some readers may want simple guidelines, minus the explanations. Here then are some highlights of what follows.

  • Grain – Soft wheat is most suitable, but possibly difficult to leaven; rye and maslin (rye and wheat) are the main alternatives to simple wheat.
  • Types of bread – Broadly, bread can be viewed as urban bread (defined in statutes or assizes) and domestic bread (made privately and most often in the country). Urban bread consisted of at least a light and a dark bread, more typically of three or four graduated qualities of bread (corresponding to different degrees of bolting). Domestic bread is more fluidly defined, but was probably most often from moderately bolted flour and somewhat bigger (to last a while); for servants, it would generally have been made of maslin or rye. In general, class was a key consideration; bread intended to reflect the diet of laborers, especially, would be dramatically worse that that reflecting an elite diet.
  • Milling – It is best to mill your own grain, ideally with an impact method, but a simple blender will serve for many purposes. Let it rest a few days after doing so. If you use commercially ground flour, do not assume that because some is marked “stone ground” it has ONLY been stone-ground.
  • Bolting – Cloth or fiber bolting is most suitable. For French municipal breads, the goal is to produce three qualities of flour, from the whitest to dark (mainly bran). A home baker, however, will probably find it very difficult to produce a white flour that yields a truly white bread.
  • Leavening – Sourdough (specifically, old dough) was the standard French method. Use the same grain for your culture as for your bread. One local proportion was 17.5% of the dough (added to the full amount); in one Paris trial, the figures are around 7%. For English bread, you will want to use yeast, but “impure”, as it long was; ideally from the ale-brewing process, possibly adding dark ale to yeast as an alternative. Add the leavening directly to the flour, or vice-versa; yeast preferments (sponges, bigas, poolishes) are not period and simple sourdough is the only preferment recorded for that method.
  • Hydration – Evidence suggests that the most common medieval French bread was minimally hydrated. But more theoretical writers mention the effect of hydration in making bread “spongy” and an argument can be made for higher hydration in bread for, say, fine households.
  • Additives – Most regions used little or no salt, except perhaps for upper class bread. Aromatic seeds such as anise and fennel can be added to domestic style bread. Sources mention the use of milk, but give no details on this.
  • Size – White breads should be around a pound in weight; eleven or twelve ounces was a common average size. Coarser breads can be much larger.
  • Shape – The familiar spherical or hemispherical shape seems to have been almost universal in the period.
  • Oven – If you are not working with a wood-fired oven, try starting at a strong heat and stepping down over time (as would have happened when the heat from burning wood inside the oven dissipated). Bread was also cooked under the coals on the hearth and under a pottery bell.


In England, sources are very thin on bread for this period. Several assizes give information on what bread was produced by urban bakers and one gives information about their comparative weights; but in general, these are light on details. Ironically, it is in England that the first actual recipes for bread appear; but these are already too late to be very useful for the medieval period. The Good Huswifes Handmaide for the Kitchin (1594, 1595) gives details on making the best kind of bread (manchet), but not the others. The next work – by Gervase Markham – comes in the seventeenth century (1615). It does provide useful notes on both manchet and cheat (a low quality) bread, but it is uncertain how much these would have applied in the centuries before this.

In France, municipal statutes provide the equivalent of the assizes. A number of these – from Paris and elsewhere – specify comparative weights for the different qualities of bread named. This is useful, if not definitive, information. What is more, a number of municipalities performed bread trialsessais de pain – to determine how much grain was needed to produce one or more qualities of bread. In some cases logs were kept of these tests, at various levels of detail. One from Paris in 1432 provides a number of useful details. By far the most detailed of these, however, comes just at the end of the medieval period (or arguably the beginning of the Renaissance). This is for a trial in Limoges in 1499. The test is specifically for white bread, though it includes some information on darker breads. It is so meticulously annotated that it is virtually a recipe (even if some points remain obscure).

Moving into the sixteenth century (the French Renaissance, essentially), a number of writers began to document subjects like agriculture and food and in the process gave at least some details on bread and bread-making. Symphorien Champier's Rosa Gallica (1514) includes some specific notes, in Latin, on bread. Bruyerin de Champier wrote de Re Cibaria in Latin in 1560; this touched on a number of aspects of bread. In “translating” Platina's De honesta voluptate et valetudine, Didier Christol added numerous notes specific to France, including some on bread (1571).

Charles Etienne also wrote a work in Latin – de Nutrimentis (1550) – which explored bread but his most well known work, l'Agriculture et maison rustique, was in French. His own first edition (1564) says almost nothing about bread, but in 1570 his son-in-law, Jean Liébaut, expanded on it, drawing in part on the earlier works in Latin, and provided the most extensive look at French, or even European, bread up to that point.

Blaise de Vigenère, in studying weights and measures, actually conducted his own bread trials, noting details of uneven interest (1583).

Earlier medical texts include some useful, but possibly idealized, notes. A text included in Arnau de Vilanova's collected Opera Omnia (1585) is in fact from Maino Maineri (14th c), who was from Milan but spent years in Paris (thanks to Sebastià Giralt for clarification on this). It has a long section on bread. Aldebrandino of Siena's thirteenth century dietetic has brief remarks on bread.

Beyond prime sources, Francçoise Desportes is one of the rare modern scholars to have studied French medieval bread closely and provides some useful overviews of her research.

