Friday, June 12, 2015

FRENCH BREAD HISTORY: Late medieval bread

This is the fourth in a series of posts on the history of French bread. The preceding post was on early medieval bread. Further information on bread history can be found on Facebook in the Bread History Lounge.

Around the time of the First Crusade (1096–1099), French mentions of bread began to reflect something taken for granted today: variety.

A modern French bakery presents a visual experience that is almost musical in its contrast of long baguettes with round pains de compagne, torpedo-lie bâtards, braids, croissants, etc. For centuries, such variety has, to a greater or lesser extent, been typical of French bread bakeries, even when the bakeries themselves were (until the nineteenth century) rustic and sparse. But this variety has developed over time and only begins to appear in records in the late Middle Ages. After the Romans and before the Crusades, mentions of bread are repetitive and monotonous and only scattered images suggest any variety at all.

This was probably all the more true in the century before the Crusades, when Hugh Capet (c. 940-966) founded the last major French dynasty. Overall, the Capetians (whose reign ended with Louis XVI) ruled a more sophisticated and settled France than the two dynasties before them. But soon after their reign began, France experienced some of the worst famines in its history. Claims of cannibalism, of human meat sold in markets, have been questioned, but there is no doubt times became dire in the tenth century. Ironically, some scholars believe that the very desperation of this period spurred improvements in agriculture. Whatever the case, even if famines and shortages would recur throughout French history, nothing like the horrors of this period are mentioned after it.

For many, simply getting bread was problematic.

A rare glimpse at what variations did exist at this point comes from the Abbey of Cluny, where monks used hand signals at meals in order not to talk. These are believed to date to the tenth century, though they are mainly documented slightly later. A round bread of one pound was indicated by thumb and index finger touching in a circle; a “better bread” (panis melior) was indicated by making a hollow with one hand, and “pouring” with the other, suggesting that this was a type of bread poached in water (like the later eschaudés). Other signs show a rye bread (turta) divided into four parts with a cross and a tortula, possibly a bread cooked on the hearth. This is a narrow range of variations and most were already known in the early Middle Ages (though this may be one of the first references since Roman times to a poached bread).

By the thirteenth century, if many records still referred to "bread" without description or qualification, bread was no longer just a generic term; there were already, as today, various breads, intended for different uses and purposes. If the soccer-ball like shape seen in innumerable medieval images remained the standard bread, the names at least of different breads were recorded and very probably some, though unknown to us, had different forms.

In the seventeenth century, 
the great philologist Du Cange (1610-1688) included a long list of breads in his Latin glossary, drawing on various histories, charters, donations, etc. He does not give dates for all of these and no doubt many of his sources have disappeared since. What is more, it is rare that his sources provide anything more than the name of each bread, so that it is often impossible to know exactly what each term corresponded to and whether different terms from different sources in fact refer to the same bread. Still, what can be gleaned from Du Cange's glossary shows that different terms for breads already appeared in documentation around the time of the Crusades.

One of his terms, panis asper, comes from Rudolph Glaber, the same historian who recorded some of the dramatic accounts of tenth century famines. Glaber writes more specifically of panemque admodum asperum; that is, very rough or harsh bread. Du Cange takes this to mean black or bran bread.
Here in a somewhat summary fashion are references from Du Cange going into the late Middle Ages:

1050 Panes Coronati [Crown Breads] – Perhaps a circular bread?
1106 Panis Ecclesiaticus [Church Bread] – Probably a finer bread
1130 Panis de Cambio [Trade or Exchange Bread] – This was a rye bread used as a kind of currency (in Italy): “In northern Italy, bread loaves of a determined size (panis de cambio, exchangeable bread) were ascribed a specific monetary value as fractions of the gold coin, thus making up for the lack of coins usable for small transactions.” (Postan)
1188, 1214 Panis Natalitius [Nativity bread] (also tourte)
1190, 1209 Panes Curiales [Meeting or Banquet Bread]
1201 Panis de Mait [“trough or tub bread”]- The name suggests one of a number of breads which were poached in water, though it might simply have been made in a tub, presumably in larger quantities.
1224 Panis de Paribus [“Equals' Bread”] - A shared bread, apparently.
1235 Panis Emendationis [“Compensation Bread”] - Given as payment to a worker. and so very likely to have been made of rye, barley or maslin.
1242 Panis Biscotus [“Twice-cooked Bread”] - Biscuit, basically (but this had already existed for some time).
1248 Panis Falsus [“False Bread”] - Not of statutory weight or form.
1248 Panis Panetarie [“Storehouse Bread”]
1248, 1270 Panis Militis/Armigerorum [“Soldier/Squire's Bread”] - This was probably an inferior, bran-heavy bread as was true of later military breads.
1254, 1270 Panis Conventualis [“Conventual or Meeting Bread”]
1271 Panes Festi [“Holiday Bread”] - Probably a finer bread.
1284 Panis de Aula [“Hall Bread”] - For the abbot or guests of an abbey; thus finer.
1291 Panis Francicus [“Frank/Frankish Bread”] - Apparently a finer bread, possibly even a pastry.
1296 Panis Medianus [“Middling Bread”] - Probably between dark and the finest bread.
1303 Panis Foliati [“Leafed Bread”] - Puff pastry, or something very like it.
1308 Panis Orationis Sancte [“Bread of Holy Prayer”] - Possibly a Communion wafer, which would later be called (among other things) a “bread for singing”.
1309 Panis Paganus ['Village or Peasant Bread:”] - Du Cange says this was bread made by peasants; probably of barley or rye.
1339 Panis Focagii [“Hearth Bread”, foccacio] – Bread made on the hearth; this was known by a variety of names – fouace, fougasse, etc. - all corruptions of the Latin for hearth bread.
1392 Panis Ferratus [“Iron Bread”] - Wafers or waffles; that is, bread made between two hot irons.
1460 Panis Moly [“Softish Bread”, corruption of pain mollet] - A very fine white bread which in later centuries would be the first to be leavened with yeast instead of sour dough.

