Molecular Gastronomy: Exploring the Science of Flavor

THE TITLE I GAVE THE ORIGINAL edition of this book was Casseroles et éprouvettes: saucepans and test tubes. Not the sorts of thing one normally expects to find together, either in the kitchen or the laboratory—or so it seemed before the creation of a new scientific discipline called molecular gastronomy. I should perhaps say a word or two about the origin of the name.

In 1988 the Oxford physicist Nicholas Kurti and I were preparing the first of a series of international workshops on the physical and chemical aspects of cooking, and we realized we needed a pithy phrase that would describe this new field of research. Brillat-Savarin’s classic definition of gastronomy in the Physiology of Taste (1825) naturally came to mind:

In this view, the humble hard-boiled egg belongs every bit as much to gastronomy as a wonderfully complicated dish such as the one Brillat-Savarin invented in honor of his mother, Oreiller de la Belle Aurore, a kind of pillow made of puff pastry and stuffed with seven kinds of wild game as well as foie gras and truffles. In both cases “intelligent knowledge”—that is, rational or analytical understanding—is needed, but it must be admitted that understanding the scientific principles of boiling an egg will be more useful to many more people. If all you have to eat is an egg, you had better know how to cook it properly.

So we had one part of the name we were looking for. But what kind of gastronomy? The term “molecular” was very fashionable at the time (molecular biology, molecular embryology, and so on), but it was also indispensable if we were to limit the scope of our enterprise. I proposed that we call our field simply molecular gastronomy, but Nicholas thought that “molecular” would too narrowly identify it with chemistry and suggested “molecular and physical gastronomy” instead. We started out calling it by this name, but after a while it seemed too cumbersome. Because the analysis of the structure and behavior of molecules obviously involves a certain amount of physics, after Nicholas’s death in 1998 I decided to revert to the shorter form in announcing our workshops, which now meet every two or three years in Sicily. And so molecular gastronomy it has been ever since.

But why not “molecular cooking”? Because cooking is a craft, an art—not a science. Nor is molecular gastronomy the same thing as the technology of cooking, because science is not technology. Furthermore, gastronomy seeks to answer a wider range of questions. For example, why does a tannic wine have a disagreeable taste if one drinks it in the company of salad that has been tossed with an acidic dressing? This question has nothing to do with cooking and everything to do with gastronomy.

What exactly does molecular gastronomy deal with? And how does it differ from the well-established field of food science? Some historical perspective will be useful in answering these questions, but generally speaking it is correct to say that food science deals with the composition and structure of food, and molecular gastronomy deals with culinary transformations and the sensory phenomena associated with eating.

Let’s begin by going back to ancient Egypt. When the anonymous author of the London Papyrus used a scale to determine whether fermented meat was lighter than fresh meat, was he doing an early form of molecular gastronomy or of food science? It depends on what the motivation for the experiment was. If he wanted to understand an effect of cooking, it was molecular gastronomy. If he was interested mainly in the properties of meat, then it was food science.

The succeeding centuries witnessed the development of chemistry. For a long time it resembled cooking and used many of the same techniques: cutting, grinding, heating, macerating, and so on. In the late fifteenth century Bartholemeo Sacchi, author (under the pen name of Platina) of a cookbook titled De honesta voluptate et valetudine (1475), made little if any distinction between chemistry, medicine, and cooking. More than 250 years later one finds much the same state of affairs in La suite des dons de Comus (1742) by François Marin. Note, however, that in the interval the French physician and inventor Denis Papin (1647–1712) had built a pressure cooker in order to recover the substance of bones in broth, hence the name of his machine: the digestor.

Marin’s views echoed those of Sacchi. “The science of cooking,” he wrote, “involves decomposing, digesting, and extracting the quintessence of meats, drawing from them their light and nutritive juices. Indeed this kind of chemical analysis is the main object of our art.” Chemical analysis! We need first of all to make a clear distinction between art, technology, and science. To ride a bicycle, for example, one has to push the pedals forward; this is a matter of technique, or art. If someone were to inquire into the difference between pedaling with the front of the foot instead of the heel, this would be a question of technology (which, as the Greek word indicates, is the systematic treatment of techne—art, craft, or skill). And if someone were to survey the surrounding landscape while pedaling a bicycle—say, in order to avoid having to climb a mountain—this would be an example of scientific investigation (or, more generally, the attempt to discover the mechanisms of natural phenomena).