Otherwise, medieval Irish bread is a tangent too far for these purposes, but if you want to look into it, there is some excellent information in this work on Munster: Early Medieval Munster: Archaeology, History and Society


Experienced bakers will encounter a number of familiar terms here. Not all of these however had the same meaning in the Middle Ages.
  • Barm/yeast – Today a distinction is made between yeast and barm; yeast is a chemically pure product used for leavening, barm is the scum formed on the top of some fermenting beverages and has a number of different uses. Until the nineteenth century however, these were the same thing: “Barm - ...Yeast; the scum or foam rising upon beer, or other malt liquors, when fermenting, and used as leaven in bread to make it swell”; (Ogilvie).
  • Sourdough/levain/old dough – Sourdough today is cultured on its own. Some Americans call it levain to distinguish it from the famous San Francisco variety. “Old dough” is literally a piece of dough from a previous batch, used in the same way as sourdough, but including any impurities from the last bread. In the Middle Ages, these all referred to the same thing (old dough that had “soured”, picking up wild yeast). - To complicate matters, the French term was sometimes used to refer to the dough which had been mixed with sourdough.
  • Bolting/sifting/sieving – By the eighteenth century, these terms were often being used interchangeably, even if their strict meanings remained. But in the Middle Ages they referred to distinct, separate processes. In France, wheat was also “cribbled” before being ground, but later English usage could refer to coarse sifting (or the coarse flour that resulted).
  • Bread/pastry – In the Middle Ages, “pastry” first referred to the shells of dough used to enclose various foods. But by the end of the period, this had already begun to include finer, if not always sweet, baked goods. These however were often considered a form of bread (and made by bakers rather than pastry-cooks). As a result, one sees references to “breads” made with eggs, milk, honey, etc. Most, but not all of these, were what would later be considered pastry. (Gateau/gastel, however, began not as a cake, but as a finer form of bread, with no additives; with the very rare exception of a "gastel kneaded with eggs"from 1381.)

    Weights and measures

    The quantities which appear in the following excerpts are those in the original texts. They should be used for relative amounts and a general order of size.

    In the unlikely event that you need precise quantities however, that will require further research into the specific values in each region for each period. Until the Revolution, French regional measurements were notoriously chaotic. A pound in one region might be equal to sixteen ounces; in another, twenty-four. These variations were not so great as to change the general scale of size; an ounce for instance was still a relatively small part of a pound. But you should not take it for granted that a pound from one region was equal to a pound in another, much less to a modern American or English pound.


    While there are no French recipes for bread in this period, there are general instructions. Here are the major ones.

    In his “translation” of Platina, Christol includes these very general directions (which are not in the original Latin text):
    Take wheat flour... and this ground moderately put through a very fine sifter or strainer and after soak this with hot water, in which will have been put salt, as is done by those of Ferrara in Italy to make the bread very savory, and be careful not to put more or less leavening in the said bread than needed because thus with too much leavening, the bread is sour and ill-flavored, just as with too little, it is heavy and of bad digestion, and unhealthy, and binds marvelously, similarly you must put water moderately so that the dough is not too soft nor either too hard: and similarly the bread must be well baked, and it is best to adjust the oven well, in which the bread will be cooked, to be neither too hot nor too cold... And if you want the bread to be well nourishing and pure, it is best to sift the flour well and let no bran remain; and if you want it to be laxative, it is best not to sift it all...
    If this seems cursory, consider that culinary recipes in the period were not always more explicit.

    Liébaut too gives general instructions on making bread:
    The most excellent and best bread of all (if one needs to choose) is that which is made of good and pure new wheat, not old, not corrupted, nor at all spoiled...; of well ground flour, well cribbled or sifted, well put into dough with a great deal of leavening, and enough river, or fountain water, rather than well water... and turned about on all sides, left to rest for a few hours, well covered, a little salted: of a modest sized and not excessive mass of dough, so that it receives the heat of the fire equally on all sides on top as on bottom: baked in an oven heated with a moderate fire... moderately baked, lest by too great and long baking the crust be scorched... or by light baking the inside of the bread remain uncooked....
    He writes about Beauce wheat specifically that the farm-woman is to:
    wet her arms and hands, and knead the dough carefully, turning it and spreading it out on every side this way and that for a long time, and let all the glutinosity and viscosity of the flour be broken and dried, so that the bread be that much more fragile, easier to chew, and not so pasty to the teeth, mouth and stomach. After such handling, she will be careful to soon shape her dough, so that it not turn to sourdough, otherwise the bread will not be so good to eat.
    Otherwise, having gone into details about leavening and the use of different temperatures of water for different wheats, he writes that once the dough has been prepared:
    divide it into orbicular portions, sufficiently large and thick, to be put in a reasonably heated oven so that the bread is cooked enough according to the size, thickness and quality of its dough... If the oven is too hot, the bread will be scorched on the crust, and will remain badly cooked inside.
    He also says that any salt or other additive should be added while kneading.

    The 1499 account of a bread trial in Limoges is unique in this period in laying out every step of making municipal bread. The trial is concerned specifically with white bread and so some details for dark bread are missing. There is also at least one obscure point on what is added at one stage. But certainly there is a great deal to be gleaned from the account.

    Here is a paraphrase of its contents:
    February 22, 1499
    The commissioners bought 1 setier of average price, weighing 94 pounds without the bag. They then had it ground and sifted and gave the miller his portion, leaving 81 pounds in flour with the bag.
    It was then sealed and left to rest for three days.
    On February 25. the flour was sifted by a pastrymaker "in his strainer" and then bolted by "an expert in doing this" [a bonetier - a hat maker], bolting the half of the bran (somp in the local patois), which was rebolted several times.
    All this done, the flour weighed 58 pounds without the bag.
    10 pounds of leavening were then put in.
    The bolted bran for making dark bread (bolent) weighed 9 pounds.
    The remaining bran, after the preceding, weighed 19 1/4 pounds.
    The next day 3 pounds 2 ounces of water were added.
    6 ounces of salt were added.
    To prepare both white and brown "besides the 10 pounds declared for leavening in the paton, 13 pounds 2 ounces" were added [meaning more leavening? more water?]
    "This flour and dough prepared", the baker and others who had prepared it judged that it was raised enough, and "half ready", it was put into pieces of dough weighing 15 ounces each.
    "From the said flour" were produced 76 white breads of 15 ounces each and 12 dark breads weighing [presumably collectively] 26 marcs [13 pounds] 3 3/4 ounces.
    This was then taken to the oven [a separate business] where straw was set on floor and after covered and after a cloth, then the weighed out pieces of dough put in pieces and two cloths and "the said cover".
    This was left about two hours, and when the baker and his team decided it was sufficiently raised and ready to put in and the oven hot enough, the dough was put in the oven and stayed there "a good piece" until everyone agreed it was ready.
    These were taken elsewhere to be weighed. The result was 27 white loaves weighing [presumably in total] 37 marcs 4 ounces, 27 at 37 marcs 5 ounces [roughly 11 ounces each] and 22 at 30 marcs 4 3/4 ounces. All 76 together weighed 105 marcs 5 3/4 ounces.
    The twelve brown breads together weighed 26 marcs 3 3/4 ounces.
    Different readers may interpret this differently, but it is about as precise an account of making municipal white bread in the period as one could hope for. It should be fairly easy to scale down for anyone who wishes.