Note that it is  relatively rare here to indicate anything particular in the making of each bread. Most of these breads are named for the circumstances under which they would be used and many may have been the typical ball-shaped bread made more or less fine. But even this variety in the names is a new development at this point.

John de Garland's Dictionary

John de Garland (Johannes de Garlandia) (c. 1190? - c. 1270?), an Englishman who lived and taught in Paris, left a curious dictionary which gives an overview of Parisian trades in the thirteenth century. He says that bakers at that point sold bread of wheat, rye, barley, oats, maslin (mixed grain) and bran, but says nothing of specific loaves. Still it is interesting that such a range of grains was used for bakers' bread; whether they were used or not, most are not named in subsequent statutes, etc. Barley and rye were more typically used for servants' bread; oats are rarely named for bread at all.

He also describes regrattiers (resellers or grocers) selling items which a period commentator glossed: placente (usually "pastries" or "cakes") said to be simenians (that is simnel, a very fine white bread still found in England much later); flammicia, or flamiches, today a tart, a regional specialty, but apparently a bread or pastry then; ignacia, said to be fouace, which originally was a hearth bread, but is sometimes cited in ways that suggest it was something finer in this period. (He does not however say which professionals made these.)

If these glosses are accurate, they show that several items were made which were still known in later centuries, but rarely mentioned in documents of this period. It also suggests that bakers solde some items not named in statutes, though Garland's observations also precede Boileau's 1268 statutes, which may simply have eliminated certain items from the bakers' repertoire.

Bread in statutes

Municipal statutes for bakers first record which breads were in standard use in cities. These above all were the ancestors of the baguette, the pain de campagne and other breads found in Paris bakeries today.

When Boileau wrote down statutes for the different trades in 1268, he began by addressing the tameliers (“sifters”) who were the first municipal bakers. At this point, the standard size for a bread seems to have a denier's worth (neither the weight nor the quality of the flour are mentioned). This is sometimes referred to as a denrée ("good", as in "goods"), though the term is also used in a way that suggests a unit or an item. The loaf twice the size, a doubleau, sold for two deniers; one half the size, a demie ("half"), sold for an obole (that is, half a denier). Normally, it was forbidden to make a bread larger than a doubleau, unless it was a gastel (cake) made for special occasions (early uses of the word gastel/gâteau imply a special bread rather than a sweet cake). It is striking that the statute also mentions echaudés, “of which one can give 14 denrées for 12 deniers” (the implication seeming to be that these were roughly the size of a standard loaf, but could be made slightly smaller). This bread/pastry has existed in various forms since this time, but above all is characterized by being poached (like a dumpling). It may be the bread dipped in water referenced at Cluny. It would already become less common under the Old Regime, but at this point seems to have been counted as one of the more common breads.

Note that there is nothing here about the weight, quality of the flour, or degree of fineness (bran extraction) of each bread. The bread in commerce, at least, seems to have been roughly standard in its make-up and the Parisian consumer initially had very narrow choices.

The statute's restrictions on price (the first since Charlemagne) assume a correspondence between weight and price, so the statute, rather than forbidding false weight directly, forbids sales under certain prices. Three doubliaux normally sold for six deniers; they could not be sold for more, nor for less than five oboles (that is, two and a half deniers). Breads normally sold for twelve deniers (presumably in quantity) could not be sold for less than eleven.

The largest of these was sold by three; the denrée was sold by the dozen or half-dozen. In giving a discount of one obole on a half-dozen or a denier on a dozen, the tamelier effectively offered a thirteenth bread for each dozen; that is, what English-speakers would later call a “baker's dozen”.

The price of breads sold in shops was carefully controlled. More leeway was allowed on price on market days, as long as the doubleau did not cost more than two deniers. This marginal freedom resulted in its being called a pote loaf, "pote" being from the Latin word for power or "being able to".

Note that these were the breads in commerce. In households, for instance, servants no doubt still got bread of rye, barley or maslin.

Boileau's statutes also mention bakers from outside Paris. The medieval French word fors (now hors) means “except” or “outside of” and these bakers were called boulangers forains. Parisian bakers objected to these “outsiders” selling bread in Paris and the statutes limit them to selling rejected or damaged bread and then only on Sundays. Some towns beyond Paris had a more specific reputation. The fourteenth century Menagier de Paris mentions bread from Corbeil (today a major metro stop) as inferior (and so dark). Bread from Chilly (today Chilly-Mazarin) was known as “bread of Chailly” [sic] and considered finer and whiter.

In 1305, Philip the Fair established statutes for the bakers. But though these prescribed punishments for making defective bread, they gave no details at all on the proper weight or price of bread. In 1305, John II (John the Good) was the first to do this. These statutes make it clear that in the decades since Boileau's statutes, bread-baking had developed in ways they did not address. Not only had pain de Chailly gained a good reputation, the term was now used for the finest bread made in Paris itself. What is more, bread was now priced by quality.