Plainly, then, cooking is not the same thing as molecular gastronomy, for craft aims at the production of goods, not of knowledge. For the same reason molecular gastronomy is no substitute for cooking because it seeks to produce something entirely different. Marin was wrong, then, in saying that cooking is a form of chemical analysis. Similarly, Papin’s digestor was an achievement of technology rather than of science.

Let’s return to the eighteenth century. In 1773 French chemist Antoine Baumé (1728–1804) devised a “recipe” for preparing “dry stocks for times of war, or stock tablets.” In order to improve the extraction of organic compounds he recommended boiling meats again, after the first extraction, then clarifying the stock with egg whites and allowing the liquid to evaporate in a bain-marie until only “perfectly dry and brittle” tablets remained. The whole problem of making stocks and meat extracts was a very important one at a time when neither refrigerators nor freezers existed to preserve food. But there were also financial considerations. It is often forgotten that the great Lavoisier himself took an interest in the confection of stocks. As fermier général, responsible for collecting taxes and supplying the hospitals of Paris with food, he understood that it was not the water in a broth that provided nourishment but the matter that had been extracted from the meat and that had undergone chemical reaction in the course of cooking. He therefore devised a densitometer to determine how much meat was necessary to feed the indigent patients of the hospitals.

Three years later, Benjamin Thompson (1753–1814), later Count Rumford (later still he married Lavoisier’s widow), published a 400-page book titled On the Construction of Kitchen Fireplaces and Kitchen Utensils Together with Remarks and Observations Relating to the Various Processes of Cookery and Proposals for Improving That Most Useful Art (1776). Rumford did a great deal of work related to food and perhaps should be considered one of the most important figures in the prehistory of molecular gastronomy, as he was interested not only in technology but also in science. It is even said that he discovered fluid convection while eating a thick soup whose viscosity prevented the inner layer from cooling, thus causing him to burn his mouth.

Also during this period Antoine-Augustin Parmentier (1737–1813), a pharmacist who had an interest in food, sought to win acceptance for the potato in France and made a study of the flours used to make bread. Parmentier’s fame came to be widespread in his native land, which probably explains why A. Viard wrote in Le cuisinier impérial, ou l’art de faire la cuisine et la pâtisserie pour toutes les fortunes (1806), “All the arts and sciences have made enormous advances in the last one hundred years, especially chemistry, which has progressed so much that a student familiar with [recent] discoveries about flavor can demonstrate theories whose existence was not previously suspected. It is natural, in these conditions, that cooking, which is a kind of chemistry, advanced at the same pace.” Again one sees the confusion between chemistry, a science, and cooking, an art or craft. The definition of artists as inspired craftsmen (proposed by Walter Gropius, founder of the Bauhaus school of design in Weimar Germany) is worth keeping in mind.

Another German, Justus von Liebig (1803–1873), who devoted much of his later scientific career to the analysis of food, made a fortune from an eponymous American company that produced meat extracts from surplus supplies of meat. The chemical theory underlying the product turned out to be wrong, but Liebig’s extract became famous throughout the world.

A more important result was obtained shortly afterward by the French chemist Michel-Eugène Chevreul (1786–1889), who analyzed fats and discovered their chemical structure. Also during this period, in Germany, Emil Fischer (1852–1919) was studying sugars, again with significant consequences for the development of chemistry. Already in 1821 Friedrich Christian Accum (1769–1838) had brought out in London a very interesting book called Culinary Chemistry: Exhibiting the Scientific Principles of Cookery, with Concise Instructions for Preparing Good and Wholesome Pickles, Vinegar, Conserves, Fruit Jellies, Marmalades, and Various Other Alimentary Substances Employed in Domestic Economy, with Observations on the Chemical Constitution and Nutritive Qualities of Different Kinds of Food. Here the question arises whether a distinction must be made between chemistry as a science and chemistry as an application of science, or technology. My own view is that the terms chemistry and science should be reserved for the scientific exploration of chemical phenomena.

