Old Polish Almond Torte

In my family, when someone (including me) has a birthday, they choose a type of cake, and then I bake it, often with the help of one or more kids. A family favorite is the old Polish Chocolate Almond Torte. This is a gluten-free cake, made of almond meal, with leavening provided by a full dozen egg whites (with the twelve yolks going into the cake, too), and a layer of home-made marzipan in the middle. My kids found a version of it in Lemnis and Vitry’s Old Polish Traditions, but the recipe was somewhat confusing and we didn’t like the lemon juice in the marzipan. We found another version online here, which pointed us to the original source as the 1931 Polish cookbook How to Cook by Maria Disslowa. Between the three sources, and experimenting across multiple birthdays, we have the following recipe which I baked for my last birthday.

Old Polish Almond Torte

Note: The recipe takes three days and is ready to eat on the fourth day, but the only thing done on the first day is freezing the chocolate chips.

Ingredients for Cake

• 284g almond meal (unblanched)

• 227g castor sugar

• 284g powdered sugar

• 12 eggs, separated

• 284g dark chocolate chips (we use Hershey’s Special Dark)

• a couple of tablespoons of unsweetened cocoa

• optional: 1 tsp almond extract (my current opinion: omit to have better taste contrast between layers)

Notes on cake ingredients: Our grocery store doesn’t have unblanched almond meal, so we just buy unblanched (i.e., with skin) unsalted roasted bulk whole almonds and grind them in a coffee grinder. Similarly, we can’t get castor sugar, so we grind granulated sugar in a coffee grinder.

Ingredients for Marzipan Filling

• 250g almond flour (blanched)

• 273g powdered sugar

• 78g (or 78mL) water

• 2.5-3.5 tsp almond extract (use 3.5 if you didn't add the teaspoon to the cake)

Ingredients for Ganache

• 320g dark chocolate chips (Hershey’s Special Dark)

• 43g unsalted butter

• 238g (or 1 cup) heavy whipping cream

Additional Topping Ideas:

• slivered almonds or sweet strawberries

Instructions

Day 1:

Freeze the 284g of chocolate chips for the cake. Do not freeze the chips for the ganache!

Day 2:

Preheat oven to 356F (= 180C). Read over all the instructions for this day.

Grind the pre-measured frozen chocolate chips into a fine powder with a coffee grinder. Since a coffee grinder can’t take all of the chips at once, keep the ones that aren’t ground yet in the freezer.

If you don’t have almond meal, grind the almonds in a coffee grinder as finely as you can.

If you don’t have castor sugar, grind granulated sugar in a coffee grinder. It should be a bit coarser than powdered sugar.

Prepare a 10-inch spring form pan by buttering the bottom and sides, and putting a parchment paper circle on the bottom. Coat the sides with cocoa (this way, the cake remains gluten-free, and it’s better than flour).

Separate the 12 eggs.

Cream the egg yolks with the sugar until a bit fluffy, and while creaming mix in the powdered sugar and the almond meal.

Stiffly beat the egg whites. Lightly mix them into the yolk-sugar-meal combination. (The egg whites provide all the lift to the cake, so the cake will be too dense if you overmix.)

Pour into the pan. Bake for one hour. Cover and leave overnight. Feel free to refrigerate.

Day 3:

Combine almond flour and powdered sugar for marzipan filling, together with the water and the almond extract. Knead into a homogeneous ball of dough. I start by using a dough hook in a stand mixer, and finish up by hand. Roughly flatten into a thick disc of the same diameter as the cake. It will be about 1 cm thick.

Remove the cake from the pan (I like to use a plastic knife). Cut into two layers and put the disc of marzipan between, making a level layer that reaches the sides.

For the ganache, heat the butter and cream together in the microwave until it is hot, with the butter melted, but before it boils. Add chocolate chips, mixing to ensure they all melt into the ganache. The last couple of grams took more time to dissolve.

Allow the ganache to cool somewhat and become more viscous. (I am usually impatient at this stage, and regret not having enough viscosity.) Pour it a bit at a time on the top of the cake, letting it drip over the sides on a very large plate or other clean platform. Use a spatula to push the drippings up over the sides again, repeating until the cooling ganache stops dripping significantly, and you have nice smooth edges.

Refrigerate overnight.

Fine moral distinctions

I find myself sometimes troubled by narrow moral distinctions in the Christian, and sometimes more specifically Catholic, tradition. Lying is wrong, but deception—perhaps including verbal deception—is not. Intentionally killing the innocent is wrong, but redirecting trolleys onto innocent people can be acceptable. Salpingostomy is wrong as a treatment of ectopic pregnancy, but salpinectomy is right. In each of those cases, of course, there is a defensible moral theory justifying the distinction, and in fact in each case I accept such a theory. But I still feel troubled.

There is an old Polish joke. After World War II, the Soviets are shifting the border between Poland and Russia. A farmer used to have a farm in Poland, but now the farm is going to be half in Poland and half in Russia. The farm is given a choice of which half he wants. He says: “The Polish half, of course. Russia is too cold.”

Of course, when you divide a continuous landmass into countries, there will be places where a step in one direction will get you into another country. And the climactic conditions are going to seem pretty similar. They will seem pretty similar, but they won’t be exactly the same.

Similarly, if you divide the space of human actions into, say, murder and non-murder or into wrong and non-wrong, one will find pairs A and B where A falls on the bad side and B on the good, and yet A and B are pretty similar. That’s just how it is. As long as we have moral objectivity, classical logic, and continuity among actions, this is unavoidable.

This does not mean that the distinctions will be arbitrary. If there is a roadside honor-system vegetable stand and a bunch of carrots is $3.50, then there is indeed a distinction between stealing by paying $3.49 and giving a fair payment of $3.50, even though the actions are very similar. Nonetheless, non-arbitrary as the distinctions are, they may not be major.

We should thus not be surprised if there are fine moral distinctions. There have to be.

Of course, we might dispute over where the boundaries lie. One might propose different boundaries: perhaps instead of saying lying is wrong while mere deception is permissible, one will say that both are permissible when needed to save lives and neither is permissible otherwise. But the alternate distinction will also have close-by cases. Why is it, on this story, permissible to lie to save oneself from death but not from torture? And what does it mean to save a life? One is never certain that a lie will save a life. What probability of saving a life is needed? There is no way to avoid boundaries between cases that will seem similar.