Part 2 of a 5-part series about those chocolate coated treats you adored as a child. This examines the cookie base.
This should be a short post because it has a relatively simple, if far-reaching, message. Most of us grew up with a single version of Chocolate Marshmallow Treats. That’s the one that will feel “right,” while everything else is an experiment. But that version is not “right” to the rest of the world. There are almost infinite ways to customize these cookies, and a great many of those variations have been popular in some part of the world even if they were unknown in yours. So when you look at the different options for the cookie base, start by considering the options and possibilities.
Chocolate Marshmallow Treats can build on almost any kind of cookie base, and shifting that part up will completely change the effect for the loved ones who are privileged to share your bounty. We really did test this, and the distinctions aren’t exactly subtle. Indeed, the best advice we can give is this: keep the amount of flavor in your cookie base down to a hint rather than a dominant feature. You want the cookie part to mesh with the marshmallow and the chocolate, not fight them for dominance.
The recipes we tested include:
Commercial Vanilla “Wafers” – on the smaller side, but absolutely delicious as a plain, almost invisible foundation. The marshmallow filling will soften them in the beginning, but time will help them to crisp back up a little.
Homemade Vanilla Snaps
Homemade Vanilla Snaps – Step up your game from “plain” to “elegant” with a homemade version.
Graham Cracker Snaps – I grew up with MallomarsTM as my personal norm. Those come with a graham cracker base that evokes a S’more. See below for an elegant homemade version that’s worth eating out of hand as well as inside a cookie.
Spice Snaps – See the master recipe below, or use your personal favorite with about half as much spice as you’d use for a free standing cookie. Full-flavored snaps aren’t partners so much as pushy, wannabe overlords.
Cocoa Spice Snaps with marshmallow only, white chocolate, and dark chocolate
Shortbread Wafers (Commercial or Homemade) – You just can’t lose when buttery goodness comes to town.
Miniature Cake Rounds (No Recipe Provided) – This is as common as can be among the commercial versions, but you won’t see a recipe here. We do cookies, thank you very much, not cake. (I might also be just about done with experimenting by now…)
No Cookie Base – Omit the base and you get a candy rather than a cookie. Same issue as the cake base. Besides, I have zero doubt that anyone who reads this will be able to figure this one out on their own.
Want to play? Lemon curd dots to go inside the marshmallow!
CONCLUSION
That covers it folks. Expand out from here. But please pause for a moment and look at how we’re already multiplying out! So many choices for just the cookie part. And believe me, there are just as many options when it comes to the marshmallow filling.
Today you get a big, fat note from Kate, because I’ve somehow become the family chocolate dipper.
Oh, we’ve all tried it, and we all have our tricks.I’m the one who gets to share the final results with you, though, because I strongly believe in the Power of Chocolate. It is, after all, its own separate food group!
[NOTE FROM SCOTT: It also has magic powers. I know this from long experience with both a wife and two daughters. “Just stand back and throw some chocolate at her” isn’t a joke, it’s a survival mechanism. The masculine life expectancy in America soared in the late 1890’s and early 1900’s. Hershey’s went into business in 1894. Coincidence? You lay your bets, and I’ll lay mine.]
We’ve been dipping various cookies in chocolate for years, as evidenced by the recipes in Volume I of the Heritage Cookies Cookbook. A careful observer may notice that the quality of our chocolate dip in some of those photos isn’t nearly as smooth and shiny as commercially made products. There is a reason for that. First, we are home bakers, not chocolatiers.
Sure, I could invest into (a few more) additional pieces of equipment. Then, because tempering chocolate is best done at volumes larger than we usually use, I could sink an inordinate amount of money into ingredients only to learn how to temper chocolate properly. But… really?
Friends, chocolate ain’t cheap. Never was. And neither is time. Therefore, I am forever grateful for every little trick, as well as for those immense bricks of good-enough, affordable chocolate from Trader Joe’s. They do a pretty good job for most our applications. For fussier cookies we’ll buy Callebaut or Valrhona, and for convenience, Scott is willing to use those commercially available chocolate melts. But that doesn’t get us into the main point of this article.
Why is high-quality, commercially produced chocolate so pretty and shiny, and how can one do that at home? I’m not a chocolate expert so I’ll give you the shorthand as I understand it.
Cocoa butter, the off-white fat that forms the bulk of the chocolate we eat, crystallizes into several different forms. Of those, the beta-crystals are the ones that produce a smooth and shiny surface. If you melt your chocolate under exacting, well-controlled conditions, working the softened product under controlled humidity, concentration, temperatures, and timing, you will end up with tempered chocolate. Such conditions can be achieved for the home baker, but they require low air humidity, a large marble or granite board, and a certain minimum volume.
Plus, of course, practice.
But there is a shortcut! Properly tempered chocolate is chock-full of beta-crystals. It snaps upon breaking, and the break is smooth instead of crumbly. The chocolate’s surface shines with that undeniable gleam of quality and good taste.
Commercial chocolate makers, who process tons of material, have their processes tightly in check. The less-expensive brands also stretch their batches with extra fillers (powdered milk and lots of extra sugar), flavorings, emulsifiers, and various fats to act as melt conditioners. You will notice that cheap chocolate has a list of ingredients as long as Twinkies, whereas high-quality chocolate will have fewer than you can count on your fingers.
Technically speaking, the final product requires only cocoa butter, cocoa, and sugar. But there is an inexpensive, readily available product that works even better if you are willing to turn a blind eye toward certain other factors.
To Palm Oil Or Not To Palm Oil, That Is The Question
Palm oil is entirely edible, tasteless, inert, and has a combination of high molecular weight and melting characteristics that allow you to easily thin chocolate enough to produce a hard, thin layer at room temperature. These properties, its low cost, and its ability to substitute the more expensive cocoa butter, makes it ideal for a whole slew of commercial applications.
NOTE FROM SCOTT: Wilton, the maker of Candy Melts™ sells a product called “EZ Thin Dipping Aid.” That is the product to use if you’re willing to go the palm oil route. It works like a charm and is the perfect solution used across the board by the baking industry. So why do we go to all this effort to avoid it? Well…
Coatings, fillings, cakes – even tortillas! Palm is an incredibly useful product for a host of culinary purposes; so useful that human greed has gotten involved to change the calculus. Entire tropical rain forests were cut down to make space for palm plantations, where brutal working conditions approaching slavery are routinely employed. And there is absolutely no way to source palm oil from only the “good actors.”
Understand: palm oil is no worse than anything else from the chemistry and biology points of view. It is a fat, with all the pluses and minuses that entails, and a particularly desirable one because of its unique characteristics. But our family will not willingly contribute toward either environmental devastation nor human degradation, and so palm oil has now become a product that we actively avoid whenever possible. You will not find it in any of our recipes until we’ve been convinced that the problems were solved.
If Palm Oil Is Out, What Then?
So what do you do when you’ve ruled out the cheap, convenient, and chemically ideal ingredient as your melting aid? We ended up resorting to the more expensive, natural and equally ideal cocoa butter. You know, the stuff that actually belongs in chocolate!
A quick online search revealed that cocoa butter is available in convenient melt discs. A purchase under $20 lasted us through great many chocolate projects, so a little goes a long way. Make sure to get the food-grade, as many less-expensive brands are intended for cosmetic use only.
What did we try before cocoa butter?
“Everything.”
It feels like it, anyway. Butter, cream, and various oils all got a whirl in our kitchen. Our younger daughter is a great fan of coconut oil as the magic ingredient for the “magic shell” used to coat ice cream. That texture is exactly what you want for Chocolate Marshmallow Treats too, but it just doesn’t work at room temperature. The coating stays solid when it’s down at ice cream temperatures, but it turns tacky on cookies and becomes downright runny if your room turns warm.
Butter and cream will also lower viscosity (i.e., the chocolate melt is thinner) but they won’t give you that shiny surface and can also remain sticky. Vegetable oil works to thin the result and it helps the melting process, but the chocolate remains even runnier because the melting point is that much lower.
We found that cocoa butter was the only alternative that really worked. It has the right combination of both melting point and melt properties to help soften the chocolate thin it for a good dip – but also gets hard enough to avoid being sticky at room temperature. Think of every other chocolate! They melt in the mouth, not on the plate.
But What About The Pretty, Shiny Part?
Aha! You want to grow beta-crystals.
Think back to that sugar crystallization experiment from elementary school. You coated a wet string in sugar crystals, put it into the saturated liquid, and then watched the crystals grow over the next few days. Those initial crystals were key to giving a platform for the crystals to form. You had to seed the string.
In exactly the same way, you can seed your microwave-melted chocolate with beta-crystals by adding solid grating of an expensive, shiny, properly tempered chocolate you bought at the store. If you fail to do that, the chocolate won’t look shiny after it sets. A day or five later, depending on how humid it is, you’ll notice the chocolate’s surface breaking into light and dark structures. The chocolate is blooming with the growth of unwanted, non-beta crystals.
Compare the two enrobed treats in the photo below:
So here you have it. You will find that white chocolate melts differently from all other chocolates, because it has a slew of various other ingredients. I had to add more cocoa butter to help it melt and got a thin coating as a result. Milk chocolate is the best-behaved and least likely to seize. Which brings us to a sticky situation:
What If Your Chocolate Seizes?
