The science of food… and of resetting your expectations

This is one of the more surreal weeks of my recent life. On Sunday, I took possession of an adorable and small apartment near Inman Square in Cambridge, fought my way through Ikea and spent the first night of my new itinerant academic existence in Cambridge. Monday, I moved into my office in the Media Lab, using a borrowed ID from a student as my ID card isn’t turned on yet. Tuesday, I met with my new masters students and other colleagues at the Center for Civic Media, then retreated to the Berkman Center to be part of their iLaw series. And then I found myself in a lecture hall in the Harvard Science Center, attending a lecture on cooking, given in part by David Arnold, one of the leading minds in haute cuisine… and a guy I used to hang out with more than twenty years ago. It feels like a very strange compression of history into a single (very long) weekend. But it was a great talk, so I thought I’d share it with you as well. (And that I’m not posting until two days later helps show how crazed the week has been…)

The lecture is a public talk associated with a Harvard class called “Science and Cooking – From Haute Cuisine to Soft Matter Science“, taught by David Weitz. It’s a science class focused on the chemical and physical changes associated with cooking. The text for the class is “On Food and Cooking” by Harold McGee, the opening speaker. McGee wrote the book in Cambridge and tells us “In late 1970s, I never dreamed Harvard would give a course on cooking – I can make a living now.”

McGee is accompanied by Arnold, who he introduces as the director of culinary technologies at French Culinary Institute in NYC and “the one guy in the world who knows the most about cutting edge tech in the modern kitchen.” Arnold insisted that the class needed a definition of cooking, and so we’re working with this definition: “The preparation of food for eating, especially by means of heat”. The term comes from the Latin coquere, “to cook, prepare food, ripen, digest”. Cooking is the application of energy and ingenuity to change foods so they’re easier, safer and more pleasurable to eat.

McGee quotes Arnold as observing that if a peach is perfectly right, the best thing you can do with it is put it on a plate with a knife. Nature, McGee argues, wants us to eat peaches so that we’ll carry seeds far and wide. What we do in cooking is, in part, trying to approach the complexity and the balance of the perfectly ripe piece of fruit.

The first stop on a history of cooking has to be fire. McGee references Richard Wrangham’s “Catching Fire: How Cooking Made us Human”. Cooking allows us to turn raw starch into something digestible. We needed these calories, Wrangham argued, to build our big brains. In that sense, learning to cook may literally have helped us become human. For us to tell if Wrangham is right, we need to see evidence of cooking much further back in history. We currently see evidence of 100,000 years ago, while Wrangham speculates we should see evidence 1 million years back.

By the Middle Ages, cooks had figured out how to make gelatins and clarify them, and how to do very complex decorative work for the courts. They’d also invented food as entertainment. We see a recipe from the 15th century titled “To Make a Chicken Sing when it is dead and roasted”. It involves stuffing a chicken with sulfur and mercury and sounds like a very bad idea… but it is amusing, and that notion of food as amusement is returning to modern kitchens today.

By 1681, we see the introduction of a very different way of cooking – the pressure cooker. In 1681, Denys Papin was a member of the Royal Society, working with Boyle on gases. He figured out that you could cook food using pressurized water and speed cooking processes. Because the Royal Society are mostly bachelors, there’s a wonderful set of literature of dinner parties where scientists brought ingredients and Papin cooked and served them.

Arnold jumps into explain that pressure cookers allow us to cook at temperatures other than what we could normally achieve. This leads to some fun discoveries. He read an influential book on pressure cooking that advised increasing use of onions in pressure cookers because the onion flavor dissipates. So he pressure cooked other similar foods, and discovered that foods like garlic lose their stink when pressure cooked. “The sulfur compounds in horseradish get totally knocked out so you can eat it by the bushel.” Mustard seeds cooked with vinegar puff up like caviar. And other effects can’t be replicated any other reasonable ways. “Pressure cookers speed up Maillard reactions – you can pressure cook an egg for an hour and get browning that you otherwise wouldn’t get without cooking for several days.”

McGee notes that Arnold hasn’t mentioned his durian experiments. Arnold sheepishly explains that this is a lesson in the importance of repetition. Durian smells bad (or wonderful, if you grew up in certain corners of Asia) because of sulfur compounds, and so you should be able to knock out the smell in a pressure cooker. “So I threw some stuff with durian into a pressure cooker and got the most incredible Durian caramel.” But he’s never been able to replicate it, with more than a month’s worth of attempts. “Don’t be a schmuck,” he tells us – document your work so you can replicate.

