The idea for the mazes began in January 1993 when I was invited to participate in a puzzle exhibit at the Atlanta International Museum of Art and Design. The exhibit was dedicated to Martin Gardner, who wrote the Mathematical Games column in Scientific American. I was invited because Gardner had twice written about my game Eleusis and had printed two of my mazes.

I designed a maze for the museum that was a grid of paths to walk on. There were twelve intersections where the paths crossed. As you approached an intersection you saw a sign on your right with one or more arrows. The arrows told you what turns you could make at the intersection—for example you might see a sign with two arrows that indicate you can only go straight or turn left.

My maze was quite popular. The opening of the exhibit was attended by some of the most brilliant scientists and mathematicians in the country. (I’m really not exaggerating here. A lot of scientists are also interested in mazes and other forms of recreational mathematics.) Many of them were willing to spend up to an hour in the maze until they found the solution. But the day after the opening was also interesting. Groups of school children arrived at the museum, and they also enjoyed the maze. Some of the brighter kids followed the rules and stuck with the maze until they solved it. Others missed a sign here and there but thought they had solved the maze. And the youngest kids just ran around the maze and happily ignored the rules. And everybody had fun. This gave me the glimmer of an idea: Maybe I shouldn’t design mazes only for brilliant intellectuals. That idea sounds obvious now, but in the past I had tried to make my mazes as difficult as possible.

I wrote about my experiences at the Atlanta museum in the introduction to SuperMazes. One of the people who read my book was Don Frantz, whose American Maze Company had started the craze of large cornfield mazes that are popping up around the country. When Don built a cornfield maze, he liked to include three or four small mazes next to it—what he called “courtyard mazes.” These small mazes served various purposes, from expanding the maze experience to providing parents with a place to park the kids while they have lunch.

A digression about small mazes: Some amusement facilities have tried to add a small maze, thinking they can provide something interesting without devoting a lot of space. These small mazes never work. A conventional walk-through maze is great if it’s large enough. But if a conventional maze is shrunk to less than a certain size, it becomes trivial. It might be okay for toddlers, but no one else will have any interest in it. I’m not sure what the minimum size is, probably something around 100 x 100 feet. A maze-with-rules, however, can be very small and provide as much interest and complexity as a large walk-through maze. My maze at the Atlanta museum was only 15 x 15 feet.

Don Frantz got in touch with me and asked me to design some small mazes he could use next to his cornfield mazes. I provided a few designs, and he picked three of them to implement. Actually each maze was implemented six times, once at each of his cornfield maze sites during the summer of 1998.