Puzzle Box

Hunting for Mysteries


You come across some letters in a waste paper basket in a mysterious apartment…

That was it: no instructions, no hint button, and no online walk-through if I’d even bothered to look. All I had was a website with some flavor text, a title for the puzzle – “Dear Reader” – and a series of letter fragments, each dated and written in a script font. For instance, dated March 6th: “Mother, I regret to inform you that I have just committed a homicide. I discharged my pistol into his skull, & the wound was most fatal; I am very young, but now I fear my future seems bleak indeed.”


“They’re songs,” my friend explained as I joined him in an online chat room. A bunch of my college buddies were engaged in some kind of weekend-long puzzle-solving competition at MIT, but I was stuck in Philadelphia for work at the time. I was curious about what my friends were up to, and all the puzzles were posted on the Web, so I logged on, hoping to help out remotely. As for the clue, my friend pointed out that the segment paraphrased Queen’s “Bohemian Rhapsody”: “Mama, just killed a man / Put a gun against his head / Pulled my trigger, now he’s dead.” That’s where my friends were stuck.

Scanning through my friends’ notes in progress online, I mentioned that there seemed to be a lot more songs by “Weird Al” Yankovic than by any other performer, and even the songs that he hadn’t written included some he’d covered. A long pause followed, as I eagerly awaited the next clue. My friend typed back, “Solved it. VICECHANCELLOR.”

What the heck did VICECHANCELLOR mean? I asked, but got no response. My friend was already rushing headlong into another enigma without me. Somehow, the solution was even more puzzling than the puzzle itself. It was an apt introduction to the MIT Mystery Hunt.

The Mystery Hunt began in 1980, initiated by former graduate student Brad Schaefer. As one of the oldest and best-known “puzzlehunts,” it attracts upwards of a thousand participants each year, drawn from both the campus population and unaffiliated problem-solvers the world over. The goal is to solve wave upon wave of puzzles, each without explicit instructions, each somehow yielding a text string as a solution, and each providing a clue toward an even larger “metapuzzle.” The Mystery Hunting process is a mix of the technological and the personal, with puzzles delivered on a password-protected website, a link on each puzzle’s page indicating when the team has a guess for a solution, and organizers calling and dropping by personally to check on solutions and progress.

A clever team doesn’t need to solve every puzzle and metapuzzle to win; after a certain point, one can fill in the blanks between solved puzzles, deducing a full solution somewhat like the moment before Vanna turns the final letters on Wheel of Fortune. Over the course of a weekend, all these solutions finally allow a team to deduce the location of a “coin” hidden somewhere on campus. The team that finds the coin first wins, earning the dubious honor of preparing hundreds of puzzles and serving as organizers for the following year’s Hunt.


That’s the way it works in theory, at least. Some Hunts run longer than a weekend; some shorter. The organizers sometimes have to interrupt everyone by email to toss some hints, and teams that are lagging often get puzzles unlocked for them before solving the earlier rounds. After all, some of these puzzles are pretty darn hard. At least the VICECHANCELLOR one had words, rather than expecting readers to piece together encrypted Braille, or recognize Xbox 360 Achievement icons by sight.

After the utterly unsatisfying experience of helping to solve a puzzle that only unlocked an even more cryptic puzzle, I thought I’d sworn off puzzle hunting. Eventually, though, I decided to give the Hunt another chance. Perhaps the experience is different in person, I reasoned. As it turns out, it is indeed different, but it sure isn’t any easier.

My new team was an offshoot from a larger team. This isn’t uncommon, as some teams get to be pretty big and unwieldy after the first 50 or 100 members or so. My new team was relatively small at 20 or 30 members, and our name said it all: The Grand Unified Theory of Love was formed for gaming, geekery, and good-natured fun. We really had no prayer of winning. We were just there to solve some puzzles.

We gathered in the classroom the Hunt’s organizers had assigned to us as our headquarters. One table was dedicated to snacks and sodas, but the rest of the room was optimized for puzzle-solving versatility: a laptop at nearly every seat, scrap paper and pencils scattered about, chalkboards featuring progress on metapuzzles, and an overhead projector displaying messages from our remote puzzle-solvers. (We hoped that last bit of equipment would discourage our long-distance friends from skulking off in discouragement like I had a couple years before.) Before we got into the thick of things, our fearless leader reminded us to continually edit the team wiki with our progress to avoid redundant work on any puzzle. We hovered at the laptops, continually refreshing Web browsers as we waited for the puzzles to appear.

