Zendo is one of my favorite multiplayer games. It was even before I played it. The reason is that I am a scientist, and Zendo is essentially a game about the scientific method. In this review, I discuss how Zendo works as a model of the scientific method.

When a child first learns about science in school, he usually doesn't actually learn science. Instead he is given a collection of trivia, and is told that the collection = science. As a result, children often come to believe that science consists mostly of the rote memorization of unchanging facts, which can be a turn off. It was for me-- I didn't appreciate science until I was older and began educating myself outside of school. Only then did I realize that the essence of science is not a bunch of facts, but rather a method by which to construct ever more accurate descriptions of observable reality. Here was a systematic way to alter my perception of the world in interesting and useful ways. How cool! The idea that by virtue of my own efforts, I could wake up tomorrow in a world that seemed fundamentally different and clearer than the one I live in today, was intoxicating, and after it insinuated itself in my head, I never looked back. I wonder how many others would have caught the same fever had their first exposure to science not been in the stultifying context of gradeschool.

Enter Zendo: the scientific method in a box.

In case you don't know the game, here is an abridged overview, in which I've taken the liberty of re-theming the game as an exercise in the scientific method:

Let's say we have three players.

1. To begin, one player (let's call him The Universe) secretly invents a law of nature.

2. Then, the other two players (let's call them scientists) take turns doing experiments.

3. Each scientist sets up an experiment. The experiment takes the form of an arrangement of pieces on a table. The outcome of the experiment is either a black or white stone placed in front of the experiment, according to the secret law of nature.

4. As the game proceeds, a bunch of experimental results build up on the table in the form of black and white stones. The more stones there are, the more information the scientists have about the law of nature.

5. Finally, scientists can earn the opportunity to make guesses (hypotheses) about what the law of nature is. When a scientist states a hypothesis, the universe must then spring upon him an experimental counterexample that disproves it, or else that scientist wins.

There is a little more to it, but this is all you need to know in order to follow my points below. In summary, scientists do experiments, observe the results, and based on those results, generate hypotheses about the law of nature, which are disproved if they're wrong.

The sequence of events mimics the real scientific method well (with one important exception, to which I will return at the end). What is nice is that a whole bunch of issues that pop up in real science also emerge in the game. Here are four:

1. The need to state hypotheses unambiguously-- Sometimes, a scientist will state a hypothesis that is not perfectly clear. In this case, the universe must ask for clarification in order to figure out whether it is possible to concoct a counterexample. This is of course one of the central problems in doing real science, and one over which scientists argue endlessly: how to describe a hypothesis that is actually testable? The game therefore provides a forum in which to practice the kind precision in language needed to do science. Awesome.

2. Superstitions based on spurious correlations-- Sometimes, because of the scientists' own choice of experiments, a pattern of white and black stones will build up on the table that all conform to one incorrect hypothesis about the law of nature. This is exactly how real scientific fields get stuck in conceptual ruts. And, just like in real science, the way to get out of a rut is to find a counterexample, at which point the whole field (all the scientists) undergo a "paradigm shift". Paradigm shifts can also happen when new investigators without the usual biases enter the field, who can interpret experimental results in a new way. This is why it is said that science proceeds by retirements (ie the older biased scientists retire and make way for new and differently-biased ones) In Zendo, the same thing happens when somebody who's not even playing walks by the table, glances at the experimental results, and points out a hypothesis that nobody has been able to see through their blinders. This is a wonderful, clear illustration of the value of fresh perspective.

3. The value of simple, systematic experimentation-- In Zendo, it helps if scientists do experiments in series, where each experiment differs only slightly from the last. This allows scientists to quickly pinpoint those variables that are important for determining the experimental outcome (white or black stone). Scientists also benefit by minimizing the number of overall variables in each experiment, so as to minimize the chance for spurious correlations as described in point 2 above. These are also essential practices for real scientists as well.

4. The value of Occam's Razor-- Scientists quickly realize that their hypotheses should be as simple as possible, because then it is easy to interpret the counterexamples that disprove them. The more parts a hypothesis has, the harder it is to figure out from a counterexample what part is wrong.

That all of these issues arise in a game of Zendo makes it a fantastic simulation real science. It actually feels like science. Most importantly, the Eureka moments that come from paradigm shifts have a little of the same wonder-inducing qualities that real paradigm shifts have.

I alluded earlier to a way in which Zendo fails to replicate real science. Here it is: in real science, the universe does not magically construct counterexamples to your hypotheses for you, and nor does it tell when you when your hypotheses are correct. As a result, a real scientist can NEVER, EVER know with certainty that any particular hypothesis of his is correct. There might always be a counterexample just around the corner, but you might be too stupid to find it, and so you may be stuck in ignorance forever. If there is one thing that's frustrating about science for me, it is that.

The fact that Zendo does not embody this possibility is actually to its credit. Zendo is a simulation of all the good stuff about science, with the bad stuff left out.