    In 1583, de Vigenère published an account of his own bread trial. This is not entirely clear, but includes some possibly useful information. After grinding and bolting a setier of wheat, the test resulted in eighteen bushels, weighing 235 pounds. (This was a very high yield; Kaplan says that in the eighteenth century a yield of 224 “raised considerable admiration”.)

    From the said finest flour can be made some fifteen dozen white breads, weighing fourteen ounces in unbaked dough, coming to twelve baked and cooled.
    In addition five or six dozen dark breads of the same weight, from three bushels of middlings...
    Both the white bread and the dark bread were weighed after having risen, before putting it in the oven, and were found to be the same weight after being baked; but still hot, and not cooled.
    To start the leaven of the said setier of wheat [the original quantity] a pound of water was used.
    For the second time to moisten the said white bread leaven, four pounds: and for the last eight pounds.
    To make the leaven for the dark bread the first time eight pounds of water were used.
    And for the second time to moisten the said dark bread leaven, nine pounds.
    To knead twenty two pounds [?]...
    The earliest true English recipe within this period (from the Good Huswifes Handmaide for the Kitchin) is one for manchet, the finest bread, from 1594.
    Take half a bushel of fine flour twice bolted, and a gallon of fair lukewarm water, almost a handful of white salt, and almost a pint of yeast, then mix all these together, without any more liquid, as hard as you can; then let it sit a half an hour, then pick it up, and make your manchets, and let them stay almost an hour in the oven.
    Take halfe a bushell of fine flower twise boulted, and a gallon of faire luke warm water, almost a handful of white salt, and almost a pinte of yest, then temper all these together, without any more liquor, as hard as ye can handle it: then let it lie halfe an hower, then take it vp, and make your Manchetts, and let them stande almost an hower in the ouen.


    By far the most common grain used in this period was bread wheat and that is very likely what you will want to use for your own productions. But rye especially is mentioned for lesser quality breads and both barley and oats, though mainly used for animals, are mentioned as being used for breads as well.

    Comet has analyzed the wheat available in France in this period and concludes that hard wheat was still relatively rare. "North of the Mediterranean, it only spread rather late in the Middle Ages, and only penetrated very little in France, if not not at all; we have have found no medieval mention of it;” "For Paris in the fifteenth century, we can be certain of the density of wheat, it is from .61 to .68, which, of course, excludes hard wheat;” "The supremacy of soft wheat lasted for fifteen centuries."

    This means that if you're trying to be as authentic as you can, you'll use a soft wheat. These days, in commercial flour, that is mainly used for pastry flour. A big problem (and something to consider about period bread) is that you'll already be using what is (in varying degrees) whole wheat, which does not leaven well. But whole wheat pastry flour is said to leaven particularly badly. So it may be tempting to compromise on this point.

    Few of the surviving details concern rye bread, but this was a common bread for servants and is mentioned in some municipal statutes as well, as is maslin, the mix of wheat and rye. Anyone trying to reproduce the full range of period breads will also want to work with these.

    Barley and oats are very much outliers for bread in this period, though in the thirteenth century Maino listed these after wheat and without mentioning rye: “The best bread is that made from wheat, then from barley, third from oats (Laudabilior panis est, quod fit ex frumento, deinde ex hordeo, tertio ex avena). (Maino was Italian, but largely worked in France.) If you specifically want to make French bread from the thirteenth century, these grains might be more likely choices. Overall, Maino wrote that “bread can be made with wheat, barley, oats, rye, millet, panic wheat and rice” (panis potest fieri ex frumento, hordeo, avena, siligine, milio, panico et riso), but he does not recommend most of these for healthy eating.

    An English assize of uncertain date mentions wheat, barley and oats, but no rye. (Wrottesley) So the situation may have been reversed in England.

    Types of bread

    In one sense, there are two types of bread in this period: municipal bread and domestic bread. Domestic bread is far less documented until the sixteenth century, but much of the writing in that period essentially describes bread made by private households, not public bakers.

    In France, local statutes typically describe three kinds of bread: white, darker, and dark. Much of the most detailed information – comparative weights, for instance – concerns these. Some statutes defined as few as two types of bread – white and dark – or even just one; others included a fourth, “whiter” type of bread, whose weight was sometimes included along with those of the standard three.

    Liébaut lays this out simply enough: "bread is made of various sorts: that is more or less white, depending on whether the second, or third, or fourth part of bran is separated using the sifter."

    One approach to recreating French medieval bread would be to reproduce these varying qualities of bread. A very approximate way to do this would simply be to use white flour, whole wheat flour and wheat bran, perhaps mixing in a little of each with the other (to mimic imperfect bolting), and incorporating the other parameters laid out here.

    Note that even the worst quality of these was made from bolted flour. But worse bread still existed, especially outside Paris. Liébaut matter of factly writes:
    Bread which is made of whole wheat flour, and from which nothing has been separated by the sifter, is suited to laborers, ditch-diggers, [etc]... also suited to them is that which does not have much leavening, which is not very baked, which is somewhat pasty and viscous...
    A specific example survives of this from 1387, when people from the Norman town of Harfleur objected to having to make the same kind of bread as in Montivilliers, a town where most residents were poor tradespeople and laborers and the bread was “coarse dense poorly baked, heavy and little risen”.

    On the other hand, those who did little physical work, says Liébaut, such as monks and more "studious" people, needed, at the least, white bread, or perhaps even “the bread of courtesans, pain de bouche (“mouth bread) "which is well risen, a little salted, well kneaded, full of holes ["eyes"], of well risen dough”.