Replicability is, of course, the essence of experimental science. In 1770, McGee tells us, Ben Franklin was spending a huge amount of time on ships, traveling between the US and France. He noticed that when the cooks threw out the waste from cooking, the wake behind the ship calmed. He later tried an experiment in Clapham Pond in London, putting a teaspoon of oil onto a pond on a windy day. The water calmed over an area of half an acre. Had Franklin made a further leap, he could have pretty easily calculated the size of a molecule based on the experiment, assuming that the layer of oil eventually was a single molecule thick.

To get a sense for the molecular scale, Arnold gives us a demonstration of Dragon’s Beard candy, a preparation seen in China, Turkey and Iran. Cook sugar to a particular hardness and you can stretch and fold it at will. Arnold takes a centimeter-thick piece of sugar, turns it into a loop, and stretches it. Folding it once, it’s now two loops. He repeats until we have over 15,000 strands, each about a micron thick. It’s flavored with cocoa, but Arnold likes to serve it with vinegar and mustard powder, with peanuts wrapped inside.

McGee would like us to take Count Rumford as seriously as we tend to take Franklin. Rumford was a Colonial New Englander who was on the wrong side of the war, so he spent much of his career in England. Amongst his many discoveries, Rumford discovered that slow cooked meat is delicious, a discovery that’s come into fashion recently with sous vide cooking. Rumford accidentally discovered the technique by trying to cook a leg of mutton in his potato drier, and left it overnight. In the morning, he encountered an “amazing aroma”. And because was scientifically minded, he replicated the experiment and tried an objective taste test. At a cocktail party, he cooked one leg of mutton over a fire and another using the slow technique and put them at opposite sides of the room, and weighed the remnants – the slow-cooked mutton was far more popular.

The opposite of Rumford was Justus Liebig, a German chemist who was a theoretician, not an experimentalist. Working only from his own “brilliance”, not from experiments, Liebig introduced a new way of cooking meat – searing it to seal in the juices. It’s revolutionary, but also really bad. Apparently he never actually tasted it.

In 1969, the British scientist Nicholas Kurti suggested that we bring scientific methods back to ordinary, everyday phenomenon. “I think it is a sad reflection on our society that while we can and do measure the temperature in the atmosphere of Venus, we do not know what goes on inside our soufflés”. His investigations were part of a movement towards “soft matter science”, a study of phenomena like soap bubbles that led to a 1991 Nobel prize.

McGee found himself investigating these phenomena in 1984 when he wrote his book on the history of food. In collaboration with scientists, he began testing a Julia Child assertion about whipping egg whites in a copper bowl – Child advocated always whipping in copper. Experiments testing whipping in copper demonstrated that it took a much longer time, but led to lighter whites. The paper was eventually accepted by Nature, though one reviewer commented, “The science is good, but the subject is fluffy.”

While much of what’s emerged in science in the kitchen, like molecular gastronomy, is fairly recent, nouvelle cuisine is very old. In 1759, a poem was published that read:

Every year nouvelle cuisine
Because every year tastes change;
And every day there are new stews:
So be a chemist, Justine.

French cooking, historically, has been far from experimental. Classic French cooking as compiled by Escoffier and others codified cuisine to the point where it was difficult to innovate, since the classic textbook offers 100 “correct” recipes for beef tenderloin. McGee cites Michael Bras as helping invert these dynamics with the melting chocolate cake, an inversion of the “correct” idea that a cake is surrounded by a sauce – instead, a cake contains a ganache. A later dish, the Gargouillou, recreated a salad as a walk through a garden, whatever ingredients were most appropriate on the given day.

Chef Jacques Maximin was influenced by these experiments and observed, ” To be really creative means not copying.” His maxim struck a chord especially with Ferran Adria, who recreated the gargouillou as an endlessly surprising salad – nothing is quite what it seems. Adria went on to thoroughly revolutionize cuisine as we know it, with techniques like flavored foams and the spherification of ingredients like melon into texturally odd balls of flavor.