Seizing refers to a phenomenon where you are melting chocolate, stirring it, and it suddenly thickens into a grainy, thick, clay-like mass. Seizing occurs when a droplet of water accidentally splashes into your melt. A steamy kitchen has been reported as a culprit in cookbooks. The cocoa butter crystallizes into the “wrong” crystals, locking its structure into something you can’t easily use.
Overheating your chocolate will make it more prone to seizing. The darker the chocolate, the easier it is for it to seize, too.
But there is a fix: Add boiling water one tablespoon at a time, stirring vigorously. The water will dissolve the structure of the crystals, and give you a chance to melt-and-seed as described above.
This being said, I have not tried to make chocolate coverture with a melt that has seized and been rescued. I don’t know how the final product would set, and whether it would bloom or not. However, this chocolate can be used for any other melted chocolate application. If you make ganache, you can save it for a cake filling, brownies, or truffles you’ll probably be making down the line.
I hope this helps. Keep playing with it and expect to have a ruined batch every so often. Since no chocolate is wasted unless you burn it, this experimentation is a low-stakes learning curve. Let us know how this worked for you, and please share tips of your own!
NOTE FROM SCOTT: As you might expect, we made a lot of coated treats in the process of recipe development over the past several months. A lot of them looked perfect when made, and perfect the next day, but mysteriously bloomed after a few days out on the table by the window (we being too lazy to store them properly). Poor Kate was beside herself, figuring that all her careful work had gone wrong. (NOTE FROM KATE: Not “beside myself.” I am entirely too cool a cat for that. More like, pissed off and frustrated. This might have something to do with the fact that I dejunked our excess storage containers 😉
Then, one late morning, en route to a work-break bit of snitching, I noticed that the cookies were positively glowing in the sunlight; and made my fingers sticky in an extremely tattle-tale way. Aha! The penny dropped. The chocolate coating was heating past the critical stage and then re-hardening in a random way without the benefit of beta crystals to guide the process! The result? A coating that bloomed with big white splotches after it rehardened.
It tasted just as good before and after. I assure you; I tested that feature thoroughly, and with great attention to detail. But it is a lesson to remember. Leave your chocolate out in the sun, and it will lose its temper, have a tantrum, and go all white in the face.
To Mold Or To Pour?
The neatest solution is to spend a few bucks on a silicone chocolate mold. Paint the inside of that with your chocolate coating, set it aside to harden, and then fill it with your marshmallow. Top that with your cookie, or some complementary jam and then your cookie. Chill a little more, and then unmold.
Perfect every time unless your coating somehow sticks to the silicone.
Part 3 of a 5-part series about those chocolate coated treats you adored as a child. This examines the marshmallow fillings.
Part 1 of this series provided an overview of this massive project, with some notes about the problems we ran into.
Part 2 listed several recipes we developed for different cookie bases: Vanilla, Spice, Graham Cracker, and Shortbread. You can also use a thin cake round, but we left that one out because this is, after all, a project devoted to heritage cookies.
This Part 3 examines the two main approaches to creating a marshmallow filling. There is a third approach too, for fruit-based Zefir marshmallows, but I’ve moved that discussion over to its own Part 5.
Piped with a round tip onto commercial wafers
Piped with a star tip onto commercial wafers
Rose, Vanilla, and Banana on homemade Vanilla Snaps. Color coded with a few drops of food dye.
The Internet offers a metric ton of recipes for homemade marshmallows, most of which work pretty well. They’re the world’s easiest candy. Basically:
Start with a thick and sticky substrate (made from the marshmallow plant in antiquity);
Heat a sugar syrup to the soft ball stage (235‑240 °F a/k/a 112-115 °C), using a combination of table sugar (sucrose) and corn syrup (glucose) syrup to prevent crystallization; and then
Whip the sugar syrup into the substrate until it balloons up to create fluffy marshmallows. These can be piped while warm, or else cooled in a pan to make cut-out eating marshmallows.
The American version uses a lot of gelatin to make that sticky substrate. The European version uses whipped egg whites with less gelatin; i.e., is a gelatin-strengthened Italian Meringue. NOTE: French meringue is egg white whipped with sugar; Italian meringue is egg white whipped with hot sugar syrup; and Swiss meringue is egg white heated to around 170-175 °F (76-79 °C) before being whipped. Swiss is stiffer and more stable than Italian, which is stiffer and more stable than French.
Problem #1: The filling was denser and sweeter than we wanted.
Think of a classic marshmallow you eat out of hand. They have a distinct chew to them, right? It’s part of the appeal. Now think of that chocolate covered marshmallow treat you grew up with. Much less chewy! And a bit less sweet, because flavor also comes from the cookie base and the chocolate topping. The The cookie would not be perfect until we got a handle on the problem.
Fair enough, but how do you change a marshmallow recipe to achieve those targets? It can obviously be done – the commercial products prove that – but we had to figure out how. And that was a major pain in the whosis, because our expertise on the baking side did not extend into this kind of candy making and sugar work.
The first and most obvious solution (cutting cut back on the amount of gelatin) proved to be impossible. Cut back too far and the structure dissolves rather than getting lighter, and I could not fidget my way into a perfect balance because kitchen scales only have so much precision. Plus the marshmallows were far too sweet for this use, and cutting back on the sugar too created other textural issues. I fought this problem for months, and ended up with a pair of solutions that work hand in hand.
Kate gets 100% of the credit for the first. She noticed that our trusty stand mixer began to really strain as the marshmallow mix cooled down. Part of this came from how the filling thickened as it cooled, but the overflowing bowl factored in too. Her solution? Shrink the recipe by a third. Huge difference. The lower volume resulted in a fluffier marshmallow because the machine wasn’t slowing down, and also because the bowl had extra room to accommodate it.
I claim credit for the second solution. I posted the question on some Internet groups/boards, and no one seemed able to help. All I got were suggestions for vegan substitutes that could be used instead of gelatin. That wasn’t the issue! Vegan shmegan; I just wanted to know what was going on in the chemistry, and how to adjust it. One long-suffering volunteer finally burst out with, “maybe you should take a food science class or two.”
Ding ding! For some reason I’d done all my Internet searches using the words “chemistry,” “physics,” and descriptions like “fluffier” and “denser.” I’d never done one using the term “food science.” Why not try? I did, and up popped this lovely article on a project for grade school science teachers: “Sweet Science – Making Marshmallows.” I urge you to read it, but here’s a quick summary in case you want to save some time. Loosely translated, the project reads as follows:
“Have your students make marshmallows with different proportions of sugar to corn syrup, and then measure the characteristics. They’ll find out that more sugar equals a denser and sweeter product, while a higher ratio of corn syrup reverses that. Then you can let the kids eat their experiment, and send the sugared-up darlings off to some class with a colleague you don’t like.”
Okay, I added that last last sentence on my own. But look at what it says! Lighter, and less sweet. Exactly what I wanted. We did the experiment; it worked just as predicted; and the results now appear in the recipes below.
Plus there’s this: Now you have the power too! By way of reference, American eating marshmallows have a sugar:syrup ratio of 6:1 by volume. That’s now at 3:2 for the cookie filling, which coincidentally works out to just about 1:1 by weight. This is just about perfect for us, but you can fuss with it to your heart’s content to find the proportion that’s perfect for you. Tastes differ. Though it must be said that we achieved a texture so “right” that our testers actually singled it out for special compliments! Yes, you really can tell.
Problem #2: Piping the Marshmallow was a P.I.T.A.
There’s no easy way to say this. I suck at piping. And that is a problem because this is a recipe where you need to work quickly. Cooled marshmallow filling stiffens, is harder to pipe, and tends to mound up unevenly instead of covering the whole cookie in a pretty layer. Nor does it help that our biggest piping bags are “mediums” on anyone else’s scale. Even the reduced recipe required us to refill the bag two or three times, which led to the filling cooling off that much faster, and becoming that much… You get the idea.
Marshmallow gets stiffer as it cools, and won’t pipe out cleanly.
In the end I found three solutions that worked pretty well.
Scott’s Solution. Rope in Kate or a daughter to cover for my sorry efforts. This works perfectly, and I recommend it highly.
The Doh! Solution. Buy a bigger bag. Maybe next time. They aren’t as easy to find as you’d think.
Solution #3. Prepare two or three of the medium-sized bags before you start to pipe the cookies out. This saves a few critical minutes, and also lets you use more than one tip.
Yes, I am overselling the difficulty a bit. Anyone with the right tools and some basic skills won’t find it as challenging as I did. The multiple-bags approach also solves a lot of the problem, and the cookies taste just as good even if the marshmallow cools enough to come out lopsided. Still, forewarned is forearmed. Be ready to pipe as quickly and efficiently as you can; and/or to laugh at your funky shapes.
Problem #3: The Infinite Variations
Continuing the theme from Parts 1 and 2, the marshmallow filling can also be varied in as many ways as your imagination permits. And it’s super simple to do!
So-called “plain” marshmallows are flavored with vanilla extract. An even teaspoon works perfectly at these amounts. “Flavored” marshmallows replace the vanilla with with any extract and/or food color that’s buried in your spice cabinet, and that same 1 tsp. will do the job just as well. Lemon, rose, rum, banana, coffee, mint, almond… there is some place in the world where that flavor of marshmallow is a beloved favorite sold in stores. I recommend some color coding to avoid frowns of confusion, but it really is that easy.