    In England, the first assizes only define white bread” (albis panis) and “whole wheat bread” (panis de toto blado) These were the two main sorts of English bread – a fine white bread and a coarse brown bread – and probably correspond to the later manchet and cheat bread. Later terms included simnel, pain de maigne/payman, cocket or tourte; and, between these, wastel (the same as the French gastel/gateau, though possibly not as fine). While these give general ideas of the relative quality of these loaves, little specific information exists on them.

    Anyone who wants to make English medieval bread could look to Markham's later recipes for manchet and cheat, knowing that some things would have changed in the intervening centuries. Alternately, using yeast in recipes based on earlier French information might also be a solution.

    Milling and bolting

    Many home bakers may essentially be obliged to use commercial flour. If you do so, and use stone ground flour in an attempt to be more authentic, do not assume that because the label says “stone ground” the flour has ONLY been stone ground:
    Stone-ground can mean anything from wheat berries first cracked on stone mills and then ground to flour on rollers to finished flour passed over a stone after it has been ground. "Or it could mean it's just a nice name," says Jeff Gwirtz, technical services director of the International Association of Operative Millers. "It's more a conceptual, warm, touchy crunchy feel."
    As for “whole wheat” flour, a number of sources say essentially the same thing as the Whole Truth site:
    When wheat is ground for commercial flour sales, the bran is first removed and the germ and oil in particular are separated out, since these spoil in a short period of time. The remaining endosperm is then finely ground, leaving white flour. In order to market “whole-wheat flour,” a small percentage of the bran is returned to the product, yet it still lacks the germ and thus is far from being “WHOLE” wheat flour.
    There are a number of reasons then that it is preferable to mill and bolt (or sift) your own grain.

    Processing your own grain by hand is laborious and time-consuming. It may not be not worth the effort for anyone making period bread regularly, but it is worth doing at least once, even a few times, if only to see how different the results will be even from “historical” flours sold specifically for making older breads. Simply put, it is very unlikely that any commercial flour you can buy will be close to what you grind and process yourself.

    In particular, you will probably find it difficult, if not impossible, to produce not only white flour, but white flour that, when baked, produces a white bread. It is very unlikely, in fact, that medieval white bread was anything as white as white bread today and this exercise will show you why. Another advantage of this exercise is that it will remind you how very much effort went into making bread and perhaps give you a new idea of what “white bread” really represented in terms of human toil.

    Ideally, anyone who wants to do this would have specialized milling and bolting equipment. As a practical issue, most bakers won't, so a variety of workarounds or “least worst” options may be required.


    For a home baker, the best option here is to have a specialized impact mill with a stone attachment. These exist, but cost over two hundred dollars. This investment may well be worthwhile if you are planning to grind your own grain regularly.

    If not, a simple blender will indeed grind grain and may suffice for many purposes. Unfortunately, “grind” is what it does – that is, it cuts up the grain with blades. This can be a problem if you are planning to sift out the bran, which will now have been mixed in with the endosperm in tiny particles.

    If you want to mimic stone grinding, one option may be to use a mortar and pestle. For centuries, after all, people have ground grain with something very like that (more precisely, a thick wooden rod or pole pounded into a hole in stone). In theory, it should be possible to pound grain, a little at a time, in a mortar. Personally, however, I have had no luck doing so. If you are more at ease with a mortar and pestle, you might have better luck.

    Another option may be to use a spice mill. Ikea sells little ones that fit on their spice jars and use porcelain blades – not quite stone, but not sharp metal either. Using these is very time-consuming, but with a little persistence (and maybe some friends helping) you can use one or more of these to grind grain in a gentler way than in a blender.

    Note one other aspect of period milling. Liébaut says that soft stone in some mills left gravel in the flour, which would “take away all the grace and flavor of the bread, and most often cause problems to the teeth.” You don't want to push authenticity to this point.


    The idea that ground wheat should be left to sit before being further processed was already known by the time of the Limoges trials, when the ground wheat was left for several days before the next step. If you are milling your own grain, let it sit a few days before bolting it, etc.

    Bolting, Sifting and Sieving

    Once you have ground grain, you will probably want to refine it into flour; various grades of flour, if you are using French bread statutes as your guide.

    If you are serious about reproducing older forms of bread, consider the following about bolting, etc:
    • These methods were essentially textile based (cloth or fibers) in the past (as opposed to the metal screens or sifters people tend to use today)
    • Bolting, sifting and sieving were separate methods in this period (already in the eighteenth century they tended to be lumped together).
    • These methods evolved and improved (i.e., eighteenth century bolting was already more advanced than sixteenth century bolting, never mind fourteenth century).
    Once again, as a practical matter, you might find it difficult to actually apply these distinctions. But they are worth considering in trying to approach something like actual medieval bread. Notably, even bolted and sifted flour was probably less pure in the medieval period than in subsequent centuries, since textiles are more subject to deterioration than metal. In eighteenth century bread trials, one item was sometimes for "upkeep of bolting cloths", showing that these required on-going maintenance. Where a baker, for lack of funds or otherwise, did not see that these cloths were kept in shape, the result was likely to be imperfectly bolted flour.

    Relatively little information exists on bolting before the eighteenth century. Yet exactly how ground wheat was converted to flour is important to understand in this process. In fact, arguably, anyone who wants to work seriously at reproducing medieval breads should do research into and experiment with the different bolting and sifting methods. It may be useful then to look at what information appears on these from a later period.