When the links on the site finally went live, we broke into groups and hit the ground running. One group assembled Lego tableaus to satisfy the requirements of a scavenger hunt. Another group left campus entirely for a puzzle, driving off to Harvard to follow up on a lead somehow related to maps and architecture. Most of us launched into any one of a number of puzzles delivered via the Web. I struggled with one that involved analyzing several hours of prison guard logs and diagrams of prisoners’ movement across multiple floors, laying out all the information in a series of composite images in Photoshop. I wouldn’t know until the solutions were released weeks later, but I got pretty close before I gave up in frustration. (If only I’d shrunk the lower floors to simulate visual perspective, those little dots representing prisoners would’ve spelled out an answer.)


Instead, I wandered over to a teammate who was stuck on her own puzzle, titled “Cluesome.” She’d already made some progress with others on the team, who had wandered off in frustration much like I just had. Their puzzle offered a series of text strings, like “WYVMLZZVYWSBTYVVMAVWWPHUV.” My teammates had made some progress, at least, determining that each string was a cryptogram encoded with a simple Caesar cipher. Shift each letter of the above gibberish seven letters down the alphabet, and you end up with “PROFESSOR PLUM,” “ROOFTOP,” and “PIANO.” It seemed be a clear a reference to Clue, but that’s about as far as anyone got. Every string was like that: shifted some number of letters down the alphabet, but referring to unfamiliar murder scenes and weapons.

Something struck me as familiar about that modus operandi, though, and inspiration struck. Wasn’t that how Christopher Lloyd killed somebody in Who Framed Roger Rabbit? I started screeching quotes from the movie, and another teammate, engaged in some other puzzle, looked up to call out: “He played Professor Plum in the movie version of Clue!”

Suddenly we were back to work. Some research online revealed that each decoded string described a murder in a movie starring someone who had also starred in Clue. We tracked down each movie and counted into their titles by the number of letters they’d been shifted to derive one letter from each. The result spelled a word: EMOTIVE.


We clicked a button online, and the team running the Hunt called us on our designated headquarters’ cell phone. The room fell silent as my teammate answered.

“Hi, we’d like to call in a solution for ‘Cluesome.’ EMOTIVE. E, M, O, T, I, V, E.” She paused – and smiled. We started to cheer even as she was wrapping up the phone call. We updated the wiki and wrote the solution on the chalkboard for those working on the metapuzzle.

EMOTIVE! We had no idea what it meant, and over the course of the weekend, we never solved enough puzzles to find out. I only helped complete two puzzles all weekend myself.

For me, the real mystery of the Mystery Hunt may be why I would go back for more – and not just once, to try it in person, but year after year, on the same losing team. We never win, or even have any chance of winning. This isn’t a Bad News Bears underdog comeback story waiting to happen. In fact, I don’t think we even want the burden of competitiveness, of trying so hard that we lose sight of the pure joy of playing a really hard game, let alone the burden of preparing an entire hunt the following year.

As a longtime gamer, there’s something exciting about facing puzzles you can’t look up on GameFAQs, quests that force you to leave the living room, and challenges that turn a bunch of nerdy strangers into a team. There’s something magical about those moments when the combined pop culture knowledge and analytical thinking skill of a room full of people fits together to unlock some carefully designed mystery. Though the MIT Mystery Hunt technically predates any videogame I’ve ever played, to me, it feels like the new “Hard Mode” in gaming. When you do finally (if only occasionally) succeed, it’s so much more satisfying for being so unreasonably challenging. And, for my team, it’s a challenge that you can’t just face alone on the couch.

When Mystery Hunt weekend rolls around again in January, I fully expect to resent the odd sleeping hours, the junk food diet, and, most of all, the puzzles we never figured out how to solve. And I fully expect to be back for more a year later.

Jason Tocci is an academic, a writer, a designer, and kind of a huge nerd. He blogs about his research on gaming and geek cultures at Geek Studies.

About the author