The new statutes established a hierarchy which would be found, under different names, throughout France and for most of the Old Regime: fine bread, the whitest sold; a good but less fine bread, typically that of the greater part of the population; and a dark bread for the poorest people. The 1350 statutes name these, respectively, as pain de Chailly, pain coquillé and pain bis. Coquille simply means “shell” and the term would later refer to a swollen crust; in the seventeenth century, Cotgrave defined pain coquillé as "a kind of hard-crusted bread, whose loafs do somewhat resemble the Dutch Buns of our Rhenish Wine houses." In this case, however, it seems simply to have referred to a bread of average quality, between the best and the worst; in later centuries this would be called bis-blanc (“dark-white”). Pain bis means “dark bread” and would be the term most commonly used for poor quality bread.

Note that echaudés are no longer mentioned, though bakers were still making these in 1438 (Bourgeois de Paris).

The price of bread would later vary for the same size loaf, but at this point it was the weight of the loaf which varied, while the price (a denier or a multiple of it) remained constant. As an example, when a septier of wheat cost forty sols, the statute specified:
The dough for a Chailly bread of one denier weighs five ounces, and baked four ounces, five estellins [20th of an ounce]. The dough for bread of two deniers [that is, the old doubleau] weighs ten ounces, and baked six ounces, five estellins.
The dough of bread of one denier coquillé weighs six ounces five estellins; and baked five ounces and a half...
The dough of dark bread of one denier weighs nine ounces and a half; and the baked bread eight ounces.
As the price of the septier varied, the weight of a one or two denier bread was to vary with it.

Apparently bakers only approximately followed these guidelines and in 1372 Charles VI issued new statutes. These were very similar, but the three types of bread were now pain de Chailly, pain bourgeois and pain faitis, also known as pain de Brode.

In the eighteenth century, Le Grand d'Aussy wrote that pain bourgeois (city-dweller or burgher's bread) was the same as the pain de ménage (“household bread”), of his own time. Faitis simply means “made”, though it can also mean “fake” or “fraudulent” (like the modern French word factice). Why the term was used for dark bread is not clear, but at any rate pain bis would remain the preferred term over time (pain de Brode may have taken on a slightly different meaning, later being made from a mix of rye with wheat flour). For over a century, however, these were the standard Parisian terms for the three qualities of bread.

An exception appeared in 1420, when times were desperate and Charles VI ordered that bakers only make two kinds of bread: white and brown. The white was to be made of wheat; the brown of wheat combined with “mixed” méteil (maslin) flour which is then explicitly said to contain either rye or barley flour. What is more, these are set at a pound, or multiples or fractions thereof, the white selling at twenty deniers a pound and the brown at sixteen. A similar statute appeared in 1420, but referring to white, brown and rye bread separately. (Rye bread was rarely mentioned in early statutes; its inclusion here shows how bad times were.)

In this same difficult period, a type of bread briefly appeared called pain armé (“armed bread”) which was made from wheat as it came from the mill, without being bolted or otherwise cleaned. Bread of this sort was made at different times from both wheat and rye flour. In 1419, a decree noted that this “bread was not agreeable to the people, because it was brown, and they were not used to using such bread” - an early note of how demanding the Parisian populace could be about its bread, even in hard times.

Otherwise, towards the end of the Middle Ages, the main breads in commerce were fine white bread (pain de Chailly), average quality bread (pain bourgeois) and dark bread (pain bis, faitis or de Brode). These were not the only breads produced in the time. But this trinity was the ancestor of the far more varied selection in today's Parisian bakeries.

Pain de bouche and trenchers

In fine households, two types of bread in particular were used: pain de bouche (“mouth bread”) and trenchers. Pain de bouche may have been nothing more than pain de Chailly (in or near Paris, at least), or it might have been an even finer bread made by the household baker; no details survive to be sure either way. Nor is it sure why the term was used. One source claims it was made at the mouth of the oven, but it seems more likely that it was used to contrast it with the bread that was NOT for the mouth (that is, not to be eaten): the trencher.

The idea of people eating off trenchers is so associated with the Middle Ages that it may come as a surprise to know that the concept only appeared towards the end of the period (and did not persist for long after it); it is very unlikely that Charlemagne or even Hugh Capet ate off trenchers. The English term is a corruption of the French tranchoir; that is, “slicer”, from the verb trancher (“to slice”); a trencher served as surface on which to carve and eat one's portion. A trencher did not have to be made of bread; it could be, for instance, a silver dish. Why or how the idea appeared of using hardened bread for this purpose is unknown.

We have a clear idea of how at least some trenchers were made, thanks to the Menagier de Paris, which says to order bread “a half foot wide and four fingers high, baked four days before” for that purpose. (Note that an order for white bread suggests that it be a day old.) Alternately, it says to use Corbeil bread, which is to say, dark and inferior bread. But a 1336 record for a regional prince says that four small one pound breads are to be used for slicing. A regional record from the early fifteenth century says that rye flour was to be used to make trenchers, but these were for prisoners and appear to have been intended to eat (Aube).