We now come to the strange case of Louis-Camille Maillard (1878–1936). On completing studies in medicine and chemistry at the University of Nancy, Maillard wrote his doctoral thesis in the latter field on the reaction of glycerol and sugars with amino acids. These chemical processes, first explained in a publication of 1912, are very important because they impart flavor to grilled meats, bread crust, roasted chocolate and coffee, and many other things. After World War I Maillard took up an appointment as professor of biochemistry and toxicology at the University of Algiers, where he taught until his sudden death in Paris almost twenty years later. Curiously, Maillard was renowned for his work throughout the world but not in France until a few years ago.

No brief survey of the prehistory of molecular gastronomy would be complete without mentioning Édouard de Pomiane (1875–1964), a biologist at the Institut Pasteur in Paris who was well known in the first half of the twentieth century for a series of popular works on what he called gastrotechnie, or gastrotechnology, an attempt to rationalize cooking similar to ones also being made in the United States and in some European countries. But these works were full of elementary mistakes, based on insufficient experimental evidence. For example, it was believed at the time that the bowl and whisk used to whip egg whites had to be made of copper and galvanized iron, respectively, to promote the formation of foam. But all this had to do with technology, not science.

Food science thus initially developed in close contact with cooking. But it soon gave way to an interest in feeding people and making more efficient use of ingredients. It is often forgotten that until recently the chief concern of people in most countries, even in the West, was having enough to eat. Gradually scientific research came to concentrate more on foods themselves than on their domestic preparation.

But what about the millions of people who cook every day in advanced industrial countries? We now have access to products that have benefited from advances in food science, but do we know how to cook them? This question has two parts. On one hand, how good are the products we use? On the other, how competent are we as cooks?

First, the question of quality. Like so many others in France today who long for the countryside they left in order to live and work in cities, I am not immune to nostalgia for the good old days. I, too, miss the chickens running freely about the courtyard; the asparagus picked just before the meal, with its delicate milky juice running out from the stalk; the peas shelled just before they are cooked; the strawberries served still warm from the sun—all this is the stuff of literature. But the countryside is also the mud that comes when it rains, the wild rabbits that visit at night to undo the gardener’s work during the day, the mice that gnaw at the food he has stored away, the aching back that rewards him for his toil.

By all means, then, let us fill our souls with such nostalgia, for they have need of it. But let us also compare. The same Alsatian wine that thirty years ago produced migraine headaches and kept for only four years has now become a nectar that no longer degrades so rapidly. Mediocre homemade yogurt has been supplanted by commercial brands in various flavors that have a perfectly regular texture. Should we reproach them for having a strawberry flavor too unlike the flavor of strawberries from the orchard? Or should we reproach ourselves instead for wanting to eat strawberries in winter? The same goes for insipid year-round tomatoes: Wait for summer!

Enough of this facile apology for progress. We would do better to accept products for what they are and recognize that the possibility of improving them lies first and foremost in submitting them to the transformations of the culinary art. If we want yogurt to be flavored, we should be prepared do it ourselves. In other words, let’s go into the kitchen and start cooking.

This is why I raised the second question, concerning our culinary skills. To answer this we need to ask ourselves how we cook, and we will have to admit that by and large we repeat what we have seen done at home, by our parents or grandparents. When we try out a new dish, one that does not belong to the family culinary repertoire, we have the same feeling Christopher Columbus had setting out to discover the New World. Why do most people find cooking so difficult? Because for most people it is a matter of repetition and habit. In Meditation 7, section 48 of the Physiology of Taste (best known in English in the translation by M. F. K. Fisher, from whom I quote once again), Brillat-Savarin gave a more detailed answer:

“Maître la Planche,” said the Professor, in a tone grave enough to pierce the hardest heart, “everyone who has sat at my table proclaims you as a soup-cook of the highest order, which is indeed a fine thing, for soup is of primary concern to any hungry stomach; but I observe with chagrin that so far you are but an uncertain fryer.