Spices work too but we found that spiced cookie bases were easier to control. Marshmallows have a ton of air, and adding enough spice to provide solid flavor can make them a little gritty.
Fruit flavors are a little different. The extracts work fine for a properly artificial flavor reminiscent of commercial products, but you can also get The Real Thing by making Zefir instead of a classic marshmallow. Talk about intense essence of fruit…! But that’s getting ahead of myself. I moved that discussion to Part 5 for very good reason, and you’ll simply have to wait.
CONCLUSION
It took some serious work and research to nail down the filling part of this cookie, but that really came down to exploring things at the food science level. [It had nothing to do with me being obsessive, no matter what my piping assistants may say!] If you follow the recipe below – either one – you will find it quick, easy, and effective. You’ve got your wheel, and there’s no need to recreate it.
So where are we at?
Make a simple cookie base, flavored in any way you can imagine.
Pick your choice of marshmallow approach, and flavor it with any extract or liqueur in your cabinet by substituting 1:1 for the designated vanilla, and color coding with a touch of food dye.
Pipe your marshmallow creation onto your choice of cookie, using your choice of piping tip.
Bing, bang, boom. You’ve got these Chocolate Covered Marshmallow Treats 70% done. All that remains is coating them in everyone’s favorite darkest-brown food of the gods.
The classic, with versions for eating out of hand or for using in other recipes
Here’s another head smacker cookie really should have gone into Volume I but somehow escaped our notice. It came up now in a sideways fashion. We’ve finally gotten around to working on Chocolate Covered Marshmallow Treats. You know that cookie. Everyone does. It’s the one you loved as a child, with a simple cookie base that has marshmallow on top, and is then coated with a thin chocolate shell.
That cookie has scores and scores of different version around the world, but the most common uses some equivalent to a Nilla Wafer® base. So that meant I should start by making… Doh! I’ve never actually worked up a homemade version of Vanilla Wafers. [Sigh]
The result was all we could have hoped for and more. In fact, we discovered that it isn’t a single cookie. It’s actually three.
Use the full amount of flavor and you’ll get a very pleasant cookie to eat out of hand, that adults will enjoy as much as children.
Use a moderate amount of vanilla, and you’ll make something very close to the commercial product. Just better in every way. Real butter and real love make a real difference. Kate says that toddlers in Prague were fed a similar cookie whenever treats were earned.
Reduce the amount of flavor all the way, and you will get the plain, invisible cookie that’s perfect for all the “other” uses these get purchased for. Banana Pudding; El Cheapo Tiramisu; Chocolate Marshmallow Treats; etc., etc., on and on, and suchforth and suchlike.
This is a simple cookie, and it’s meant to be a simple cookie, so I’m going to try to keep the discussion brief. There weren’t many issues that needed to be addressed. It was really a matter of focusing on what we wanted the cookie to be.
First the Eggs
I wanted this to be a light, crispy cookie with a fair amount of air, that would taste a little buttery without coming off as rich. Most recipes in our books and online resources called for one whole egg to go with each stick of butter. I tested that, along with one egg white, and one egg yolk.
The whole egg won handily, both for ease and flavor. Yes, you can make a good cookie with both egg white and egg yolk, but the whole egg is simpler and, for this use, better.
Wait; Haven’t We Made This Before?
For crying out loud! Butter cookies. Sugar cookies. Shortbread. Alfajories. Spritz Cookies. Snaps. They’re all pretty much the same, right? Just butter, sugar, flour, starch, and egg in slightly different proportions, plus a hit of flavor. Was this one really different, or just a variation on one of our existing and already-tested recipes?
Well… no. I went back and crunched the numbers. These are in the same family but the proportions are distinct enough to earn their own name. Same for the classic French cookies we plan to work up someday, such as Langue du Chat (“Cat’s Tongues” a/k/a Homemade Milanos®), and Tuiles. Those use several egg whites to create a different species within the same genus.
What’s In A Name?
All that recipe comparison work did clarify one thing that was bugging me. These are not a “wafer”!
I am a lawyer in my real life, and have been for 30 years. So I can speak for my profession when I tell you that virtually every non-criminal problem I see comes down to a variation on the theme of frustrated expectations. Person X thought s/he deserved or would get an “A,” ended up with a “B”, and grew angry at that result. So Person X goes back to Person Y to complain, “where is my A?”, and the situation then grows worse because Person Y insists that s/he meant “B” all along. And most of the time both interpretations fit the actual words.
The solution? Define your bloody terms so you can at least know where you really disagree! Try it the next time someone wants to argue about politics. “Exactly what do you mean when you use that word?” You’ll be shocked at how much of your fight disappears, or at least settles down to the amicable question of how words ought to be used.
In any case, I have spent a lot of years getting frustrated over that stupidity, in both my profession and in politics, so I get a little [ahem] picky about calling things by their proper names. A “wafer” is an extremely thin, usually broad cookie like our different Snowflake variations, Tuilles, or Lace Cookies. You can get to that result in many different ways, but that’s the definition.
This cookie does not meet that definition, and thus it isn’t a “wafer.” This cookie is a “Snap”. Harrumph.
We define Snaps as any plain, simple, round cookie that features a clear and focused flavor. Cardamom, Anise, and Black Pepper Snaps feature a single spice. Ginger snaps feature a spice mix with a touch of molasses. But that is the family group, and that is where this cookie belongs. Harrumph I say again! And harrumph for good measure. Harrumph. Harrumph. Harrumph!
Alas, but the outside world does foolish things on occasion. Sometimes it will even [gasp!] ignore it when some large company mislabels what ought to be a Snap as a Wafer. And a wise man has to know his limits. So let us agree on a simple formula, okay?
First, I am in the right. Harrumph. And that means the world is wrong. And thus equity, justice, and fair play are unanimously on my side.
Next, I am not being petty no matter what certain females in my household may choose to imply. Nor childish. Not, not, not!
And no, this isn’t a fight I’m going to win. So how about a compromise?
Let’s agree to call the eating version of this cookie a Vanilla Snap. That’s the tasty one with a full tablespoon of vanilla, which is going to make people sit up and say, “I never thought I’d say it, but these things are capital-G Good.” The weaker versions, which are intended for use in other recipes anyway, can be called by that other name for the sake of allowing the rest of the world to save some face.
Back to your regularly scheduled programming.
Confirmed Again: “Other” Starches Work Better Than Corn Starch
Is the cookie better with a bit of gluten-free starch to supplement the flour? Yes, it is. Cake flour might work too, but we didn’t test it.
Would corn starch work? Yes… but it isn’t as reliable as rice flour, tapioca flour, arrowroot, etc., because you’ll occasionally get a less-done piece that leaves a gummy sensation in your mouth. It’s an effect we wrote about in our book Heritage Cookies of the Old and New World. Corn starch breaks down at a higher temperature than those others, and thus cookies made with corn starch are more likely to have leftover “uncooked starch” in the final product. That’s the underlying problem; it was confirmed again in these experiments; and there’s simply no reason to fight the science. Vanilla Snaps cook until the bottoms brown, which gives a margin of error, but why make use of that margin if you don’t have to?
We usually bake with tapioca starch because it is cheap and readily available in large quantities at our local oriental market. We have tested rice flour and arrowroot too, and found them to be identical. We have read that potato starch fits in too. Corn starch is the only one that requires a higher baking temperature.
Would A Drop Cookie Work?
We always prefer to write up drop cookie recipes when possible, just because they’re so much easier for new people. Refrigerator slice-and-bakes come in second, followed by roll-and-cut cookies, and then by ones you pipe out. What was the best way to go for this one?
The answer depended on what we wanted to do with them. Drop cookies worked fine from the baking perspective, but I simply couldn’t get them to come out even enough to use as a consistent base for the other project. So here is my verdict:
If you’re after the eat-out-of-hand Vanilla Snap version, make a simple drop cookie and be done with it. You don’t need to do it that way, but you can. Kate swears that they also came out a little lighter and crispier than the rolled-and-cut version, though I couldn’t tell the difference.
If you’re after a cookie to use in some other role (like the base of a Chocolate Marshmallow Treat), roll them out and use a cookie cutter. These are dead consistent and thus better as an ingredient. The dough is a little on the sticky side, but works fine if you simply dust with flour as needed.
If you’re trying to jazz things up, add a bit of milk to thin the dough out to a batter, and then pipe out something fancy. You could even sprinkle it with a bit of sanding sugar. It ought to work with a spritz gun as well.
Bake Until The Bottoms Are Good And Tan
Part of the flavor in this cookie comes from actual browning on the bottom. Most cookies want to come out of the oven when they’re still pale, and the edges feel barely set. Leave these in for the full 12 minutes, or until you can see browning around the edges.
CONCLUSION
That’s about it. One forgets that plain old vanilla isn’t that plain at all. It’s a really delicious spice, and it fully deserves to be just as celebrated in a Snap Cookie of its very own. They are as “heritage” as you can get, and well worth making.
This is a sample drawn directly from our book Heritage Cookies of the Old and New World. The first part is our favorate “Cookie Nerd Moment” of the entire tome. This is where we stop and take the time to discuss particular tricks, tips, and historical bits that provide a lot of the “why” behind the recipe’s “what.”
Anyone who makes 20 or 30 different cookies each Christmas gets used to two common questions: “What are your favorite cookies” and “What is the favorite cookie in general?” There is no answer. Everyone has his or her own favorite and there isn’t a single cookie that would touch the 10% mark if we kept track of who voted for what.