    In his classic French work on baking and milling (1767), Malouin included a brief history of bolting:
    First these light cloths were used called canvas; and horsehair sieves were also used for that: also a type of sifter made of prepared and punched skins has been used. Various sifters were called sas, from the name seta, silk, because some were formerly made with the bristles [“silk”] of pigs and boars.
    Since then finer sieves have been made from wool, goat's hair and silk.
    In England, in 1774, a commission had reason to look closely at bolting and the quality of wheat it produced. In France at this point bolting cloths were designated by the number of threads, so that, says Malouin, a number 11 had eleven threads, a number 44, forty four. In England the rating was by the cost of each in shillings. The measurements here then are for the coarsest (8 shillings) cloth to the finest (21 shillings).
    21 s. Cloth 64 Threads to the Inch in the Warp
    52 “ “ “ “ in the Shute
    18 s. “ 52 “ “ “ “ in the Warp
    44 “ “ “ “ in the Shute
    16s. “ 44 “ “ “ “ in the Warp
    40 “ “ “ “ in the Shute
    14 s. “ 40 “ “ “ “ in the Warp
    13 s. “ 32 “ “ “ “
    13 s. “ 32 “ “ “ “ in the Warp
    28 “ “ “ “ in the Shute
    8 s. “ 17 “ “ “ “ in the Warp
    16 “ “ “ “ in the Shute
    Report from the committee appointed to consider of the methods practised in ...1774

    Thirteen shilling cloth seems to have been the low-end standard. In 1796, bolting cloth was limited to this size during a wheat shortage, to prevent wasting wheat:
    they be empowered in like manner to prohibit, if they think proper, any flour purporting to be of a superior quality, and sold at a higher price, than the whole flour of wheat from being made for sale, or sold, except for the purpose of making such small bread as may then be allowed, by licence under the hand and seal of such magistrate, or for the purposes of pastry of confectionary; and that the said magistrates be empowered to order that no miller or mealman use, during the continuance of such their order, any bolting cloth finer than one 6 feet long by 7 feet broad at the head, and 6 feet broad at the tail, composed of woollen yarn, and weighing one pound when new, having 32 threads to the inch in the warp, and 28 to an inch in the shoot; and which is at present known, and commonly called a 13s. cloth, nor any finer wiresieve, or machine, than that which consists of 41 wires to an inch both ways, and the weight of 6 inches square of which is 1 ounce and 1 dram.
    This is all well after our period and it is always possible that bolting cloths were made within a different range in earlier times, or otherwise varied. But this is as close as we are likely to come to having some idea of what these were like.

    Note that in the Limoges bread trial the bolting (as opposed to the sifting) was done by a bon[n]etier – that is, a hat maker.

    The other methods were to sift (sasser), sometimes in a bag (sachet), and use a strainer (etamine). The distinctions between these are not clear, though both fibers and holes were used in them to let flour pass; Cotgrave's 1611 definitions are almost identical for both. At the least, they seem to have been gentler methods. Liébaut says: “It is better to bolt than to sift maslin and rye flour, because the bolter through the work of the arms forces the bran to loosen its flour: which neither the bag nor the strainer do, especially since they only move the flour from one place to the other.”


    The focus here is on sourdough. That was the main method in France for most of the Old Regime and also used to some degree in England (though most documented examples include yeast as well).

    Before looking more closely at that method, here are some observations for those who want to make something more like English bread. There are strong indications that this was mainly made using yeast (barm). But the yeast of former times was not like the pure chemical product we know today. Until the nineteenth century, it was essentially the foam floating on brewed ale.

    One reason efforts were made to purify it was that this foam was tainted with impurities – such as hops – from the brewing process itself. Conversely, if you want to use yeast as it was in this period, you will want to either use a similarly impure product or add impurities to modern yeast. Ideally, you would make your own ale or know someone who does. Since few people will be in this position, one option might be to mix dark ale with modern yeast before using it in bread, or perhaps after letting it develop in the flour. You can also use brewer's yeast to make bread, but while the flavor is said to differ somewhat from using baker's yeast, it would still need some traces of actual beer to come close to the effect of earlier centuries (note that one of these effects would have been to make the bread more bitter; there are good reasons people tried to improve on this method).

    Champier also describes the Flemish using yeast and says that they produced it by boiling grain and then using the foam which came from this. This process is hard to envision, but at the least it would have created a somewhat purer form of yeast than using that from ale, should you care to try it.

    Though the first leavening mentioned in Gaul (by Pliny) was yeast, the Romans (who ruled Gaul) mainly used sourdough and a number of the sources here specifically define leavening as the soured dough from a previous batch of baking.

    Pliny already described this as the main Roman leavening method:
    the leaven is prepared from the meal that is used for making the bread. For this purpose, some of the meal is kneaded before adding the salt, and is then boiled to the consistency of porridge, and left till it begins to turn sour. In most cases, however, they do not warm it at all, but only make use of a little of the dough that has been kept from the day before.
    This differs little from descriptions in the Middle Ages:
    Ferment means, what increases agitation: it is for making flour rise, like boiled porridge, before the salt is added, and it remains until it goes sour; commonly using some matter kept from the wheat used the day before.
    Fermentum dictum, quod fervendo crescat: est enim farina quae subigitur, ad pultis modum decocta, priusquam addatur sal, et relinquitur donec acescat: vulgo atuem pridie tantum asservata materia utuntur frumento.
    Champier says that bread with a lot of leavening is "found finer and more subtle" (magis tenuis et subtilior invenitur). Liébaut:
    Leaven, called in Latin Fermentum, because it swells and rises over time, is a piece of dough left from the last bread-making, covered and enclosed in flour, which is soaked, to remove the excessive glutinosity and viscosity from the flour one wants to use to make dough for bread: this leavening takes on a sourness over time which brings a grace and a better taste to the bread, and so we see that the more breads have leavening, the more pleasant and healthy they are than those that have less leavening.
    Though he too defines it as old dough, he does touch on how to make it from scratch:
    We make it of wheat dough, to make breads of wheat, of rye dough to make rye breads: some add salt to it, others vinegar, several of verjuice from grain, or from wild apples.... The sourdough which bakers and baker-women use to make their bread can be kept fifteen days, and no longer, otherwise it spoils and corrupts: the best is to not keep it so long: to keep it, one must work the dough into a round shape, cover and wrap it in flour, even in winter cover it with many clothes [sic] in the trough. When the farm-woman wants to bake her dough, two or three days are needed, or at best the day before, soak one's sourdough with hot water, or even cold water, according to the weather and the type of wheat she uses for her bread.
    He follows this with details on the various types of wheat from different regions which, by his account, required different types of leavening and different temperatures of water. None of this is likely to be useful to a modern baker, but some readers may want to know that this information is available.