“Yesterday I heard you moan over that magnificent sole, when you served it to us pale, flabby, and bleached. My friend R. … threw a disapproving look at you; Monsieur H. R. … averted his gnomonic nose, and President S. … deplored the accident as if it were a public calamity.

“This misfortune happened because you have neglected the theory of frying, whose importance you do not recognize. You are somewhat opinionated, and I have had a little trouble in making you understand that the phenomena which occur in your laboratory are nothing more than the execution of the eternal laws of nature, and that certain things which you do inattentively, and only because you have seen others do them, are nonetheless based on the highest and most abstruse scientific principles.

“For instance, you could dip your finger with impunity into boiling spirits-of-wine, but you would pull it out as fast as you could from boiling brandy, faster yet if it was water, and a rapid immersion in boiling oil would give you a cruel injury, for oil can become at least three times as hot as plain water.

“It is because of this fact that hot liquids react in differing ways upon the edible bodies which are plunged into them. Food which is treated in water becomes softer, and then dissolves and is reduced to a bouilli; from it comes soup-stock or various essences: whereas food which is treated in oil grows more solid, takes on a more or less deep color, and ends by burning.

“In the first case, the water dissolves and pulls out the inner juices of the food which is plunged into it; in the second, these juices are saved, since the oil cannot dissolve them; and if the food becomes dry, it is only because the continuation of the heat ends in vaporizing their moistness.

“These two methods also have different names, and frying is the one for boiling in oil or grease something which is meant to be eaten. I believe that I have already explained that, in the culinary definition, oil and grease are almost synonymous, grease being nothing more than solid oil, while oil is liquid grease.”

“By means of this surprise, a kind of glove is formed, which contains the body of food, keeps the grease from penetrating, and concentrates the inner juices, which themselves undergo an interior cooking which gives to the food all the flavor it is capable of producing.

“In order to assure that the surprise will occur, the burning liquid must be hot enough to make its action rapid and instantaneous; but it cannot arrive at this point until it has been exposed for a considerable time to a high and lively fire.

“The following method will always tell you when the fat is at a proper heat: Cut a finger of bread, and dip it into the pot for five or six seconds; if it comes out crisp and browned do your frying immediately, and if not you must add to the fire and make the test again.

“Once the surprise has occurred, moderate the fire, so that the cooking will not be too rapid and the juices which you have imprisoned will undergo, by means of a prolonged heating, the changes which unite them and thus heighten the flavor.

“You have doubtless noticed that the surface of well-fried foods will not melt either salt or sugar, which they still call for according to their different natures. Therefore you must not neglect to reduce these two substances to the finest powder, so that they will be as easy as possible to make adhere to the food, and so that by means of a shaker you can properly season what you have prepared.

“However, do not forget, when you are confronted with one of those trout weighing barely a quarter-pound, the kind which come from murmuring brooks far from our capital, do not forget, I say, to fry it in your very finest olive oil: this simple dish, properly sprinkled with salt and decorated with slices of lemon, is worthy to be served to a Personage!

“My two prescriptions are founded, again, on the nature of things. Experience has taught us that olive oil must be used only for operations which take very little time or which do not demand great heat, because prolonged boiling of it develops a choking and disagreeable taste which comes from certain particles of olive tissue which it is very difficult to get rid of, and which are easily burned.

The dissertation of the Professor (as Brillat-Savarin styled himself) is full of errors from the scientific point of view. First of all, there is a difference between boiling temperature and heat capacity. Boiling temperature is the temperature at which a liquid boils. Oil, which begins to decompose before coming to a boil, has a higher boiling point than that of water (100°C [212°F]), which in turn is higher than that of ethanol, the alcohol found in liquor (78°C [172°F]). Heat capacity is something quite different: the quantity of energy needed to increase the temperature of a compound by 1°C (1.8°F).