You probably could have guessed those two if you’d given it a bit of thought. But the third most common question might surprise you: “Do you know how to make [Nut Horns]?” We put that in brackets because the same basic cookie is loved with fervor and devotion across a huge part of the world with very minor changes. In Central Europe the cookie’s known by some word that translates as either Nut Horns or Nut Crescents (Kipferl in Austria and Hörnchen in Germany are two examples). Eastern European bakers shape them into balls, often with a heavier percentage of walnuts, and call them Russian Tea- or Wedding Cookies. Move south and west and the name turns into Italian Wedding Cookies, also ball shaped but with a tilt toward almonds. The south and east features ball-shaped Greek Kourambiedes, which have slightly fewer nuts but include some brandy. The New World knows them best as Mexican Wedding Cookies, which usually feature pecans. We could go on.
Simplicity plays a big part in the cookie’s charm. It’s all about the nuts, with vanilla to heighten their impact. Powdered sugar goes on top – a heavy coat that may also have a hit of vanilla. The Greeks stand out by adding a single whole clove for decoration.
There’s also a common structural theme: you’re after a cookie so tender that it will shatter to fragments at the slightest excuse, and will dissolve in your mouth with a burst of buttery, nutty goodness. That’s the hard part – getting them just tough enough to wait for an excuse without losing that meltiness.
We spent many years fidgeting and playing with this recipe to get it just right. The final product incorporates a number of subtle tricks that will add up to make the best [Nut Horns] you’ve ever eaten. Better even than Grandma’s!
TRICK #1 (for flavor): Toast the nuts in advance. It gives you a longer, deeper flavor.
TRICK #2 (for texture): The textures from your nuts matter as much as the flavors. We strongly prefer to mix that texture up to give the full experience from meltingly soft in most parts to recognizable bits sprinkled here and there. We always start by using a rotary grater to reduce at least half the nuts to a point just short of where you’d start calling them “nut flour.” Maybe even a little short of “nut meal.” The other half should be a mix of textures ranging from “finely ground” to “finely chopped”. If push comes to shove we suggest a 50/50 mix of nut flour for the first half and ground nuts for the second. Grandma may have ground all the nuts for convenience but even she would agree that mixing the textures adds something extra.
TRICK #3 (for flavor and texture): We also favor a mix of nuts. This particular trick may cause Grandma to start muttering about upstarts and heretics when you mention it, but trust us: a little hint of walnut will make your pecans taste more “pecany”, just as some almond and hazelnuts will make your walnuts more “walnutty.”
TRICK #4 (for flavor): A hefty pinch of salt also enhances the nut flavors. Just make sure to use so little that no one can tell it’s there.
TRICKS #5.a to 5.c (to counter fragility): Our recipe hovers right on the edge of falling apart into a 50% casualty rate from being too tender. There are several tricks and options designed to pull it back from that horizon. You’re going to need at least one of them:
A spoonful of water in the flour adds just enough gluten to hold it together. This trick doubles as a way to customize for your particular kitchen and flour.
An egg yolk in the dough works too and is more traditional, but can make the cookies caky if it goes too far. This was the “cheat” that Grandma used in secret because her best friend, the one she wanted to bake like, didn’t need to.
Working the dough as you form the crescents can also add a little gluten, but it comes with a real warning: the heat of your hands can melt the butter in your dough, which turns the cookies oily and guarantees they’ll be too tender.
TRICKS #6.a to 6.c (to counter toughness): Our recipe uses so many nuts that fragility tends to be the bigger problem, but sometimes you have to go the opposite way. We ran into this when we lived in Boston and used the high protein King Arthur as our house default. Here are some techniques that will add tenderness without changing the basic formula:
Remove any extra water or flour from your dough.
Refrigerate the cookies on their tray for 30 minutes before baking. This will relax any built-up gluten.
Remove a tablespoon of flour from the recipe.
TRICK #7 (for flavor): It’s a little counterintuitive but this particular cookie actually tastes better if you narrow the “horns” of your crescents toward the ends. Yes, the easiest way to make a crescent is to form the dough into a rope and then bend it in the middle. Yes, that is a typical way to ensure even baking by creating an even thickness. Yes, you can even treat this as a roll-and-cut cookie with fine results. But this isn’t a typical cookie. Having some parts toastier and others tenderer will only add to the sense that you’re eating something made from [nuts!!!].
TRICK #8 (the post-oven danger zone): You must cool these cookies on the tray until they are only warm. They are ridiculously fragile when hot from the oven and will fall apart if you even stare too hard.
TRICK #9 (for a final touch): We aren’t big believers in fancy chancy “touches” like scenting the powdered sugar for a few days with a more-expensive-than-gold vanilla bean bought for the purpose. This is the one recipe where we sometimes make an exception. Not always… but sometimes. There’s something about the wafting smell of vanilla that seems to stimulate a person’s specialized nut detector taste buds. [What, you never heard of those? Try a proper Nut Horn and you’ll see the proof for yourself!]
TRICK #10 (coating with enough powdered sugar): Warm cookies are fragile cookies, but nevertheless you must coat them with the powdered sugar while they are warm because the outgassing steam helps the sugar to stick. Kate likes to move the whole sheet of parchment onto a cooling rack, dust them there, wait 5 minutes, and then turn them over to dust what had been the bottoms.
A buttery, hazelnutty, spice cookie dunker from the Netherlands’ Viking province that has become our new favorite to pair with a cup of good coffee.
It’s been a busy few weeks and that’s cut into cookie testing, but to be entirely honest the delay also comes from my perpetual fidgeting with a new recipe for the Ultimate Butter Cookie. Another batch or two and I’ll have it out… but all of that is just an explanation for why I’m now publishing something totally different. [Talk about a teaser!]
That butter cookie comes from the Middle East, so I wanted to go back to Europe for a break. Clearing my head for the final assault on the ramparts, as it were. I keep an ongoing “cookies to test” list, so I pulled that out and asked Kate to choose one that caught her eye. She marked off a few possibilities and then paused.
“Frisian Thumbs? What in the world?”
I smiled. “Top 10 on the great names list, right? It’s a Kate Cookie too – they describe it as a dunker made with a lot of hazelnuts and some spice. Butter too, I’m afraid.”
She asked to see the initial recipe I’d downloaded, approved both the idea and the incursion into her monthly butter allowance, and then waved me away to work some magic. Magic done; mischief managed. This is a good one.
By this point I have a pretty well established pattern for working on new cookies. I usually find them in the first place by wandering through the Wide World of Google doing searches like, “regional cookies of [This Place]” or “local baking traditions of [That One].” E-tourism, cookie style. Potential targets pop up in various locations: regional promotion sites, historians looking through an oddball lens, memoirs from tourists, émigré groups sharing memories, other bakers, etc. I mark down the names, look up the recipes, pick an example for the files, and then put it on the “cookies to test” list with a brief description for later.
When it’s time to look deeper – now, in the case of Frisian Thumbs – I go to the next level in two different ways. First there’s the recipe. I start on that by searching out every possible version the Internet contains. This gives a firm idea of the parameters, from which I’ll write something up using round figures and my experience with other projects.
Then there’s the heritage part. That also begins with the Internet, but then moves on to a search for actual humans who know the cookie, and can provide some context. Those two steps provide the raw ingredients: a recipe to work from, and a heritage to explore. And the follow up question: How did these two meet and marry? Which is where the test batches come in, followed by gifts for my sources to sample, and adjustments to account for both my reactions and theirs.
ASIDE: There are few things more satisfying than developing my version of a beloved childhood treat for someone who knew it back in the Old Country. Life doesn’t have many perfect moments. Two strangers exploring a cookie together, on truly neutral ground? That is one.
STEP 1: Creating the initial test recipe
So, off I went and found no less than eleven (11) different recipes, most with photos and descriptions. These taught me a lot about the underlying issues. Here are my notes:
Almost every recipe centered on anise, supported by a background of cinnamon and ginger in roughly equal amounts. What an odd combination! But it works. Taste it in your head and you’ll see.
The hazelnuts bring it all together. Remember to toast them first. No two ways about it.
The cookies bake low and long. 325° for 20 minutes. Interesting.
The recipes are pretty close together after normalizing…
“Are you done yet?” I stop and look up. It’s Kate, back from her writing lair. “Those thumb cookies sounded really good, and I’m going to make a pot of coffee. Any chance you’ll have some ready for me to try?”
My friends, this never happens! Don’t get me wrong. Kate is a certified cookieholic. She’s at least as bad as I am. But her addiction gets triggered by proximity. See cookie; snitch cookie; savor cookie; repeat. I’m the one who gets entranced by recipes alone. But not this time, apparently. I wasn’t the only one getting hit in the, Oh that sounds good! spot.
I promised to get her something in the next day or two and then went back to my research.
… normalizing to a single stick of butter (½ cup = 115 grams). Which is a surprisingly small amount; most cookies start with two sticks. Hmmm. One stick of butter yields three dozen cookies? Definitely on the small side.
The big difference between the recipes comes down to the amount of flour. Anywhere from 130 grams all the way up to 250. What’s going on? Aha! The ones with more flour have some baking powder added, and the ones with very little look slumped in the photos.