    Texts warn against using too little leavening or too much (which could make the bread bitter). The Limoge trials use roughly one part leavening to almost six parts flour or roughly 17.5%. For the bread trial in Paris in 1452, about 7.7% or one part for 13 was used for the white bread, about 6.4% or one part for 15.6 for maslin and about 7% or one part for 14.17 for dark bread (made from a mix of maslin, middlings, etc). The range then for white bread was from 7.7% to 17.5% leavening per quantity of dough. But the Parisian range was around 7% for all.

    These proportions are much smaller than those Parmentier gives for the eighteenth century, roughly a quarter in the summer and a third in the winter. But a number of other parameters, including the development of leavening, also differed in his period.

    The 1452 Parisian leavening was put in at 10 a.m. and the dough kneaded and shaped at 3 p.m; it was put in the oven at 8 p. m. The Limoges account gives no times, but states that the leaven was left overnight before being used. After the dough was kneaded and shaped, the loaves were set to rise for about two hours before being baked.

    The first times here – for what might arguably be considered autolyse– are very long by modern standards. This may be a function in part of the quantities involved, but given that this was soft flour and more or less whole wheat it may also reflect the extra time needed for it to rise.

    The Parisian bread trial essentially reflects Liébaut's principle that one should use the same flour for the leavening as for the dough, though leaven from dark flour is used for two different types of it. De Vigenère too matches his leaven to the flour in the dough. (If you are laboriously milling and grinding your own flour, you may want to use commercial whole wheat flour to start the culture, then increasingly feed it with your own flour as it reaches full strength.)

    Pre-ferments, etc.

    Leavening was added directly to the dough (or flour to the leavening). No yeast pre-ferments (sponges, bigas, poolishes) are mentioned in this period or even immediately after it. Sourdough itself is considered a pre-ferment, but a sourdough biga, for instance, is a separate type of pre-ferment; no such practice is noted in France in the period. Nor was there any concept of generations of sourdough (levain de chef, levain de première, levain de seconde and levain de tout point) as would already exist in the eighteenth century (though de Vigenère's uncertain reference to refreshing the leavening three times may be referring to this in practice if not by name - see "Refreshing".)

    In his early seventeenth century instructions for making cheat bread, Markham does include a step beyond simply using old dough:
    take a sour leaven, that is, a piece of such like leaven saved from a former batch, and well filled with salt, and so laid up to sour; and this sour leaven you shall break into small pieces into warm water, and then strain it.
    So in England it may be that such a process was already used earlier for coarser breads. (The sponge on the other hand would not be documented for some time.)


    The practice of mixing leavening and flour and then letting it rest before proceeding is considered a modern innovation, credited to the great baking teacher Raymond Calvel: “Raymond Calvel has been called the teacher of bread teachers and is widely considered to [be] the expert on French breads.,,,,One of his innovations is the autolyse, a resting period between the early mixing and kneading phases.” (Artisan Bread Baking)

    But note that in the Limoges trials, leavening was put in the flour (though apparently no water) and the mixture was left overnight, before both water and salt were added the next day. Presumably this would have had a similar effect. In the Paris trial, not only was the mixture left for several hours, the baking troughs were roped and sealed (the 1452 record does not track additions of water, which may or may not have been added here as well). So an argument can be made for using this “modern” technique for medieval bread.


    Eighteenth century texts describe earlier bread as harder (“firmer”) than that then popular. This implies that it was less hydrated. The Limoges trials use relatively little water. Pipponier, reviewing fourteenth century bakers' equipment inventoried in Bordeaux, notes a broie, that is a kind of wooden stick used to knead extra-hard dough. This suggests that they were working with minimally hydrated dough. Even pain de Chapitre, the finest bread in the late sixteenth century, was said to be so hard it had to be kneaded with a broie (or even the feet).

    Two other points strongly suggest that bread of the time was much harder than it would later would be. One is that the crust was hard enough that in fine households (per Liébaut) it was grated off. The other is that for a very long time Parisian bakers, at least, were supposed to stamp bread with an identifier - something which became much harder as a softer ("bastard") dough was later adopted.

    This did not necessarily jibe with more abstract advice. In the thirteenth century, Maino wrote that “bread must be tempered with an amount of water, such that it be neither too soft, nor too thick, or hard.” (debet esse panis cum quantitate aquae temperata, sic ut non sit nimis mollis, nec nimis spissus, seu durus). Christol's version of Platina says “You must put in water moderately such that the dough be neither too soft nor too hard”.

    Specialists knew that adding more water would make bread lighter and spongier. Champier said to “make it with a lot of water to make it spongy” (conficint cum multa aqua ad hoc ut fit spongiosus) (Symphorian Champier). Bruyerin said that a great deal of water was to be poured over flour “so that the spaces in a bread's sponginess admit a great deal of air” (ut panis sua spongiosa inanitate multum admittat aerem) (Bruyerin de Champier). It is not clear where this knowledge was applied, but conceivably some of the better private households might have used lighter, more hydrated bread; the best breads (such as pain de bouche) are described as “full of eyes” (that is, holes in the crumb).

    A modern baker then can justify either choice, though for municipal bread it seems likely that the dough was not very hydrated; the strongest evidence suggests that a “firm” dough was most used in public bread in this period. Bread for the “better sort” was probably better hydrated.


    Note first of all that the term levain (leavening) is sometimes used in ways that suggest it is referring to the dough with leavening added, as well as to the sourdough itself. In the 1452 Paris trial, for instance, after leavening has been added to the flour, the text reads “and the levains made”, referring to the resulting mixture.

    Liébaut says for Beauce wheat specifically that leavening should be "refreshed" with cold water at noon, then at five, then, for the last, at nine. It is possible here however that he is using the word levain in the larger sense.

    The account of the Limoges trials says that the day after the leaven had been put in the dough, the paton (the prepared dough) was “refreshed”, meaning that water was added to it.

    De Vigenère specifies several additions of water to the levain, but again may be using the term in its secondary sense. If he simply means the original leavening alone, it is not clear what quantity of this he is referring to. It is possible too that he is referring to the different stages of leavening which would later be called levain de chef, etc. 