Furthermore, it is not true that putting something in boiling water always leads to softening, dissolving, and the formation of a sort of purée or mush. When you put egg white in boiling water, for example, it hardens. And although it is true that collagen—the tissue that holds muscle fibers in meat together—dissolves slowly in boiling water, protein coagulation inside these cells produces a tough material.

Nor is it true that the crust formed during frying prevents oil from penetrating meat and helps to concentrate its juices. The crust is full of cracks, through which vapor bubbles escape during the course of frying (they can be seen with the aid of a microscope or in some cases with the naked eye) and through which oil enters. Furthermore, the notion that the juices of a piece of meat go to the center during cooking rests on a misunderstanding. Because these juices are mostly water and therefore not compressible, the effect of cooking a piece of meat (even by boiling in water) is to force them outward, away from the center. What Brillat-Savarin thought was concentration is actually expansion.

Just the same, there is a certain amount of truth in what the Professor says, some of it useful. This is why it is so important to distinguish what is right from what is wrong—to rationalize the old “chemical art” of cooking. Above all we want to retain the idea that “certain things which you do inattentively, and only because you have seen others do them, are nonetheless based on the highest and most abstruse scientific principles.” In other words, culinary phenomena—the phenomena that generate transformations in food—are at bottom nothing more than chemistry and physics. To cook well, at least from the technical point of view (art, as I say, is another story), we have to know both these sciences.

This is precisely why Nicholas Kurti and I sought to promote the notion of molecular gastronomy: Chemistry and physics, judiciously applied, can tell us how to preserve the tenderness of meats, how to master the chemical reactions that give the crust of roasted meat its wonderful flavor, and how to avoid the failures that are commonly encountered in making a variety of sauces, from mayonnaise to hollandaise, béarnaise, ravigote, and many others. Do we dare make the leap? We hesitate because as human beings, which is to say primates, we have a fear of new foods or of foods that are unfamiliar to us. Jonathan Swift famously said, “He was a bold man that first eat an oyster.” No matter that each new recipe may be likened to the discovery of a new continent, science is there not only to guide us but also to help us exercise our innate capacities for discovery and invention. The application of eternal laws of nature both informs and stimulates culinary innovation.

Is it enough to read cookbooks? Certainly not. They are generally little more than collections of recipes, which is to say protocols that relegate cooks to the status of mere executors. Moreover, they contain a great many doubtful instructions: Steak ought to be seared, because this causes an impermeable crust to form that retains its juices (a sound practice, as it happens, although the reasoning is wrong); in making stocks, meat should be placed in cold water to begin because this will cause “albumen” to coagulate and so prevent the loss of juices; mayonnaise will break if women try to make it when they are menstruating; egg whites will not stiffen if one changes the direction in which they are whisked; and so on. Clearly a bit of science must be brought to bear if we are profitably to explore our culinary heritage, as I propose to do in the pages that follow.

And a bit of poetry as well? Its absence will be regretted only by those who prefer the catastrophes that have routinely been courted by respecting the traditional method for making soufflés rise, for example. It is a mistake to suppose that in understanding physical phenomena we lose the ability to take pleasure in culinary art. Besides, one finds poetry where one may: Are the names “ionone” (for a molecule with a delicate violet odor in dilute alcohol solution) and “hexanal” (for a molecule that imparts a fresh herb flavor to virgin olive oil) any less beautiful than “cauldron” or “knife”?

Poetry aside, let us consider efficiency. Time-honored maxims, proverbs, old wives’ tales, folk beliefs, and culinary rules are millstones round our necks that weigh us down when they are false and wings that carry us aloft when they are true. Hence the importance of molecular gastronomy, whose primary objective is first to make an inventory of such rules and then to select those that have withstood careful analysis. Culinary art has everything to gain by separating the wheat from the chaff of empirical observations.

In the first part of this book, then, we will consider the rules that have long guided the preparation of a variety of familiar dishes: stock, hard-boiled eggs, quiches, quenelles, gnocchi, cheese fondue, roast beef, preserves—almost twenty dishes in all.

