expertise and cooking

There were two essays in Best Food Writing 2004 that got me thinking about expertise and how people become expert cooks.

The first essay, entitled Learning to Cook, Cooking to Learn, was by Matthew Amster-Burton on eGullet.org. Follow the link to read the whole thing. He writes about recipes, and the difference between a recipe for beginners and recipe for experts.

There is a place for the highly specific recipe: beginner's cookbooks and recipes for things generally considered tricky (pie crust) come to mind. A helpful hint is never out of place (cook the pancakes on the first side until the bubbles on top start breaking)...

The problem with detailed recipes is that they reinforce a bad assumption: that everything you need to know about making a dish can be encapsulated into a recipe, and that learning to cook is just a matter of picking the right book and following the instructions carefully.

Being a good cook is not merely the skills needed to follow a recipe, it's also the skills needed to fill in the gaps in a recipe, to deal with surprises and to improvise. Although recipes can certainly inspire new ideas and are great ways to pass on knowledge quickly, they can be a bit of a crutch when trying to become a better cook.

Christopher Kimball, editor of Cooks Illustrated and host of America's Test Kitchen, writes an essay that continues on a similar vein.

Most cooks I know are constantly looking for new recipes the way some folks are constantly on the lookout for antiques, clothes, computer software, or specials down at Price Chopper. There is nothing wrong with living life vicariously through recipes -- we all do it to some extent -- but the problem with most home cooks is that they have too many recipes rather than too few.

Two hundred years ago, good cooks generally had a relatively limited repertoire, extending to perhaps 50 key recipes. They could make a roast, a cake, a loaf of bread, a few casseroles, etc. As a result, they became experts at what they cooked -- they could make buttermilk biscuits, a pot roast, wax beans, or chocolate cake from memory, even with their eyes closed. The point? Well, they soon relaized that details mattered, that small variations in pereparation yielded different results, and that baking a pie in July was quite different from doing so in February, because the dough was more likely to heat up and become unworkable.

Kimball suggests something that I think is radical, but also very intruiging. Instead of collecting cookbooks and trying everything under the sun, instead of trying to become good at eight cuisines all at once, pick 25 recipes, and cook them repeatedly until you become an expert. "You will soon learn what a bisuit, a roast chicken, or a chicken soup is really supposed to taste like."

Another thing to think about is how much time it takes to become an expert cook. Think about a chef. They spend two years in cooking school, spending, literally, 10 hours a day in the kitchen. Then they spend several more years apprenticing, working as a line cook, 40 hours a week. At that point, the budding chef will have spent something like 10,000 hours chopping vegetables and roasting meat. That length of time, 10,000 hours, is the general rule of thumb for how long it takes to become an expert at something. I probably spend 5 hours a week cooking these days. That means that if I keep it up, around the time I become too old and feeble to hold a knife, I'll have something like the expertise of a 25-year-old professional chef.

Psychologists have studied expertise for quite a while now. That 10,000 hours it takes to become an expert makes you perform qualitatively differently from a novice, whatever the task. Here are five differences between experts and novices, from a web page by Jan Armstrong at UNM, with my food-related comments:

1. Novices rely on formal rules and procedures to guide them. Experts rely to a greater degree on their accumulated experience.
   This is why, and how, novices like myself use recipes. When I recently learned to make a beurre blanc, I looked at that recipe every 30 seconds to make sure I did it right. Everything was measured precisely. An expert would just dump the ingredients into the pan until it looked right.
   
2. Novices are highly conscious of the task performance process. This is a distraction and creates additional "load" on cognitive processing. As expertise grows, performance of the task becomes automatic. This cognitive phenomenon is called "automaticity."
   An expert can cook 15 dishes at once. If I concentrate, I can do three simple dishes simultaneously, or one new and complicated dish.
3. As expertise is acquired, the learner's cognitive processing system becomes more efficient at processing new information. As a result, experts can see the whole picture. They are also more aware of the specific circumstances in which they are working. They have good self-monitoring skills. Experts can make even very complex, difficult tasks look easy.
   Last year I spent 3 or 4 months making almost nothing but Indian cuisine. By the end of that time, new recipes and techniques were much easier to pick up than the first few, as I could relate those new processes to old ones I'd previously learned.
4. The expert has a larger number of strategies, and more effective strategies, for performing the task. This may be the most critical difference between the expert and the novice. Experts know how to get out of trouble because they have multiple strategies for dealing with the unexpected.
   Two words: pie dough. (My nemesis.)
5. Experts are more flexible than novices. They rely on intuition in ways that novices find difficult to comprehend.
   I, for one, have never seen a measuring cup used on Iron Chef.

Another Psychologist of experise, Anders Ericsson of Florida State, says this about his recent work:

[T]he difference between experts and less skilled subjects is not merely a matter of the amount and complexity of the accumulated knowledge; it also reflects qualitative differences in the organization of knowledge and its representation. Experts' knowledge is encoded around key domain-related concepts and solution procedures that allow rapid and reliable retrieval whenever stored information is relevant. Less skilled subjects' knowledge, in contrast, is encoded using everyday concepts that make the retrieval of even their limited relevant knowledge difficult and unreliable. Furthermore, experts have acquired domain-specific memory skills that allow them to rely on long-term memory to dramatically expand the amount of information that can be kept accessible during planning and during reasoning about alternative courses of action.

Unpacking this a little bit, the key bit of what Prof. Ericsson is saying is that experts can use their memories of their knowledge and experience much more efficiently. Rather than having to consciously recall a rule (how many minutes to hard-boil an egg again?), they are able to retrieve that sort of information from their long-term memory without running into the inherant limits of short-term (working) memory. It's not that an expert cook can think about more things at once than a novice cook, it's that they don't have to think in order to cook!