He’s had many followers. Joan and Jordi Roca use rotary evaporators to separate aromas from ingredients – this makes possible a dish of foods that are shades of white which have flavors usually associated with visually dark ingredients. Jose Andres experimented with a chemical most often used to make cough drops, offering a bonbon of liquid olive oil within a clear shell. Wylie Dufresne uses an enzyme called “meat glue” to offer a chicken nugget that’s white meat wrapped in dark, wrapped in skin. And now the field has been exhaustively documented by Nathan Myrvold, who’s published a massive, five-volume book on Modernist cuisine.

At this point, McGee gives the reins to Arnold, who offers a rapid-fire walk through some of his favorite techniques and his creative process. He shows us a Japanese ring that features a wavy woodgrain effect, produced by beating two different metals together. Arnold achieved something similar using fish as a way of persuading Hobart, the cooking machine company, to give him a really badass slicer. Using meat glue and casein, he glues salmon and fluke together and slices them into a thin sheet that looks a little like mortadella and a bit like wood grain. It’s served with creme fraische seasoned with nitrogen-frozen herbs, a fennel apple salad infused with curry and pressure cooked mushroom seeds, a veritable tour of modernist technique on a plate.

(The nitrogen chilled herbs allow fresh herbs to be broken into very small pieces, as you would break up a dried herb, but maintain the fresh flavor and texture. Arnold recommends you blanch your fresh herbs, flash freeze in liquid nitrogen, shatter into tiny pieces and pass through a chinoise, using only the tiny bits that escape the mesh.)

Using agar, a gelling agent made from seaweed, Arnold produces a concord grape jelly, a thick, stiff substance. He points out that it cuts cleanly and can’t be put back together. But if you break it violently – in a blender, say – you get a different effect: a microgel or fluid gel. It looks like a puree on the plate, but tastes like juice in the mouth.

Agar works well as a clarifier too, in lower concentrations. Arnold makes a loose gel of lime juice, then uses a whisk to separate it into “whey and curds”. He passes this through cheesecloth, making rude comments about “gently massaging the sack”, before producing a liquid that looks very much like water, but turns out to have intense lime flavor.

We clarify liquids, he tells us, because then we can infuse them into other foods. “We can make a cucumber better by adding liquor to it… we can make a lot of things better by adding liquor to them.” Injection techniques work better with clear liquids, and Arnold shows us how to infuse a cucumber with lime and sugar in a vacuum machine. The vacuum pulls air out of the cucumber, and rapidly threatens to boil it, as liquids boil at lower temperatures in vacuum. (Arnold recommends you heavily chill your ingredients as you vacuum infuse…) While the air is sucked out, the liquid is incompressible, and as air floods back into the chamber as he turns the vacuum off, liquid infuses into the cucumber in a flash, turning the vegetable into something that looks like stained glass. “It’s one way to get something that looks cooked, but still has crisp, clean lines to it.”

You can rapidly infuse using pressure as well. Arnold puts vodka and coffee into an ISI whipped cream maker, and uses nitrous oxide to force the coffee into the vodka. What results is heavily flavored, but not carbonated – the tingle of carbonation comes from carbon dioxide escaping from solution. Nitrous oxide offers pressure and fluff without carbonation.

Arnold offers his advice on carbonating some of his favorite things. As with infusion, clarified liquids work better. “If you’re going to carbonate liquor – which I highly recommend – you’re going to need more pressure than carbonating water because CO2 is more soluble in alcohol than in water.” You can force carbonate a wine at 30 psi, sake at 35psi, and liquors at about 40 psi.

Why would you infuse vodka with coffee? “The flavors you pull out of a product are dependent on time, temperature, pressure.” You don’t just get yummy coffee vodka – you can get different flavors than you’d ever experience through conventional means.

It must be fun to have a kitchen where liquid nitrogen is as common as hot water. Arnold chills a glass with liquid nitrogen, pointing out that it’s cold only on the inside, and doesn’t generate condensation. He pours himself a carbonated gin and lime concoction as the audience is served marshmallows frozen with liquid nitrogen. McGee returns to explain the history of the marshmallows – they were served at The Fat Duck as both a palate and “mind cleanser”. The chef responsible wanted to reset his diners’ expectations, so he served them a marshmallow flavored with lime, tea and vodka and frozen. The heat of your mouth melts the treat and you find yourself with vapors pouring from your mouth and nose. We have a similar experience with the frozen marshmallows, and like the Fat Duck diners, find ourselves laughing, our expectations reset.