The flour also ties in with eggs. Most of the recipes include an egg on the inside, excluding only the ones with very little flour. The recipes with both an egg and some milk have the most flour. Got it.
But there is not a lot of flour in any of these by the standard of other dunkers! Corsican Canistrelli have a good 50% more in the flour to fat ratio, plus a cup of chestnut flour to push the ratio even further. The hazelnuts in this recipe are chopped, not ground. They won’t act the same. So Frisian Thumbs may be dense, but they shouldn’t be hard, dry, or crispy.
And that flavor profile! Hazelnut, hazelnut, spice, and hazelnut. That’s what we’re after here! No extract, either. Very elegant even if they are kind of plain to look at. One for bakers rather than bakeries.
Time to check in on the heritage part of the equation.
STEP 2: What’s the story behind all this?
I knew going in that the Frisians were a big deal back in the Viking Age. Why would someone name a cookie after their thumbs? It turns out there’s a string of small islands off that coastline called – look at that! – the Frisian Islands, with the modern Netherlands province of Friesland right in the sheltered area behind them. The name “Frisia” goes back to Roman times, but modern Frisians descended from a Viking tribe who settled in the region after a series of environmental disasters had emptied out most of the original inhabitants. These new Frisians (they adopted the name) were closely related to the Angles and the Saxons who settled in England, only to be conquered in 1066 by the even fiercer Norman Vikings who had settled 400 miles further south.
The Frisians who didn’t travel to England with their Anglo-Saxon cousins settled down and became famous for breeding the finest warhorses on the continent. Including, ironically, the mount used by William the Conqueror. Cool! Their descendants maintain a fierce pride about their particular roots, with a regional dialect that is sort of like ancient English mixed with Dutch and German.
There’s no way I was going to find a group of Frisian émigrés here in Pittsburgh! But as it happens, I do know several Dutch bakers through a Facebook group devoted to the molded cookie Speculaas. Thanks go out to one and all. And to Google Translate, since the recipes they pointed me to were written in Dutch, with notes on particular Frisian words.
The material from my online friends brought me up to 14 versions, which led to a modified master recipe while also answering a lot of questions. First, Frisian Thumbs are a true specialty of modern day Friesland, and little known outside its borders. They are not an extension of other Dutch cookies. Second, Frisians are a lot more culturally distinct than Americans would tend to assume. So much so that road signs in the province appear in both Dutch and Fries. Thus it may be inevitable that Frisian bakers will start from normal Dutch ingredients when it comes to flour and sugar (see below), but they wouldn’t be caught dead using a standard spice mix from the store with some extra anise.
Third, the photos make it clear that these cookies are supposed to be small, thick rectangles, not oblong drop cookies. They must have enough flour to more or less hold their initial shape. That evocative description about, “the size of a child’s thumb” actually makes a good deal of sense. Pull out the ruler and… ½” thick x ¾” wide x 1¾” long.
Just roll out a slab and slice to shape. Nothing fancy. This would be the point to put anise seeds on top as fair warning. They are all buried inside this particular dough.
Fourth, not a single one of the Dutch photos contain the thumbprint mark. They’re all plain old rectangles. But I did see several others that had the dent, and you know what? It creates an effect that looks a lot like a thumbnail! I like it.
What’s a thumb without its nail? They are just as tasty plain as they are with the decoration, but it does add a certain distinction. Press the shape in when they are warm and fresh from the oven. I was too eager and got a little burn. Next time I will press in using the tip of a spoon.
Next, I wanted a chewy and somewhat dense texture, but not excessively so. Too much flour, especially with no leavening, would make little bricks. Balance is everything.
Speaking of which, those websites convinced me that balancing the spices had to be job one. Spice cookies are normal for the Netherlands, and don’t raise eyebrows over there like they often do in the U.S. On the other hand, Dutch cooking incorporates a very particular sense of poise and restraint. It shows up on the savory side too. Any cookie that Amsterdam writers would praise as “well rounded” had to be exactly that. So my target was clear: a nut cookie enhanced by anise, rounded out with cinnamon and ginger, chewy but not dense, and ideal when paired with a cup of coffee.
DIGRESSION: A primer on Dutch ingredients
Dutch ingredients can differ a bit from their counterparts here in America. To start with, bakers in the Netherlands have a different type of flour than you find in the U.S. Here in America we grade our flour according to protein content, going from low protein cake flour all the way up to high protein bread flour. On the Continent they grade their flour more by grind size. The best Speculaas, e.g., use something called TIP-00 flour, an Italian product that is ground to a spectacularly fine level “similar to baby powder.” That doesn’t matter for the recipe at hand, but it makes a real difference for a molded cookie where the features are a key difference between good and better.
In this case we need to look at the sugar. U.S. sugar is made by grinding sugar cane to a pulp, extracting the sweet syrup, and then removing the impurities to leave only pure sucrose. Those impurities are sold separately as “molasses.” Mix a generous tablespoon of molasses back into a cup of white sugar, and you get “light brown sugar.” Mix in two tablespoons and you’ll have “dark brown sugar.” Half a tablespoon yields “turbinado sugar.” [NOTE: you technically want to add a little bit more than just 1 tbsp. Hence the word “generous.”]
Not so in the Netherlands. Their basic ingredient comes from sugar beets rather than sugar cane, which creates some differences you’d notice if the two were used side by side. Speculaas use something called “Belgian Brown Sugar,” which is made by taking regular sugar, gently caramelizing a small portion, and then mixing it back in with the white kind. Stella Parks published a great essay and recipe on this at SeriousEats.com if you want to learn more. This goes to the second part. Then there’s a “brown sugar” made by mixing some actual cane molasses into beet sugar; and another “brown sugar” that’s made by mixing in a cooked sugar syrup that Dutch families use on pancakes instead of the U.S. maple syrup. And…
You get the point. The recipes all used “brown sugar,” but those words don’t mean the same thing in the Netherlands as they do in America. Fortunately enough, I’m told by people who have lived on both sides of the pond that these differences are real, but subtle, and not enough to alter the cookie’s character. The recipe you’ll find below starts with white sugar and adds just enough molasses to turn it into turbinado (i.e., identical to splitting the total 50/50 between white sugar and light brown sugar). Add more molasses to taste. That’s a great spot to customize.
STEP 3: Putting it all together to reveal a Heritage Cookie
After all that research, and a second “are we there yet?” episode from my beloved, the final test recipe came together quickly. I stayed up late to make a batch for the morning coffee, went out of the kitchen to head upstairs… and discovered She Who Must Be Obeyed waiting on her reading chair, drawn from her aerie by aromas of cooking hazelnuts, butter, and spice.
“They smell fantastic. Are they ready for testing?”
“It’s a dunker. They’ll be all soft and warm fresh out of the oven. It won’t be a fair way to judge.”
She cocked her head and gave me a look. “How many are there in a recipe?”
“Almost 40 for this batch. They’re actually kind of small.”
“Then there will be plenty left for tomorrow, won’t there?”
“I don’t know. My photographer gets awfully picky…”
“I’m the photographer! If you want to see picky, keep standing there in the way and you’ll learn about picky!”
I am a man. A mere man. I can’t even translate that sentence. I certainly didn’t want to experience it. So I stepped aside. But I spoke up even as I deferred. “I do want to press in the thumbprints, at least on some. Then we can sample any ugly ones.”
And so we did. I burned my thumb. She did not. I will make my thumbprints with the tip of a spoon next time. She will laugh when I do. We both think it’s a great look. And yes, we both had a taste. And then another. They were goooooood.
FOLLOW UP NOTE: This batch produced 39 cookies. We ate two that evening, and Kate put ten aside for the photography. There were exactly 20 left when I put them in a tin for my “get thee out of the house” sharing trip. Cookie gremlins are a thing in this house, and I can’t exactly blame them. Frisian Thumbs are goooooood.
NOTE FROM KATE: Scott is a talented writer of fiction. I categorically deny all of the above. Except the part about Frisian Thumbs being good. No one can deny that.
Final touches
Some cookies take a ton of fidgeting. Others just slide together all on their own. This was one of the latter. They were great warm, and the ones that survived our gremlins were even better the next morning. I’m noshing on some of the rapidly vanishing leftovers even as I type. So there weren’t a lot of changes in this one. But there are a few decision points beyond the recipe that our readers should know about in advance. Knowledge is power, and this way so you can make your own choices.
About the flavor. The first thing you taste is a combination of butter and hazelnut, which changes as the ground spices kick in, and then changes again when you crunch through one of the whole anise seeds. It’s definitely a spice cookie, and definitely a hazelnut cookie, and definitely unique, but at the same time there’s nothing here that is “strange”. We’d have no problem serving this to kids of all ages with no warning other than, “Here’s a cookie you’ve never had before…”
About the anise. No warning unless you dislike anise, that is. Some people don’t. My Dad hated licorice flavor, and our younger daughter is lukewarm at best. This isn’t a cookie for them. No, it isn’t an “anise cookie.” Anise is just a component flavor. But it is a clear and distinct component.
About the whole anise seeds. Whole seeds can get annoying when a hull gets caught in your teeth. It isn’t a big issue, and the whole seeds add a late-hitting pop of flavor that both of us enjoy. Kate would keep them buried inside the dough for the sake of simplicity, but you’ll see that I wrote the recipe in a way that sets half of the seeds aside to be placed on top. I only did it to remind people that the seeds are there, and to expect the occasional chewy bit. Don’t bother if you don’t care.