    The most obvious additive is salt.

    Salt had been used in bread for a long time. Note Pliny's mention of it above. In the thirteenth century, Aldebrandino wrote that bread should be well risen and "a little salted". Maino said the same thing.

    Champier writes that “bread becomes better when it has a lot of salt” (melior redditur quando multum salis habet). Bruyerin says that “courtiers, nobles and those in the cities who take pleasure in a purer and more refined life eat bread with salt. The common people because of the lack of salt and its cost do not use it.” (Aulici proceres, et in urbibus, qui nitidiori atque elegantiori genere vitae delectantur, panibus sale conditis, vescuntur. Vulgus ob salis inopiam et caritatem, non usurpat.)

    But salt had also been relatively cheap – being a native French product – until the fourteenth century, when French kings began to tax it. This may be one reason that some regions in France used no salt at all in their bread. The statutes for one town in Normandy mention both salted and unsalted bread. The Limoges trials include salt, but a relatively small amount.

    Desportes, having studied a number of bread trials, writes: "the absence of salt among the ingredients necessary in making bread dough is flagrant in all the accounts of bread trials which I have studied in numerous cities in the north of France for the period 1350-1570”.

    In the Limoges trial less than 1% of salt was added. In the 1432 Parisian trial, no salt is mentioned either in the steps, nor in the itemized costs for each item (which are very detailed). While adding it might have been passed over in the steps, the lack of any expense for it suggests it was not used at all. De Vigenère does not mention it either. Since he was mainly concerned with weight, at the least this indicates that no significant volume of salt was added.

    Michel de Montaigne (1533 – 1592) could certainly afford salt, but had his baker make bread without it which was, he writes, "against the custom of the region" (near Bordeaux).

    One can argue then for using it or not. But even where it is used, for standard bread it should be used sparingly, at least for any bread referencing the fourteenth century or later, though bread for the upper classes might use more. (We know very little about earlier medieval bread – that is, bread for most of the Middle Ages – but salt was cheap enough to be used by, for instance, hermits in that period, so it is likely that it was far more commonly used in bread.)

    The situation is less clear for English bread, but the first recipes to appear use salt freely.

    Several writers also mention adding things like anise or fennel to bread. This was almost certainly not done for municipal bread, but it seems to have been at least a familiar practice for country bread. It may be that it made up for the lack of salt (and a modern baker especially might want to use it that way), but it also may simply have been a function of what was available in each region. Where an herb or seed was common in a particular region, one could arguably add it to the bread; at the least, anise and fennel seed are solidly documented choices.

    Several texts mention using milk in bread, but with no idea of how or for which breads. In later centuries it was used for finer breads, and this seems likely to have been the case in this period, but even descriptions of specific better quality breads do not mention it. A modern baker can justify using it, but will be “flying blind” in doing so.

    Bruyerin goes further and says that “other than fine flour, other than wheat, the work of the baker consists in kneading butter, eggs, cheese, milk, honey and sugar” (alia ex similagine, alia ex polline tritici constant opera pistoria butyro, ovis, caseo, lacte, melle aut saccaro subigentes). But it is not clear that he is distinguishing standard bread from pastries (which were still often considered a form of bread). Similarly, Desportes notes that in Rouen butter, eggs and cheese were put in preparations considered "white bread"; but again the specific names for these suggest they were probably pastries.


    In later centuries, the most common size for a Parisian loaf seemed to be four pounds. But medieval bakers' loaves were generally smaller both in Paris and elsewhere. While the size varied for a denier's worth, depending on the price of wheat, the average white loaf hovered just under a pound in weight. The loaves in the Limoges trial weighed a little over eleven ounces each. De Vigenère's trial specifies twelve ounce loaves for both dark and light bread. Otherwise, there were exceptions outside Paris, but even these did not go far above a pound.

    Basically, for a modern baker, a pound or just under is a credible size for a period loaf.

    Coarser breads were sometimes made in larger sizes; one loaf sometimes lasted a week. In Poitiers towards the fifteenth century, a large wheat bread made for households could weigh from three pounds two ounces to twelve and a half pounds. Maslin bread could weigh even more. In Nantes, coarse bread (mainly bran) could weigh twenty seven and three quarters of an ounce, whereas a fine bread (fouace) weighed eighteen ounces. In Limoges, a big rye tourte was to weigh thirteen and a half pounds.

    A Parisian bread trial from 1432 mentions dark bread weighing forty ounces unbaked. (No baked weight is given, but the baked white loaves weighed sixteen ounces and dark bread weighing thirty ounces unbaked weighed twenty-seven baked.) In a 1477 trial, baked white bread weighed twelve ounces, dark-white bread two pounds and dark bread three pounds.

    Here are some other sample weights from Paris:
    1350 statutes
    Different weights were made for one and two deniers depending on the price of wheat
    Note that these breads were made in much smaller sizes than later.
    Good white bread from five ounces and ten ounces up to nine ounces and fifteen and half ounces
    White-dark bread from almost five ounces and eleven ounces up to nine and a half ounces and nineteen ounces
    Dark bread from eight and sixteen ounces up to thirteen ounces and twenty-six ounces
    1415 bread trial Dark-white bread twelve ounces
    Maslin (wheat and rye) bread eighteen ounces
    Rye bread with all endosperm and bran twenty four ounces
    1419 statutes All types: half pound, one pound and two pounds
    1421 statutes White and brown bread thirteen ounces
    Brown bread was also made in twenty-six ounces
    Rye bread thirteen and twenty-six ounces
    1431 diary note The Bourgeois de Paris complained that “very black white bread” did not weigh more than twelve ounces
    1441 diary note Double size white bread weighed twenty-four ounces
    Double dark white bread weighed thirty two ounces

    In the British Isles, things were similar. In Aberdeen, "a norm of around 15 ounces in the second half of the fifteenth century falls to around 10 ounces in the first half of the sixteenth century." (Gemmil) English bread might have been slightly bigger, but still near a pound in weight. In an assize from the time of Henry II: “the size of the loaf when corn was sold for four shillings and sixpence; it was to weigh 30 shillings, each presumably of twelve pence, and the pennies of twenty to the ounce”. A pound then weighing twenty shillings, the English loaf weighed (with variations) a pound and a half. (The use of coins for weight was not common, but appears to have been used here.)