The concept then was varied, but in general referred to hard plates of bread used for slicing. Exactly how hard is open to question; until the late sixteenth century, French bakers generally used sourdough as a leavening and sourdough leavened bread can stay fresh far longer than yeast-leavened bread. (The Menagier specifically contrasts the leavening used for beer – that is, yeast – with the leavening used for bread.) So even a four-day old bread might have remained somewhat edible.

Describing a rich man's meal, a poem from 1475 says that the poor:
get the trenchers of bread
which remain upon the table.

Ils ont tranchouers
Qui demeurent du pain dessus la 
Martial d'Auvergne, Les Vigiles de Charles VII
This brief verse may be the source of the idea that trenchers, soaked in sauce from the food, were given to the poor after the meal. While this was ultimately true, the trenchers may first have been gathered in an alms plate and stored before being distributed. Also, if the breads were indeed as hard as is sometimes thought, it is not clear that they would have imbibed much of the sauces on them.

Shells, wafers and pastries
In this period too, French pastry began to appear as a separate product. As references to things like gastel (now gateau) show, bread-bakers at first made products later associated with sweet baked goods. The Romans and Greeks had had pastries (often made with cheese and/or honey), but the idea seems to have faded in France long before the Crusades. It reappeared from two very different directions: making wafers and making pasties.

Pasties were nothing more than shells of dough which held food. In England, these were known as pies, even if the French word pasté (meaning “doughed”) also became “pasty” in English. For those who know this period, they are so universal a feature of medieval cooking that a question may never arise: how did Europeans have the idea of making shells out of dough? It is tempting to trace this innovation, like so many after the Crusades, to the East. But it is also true that by this period the flan (which had begun as a flatcake) was essentially a pastry shell holding cream or cheese and there is some evidence that over time the flat flan had initially gained low sides (like a pizza) which later rose to form a bowl.

That history remains obscure, but what is clear is that at some point making pastés became known as pastisserie – that is, pastry – and those who made them pastissiers (pastry-makers; or, in English, pie-men). But for a very long time pastry-makers did not make sweets, except incidentally; they made anything (usually meat or fish) served in a pastry shell.

Yet a light (if not necessarily sweet) baked item did exist in this period, and had since before the Crusades: the wafer. This frequently appears on menus, along with or in the place of fruit, as dessert. What's more, by now what had been a simple secular variant of the Communion wafer was made in varied forms which are also cited on such menus. The maistre (“master”) may have been a larger wafer, made with white wine in the dough or it may have simply been a service of wafers. The nieule – known in Latin as a nebula, or cloud – was a lighter wafer (though some texts make it synonymous with the latter).

These humble waffle-like treats (they were made between hot irons) along with the more dessert-like pastry shell dishes such as a flan were the ancestors of what would, over time, become French pastry. But at this point it was the oubloyeurs (wafer-makers) who made oublies (wafers) and so their role in the period was closer to that pastry-makers would hold in the future.

Other innovations
Two references from this period shows that the French were already making what they call “leafed dough” (pâte feuilletée) and what English speakers call “puff pastry” (or, more technically, “laminated dough”). Recall the 1303 mention cited by Du Cange of Panis Foliati. A charter from 1311 mentions gasteaux feuillés. In modern French this would be a “leafed cake”, but gateau still had a more ambivalent meaning at this point and this may still have been a kind of bread. Various explanations have been offered for how this technique – so central today to products like the croissant - first came to France. It might have been an Arab technique which made its way up from Spain. A thirteenth century Andulusian recipe book includes this recipe for “leafy” (muwarraqa) musammana:
Preparation of Musammana [Buttered] Which Is Muwarraqa [Leafy]
Take pure semolina or wheat flour and knead a stiff dough without yeast. Moisten it little by little and don't stop kneading it... until it relaxes and is ready and is softened so that you can stretch a piece without severing it. Then put it in a new frying pan on a moderate fire. When the pan has heated, take a piece of the dough and roll it out thin on marble or a board. Smear it with melted clarified butter or fresh butter liquified over water. Then roll it up like a cloth until it becomes like a reed. Then twist it and beat it with your palm until it becomes like a round thin bread, and if you want, fold it over also. Then roll it out and beat it with your palm a second time until it becomes round and thin. Then put it in a heated frying pan after you have greased the frying pan with clarified butter, and whenever the clarified butter dries out, moisten [with butter] little by little, and turn it around until it binds, and then take it away and make more until you finish the amount you need. Then pound them between your palms and toss on butter and boiling honey. When it has cooled, dust it with ground sugar and serve it.
But in France the brief mentions above only show that it was known, not how it was made or came to be.

In 1365 a church council at Angers forbade the use of milk or butter at Lent, whether in bread or vegetables. This early mention is striking, since the use of milk, in particular, in bread is rarely mentioned until later centuries, when it was associated with luxurious breads and, especially, in the nineteenth century, Vienna-style breads. Unfortunately, this tantalizing reference is the only one of the sort in the period, so we do not know who was making such bread or if it was considered to have a special status. (While it is just possible too that the reference is to using butter on bread and dipping bread in milk, the Latin is specifically in pane.)

A decree of February 17, 1436 mentions several fine breads of “similar whiteness” including “cakes” and échaudés, but also brioches. This may be the first mention of this in French baking history, though there is no way of knowing if it was already the rich bread later known.