About the salt. Using the full ½ teaspoon of salt gives this cookie a noticeable kick that I enjoyed, but I also prefer salted nuts over plain for eating out of hand. The salt level might have been a tad too much on some other day. The cookie definitely wants some salt to bring out the nutty flavors, but we recommend that you start at the ¼ teaspoon level and then add more as desired. Make a note of your preference on the recipe.
About the coffee, and the dunking. I have never come across a cookie that pairs better with a cup of coffee. Not ever, anywhere, in any form. It’s almost stunning to write those words, but they are true. At the same time, it really is a “pairing” rather than a case where one side needs the other. Together, they are more than the sum of their parts. But separated they’re still darned good on their own. Frisian Thumbs have enough texture to hold up well as a dunker, but they also have enough moisture and suppleness to eat out of hand. Frisian Thumbs are goooooood.
That’s about it. Sorry to run on so long. This is an easy cookie to fall in love with, and both of us have. Head over heels. Let us know if you feel the same.
Here’s a recipe that many readers will know because they’re very popular at certain upscale supermarkets. They have a lot going for them, starting with a very pure hit of almond and chocolate (two flavors that love each other), and continuing through the fact that they’re naturally gluten- and dairy-free, which suits a lot of people with allergy problems. They’re also very generous and fairly dense, which means that anyone who’s willing to eat baked goods for breakfast can get by with 2-3 “cookies” and a nice cup of Joe. (A meal that Scott has admittedly enjoyed as he writes this article).
The other benefit? Almond Horns are incredibly easy to make. The basic recipe couldn’t be simpler: (1) put everything in a food processor; (2) grind until it’s smooth and begins to ball up; (3) roll each piece in almond flakes; (4) bake; and (5) coat the horns or bottom with chocolate. The issues come up with the little details that avoid problems in each of those steps.
Classic European recipes typically start from Marzipan rather than working from scratch. Take Marzipan, blend with egg white(s) to loosen it up, and you’re basically there. Convenience cookies!
Marzipan isn’t as common in the U.S., costs more, and the different brands vary in their dedication to the cult of sugar and the abuse of almond extract. We’re also a nation of people who take food processors for granted. That tool converts Marzipan from a painstaking process into a walk in the park. If cooking from scratch is cheaper, just as easy, more reliable, easier to customize, and lets you use a little less sugar… well, cook from scratch.
We recommend that you start from almond meal because it’s become easy to obtain and ensures an even result. In the ideal world you’d start from whole or slivered almonds, toast them to enhance the natural flavor, and then reduce that to meal during the grinding stage. It’s really up to you. But from one obsessive purist to another? We just get the almond meal. There’s so much extract that the toasted flavors get lost anyway.
Step 2 – Grind It All Together
No kidding. That’s all you do. Put everything in the machine except your coatings (toasted almond flakes and chocolate). Process, scraping down the sides from time to time, until it comes together into a smooth dough that balls up around the blade. Done.
That couldn’t be any easier, but there are a couple of tricks because the dough changes with (a) your kitchen’s level of humidity, and (b) whether your Size Large egg whites are closer to mediums or closer to XL’s. They do vary within each range you know. It isn’t a problem if you keep the target in mind: “Dough that is just sticky enough to hold on to toasted almond flakes.” Too sticky and it will be hard to work with. Not sticky enough, and your coating will fall off.
If the mixture seems too dry, add in a tablespoon or two of water. This happens every time in a Pittsburgh winter, and almost never in our summers.
If the mixture seems too sticky, add a tablespoon or two of extra almond meal. Extra sugar works too.
For what it’s worth, we actually get a kick out of these little customizing steps. They make you feel like a Real Expert who knows the secret cheat codes.
Step 3 – Form the Cookies
A few of those secret codes make this step just as easy.
First, remember that most people make these into large, 4-5 bite treats. One recipe makes enough for 16.
That’s not a coincidental number because you get there by cutting the dough in half, then each half in half four more times. If you want smaller cookies, cut the initial piece into thirds and then proceed to halve them over and over. This will get you to 24.
Second, moisten your fingers and palms before you form each cookie. That will keep the dough from sticking to your hands, and also make the surface extra sticky for holding on to the almond flakes.
Better for you, and better for the cookies. You’ll need to wet your hands several times before all the pieces are done.
Third, form each dough piece into a short, fat sausage as thick as you want the middle of your cookie to be. Roll this cylinder in the toasted almond flakes, and then pinch/curve the ends to form your crescents.
Cylinders are much easier to work with at the rolling stage, and you do not need as many almonds on the ends anyway because they will get coated in chocolate. Yes, you can just make a longer, thinner sausage, roll that, and bend it into a “C”. And yes, it is a little easier. We prefer the result from crescents; they have a texture that changes as you nibble through each treat.
Step 4 – Bake
375° is just hot enough to get a few of the almond edges extra toasty. A nice touch. As always, you want to pay more attention to the aroma and texture than the actual time. They are done when the sides firm up and the bottoms have turned light tan. It’s taken a consistent 13-14 minutes during winter testing in Pittsburgh with our oven. Please let us know if your results vary.
We also encourage you to spin these trays halfway through to ensure even cooking. We usually don’t bother with that step unless we’re dealing with wafers, but for some reason this particular cookie has been sensitive to hot spots.
Step 5 – Coat The Ends With Chocolate
This is the one step in the recipe that creates some frustration. It comes from two problems. First, the chocolate will harden as you work. And second, you won’t have enough chocolate in your bowl to dip the ends when you get to the final few pieces.
You can solve the chocolate problem by either keeping it warm, or mixing in a small amount of fat. Scott prefers to use plain chocolate, and to keep zapping it in the microwave for very short bursts as needed. Kate, who did most of the chocolate work, suggests adulteration – with the proviso that you must use a fat hardens at room temperature. Coconut oil works but can add some flavor. Palm oil has the perfect characteristics but is evil. She recommends butter or shortening, whisked into carefully melted chocolate. The resulting chocolate dip can be extra sticky at room temperature and takes a while to harden, but it is easy to work with. Your mileage may vary.
The problem of what to do toward the end takes more thought, and there is no really good answer. If you want the chocolate supply to end on the final cookie, you won’t have a deep enough pool to actually dip the horns of the last two pieces. Careful work with a spoon can help, but it only goes so far. Either you need to make/waste extra chocolate dip, or you need a cheat code.
Scott offers this workaround: Don’t try to dip the horns on your final piece. Coat the entire bottom with your leftover chocolate instead. It creates a more uniform treat that’s a little less intriguing to look at, and has less contrast when you eat it, but it does have the benefit of giving every bite a little bit of chocolate. Brush all the bottoms with chocolate if you want every piece to be the same.
Kate rolls her eyes at that, and says you should simply make enough chocolate to do the dipping and be done with it. Leftover chocolate makes a wonderful ganache for truffles, or at worst a special treat for the first lucky passerby who can find a spoon. You, of course, are a mature, fully grown person with the control to inflict those treats on someone else.
Cool and Consume – The End… Until Next Time
Mandelhörnchen are one of those treats that become a baker’s signature offering to guests, grandchildren, and drama club bake sales alike. They’re so very big! Physically, in taste, in texture, and in ways that combine to be more than the sum of their parts. They also show off familiar flavors, and they taste exactly like people think they will, except more so. There’s a huge bang for the buck factor.
Plus they’re fun and easy to make once you internalize those little tricks: keep your hands moist when shaping, and accept that there’s only so much chocolate to go around.
This is a marvelous recipe that you’re really going to treasure. We Promise. Enjoy!
Dairy free and delicious, what’s not to like? This cookie offers a unique flavor profile with a familiar texture somewhere in the slice-and-bake or sugar cookie range. The only unusual part is that title ingredient, and it’s hard to say if that’s really as unusual as people might think.
EXTRA SPICE FROM KATE: Okay, hold the phone a sec. My husband’s version of “health consciousness” can sometimes cross that invisible line into “health denial”. Before he continues with his ode to pork, we want to point out that butter, lard, shortening, and all other saturated fats – even bacon [gasp!] – are not good for you. That is why cookies are a treat, not a staple. No, bacon fat is not really any worse than any of those others. Granted. And he’s 100% right that using it in a cookie is “heritage”, not “weird.” But keep the enthusiasm within bounds. This is not a physics-optional cartoon. You may proceed now.
Animal fat cookies are rare in the U.S. for sure, but that is only because we tend to be wastrels and spendthrifts when it comes to leftover fats and drippings. Americans are taught to reach for purified, store-bought fats like butter, shortening, or oil when we cook. But that isn’t how things used to be, and it still isn’t true in professional kitchens, or in any part of the world where people pinch their pennies a little harder. In all those situations the leftover fat, such as drippings from a roast, are treasures to be saved, not waste to be discarded.
That isn’t to say we are holier than thou. Here is our standard method for cooking a few slices of bacon for breakfast: place some slices between several layers of paper towel, put it on a plate, and then microwave for 3-5 minutes until it gets as crispy as you desire. Discard the dirty paper with all that nasty fat.
What could be better? Well… how about not wasting all that fat, which qualifies as “delicious” rather than “nasty” in any other time and place the world has ever known? Sorry to sound all rant-y. We really do use that paper towel trick, and it really does work. But it is sort of embarrassing when you consider the historical perspective.