Because of all this, if I had a child, and I wanted to teach her science, I would play Zendo with her A LOT. If you have a child to whom you would like to teach science, you should too. If you need to throw out all those text-books to make time, do it. The facts are fish. Don't give your kid a fish. Instead, teach her how to go fishing.

While I doubt I will convince people to switch from "has buddha nature" to "fits the experimental results" any time soon, the point is interesting in that the thought processes needed for Zendo *do* mirror those required for scientific inquiry.

If only the universe would hand out green stones.

Nick, THANK YOU!!!! As a homeschooling mom who already owns this game AND runs a game night for the kids once a month, I'm thrilled to be given this perspective and plan to introduce this game in this manner at our next game night now! Thanks!

Great! It makes me very happy to read this.

Now that I'm thinking about it, if the designer wanted to do some real good for the world, he could retheme the game, and sell it to schools as as a learning product (one that actually improves education!).

One note:
In order to make the game easy for kids to begin with, do the following:

1. have the universe choose the simplest rule he can think of.
2. at the beginning of the game, describe just a couple of possible rules to the scientists, tell them that the real rule will be one of them, and their job is to figure out which it is. This will give them a direction that they would otherwise have to grope around for.

I like milomilo122's theme a lot more than the one in the rules.

I run a molecular genetics lab and I used Zendo to demonstrate the danger of positive results and the power of negative results.

You can create a model that conforms to all current observations and still be proven incorrect on future experimentation. The negative results are conclusively negative.

Ruling out possibilities: the way science should be done.

Well done. One of the reasons I rate Zendo so highly is for it's simple yet thought-provoking depiction of our search for truth. I applaud you for writing this review; thank you for your efforts.

.w.

Fantastic review!
As a game-playing student teacher I am in the process of collecting information about bringing games into the school, including into the curriculum. How awesome is it that a fun game like Zendo so nicely resembles an important aspect of the Science curriculum.
MANY thanks.
Now to add this to the list

Great review! This one inspires me to play it with my friends. If only I could find a cheap source of Icehouse pieces. Oh well, maybe I'll stick to LEGO bricks, of which I have plenty.

Have you tried the cardboard Icehouse pieces (either DIY or buy)?

I was thinking about trying Zendo before, but now I must absolutely play this game.

Thanks!

Fantastic review! It compelled me to tip you some GG

I first knew of this game some two or three years ago and it looked curious to me but then I wasn't prone to spend the money for it. With the present rate of exchange between euro and dollar and your explanation I feel it would be an investment beyond simple entertainment. Ach, molecular biologist professional deformation, I guess.

Me too!

How would you formulate an appropriate expression for a set of beginner cards?

i.e. instead of saying "A koan has the Buddha nature if..." would you say "An experimental result will be white if..." ?

That's basically how I would say it.

I was thinking of getting several additional colors so that the law of nature could have several parts.

"the experimental outcome will be white if ... On the other hand, it will be red if ... On the third hand, the outcome will be white *and* red if ... etc.

Hehe. Now we're starting to get into classical genetics, and aspects of codominance and epistasis... ;-)

How about a quantitative genetics version? Polygenic traits? QTL mapping? Oh heck, might as well just go back into the lab and do it for real! :-)

I will soon be going to India for four weeks to help train people to troubleshoot a complex software product at a new support center there, and I was thinking about bringing Zendo as a teaching tool. Although these Indians are highly educated, in part because their culture values education more than we do in the U.S., their entire educational system is built around rote memorization. In general, Indians are very poor at asking questions and thinking outside the box--skills that are crucial for technical support.

But I do have some concerns about whether the mild Buddhist references in the game could offend either Buddhists or non-Buddhists in India.

Perhaps I should change the language of the game too. One player could be the customer with a problem. The problem is an unexplained result, represented a pattern of pyramids. The other players could be the technical support engineers. They could propose different ways to try to reproduce the problem, and the customer could indicate whether the proposed state reproduced the problem or not, with white or black stones. And they could also try to guess at the root cause....

Uh... Not quite right. Still needs work. Anyone want to try and help me out here?

Since the icehouse pieces are thematically neutral, you don't need to make any reference to Buddhism or whatever. You can just describe it as I have, as an exercise in the scientific method, or any other way that you want.

Also, you don't need icehouse pieces. The easiest thing to do is use use sentences on a whiteboard or a sheet of paper. I actually enjoy the game more this way. It's way less fiddly and there's much more room for different laws of nature. And who wants to lug icehouse pieces to India?

BTW: I recently found another (rather different!) educational use for Zendo. I teach English as a foreign language, and I played Zendo with some students. Language teaching often doesn't focus on expressing very logically precise concepts like Zendo koan rules, so this was a fun game-playing way for the students to clearly express things very clearly, e.g. carefully saying something like "there is exactly 1 red piece and at least 2 blue pieces" instead of something easier but more ambiguous like "there's 1 red piece and 2 blue pieces".