In Search of the Perfect Proustian Madeleine

Madeleines have roots in the French town of Commercy and 18th century nuns supposedly sold the treats to support themselves and their schools. One legend has it that the little cakes take their name from a young servant girl named Madeleine who made them for the deposed king of Poland, Stanislas Leszczynska, when he was exiled to Lorraine. Another version of the origin story states that Jean Avice, considered the "master of choux pastry," invented the madeleine in the 19th century.

Whatever you believe to be the true birth story of the madeleine, trust me when I say just about every culinary expert and food blogger on the planet has a different idea about what makes a madeleine a madeleine. But the most characteristic and controversial feature of the snack cake has to be the hump. That stupid hump I quite honestly never noticed until my boyfriend pointed out that it's what had kept him from using those pans in the cupboard: Madeleines are supposed to have a hump and attaining the hump is hard work.

Some say the hump is all about creating a sudden drop in temperature, so chilling the batter and even freezing the pans before baking is crucial. Others are adamant that the hump (sometimes called a "belly") is only achievable through the use of baking powder which is apparently "frowned upon by purists" who say the hump is a natural product of thoroughly beaten eggs. As a completely amateur baker with no other motive than to escape from the horror show that 2020 has become, I set out on a small quest to achieve madeleine greatness.

The Kitchn is a solid go-to source for cute, Instagram-worthy desserts, so when a quick Google search turned up their recipe for "classic French madeleines," I thought maybe I'd found victory in my first batch. Their recipe called for all the essentials — unsalted butter, sugar, flour, eggs and vanilla — plus a heavy dose of lemon zest and lemon juice. Baking powder was conspicuously missing from the list, backing up my early research that "classic" and "authentic" madeleine makers simply don't use it. They also instructed to let the greased pans sit in the freezer for an hour before baking and to turn the pans midway through baking. I figured all that work would result in magically humpy madeleines, but when they emerged from the oven, my first batch of madeleines — while beautiful — were pretty flat. My boyfriend and I agreed that the extra lemony essence was a great bonus, both flavor and aroma-wise, and the overall taste was lovely. They just weren't as cakey as we'd expected. On to batch two.

If humps were what we were going for, humps were what I would seek out. So I did. I literally Googled "Madeleines, humps" and came upon Baking Like a Chef's recipe for "perfect French madeleines." The humps in the photos were almost vulgar — these madeleines looked wildly different from the ones I'd initially baked, and the recipe author claimed she could create humps as tall as an inch and a half (4 centimeters). I had to give it a shot. This batch called for baking powder, and not an insignificant amount: 2 teaspoons, which in the baking world, can make a big difference. The recipe also called for twice the butter of the first batch and several more eggs (even after I scaled it down to make 24 cakes, not the 36 it was intended to produce). And while baking powder might be "frowned upon" by purists, holy moly, you guys — the humps were no joke. This recipe also called for the madeleines to sit in the oven for up to 10 minutes after a very short initial bake time (three minutes), and then to bake for an additional five minutes. Whether it was just the baking powder or a combination of all those elements, the result was airy, spongey, cakey, humpy goodness and my boyfriend said these were "95 percent perfect" (he missed the extreme lemoniness of batch one) which frankly, I'll take as a success.

I figured if I was going to give it one more shot, I'd look to the icons. After taking an informal Instagram poll, my small but opinionated following made it clear that they wanted to see the results of Julia Child's recipe over Martha Stewart's, so I set out to replicate Julia's recipe, via Hungry Sofia. This one had no baking powder, a fair amount of lemon, and a relatively long baking time (15 minutes), not to mention it came from Julia freakin' Child, so I figured this would be the one to get me to perfection. I refrigerated one pan before baking, as the recipe suggested, but stuck the other in the freezer because I guess I suddenly fancied myself a scientist running an arduous and delicious experiment. The refrigerated pan batch (i.e., the one that followed the exact recipe) was less than optimal. They looked slightly anemic compared to the humpy giants of batch two, and they were browned to the point of looking overcooked. The frozen pan batch turned out prettier, but the humps were still less than impressive. And while I list off all the things we found wrong with this batch, it's important to remember that we still demolished nearly two dozen in less than a day.

In the end, I learned that despite what any culinary expert or Proustian text may tell you about what defines a "good" cake or cookie or whatever you're mixing up, the process of creating, consuming and sharing those treats is really what makes them so sweet. Also, loads of sugar.