Try this method instead if you have any plans to use the fat: Place your slices on a rack over a sheet pan (spiced as you desire) and then cook at 375° for 15-25 minutes, depending on thickness. Remove it from the oven when it hits the level of crispy you want. Flip it over with tongs to let any pooled fat drain into the pan, give it a minute, and then dab off any excess with paper towels. Super easy, super reliable, and nothing wasted. Use 350° for chewier bacon and 400° if you like more char and you’re willing to tolerate the smaller safety window before it burns.
FWIW, this recipe calls for ½ cup of fat, which is almost exactly what you ought to get from rendering a pound (16 oz = 450 g) of bacon. A perfect and thrifty method for whenever you cook large amounts of the Mighty Meat.
Comparison Recipes
Lard (pork fat) has to be the most common cooking fat after butter, shortening, and oil. U.S. bakers adore it for pie crusts and biscuits, but lard is also a common cookie ingredient; particularly in Spain and its colonies. Bizcochitos, the official state cookie of New Mexico, are the best known variation (see Heritage Cookies Volume I for a recipe). Spanish Mantecados are a more tender example, with less flour and a little olive oil added to create a shortbread effect. Look for it in a future blog post.
Bacon fat is nothing more than lard with extra flavor, if you look at it sideways. We’ve found many more savory recipes than sweet ones in the world of baking. There is a German side dish called Speckkuchen (“Bacon Cakes”) that’s definitely worth your attention if this interests you. But this article is about the sweet-treat, cookie versions and they are far from unknown.
Tricks From The Testing
This turned out to be a pretty easy recipe to perfect because the U.S. went through a bacon craze in the early 2000’s that hasn’t quite ended yet. People put bacon into everything! And it kept working, and working, and working again. As we write this in 2021 there’s still a common catchphrase that, “Bacon makes anything better.” Okay, that probably isn’t true in the literal sense. Spend some skull sweat and you’ll probably come up with a few exceptions. But the basic point stands: savory, sweet, smoky bacon goes very well with many, many flavors.
We chose a classic Christmas Ham profile for this recipe, using cinnamon, ginger, cloves, and molasses. Feel free to substitute any spice combination you prefer. Just think of it as something to go with bacon rather than something meant for a cookie. Scott is halfway toward eager in his plans to make an extra-smoky version the next time we have a big, multifamily brunch to cook for.
The cookie itself was just as easy. The proportions fell dead center into the classic sugar cookie and drop cookie range. All those charts in the book before each section? This is where they come in handy.
The only complication came with narrowing things down to the final version. The two of us went back and forth on whether the cookie wanted a bit of crispy texture inside. Translation: should we add in some of the bacon we had lying around from rendering the required fat? In the end we left that as an option. Somewhere around two slices (≈ 2 tbsp. when chopped) is the right amount if you have any extra.
But it still wanted… something. That turned out to be bits of dark chocolate.
You’re in good company if that sounds wrong. Kate almost blanched at the idea, and Scott has to agree that chocolate-dipped bacon was one of the less appealing excesses of that bacon-with-everything craze we described above. But this is the exception that proves the rule. The chocolate sort of widened the flavor, while the textural contrast rounded things out.
CONCLUSION
Bacon tastes good. Everyone agrees on that, and we waste never a crumb.
Bacon fat also tastes good. Everyone agrees there as well, but we waste it all the time. There’s no good reason to do so, and people never did until we grew wealthy enough to value convenience over thrift and flavor. This is just one example of the many, many ways in which you can use this overlooked treasure to make treats that will make you sound like a genius. We’ve seen, tasted, and enjoyed Bacon Ice Cream! If that can work, and we assure you it does, then cookies should be taken in stride.
Welcome to another long article, which obviously has nothing to do with a wordy author! Blame it on the combination of a stubborn cookie with a lot of noteworthy tricks and a fascinating tradition. Many thanks up front to Mirtill Seregély, a wonderful Hungarian baker we met online. She’s the one who published the original recipe, and was also our primary source for the cultural background that turned out to be essential. Essential how, you ask? Read on.
Kvircedli are a rectangular molded cookie that holds good impressions but does not get painted. It has a texture like crunchy, chewy sponge cake, a nondescript but familiar flavor that will remind you of Ladyfingers, and an almost unique profile that displays the airy texture inside.
The recipe itself looks like a classic Springerle, with 25% less sugar, lemon as the only flavor, and a whopping eight (8) times as much leavening. That last is the elephant in the room. Normal Springerle use ¼ tsp. of Hartshorn (baker’s ammonia) for a 3-egg recipe. Kvircedli use a full 2 tsp. Two bloody teaspoons!
That enormous amount of leavening requires a lot of little tricks that are hard to figure out, but turn out to be directly tied into the cookie’s history, tradition, and purpose. So let’s turn to that front next.
Kvircedli History and Tradition
The former Hapsburg Empire left behind a legendary baking tradition that pervades every one of the varied corners to which it stretched. Austria in the west? Ever hear of Linzer Torte or Strudel? The Balkans to the south? They may hate each other politically, but the Serbs, Croats, Bosnians, Albanians, etc. have nothing but respect for their respective baking traditions. We do too. The Czechs and Moravians to the north? Awesome. The Czech tradition of Christmas sweets is what drove us into baking so much in the first place.
The Hungarians in the east use their ovens every bit as well as their former Imperial brothers. Kifle, the filled flaky cookies we called “Kolachy” in Volume I; the famous Esterhazy and Dobos Tortes; the Poppy Seed and Nut Rolls found in half the bakeries of America; it doesn’t end. You could probably list dozens of absolute classics that either originated in Hungary, or get made there with as much skill as anywhere else in the world. So when we saw the Hungarian folk recipe that Mirtill shared, which took dense Springerle in a whole new direction, we knew it had to be tried.
You’ll read below about some of the troubles we ran into. Those didn’t get solved until Scott went back to Mirtill for a lesson on the sort of deeper background that often gets ignored.
The Hungarian people go back to Attila the Hun and his wild tribesmen from the vast steppes of Asia, who invaded the Roman Empire more than 1500 years ago. After his death in 453 they settled in the land that now holds their name, with a hauntingly beautiful language related to nothing in Europe except Finnish and Estonian – three peoples who trace their history to common roots.
The following centuries naturally saw a certain amount of immigration from other areas, leading to several ethnic minorities. The German speaking minority are known as the Swabians. Somehow or other that group (apparently with the aid of some Capuchin monks) became big players in Hungary’s winemaking tradition, with a major center in the little town of Mór (population just under 15,000). Think of it as Hungary’s Napa Valley and you’ll get the idea. Other parts of the country specialize in rich reds or aromatic dessert wines, but Wikipedia says that Mór built its reputation on dry white wines similar to sauvignon blanc.
Kvircedli come from that region, and are best known as the treat that winemakers still serve whenever they open the doors for a tasting. Mirtill tells me the cookie also shows up at special events like baptisms and weddings, though I suspect you’d get into a flavored variation for those. More on that below.
Stop and think about that. This is a cookie designed and used in a small, immensely scenic part of the world for the express purpose of vineyard tours and celebrations. How awesome is that? But it gets even more quaint if you dig a little deeper!
Legend holds that local women created Kvircedli to get around a religious law that forbade them from drinking alcohol. Some ignorant, almost-certain-to-be-a-bachelor really thought he could expect the wives to sit there quietly while their husbands quaffed the local product? Even in the name of religion? Riiiight. The ladies responded by inventing this airy, absorbent cookie with a convenient rectangular shape that’s perfect for dipping into unwatched cups. And thus they “ate” their wine instead of “drinking.”
Problem #1: Is the Flavor Too Bland?
The history and the story both help to explain the flavor profile. Traditional Springerle have a distinct anise flavor that would not play well at a wine tasting. Okay, it might not ruin a well balanced red, but after that you’re looking at some potentially ugly clashes. A nice, neutral, vaguely lemon profile makes a much better palate cleanser, while the crunchy-outside, chewy-inside texture does the exact same thing for your other senses. It makes an ideal reset button to move from one wine on to the next.
Pretty smart! If you’re a winemaker looking for a fantastic and proven gimmick, I offer Kvircedli at your tastings as suggestion #1.
But what if you just want to enjoy the cookie in some other setting, like one of those baptisms or weddings? Well… I guess my first answer would be that celebrations often involve a lot of wine, and simple flavors tend to be child friendly too. That said, it takes almost nothing to jazz this dough up with whatever interesting flavor may strike your fancy. We did a batch with that Croatian flavor profile that worked so well for Dalmatinski Rafioli, and the result was all we could have hoped. A spice or spice mix would work every bit as well. We put a suggested proportion in the recipe if you’re interested.
Problem, if it is a problem, solved.
Problem #2: Stick to Simple Patterns
Springerle taste fine and all, but their glory lies in the amazing, painted patterns that make people say, “That looks way too pretty to eat.” We went into this hoping that Kvircedli would offer a similarly result with a very different texture. The strong similarities between the two doughs gave us a lot of hope in that regard. But it was not to be.
It took a while for us to realize it, but that rectangular shape is actually a major key to making this recipe work. Here is a picture of some traditional molds that Mirtill Seregély shared:
Traditional Kvircedli molds carved from alder wood by Zsuzsanna Sáriczné. Note how the impression comes from rolling the mold over the dough, side after side so the pictures end up touching. Read below and you’ll see why that matters! Photo Copyright 2021 by Mirtill Seregély, and used with permission.