The list of words from Du Cange includes a reference to Panis Moly, a corruption of pain mollet. The tax roles for 1292 mention a man named Jehan (John) Pain-Molet [sic].  These sparse references show that the term ("softish bread") existed by the thirteenth century (possibly before, for it to have become a family name). It would remain rare until the sixteenth century when bakers began to use yeast (instead of sourdough) to leaven it, provoking memorable upheavals in French bread-baking. All that can be said about it in earlier times is that it was no doubt a very fine bread, but very likely leavened with sourdough at this point.

The word miche first appears now as well, but its meaning is unclear. Even in recent centuries, this word's meaning has shifted; it typically means the most common large loaf in a given time and region (and so in Paris today it refers by default to large round breads). The word itself is derived from the Latin mica (“crumb”), which also became mie in French. It has been traced back to Chrétien de Troyes' twelfth century Perceval le Gallois: “Mais n'a çaiens que v mices”, though this could arguably refer to “five crumbs”. But the fourteenth century poet Eustache Deschamps (1340-c 1404) has a character say:
You can only give a miche
Or a piece of dark bread.

Tu n'as pouvoir fors d'une miche
Ou d'un morsiau de pain faitis
And a fourteenth century Breton poem, Le combat de trente, says of a man killed in combat: “Jamais ne mangera de miche ni de gastel.” (“Never would he eat miche nor cake.”) The term then was established by the fourteenth century, even if its meaning is uncertain.

As the medieval period ended, then, a number of concepts had appeared which are now taken for granted in bread-making. Not the least of these was the public bakery itself. These had existed under the Romans, but only appeared fitfully just before the Crusades. It was after them that they became standard in cities, as shown by numerous statutes. The idea that they would sell three qualities of bread was established and would endure throughout the Old Regime, even if other breads were to join these later. The regulation of price and quality began in this period and would remain, in shifting forms, well past the French Revolution (and was bitterly resented by bakers).

The variety in types of bread which became typical of French bread began in this period and already went beyond the three official grades of bread. Biscuits and wafers already existed by the Crusades; wafers would grow more varied, forming the major dessert of the period. Pasties, first used primarily for savory dishes, were the first pastry (by that name) in France. Over time, wafer-makers and pastry-makers would join to create what became French pastry.

Several terms came into use during this period, though they may have different meanings than they later would: brioche, simnel, flamiche, fouacemiche, pain mollet, puff pastry, cake. Trenchers of bread appeared but would not remain in use for long after. The term pain de bouche would last longer, though other terms for very fine bread would displace it.

Milk and butter seem to have been used in bread in this period. Soon after, these and other additions like eggs and honey would more typically be used in pastry; but as the Old Regime ended they were again found in bread.

Overall, the fundamentals of French bread and bread-making were established in this period, even if these would evolve, and often in distinct phases, over the centuries to come.


Ayzac, Félicie d', Histoire de l'Abbaye de Saint Denis VI 1860

Migne, Jacques-Paul, “Consuetudines Cluniacenses - Udalrici Cluniacensis Monaci”,
Patrologiæ cursus completus: sive bibliotheca universalis ..., V149 1853

Du Cange, Charles Du Fresne, Glossarium mediae et infimae latinitatis. T6

Glaber, Rodulfus, Raoul Glaber: Les cinq livres de ses histoires (900-1044). 1866

Postan, Michael Moïssey, The Cambridge Economic History of Europe: Trade and industryin the Middle Ages, 1966

"The Dictionaries of John de Garlande", Mayer, Joseph, A Library of National Antiquities, 1857

de La Mare, Nicolas, Traité de la police où l'on trouvera l'histoire de son établissement, les … 1713

Cotgrave, Randle, A Dictionarie of the French and English Tongues, 1611

Ordonnances des roys de France de la troisième race, recueillies v2, 1729

de Coussy, Mathieu, Journal d'un Bourgeois de Paris, ed. Buchon, Jean Alexandrey 1838

Ordonnances des roys de France de la troisième race, recueillies v11, 1769

Le ménagier de Paris: Traité de morale et d'économie domestique ..., V2, 1846

Archives départementales de l'Aube, “G. 283 Registre – 1417-1418”, Inventaire sommaire des Archives départementales antérieures à1790: Aube ... 1873

An Anonymous Andalusian Cookbook of the 13th Century

de Troyes, Chrétien, Percevalle Gallois ou Le conte du Graal. V2 1866-1871

Deschamps, Eustache, Œuvres complètes de Eustache Deschamps, V52 1891

La Poix Fréminville (chevalier de), Christophe-Paulin, Le combat des trente: poëme du XIVe siècle 1819

Saturday, May 2, 2015

The Chinese origin of the Baked Alaska?

The most famous “origin” claim for the Baked Alaska is that it was invented in 1867 by Charles Ranhofer, a cook at Delmonico's, to celebrate Seward's purchase of that territory. The term would not appear however until over a decade after that and no reference appears to this event in the period itself. As with so many famous foodstuffs, most of what is said about its history is at least doubtful and sometimes plain wrong.

In seeking out solid information about this dish of cold enclosed in hot, some might be surprised to find themselves in.... China.

