We talk a lot about visualizing mathematics, and we can even listen to it sometimes. But it can be hard to get the other senses involved, especially taste. Last year, I was delighted with Andrea Hawksley’s tasty and attractive Fibonacci Lemonade, which makes the Fibonacci numbers and golden ratio tastable. Her post about Fibonacci lemonade starts like this: “How would one make mathematical cuisine? Not just food that looks mathematical (like math cookies), but something that you truly have to eat and taste in order to experience its mathematical nature.”

I recently ran across a similar idea from Nathan Yau at Flowing Data. “Data plus beer. Multivariate beer.” (By the way, if you don’t already follow Flowing Data, you probably want to rectify that immediately.) Fibonacci lemonade has two variables: lemon juice and sugar. Beer has a few more degrees of freedom in the types and amounts of grains, hops, and malt.

Many of Yau’s data visualizations involve maps and demographics, so it’s not a surprise that for his first foray into mathematical libations, he chose to make beer recipes based on statistics such as the ethnic makeup, population density, and education levels of different counties. In the end, he brewed batches that represented Aroostook, Maine; Arlington, Virginia; Bronx, New York; and Marin, California. He writes:

Here’s what I eventually settled on.

1. Population density translates to total amount of hops. The more people in a county, the hoppier the beer tastes.

2. Race percentages translate to the type of hops used. For example, a higher rate of white people means a higher percentage of the total hops (determined by population density) that are Cascade hops.

3. Percentage of people with at least a bachelor’s degree translates to amount of Carapils grain, which contributes to head retention.

4. Percentage of people with healthcare coverage translates to amount of rye, which adds a distinct spicy flavor.

5. Median household income translates to amount of Crystal malt, which adds body and some color.

Did it work? Yau didn’t run a randomized control trial, but he says the beers definitely tasted different, and he had some tasting notes notes relating to the population density, healthcare coverage, and median income of the counties the beers represent.

I am coming to this idea from the point of view of a mathematician rather than a data journalist, so something I love about the idea of multivariate beer, Hanna Kang-Brown’s census spices, and other data gastronomification, as Tom Levine calls it, is that it is a natural way to explore the idea of dimension without going the *Flatland* route. (*Flatland* is great, don’t get me wrong, but it’s good to have extra tools at our fingertips.) It seems that most practitioners are interested in the way such concoctions can help people understand real-world data, but I like the potential for use in the strictly mathematical realm. Who knows? Flavors that represent shapes or polytopes? Could you taste the prime factorization of a number?

Yau says he is out of the multivariate brewing game, but if anyone is interested in doing an experiment in mathematical flavor, I’m a willing and able taste tester.