But anyone who cooks rationally, relying solely on the laws of physics and chemistry, will soon run up against the limits of these two sciences in the kitchen. Take meringues, for example. You want them to rise? Then place them in a glass vacuum bell jar and pump the air out; the air bubbles dilate and the meringues swell and swell, to the point that you with left with “wind crystals.” Nothing to chew on there—a culinary disaster.

This leads us to ask ourselves what we like to eat and why. Further questions immediately arise: Why do we stop eating? How many tastes do we perceive? Is flavor modified by changes in temperature?

Modern physiologists of flavor have studied these questions, carrying out experiments suggested to them by their expertise in a particular field of research. They have thrown valuable light, for example, on mastication, the almost unconscious act that, to the civilized mind, separates gluttons from gastronomes. Brillat-Savarin thought this distinction so important that he devoted the first part of the introduction of his book to it. Immediately after the celebrated aphorisms of the preamble, he reports (or possibly only imagines) a “Dialogue Between the Author and His Friend”:

FRIEND: Good Lord, bachelors are as much victims of the rule as the rest of us, and sometimes to our great disadvantage! But in this case even celibacy can’t save you: my wife is convinced that she has the right to dictate to you about the book, since it was at her country house that you wrote the first pages of it.

AUTHOR: You know, my dear Doctor, my deference for the ladies. More than once you’ve complimented me on my submission to their orders. You were even among those who once said that I would make an excellent husband! Nonetheless, I refuse to publish my book.

FRIEND: Believe me, heirs will have plenty to cope, what with the church, the law courts, the doctors themselves! Even if they do not lack willingness, they’ll have little time for the various worries that precede, accompany, and follow the publication of a book, no matter how long or short it may be.

FRIEND: The single word gastronomy makes everyone prick up his ears. The subject is always fashionable. And mockers like to eat, as well as the rest. And there’s something else: can you ignore the fact that the most solemn personages have occasionally produced light works? There is President Montesquieu, for instance!

AUTHOR: By Jove, that’s so! He wrote THE TEMPLE OF GNIDUS … and one might do well to remember that there is more real point in meditating on what is at once necessary, pleasant, and a daily occupation, than in what was said and done more than two thousand years ago by a couple of little brats in the woods of Greece, one chasing, the other pretending to flee …

AUTHOR: Me? I should say not! I simply showed myself as an author for a minute. And that reminds me of a high-comedy scene from an English play, which really amused me. I think it’s in a thing called THE NATURAL DAUGHTER. See what you think of it.

A stupid lout, finding himself the Quaker’s rival in love, and emboldened by this ascetic exterior and the nature it apparently hides, teases and taunts and ridicules him, so that the young hero grows increasingly furious and finally gives the fool a good beating.

FRIEND: It simply can’t be done. You admit that you have shown yourself as an author for a second or two. I’ve got you now, and I’m taking you to the publisher’s. I’ll even tell you that more than one friend has already guessed your secret.

AUTHOR: What I shan’t say is that our native land prides itself on having produced you; that at twenty-four you had already published a textbook which has since become a classic; that your deserved reputation inspires great confidence in you; that your general appearance reassures the sick; that your dexterity astounds them; that your sympathy comforts them. All this is common knowledge. But I shall reveal to the whole of Paris (here I draw myself up), to all of France (I swell with oratorical rage), to the Universe itself, your only fault!

The latest studies of the present generation of physiologists of flavor are reported in the second part of the book. These studies, which, unlike Brillat-Savarin’s literary investigations, constitute the true physiology of flavor, have a direct usefulness in the kitchen for those who are bold enough to apply them. Yet a more complete science remains to be built on this base. Recall the definition I gave earlier of molecular gastronomy. Only the dictums, proverbs, and rules that have been shown to be sound can be placed at the heart of the new science that is needed. This means that culinary practice must henceforth be based on a genuine physiology of flavor. In asking how are individual dishes to be prepared, however, we are dealing with something rather different: the modeling of culinary operations. And this modeling must be based on an understanding of the physical transformations to which foods are subjected in cooking.