A fuller array of traditional Kvircedli molds carved from alder wood by Pál Szarka. Photo Copyright 2021 by Mirtill Seregély, and used with permission.
That rectangular shape turned out to be important.
Kvircedli stray from the traditional Springerle recipe by using an enormous amount of leavening to change the internal structure of the cookie. That’s the core idea. When we used a traditional Springerle mold, particularly a round or oval one, the pretty surface patterns held up fine, but the cookies baked into uneven, lopsided shapes tilted as much 45º off the level. The dough puffed out randomly to one side or the other, and occasionally even heaved up unevenly in two directions.
Looking a little closer at the recipe, and at some online examples, showed us why. One of the key features in a traditional Springerle is the “foot”; a ring of dough that puffs out sideways as the cookie bakes. You see this in Macarons too. The cause is the same in both cases.
First you make the dough, and then you stamp it with a fancy shape (for Springerle) or pipe it into a neat circle (for Macarons). Then you let the dough rest on a counter long enough for the outside surface to dry out into a skin. That takes up to 2 hours for Macarons, and 24 hours for Springerle. The bottom, of course, does not dry out because it is resting on a parchment-lined tray.
When you bake the cookies, the bottom and core expand as heat turns the inside moisture to steam and/or activates the leavener to create CO2. Boyle’s law says those gases need to expand in all directions, but they can’t go up through the dried out skin. So instead you get extra vertical rise combined with a horizontal “foot” as dough puffs out around the bottom.
A pan of Kvircedli packs the unbaked cookies close enough together (about 3-4 mm = ⅛” apart) for the feet to expand into each other. The result? Each cookie supports its neighbors from the side as they bake, adding a little extra rise while also preventing a tilt in any one direction. Have a look at the tray we just baked for this article (stamped with the bottom of a crystal knife rest since we do not have one of Mirtill’s special molds to play with):
There are would-be cookie molds all over the house!
Soldiers lined up and waiting to bake. The dough is ⅜” thick after being impressed with the pattern, but not yet puffed.
That is the raw dough.
See how the ones on the edges have puffed unevenly?
Here is the same tray after baking. The middle ones are nice and even, and have puffed from ⅜” up to a full 1″ tall. The ones on the edge got lopsided because they received no support from a neighbor. FWIW, this process also results in a much desired side effect. Kvircedli are a breakaway cookie that gets divided along those pre-inscribed lines. Doing so reveals the uniquely beautiful, spongy interior.
Showing the lovely sponge cake texture. Just be careful not to crush your pattern as you pull. There is nothing wrong with using a paring knife to get things started if the cookies started out a little too close together on the tray.
Problem #3: The Evil, Cat Box Reek of Ammonia
But there was a joker in the deck! Hartshorn a/k/a Baker’s Ammonia a/k/a Ammonium Carbonate has yet a fourth name: Smelling Salts. And believe me, they work. If your beloved should ever swoon in a most dainty and Victorian manner, one good whiff of this stuff will quickly break the trance. Blech.
Hartshorn works by breaking down into ammonia and CO2 when heat is applied. The ammonia bakes off, while the CO2 provides the lift you wanted. Springerle recipes use a bare ¼ tsp. in the dough for this volume of eggs and flour. That amount is enough to leave a foul odor in your oven, but won’t be noticed in the final product over all the anise notes. With Kvircedli you’re going to wince when you open that oven, and sticking your nose against the fresh baked cookies is not going to make you happy. Indeed, our recipe contains an explicit warning to avert your face when you open the oven door in order to spin the baking trays, and again when you remove the finished cookies.
Some of that odor inevitably ends up embedded inside the freshly baked cookies.
We tried a few things to fix this. First, we cut the amount of Hartshorn in half and replaced that portion with baking powder at our normal ratio of 3 parts baking powder for every 2 parts of Hartshorn being replaced. That was the batch mentioned above, where we used proper Springerle molds in addition to the knife-rest stamp. The results were uneven. Literally, and sometimes at multiple angles inside the same piece. It was worst in the oval Springerle that baked in splendid isolation, but the phenomenon showed up in the square ones we baked side by side. So was it the leavener? The much more varied heights of the pattern? Or the fact that “side by side” meant a nice even row, that only gave mutual support from one direction?
Bad science on our part. There’s a reason why each experiment is supposed to change just one variable at a time. But we never did go back to finish looking at the baking powder idea because Scott paused to talk things out with Mirtill. Understanding the history and purpose showed us that Kvircedli were meant to have a plain flavor, and were not a medium for display like Springerle. Who would want to dunk a work of art? Why would you want to distract from the wine you’re sampling, especially with brilliant, jewel tone colors? And if the rectangular shapes served a purpose, that meant the breakaway effect was deliberate, which in turn…
You get the idea. It made more sense to go back and tweak the original recipe for a longer and lower cooking time that would allow more of the ammonia to bake off. The original formula called for 12-20 minutes at 150º C, which is 300º F. We reduced the temperature to 140º C (285º F) and left them in for the full 20 minutes, with a single spin at the 10 minute mark to avoid hot spots.
The result? A little less browning on the bottom, which we missed; the exact “yellow” tone described in the original recipe, which hadn’t been as clear before; and about 75% less reek in the finished cookies. Mirtill had also told us that she’d done enough casual research to believe that true Swabian bakers solve the problem by simply leaving the baked cookies out for a day before eating. Another baker we met on Facebook confirmed this in a conversation with her Hungarian relatives. And, sure enough, that leftover smell disappeared after a few hours on a plate once we instituted the longer baking time.
And so you get to our final verdict. For shaped and dried cookie dough that is ⅜” thick (1 cm), bake at 285º for 20-25 minutes instead of 300º for 12-20. That’s what you will read in the recipe below. Here in private we will use less exact numbers and call it 18-30 minutes to allow for differing ovens and the erratic moods of your household gremlin. We’ve all got one, and they have a notoriously fickle sense of humor.
Problem solved, at least well enough to avoid going back for a new round of experiments with baking powder. Let us know if you try it, however. We’d love to know if there is a way to make the recipe without having to deal with the cat box effect.
Conclusion
Molded cookies involve some extra work to stamp in the pattern, and extra time because of the overnight rest that forms the required crust. There is no getting around that. But if you are the sort of baker who likes that process, we 100% believe you ought to give Kvircedli a try. That goes double if you happen to own a traditional 4-sided mold that can be conveniently rolled-and-pressed, or one of those patterned rolling pins. Either way, target a final thickness of around ⅜” (1 cm), which will be cut into nice, rectangular shapes. Array on your tray about ⅛” (3-4 mm) apart, and then proceed.
Go with the classic lemon-only profile if you want to stick with tradition and/or use them in a wine tasting. They really work that way, and yes: we did make sure to test it. [Hic!]
Add flavor to the dough if you want to eat the cookies out of hand (although Kate loves everything lemon and adores them as they are). Flavored, eating cookies can be painted if you so desire, but do not do so for dunkers. It serves no purpose and distracts from the wine itself.
Here is a final photo, and a copy of our updated recipe. Enjoy!
It should be a dry white wine, of course. That is the style of Mór, whence Kvircedli originate. The photographer, however, responds that white wine may be correct, but red is more photogenic. That answers that.
A few weeks ago we put up recipes for a set of Five macaroon recipes we’ve come to love – which could be described as marzipan suspended in an egg white meringue. And I am now working busily on a special Croatian treat called “Rafioli”, which is basically an exotically flavored almond paste wrapped inside a baked pasta shell. [Watch for it!]
German Mandelhörnchen and a Greek equivalent are also on our “to do” list after Kate took an initial whack that wasn’t too far from being right. Those amount to a flavored marzipan or almond paste lightened with fewer egg whites than you’d use for macaroons, formed into crescents, and then dipped in chocolate.
EXTRA SPICE FROM KATE: We have no photos of the Mandelhörnchenbecause a mysterious cookie gremlin raided our kitchen before anyone could add the chocolate part. Skeptical spouses have wondered if “doppelganger” would be a more accurate than “gremlin,” but I admit nothing. He only suggests such an outrageous thing because they are one of my all time favorites, both tasty and toothsome.
And then… Hmm, a pattern seems to have formed! There are dozens, if not hundreds of baked goods that start with a base of finely ground almonds, mixed with sugar and then bound by egg whites. That ingredient is called “Almond Paste” when left loose and a little sticky, or “Marzipan” when you add enough extra almonds to make it solid. We wouldn’t call either one a “cookie” all on its own. Commercial marzipan is actually marketed more like a not-so-sweet candy (with U.S. and English versions tending to be sweeter), and almond paste as an ingredient. But the basic concept gets used in so many true cookies, and cookie fillings, that it’s only fair to give our version.
Besides, Valentines Day is coming next month and you can earn some major brownie points by making Marzipan treats and forming them into something romantic. And double-besides, the recipe for Marzipan contains a couple of surprises that many people don’t know about, such as a tiny bit of rosewater used as everyone’s not-so-secret ingredient.
Anyway, here is our version for your education, edification, and delight.
This lovely photo of Marzipan Fruits at the Nuremberg Christmas Market is available under a Creative Commons Attribution-Share alike 2.0 Generic license at the linked website. Many thanks to the author for his or her generosity.