In the eighteen forties, several writers mentioned a Chinese specialty: roasted ice.
A traveller who visited Pekin, says, that a favourite dish in that city is roasted ice, which is enormously dear, as very few cooks possess the skill and dexterity required for its preparation. A lump of ice is taken upon a sieve, and after being quickly enveloped in a sort of paste made of sugar, eggs, and spices, is plunged into a panful of boiling pork fat or lard. The grand point is then to serve it up before the ice has time to melt. What may be the peculiar attraction of this dainty dish it would be hard to say, for though frozen inside, it burns the mouth when first tasted.
(“Recollections of Peking”)
In 1859, it was well-known enough for an American short story to include a reference to: “'Baked ice a la Ching-ki-pin,' — which was highly esteemed. The ice was enveloped in 'a crust of fine pastry, and introduced into the oven ; the paste being baked before the ice — thus protected from the heat — had melted" (The Atlantic Monthly).

Decades later still, Heinrich Heine used it (in German) as a metaphor:
This cold passion, which is served up to us in such blazing figures of speech, always reminds me of the roasted ice which the Chinese prepare so artistically by holding a lump of something frozen, wrapped in a thin coat of dough, for a minute over the fire. It is an antithetic dainty, which must be swallowed at once, and which, with its hot rind, burns the lips and tongue while it cools the stomach
The principle then was known in the West. And in fact it had already been applied at the start of the century; in 1802, Dr. Samuel Mitchell wrote that Thomas Jefferson served: “ ... in the form of balls of the frozen material inclosed in covers of warm pastryexhibiting a curious contrast, as if the ice had just been taken from the oven.” But this method, very like that used by the Chinese with ice, seems to have been instantly forgotten. (Mitchell's letters were not published until 1879, after other variants of the concept already existed.)

Nor was the idea of such a contrasting dish recent; this is from the 15th century version of Taillevent's Viandier:
Fried Fresh Butter

To make fresh butter in a frying pan, take stale white bread, and crush up the crumbs finely. Take two ounces of starch and two ounces sugar, together with the butter, and soak the dough in eggs and sugar, without any liquid. Then make it as fine as a sheet of paper, and sprinkle the dough with egg yolks. Then wrap the bar of butter in it, and fry in the frying pan with other beef [sic]. After put in dishes and serve.
But did this forgotten method make its way to later American kitchens?

A number of sources cite an article written by Baron Leon Brisse in 1866 and published in the French newspaper Liberté. The article itself is not available on-line, but luckily the Baron included it in his own publication:
During the stay of the Chinese Mission in Paris, the master-cooks of the Celestial Empire have exchanged civilities and teaching with the chefs of the Grand Hotel. The French entremétier*
is very happy about this. He has learnt from his Chinese colleague how to bake vanilla and ginger ices in the oven.
Here is how one proceeds to this delicious preparation. 
One ices hard,one wraps each ice in a crust of very light pastry and puts in the oven. The pastry bakes before the ice melts, the wrapping preventing the heat from reaching it. This phenomenon is explained by inconductibility of certain substances. Gourmets can thus obtrain the double pleasure of biting into a burning crust and cooling the palate at the fragrant contact of the ices.
*This is sometimes mistranslated as "desserts chef", but in a classic brigade de cuisine, the entremétier is actually responsible for openers like soups, etc.
This confirms then the introduction of the method to Western cooks by Chinese. Strangely, however, the same tale is referenced decades later (see below).

Even if this is how French cooks learned to make the Chinese dish, how did it then make its way across the Atlantic?

In 1876, Mary Foote Henderson published this recipe:
German Steamer Baked Ice-cream.
This dish was at least a curiosity, served at the table of one of the German steamers. A flat, round sponge-cake served as a base. A circular mold of very hard frozen ice-cream was placed on this, and then covered with a meringue, or whipped white of egg, sweetened and flavored. The surface was quickly colored with a red-hot salamander, which gave the dish the appearance of being baked.
The gentleman who told me about this dish insisted that it was put into the oven and quickly colored as the egg surrounding the cream was a sufficiently good non-conductor of heat to protect the ice for one or two minutes. However, there is less risk with a salamander.
As it happens, she also mentions ice cream from Delmonico:
Delmonico Vanilla Cream.
Ingredients: One and a half pints of cream, one ounce of isinglass, one pound of sugar, yolks of eight eggs, half a pint of milk, vanilla powder.
Scald the cream only; then add the isinglass dissolved in the milk, and pour it on the sugar and eggs beaten together to a froth; add the flavoring. Strain, cool, and freeze it; then pack it for three hours and a half at least.
But nine years after the Baked Alaska's supposed invention, she makes no connection between Delmonico's and “baked ice cream”.

It is certainly possible that the cook on the steamer had learned the cold in hot method from a French cook; he may even have been French himself (as many elite cooks were).