This is the subject of the third part of the book. However, note that the “intelligent knowledge” the reader will find there must be judged with reference to the peculiar situation in which cooks find themselves. Sutor ne supra crepidam iudicaret: Let not the cobbler criticize [a work of art] above the shoe. But cooks have no choice but to judge above and beyond the pan, for they know that this work is not for the stomach but for the heart and the soul. And this is why seemingly useless explorations find their place here—for love of the beauty of pure knowledge.

Consider the egg yolk—the ordinary egg yolk, which generations of cooks have used without looking at it any longer than was necessary to prevent it from being broken or spilled. Yet it possesses a complex and unsuspected structure, which science has disclosed to us. No longer can the sight of an egg yolk be thought uninteresting. For boredom is born not of uniformity but of a certain offhandedness, a lack of deference. Thanks to science, which teaches us that even the yolk of an egg deserves to be the object of curiosity and admiration, we have no reason to be bored in the kitchen again.

Obviously molecular gastronomy does not aim solely at attaining pure knowledge of this sort, because it seeks also to give practical knowledge a sound basis by explaining why successful recipes work and why mistakes occur. Thus, for example, if you ask why lumps form when flour is placed in a hot liquid, you will at once be led to useful conclusions that allow certain culinary techniques to be rationalized and refined.

I propose a new article of faith: Whoever understands the reasons for the results he or she obtains in the kitchen can improve on them. This is why I devote so much attention at the outset to criticizing traditional recipes. What if one follows a recipe for making mayonnaise to the letter, but the sauce breaks? As a hostage to the recipe one will have no alternative but to throw out the offending egg, mustard, and oil. But the cook who understands that mayonnaise is an emulsion—a dispersion of oil droplets in water (from the yolk and vinegar)—will be able to save the sauce, not by adding another egg (as the leading authorities advise) but by decanting the oil and once again dispersing it in the watery ingredients.

The whole third part of the book therefore is concerned with improving recipes and preparations. Reasoned analysis, allied with the ideal of perfectibility, is what gives cooking its soul. The spirit of Brillat-Savarin lives on.

In exploring physical and chemical mechanisms of cooking we will find ample grounds for modifying classic recipes. Consider once again the soufflé. First we test the maxim that the whipped egg whites must be firm; then we analyze the way in which we perceive the light, airy texture of the dish. From this we draw conclusions about what a soufflé ideally ought to be: Rational analysis of the classic recipe shows that a soufflé should be heated from below. But now we find ourselves confronted with an awkward situation. Recent studies—experimental and theoretical alike, because the two obviously go together—have shown that the classic method of cooking soufflés in the oven is not indispensable. What are we to do? Do we abandon the oven, out of distrust of tradition, in order to produce a better soufflé? Or do we go on following the teachings of the old masters, forgetting that things such as mayonnaise and puff pastry, centuries ago, were themselves innovations?

The fourth part of this book frankly rejects conservatism and resistance to change in the name of another tradition: intelligent knowledge. It is in the name of this tradition that we undertake to devise new chocolate mousses, that we resolve to abandon the useless clarification of stocks, that we generalize from the traditional aioli (a garlic emulsion) to produce a new class of flavored mayonnaises, and that, despite opposition from defenders of tradition who fear the temptations of novelty, we dare to conduct chemistry experiments in our own kitchens.

The fact of the matter is that we do both chemistry and physics whenever we make an emulsified sauce or grill a piece of meat. Nonetheless, we are like Molière’s Monsieur Jourdain, not realizing that we have been doing chemistry and physics all along. What is more, satisfied with what we have achieved, we do not look for ways to achieve something better. In the fourth part of the book it is therefore the soul of cooking that I insist on. For in seeking to understand the reasons for what we do in the kitchen we seek not to poison ourselves but rather to enjoy flavors that until now we have only dreamed of. Let us go about our cooking, then, with full knowledge of what it actually involves.