    Liébaut specifically mentions shaping dough into “orbicular” pieces, matching the generally spherical shape of breads seen in images. Some show a slightly flattened hemisphere as well.

    Though long breads would soon be mentioned and one such (very baguette like) bread appears in a famous German image, a spherical or at least round shape seems most period-appropriate.

    Note too that these were sourdough-leavened, essentially whole wheat, breads made with soft wheat. They would not have risen as robustly as yeast-leavened bread made with whiter flour from harder wheat.


    Obviously the best oven for this kind of baking is not only a wood-fired oven, but one where the fire is built inside the oven itself, then raked out. Neither is a likely option for most home bakers. One option I've seen suggested is to reproduce a similar effect in a home oven by starting at a high heat, then reducing it.

    How high a heat is another question. Some texts here warn against too high a heat, since it will burn the bread and also create a crust too soon to allow the inside to bake properly. While no temperatures are recorded for this pre-thermometer data, one method makes it possible to estimate the temperature used. In the eighteenth century, Malouin reported a method used by English bakers to test if an oven was ready:
    to try the heat of the oven, one puts a pinch of pinch of flour at the entry; if it reddens at once, the heat of the oven is just right; if the flour blackens, the oven is too hot; finally if it retains its whiteness, the oven is not hot enough.
    While a modern baker can set the temperature easily enough, doing this (probably with a light foil pan to avoid a mess in the oven itself) might be one way to determine a temperature that an earlier baker would have found adequate. Another eighteenth century English method was to use a stone: "The oven is reckoned hot enough, when a stone that is plac'd in the middle for a trial ceases to look black." (Bradley) (In other words, it turns white hot.)

    While these are late for our purposes, they do give some idea of how to determine the right heat without a thermometer, and it is credible that similar methods were used earlier.
    Nor does one necessarily have to use an oven. Liébaut says that the oven is the best place to cook bread, because it heats it on all sides, but he expects that some will make bread on the hearth or on a grill, though that leaves only one side cooked. Other sources too refer to cooking bread under the coals or under a pottery bell. A modern cook then might try cooking the dough (possibly wrapped in cabbage leaves, as noted in other cultures) under hot coals or under a bell-like covering (effectively, a miniature oven).


    In envisioning the bread of former times, it may be natural to think of one of those large, crusty, loaves with a chewy crumb sold as traditional. And some period descriptions of the best bread do suggest something, at the least, appetizing: “well risen, a little salted, well kneaded, full of holes”. But right off note that one word is missing here, a word regularly found in later descriptions of good bread: “crusty” (croustillant). This has long been a characteristic of good French bread. But not only is it never mentioned in medieval descriptions of good bread, Liébaut says specifically that in great households the crust was grated off. The very fact that the crust was hard enough to grate off suggests it was thicker and hard than most good crusts today (even if Liébaut does grant that the crust, though considered unhealthy, tasted better than the crumb). This alone should give some idea of how different even the best bread was in the time. Add to this the fact that even the best bread was made of softer wheat and often of minimally salted, hard dough and was typically far smaller than today's loaves, and the result was likely to be very different from anything sold as “traditional” today.

    As a practical matter, this means that any baker who wants to make “medieval” bread may initially work strictly within the parameters cited here, but might find the result too far from our current idea of good bread to produce on a regular basis. How many of these parameters can one ignore and still end up with a loaf that has any claim to be medieval? No doubt different bakers will answer the question differently. But hopefully the information provided here will at least offer a solid basis for each decision a baker makes in working towards that goal.

    APPENDIX: Milling and bolting trials

    With limited means and time, I have only tried to apply some of the above principles, mainly in regard to milling and bolting. Having first tried grinding hard wheat berries in a blender and, after bolting them to three qualities, obtained the same (brown) color in the baked results, I tried to come closer to stone grinding by using a spice mill (from Ikea) in hopes it would leave more whole bran to separate. In both cases, I managed to produce a very white flour (along with a less white and dark one) by bolting the ground flour first with a plastic mesh and then (for the finer flour) a close-woven straw hat. This laborious process left little flour to use for culturing sourdough, so I did that with commercial whole wheat flour.

    As with the blender-ground version, however, all the resultant breads were about the same color, even that from very white flour. Here are the flours with the crumbs they produced:

    It seems likely then that even the very white flour still contains more bran than is apparent. What would it take to mill the grain and then sift it to produce, if not a truly white bread, a noticeably light one? That is one avenue for adventurous historical bakers to explore.


    Livius, Titus, Blaise de Vigenère, Les decades qui se trouvent de Tite-Live, 1583

    Comet, Georges, “Dur ou tendre ? Propos sur le blé médiéval”, Médiévales 1989  Volume   8  Issue   16-1  pp. 101-112

    Gemmill, Elizabeth, Nicholas Mayhew, Changing Values in Medieval Scotland 2006

    Weise, Elizabeth, “The hard truth about stone-ground flour”, USA Today, March 13, 2006
    “Wheat FAQ: What about whole-wheat bread and flour?”, The Whole Truth site

    Ogilvie, John, The Imperial Dictionary of the English Language, Vol I 1896

    Parmentier, Antoine Augustin, Le parfait boulanger ou traité complet sur la fabrication & le commerce du pain 1778
    Pipponier, Francoise, L'equipement des boulangers bourguignons a la fin du Moyen Age
    de Montaigne, Michel Eyquem, Les Essais 1657

    Desportes, F., “Le pain en Normandie à la fin du Moyen Âge”Annales de Normandie  1981

    Rusticus, "Remarks occasioned by the Scarcity of Wheat", The Gentleman's Magazine, Volume 65, Part 2, November 1795

    Pliny (the Elder), The Natural History of Pliny, Vol 4 1856

    Bradley, Richard, Husbandry and trade improv'd, 1727