The first mention of a similar dish at Delmonico's comes from 1883, though retrospectively:
I remember that at Delmonico's restaurant, at the corner of Fifth Avenue and Twenty-sixth Street, New York, they served us, on New-Year's Day 1880, with a baked ice, appropriately styled an 'Alaska,' The core of this 'torridofrigid' preparation was a very firm vanille ice. Round it was a souffléor a whipped cream, I forget which. Then the preparation was lightly baked, or else browned with a salamander. It was strangely good. The soufflé was quite hot and the ice was quite cold; and we were not, immediately afterwards, taken to, the Bellevue Hospital to be treated for indigestion.
Curiously, this statement itself is a comment on the story of the Chinese cook in Paris, but treats it as a recent news item:
A paragraph going the round of the papers to the effect that the following recipe for baked ices has been acclimatized at Paris by the chef of the Chinese Ambassadors:—' Make your ice very firm; roll out some light paste thin, and cut it into small squares; place a spoonful of ice in the centre of each piece of paste, and fold it up carefully so that no air may get in, and bake. The paste will be cooked before the ice can melt.'
In 1894, Charles Ranhofer – who supposedly originated the Baked Alaska – published the Epicurean in which he provides this recipe for:
(3538). ALASKA, FLORIDA (Alaska, Florida).
Prepare a very fine vanilla-flavored Savoy biscuit paste (No. 3231). Butter some plain molds two and three-quarters inches in diameter by one and a half inches in depth; dip them in fecula or flour, and fill two-thirds full with the paste. Cook, turn them out and make an incisional around the bottom; hollow out the cakes, and mask the empty space with apricot marmalade (No. 367.5). Have some ice cream molds shaped as shown in Fig. 667, fill them half with uncooked banana ice cream (No. 3541), and half with uncooked vanilla ice cream (No. 3466); freeze, un-mold and lay them in the hollow of the prepared biscuits; keep in a freezing box or cave. Prepare also a meringue with twelve egg-whites and one pound of sugar. A few moments before serving place each biscuit with its ice on a small lace paper, and cover one after the other with the meringue pushed through a pocket furnished with a channeled socket, beginning at the bottom and diminishing the thickness until the top is reached; color this meringue for two minutes in a hot oven, and when a light golden brown remove and serve at once.
If Ranhofer had indeed named the dish to celebrate Alaska (only), he would hardly have used this name, much less years after the supposed event. Rather, it seems clear here that the two states represent, like the dish itself, extremes of hot and cold. (Nor is his recipe for a modern Baked Alaska; it is more like an ice-cream filled biscuit with a meringue topping.)

Further, Ranhofer includes a number of Delmonico's menus in the same work, including some from 1867. None of the latter mention either the Baked Alaska nor the Alaska Florida. (He does mention the latter in some undated menus at the start.)

In his account above, Sala calls the dish an “Alaska”, not a “Baked Alaska”. It is very likely that he was simply contracting Ranhofer's title for the dish. In another work from the same year, he describes the dish as “a baked ice”. It was a short step then to call the dish a “baked Alaska”. But there is no evidence that, at that point, this was its official name.

Fanny Farmer's 1896 version of her cookbook includes this recipe, which may be the first mention in print of the modern name:
Whites of six eggs 2 qt. brick of ice cream
6 tablespoons powdered sugar Thin sheet of sponge cake
Make a meringue of eggs and sugar; cover a board with white paper, lay on sponge cake, turn ice cream on the cake (which should extend one inch beyond the cream) cover with meringue and spread smoothly; place on oven grate and brown quickly in hot oven. The board, paper, cake and meringue are poor conductors of heat and prevent cream from melting. Slip from paper on ice cream platter and serve.
In 1899, a reader wrote “Mrs. Lincoln” at the Everyday Housekeeping magazine, asking: “Can you give in the next number of your magazine, a recipe for the new ice cream called 'Baked Alaska.' It has a hot meringue on the outside, yet the inside is ice cream perfectly cold and hard." Mrs. Lincoln cited Mrs. Henderson's “Baked Ice Cream” in her response and provided her own recipe for “Ice Cream en Déguiser” (approximately, if incorrectly, “ice cream in disguise”):
Make two quarts of ice cream, and when frozen remove the beater and pack it well in the freezer can. Let it stand till hard. Just before serving make a meringue by beating the whites of six eggs till stiff, then beating in, gradually, six rounding tablespoonfuls of sifted powdered sugar. Put a thin, round sheet of sponge cake on a plate suitable for serving, and turn out the mold of cream on the cake. Pile the meringue thickly round the edge and top of the cream, but do not smooth it. Place the dish on a wooden box cover and brown the meringue quickly in a hot oven. Serve at once. The plate should be larger than the cake and the cake larger than the bottom of the can. The cream will not melt, for the wood and the meringue serve as non-conductors of the heat. This is recommended chiefly for its novelty.
Any recipe for ice cream may be used and it may be molded in a brick mold if preferred, in which case a board a little larger than the mold may be covered with white paper, then with the cake and cream. After browning it in the oven, slip it off from the paper on to a platter or ice cream dish.
The fact that the query refers to “the new ice cream” suggests that, even after Farmer's recent mention, this was still coming into public awareness.

And so a dessert which had been at the least hinted at at Jefferson's table, then revived, very likely via a Chinese dessert introduced into Europe, finally found the name we give it today. But that is not all there is to the story.

Count Rumford and the Omelette Surprise
Another frequent claim about the idea of a hot food around a cold one is that Benjamin Thompson, Count Rumford, happened on the idea while experimenting with materials for insulation. By this account, in 1804 he discovered that egg whites did not conduct heat and then proceeded to invent the Omelette Surprise, a precursor to the Baked Alaska.

Two simple facts argue against this version. One is that Thompson himself does not mention it in any of his works available on-line, nor does he say anything about egg whites and insulation. Here he is on ice being protected from heat by... steam:
Experiment shewed that steam is in fact a non-conductor of Heat; for, notwithstanding the cold body used in this Experiment was very large and very cold, being a solid lump of ice nearly as large as an hen's egg, placed in the middle of the hollow cavity under the bottle, upon a small tripod or stand made of iron wire; yet as soon as the clouds which were formed in consequence of the unavoidable introduction of cold air in lifting up the bottle to introduce the ice, were dissipated, which soon happened, the steam became so perfectly transparent and invisible, that not the smallest appearance of cloudiness was to be seen any where, not even about the ice, which, as it went on to melt, appeared as clear and as transparent as a piece of the finest rock crystal.
Yet in the same passage he says nothing about egg whites.

UPDATE 5/22/2015: Note that you will sometimes see this supposed "quote" from Thompson on various Web sites: "Omelette surprise was the by-product of investigations in 1804 into the resistance of stiffly beaten egg whites to the induction of heat." You will not however find it (nor the term "omelette surprise") in Thompson's own work.

The other argument against the tale is simply the fact that the term Omelette Surprise cannot be found before the twentieth century, when, in 1903, Escoffier provided several recipes under that name, one with the alternate name of norvégienne (“Norwegian'). The latter is the more common term today for the French version of a Baked Alaska.
Omelette en Surprise, or Norvégienne
Put on the omelet platter an oval genoise base with a length proportional to that of the omelet and two centimeters thick. On this base, erect ice cream in the required flavor in a pyramid, vanilla, lemon, coffee, etc, or even several alternate ice creams, and cover the preparation with omelet soufflé. Smooth out the tower, decorate it with a decorating bag and put it in a very hot oven, so that the baking and the coloring occur quickly, and without the heat reaching the ice cream inside.
He also offers these versions: Creole, Jamaican, Elizabeth, Moka, and Icelandic. The last is flambé:
A pad of genoise placed on the round platter with, in the middle, another rolled up pad, stuck with cooked apricot to the round pad. The latter is hollowed at the top, like a bucket, and this cavity is lined with a preparation Condé-style [poached in syrup], which is dried in the stove beforehand. Surround the central pad with ice cream; cover the mixture with omelet soufflé, making it slightly spill over the hollowed pad. Smooth out, decorate, and bake like the Omelette surprise. Just as you serve it, pour a glass of hot rum into the hollowed cavity, and light it.
At the start of World War I, Paul Wentz gave this earthier account of the same dish: "I know what an omelette surprise is – I ate one once, I don't recall where. It is an omelet with rum which is brought to your table, flaming its bluest, and inside which is found pistachio ice cream."

He then went on to compare the French in wartime – excitable on the outside, but stoic on the inside – to this dessert.

Glace/é au Four

In 1868, the Baron de Brisse published his own cookbook (in the form of a calendar) and included the Chinese ice recipe he had reported earlier. He called this Glace au four ("ice in the oven").

In 1927, Heller's Guide for Ice-cream Makers included this note: "Glacé au Four (Fr. four meaning "oven") - Small pieces of Ice-Cream folded in paste and baked." This term (intentionally or not) means "iced or glazed in the oven". This is very reminiscent of both Jefferson's dessert and the Chinese preparation. The fact that it was included in a trade guide suggests that it must have been relatively common, but no other source of the period mentions it.

With its variants, this term can be considered another one for an Alaska-like dish. But its use is very are.

The above gives an overview of what is known about the Baked Alaska and its kin. Some common claims seem to be plainly wrong: that Ranhofer invented the dish to celebrate the purchase of Alaska; that Count Rumford had anything to do with a similar dish, much less the Omelette Surprise. Neither the latter term nor today's Omelette Norvégienne are found until the Twentieth Century, making mentions of either before then unlikely.
Despite the similar method used by Jefferson's cook, it is almost certainly the Chinese method that made its way via the elite chefs of Europe to America, very probably via a German steamer. But it is always possible too that Ranhofer, for instance, read of the Chinese “roasted ice” and tried to imitate the result. Whatever the case, he never himself seems to have called it a "Baked Alaska", a term which at any rate had to contend with several variants over the decades.


“Recollections of Peking”, The Monthly Chronicle, vol 3 1842

Taillevent, Guillaume Tirel, dit, How to Cook a Peacock: The Viandier, tr. Jim Chevallier 2008

“Mien Yaun”, The Atlantic Monthly, Volume 3, June 1859

Heine, Heinrich, Charles Godfrey Leland, The Salon, 1893 

"Dr Mitchill's letters from Washington: 1801-1813", Harper's New Monthly Magazine, Volume 58, 1879

de Brisse, Baron, Le Baron Brisse June 23, 1867
Brisse, Léon, Les 366 menus du baron Brisse 1868

Henderson, Mary Foote, Practical Cooking and Dinner Giving, 1876

Ranhofer, Charles, The epicurean, 1894

Sala, George Augustus, Living London: Being "Echoes"Re-echoed, 1883
Sala, George Augustus, America Revisited: From the Bay of New York to the Gulf of Mexico, and from … Vol 1, 1883

Farmer, Fannie Merrit, Original 1896 Boston Cooking-School Cook Book

“From Day to Day”, Barrows, Anna, Estelle M. H. q (Estelle Minerva Hatch) Merrill, Mary Johnson Lincoln, Eunice B. Littlefield, Winfield S. Nevins, Everyday Housekeeping: A Magazine for Practical Housekeepers ..., Volumes 11-12 1899

Thompson, Benjamin von Rumford, Essays, Political, Economical and Philosophical, Volume 2, 1798

Escoffier, Auguste, Le guide culinaire, 1903

Wenz, Paul, "Le Cocher de Reims", Revue France, May 25, 1918

Heller's Guide for Ice-cream Makers,1927