The Answer Men

18 minute read
Lev Grossman

Kraljevica is a tiny Croatian shipbuilding hamlet on the eastern edge of the Adriatic Sea. If you’ve heard of it–which you almost certainly haven’t–it’s probably because its shipyard, which had been in continuous operation since 1729, went bankrupt last June, a casualty of the pan-European financial crisis. Kraljevica (pronounced Krall-yay-vi-tsa) sits on a shatteringly beautiful, eerily calm bay, across from the stark and cruelly vowelless island of Krk.

The views are spectacular. But for three clear, sunny days in October they went unappreciated, because Kraljevica was hosting the World Puzzle Championship. Not only was nobody outside, nobody was even looking out a window.

You probably haven’t heard of the World Puzzle Championship either, but you should have, because it’s the most extreme test of pure logical-reasoning power on the planet. It has taken place annually for the past 21 years; in 2012 it drew 145 contestants from 26 countries, many of them people who eat, breathe and, on the rare occasions when they sleep, dream about puzzles full time.

These guys are outliers, honest-to-God geniuses. But they’re just the most illustrious representatives–the apex predators–of a vast, worldwide puzzling population that numbers in the hundreds of millions. Puzzles have always been a ubiquitous but unassuming and peripheral presence in our lives, folded meekly into the back pages of magazines and newspapers. But with the rise of the Internet and mobile devices, they’ve moved closer to center stage and become a not insignificant part of global culture, almost as pervasive, in terms of their reach and the number of person-hours they consume, as television and movies. No one knows exactly how many Americans do puzzles, but everyone agrees that the number of crossword puzzlers runs into the tens of millions. Estimates for the number of sudoku players run up to 80 million worldwide. And that’s to say nothing of puzzlish and puzzlesque activities like Scrabble, Words with Friends, bridge, Tetris, Rubik’s Cubes (over 300 million sold) and Angry Birds (260 million users at last count). Now that puzzles have escaped from newspapers and migrated onto our phones and tablets, no idle moment is safe from them.

At the elite level, puzzle solving is in no way a casual activity. It’s as intense and arcane and competitive as chess, in some ways more so. But events like the World Puzzle Championship, or WPC, tell us a lot about why ordinary people solve puzzles, and always have, and probably always will.

Completely in Control

The reasons aren’t self-evident. You’d think we have enough problems in our lives without making up more of them. Puzzles demand concentration and intellectual effort, to the point where they resemble work more closely than they do play–the main difference being that at least with work you tend to get paid for your trouble. So why bother?

Because puzzles are actually neither work nor play; they’re something else. “We’re faced with problems every day in life, and we almost never get clarity,” says Will Shortz, the crossword editor of both the New York Times and NPR. “We jump into the middle of a problem, we carry it through to whatever extent we can to find an answer, then we just get on with things and find the next thing. Whereas with a human-made puzzle, you have the satisfaction of being completely in control: you start the challenge from the beginning, and you move all the way to the end. That’s a satisfaction you don’t get much in real life. You feel in control, and that’s a great feeling.

“And as with any human activity,” Shortz goes on to say, “you want to know how good you are compared with other people.” Hence the WPC.

For many years, the major international competition on the puzzling calendar was the International Crossword Marathon, a tradition that began in Poland in 1984. (If you’re noticing an East European theme to this article, it’s because that part of the world has a highly robust puzzling culture, dating back to the economically straitened, media-poor period in the 1970s and ’80s, when a lot of Eastern Europeans had only pencils and paper for entertainment.) The object of the International Crossword Marathon was to create the largest possible crossword puzzle–25 squares wide and as long as you could make it–in 24 hours.

Shortz went every year, but privately he thought it was a little silly. Participants used their own languages and their own national crossword rules, and as a result it wasn’t a very good basis for comparing raw puzzle-solving skill across cultures. Plus it was exhausting. Shortz imagined something that would be closer to an Olympics of logic. “My idea,” he says, “was that we would have a true competition that would be equal for all countries. It would not involve word puzzles like crosswords. It would involve things like–nowadays–sudoku, KenKen, number puzzles, logic puzzles, picture puzzles. Things that everyone can do equally, no matter what their language and culture.” The first WPC was held in New York City in 1992. It’s been held in a different city every year since: Minsk, Rio de Janeiro, Cologne, Utrecht. And Kraljevica.

Unbeknownst to many of its inhabitants, the U.S. is a powerful puzzling nation, having won world championships in 15 of the past 21 years, either as a team or as individuals. We sent two four-person teams to Kraljevica, an A team and a B team, drawn from the top finishers at the U.S. Puzzle Championship, an open event that takes place every summer. This year the American teams were composed entirely of men; in terms of the gender imbalance, the U.S. puzzle establishment is shamefully retrograde. (Women are better represented on the teams from Europe.)

High-level competitive puzzlers tend to be single (at this level, puzzles have a way of eating your life and leaving no room for anything else), tend to have a math or computer-science background and tend to be young. The elder statesman of Team USA, at 37, is Wei-Hwa Huang, a former Google coder who quit to design and solve puzzles full time. Calm and apparently infinitely knowledgeable, Huang has been at every WPC except the first. He’s won it four times, though the last time was in 1999.

The team’s current reigning powerhouse is Thomas Snyder, 33, a rapid talker with boyish sticky-up hair and a quick smile that alternates with a perennially worried expression. Snyder has been the U.S. puzzle champion six times, and he’s generally acknowledged to be one of the top two or three puzzle solvers in the world. Atypically for competitors at this level, he has a serious career outside puzzles: he has a Ph.D. in chemistry from Harvard, and he was, until recently, the lead scientist at a biotech firm called ImmuMetrix. “What I’m very good at is observation puzzles,” he says. “You’ve got a grid with tons of numbers, but you’ve got to pick out the few that are critical to see and use them. It’s sort of what I do in my day-to-day job as a scientist. I look through gigabytes of data and find the few pieces that are relevant.” There are those who think the distraction of attempting to solve the world’s immunological woes is why Snyder is the puzzling world’s Ahab: for all his success, he has never won the individual world championship. It’s something of an obsession.

Rounding out the A team are Jonathan Rivet, 37, an affable computer programmer from Chicago, and Palmer Mebane, the team’s prodigy. He’s only 23, but Mebane–who by day works for a website for advanced math students called Art of Problem Solving–has already done something Snyder hasn’t: at the 2011 WPC in Hungary he stunned the puzzling world by winning the whole shooting match, defeating even the German juggernaut Ulrich Voigt, who has won it more times than anyone else. (Snyder came in third; Huang was fourth.) Mebane is intense, jittery and as skinny as an addict. Which he is, though drugs are not his addiction. “I’m a pretty introverted person,” he says, “so math and puzzles are pretty much all I do in life.”

Pure Logic

The word puzzle means a lot of different things to a lot of different people. Take sudoku, for example. It’s generally believed to have been invented, at least in its modern form, in 1979 by Howard Garns, a retired architect from Indiana, who called it Number Place. It got renamed sudoku in Japan in the 1980s, and then it spread back to the West via the London Times in 2004, at which point its popularity exploded–it’s the kudzu of the puzzling world. There’s a World Sudoku Championship, which is generally held right before the WPC and in the same venue. Snyder has won it three times, though he sat it out in 2012 to save his strength for the WPC.

WPC puzzles are different. They’re pure-logic problems. The principles involved are simple. There’s no trigonometry, no higher math. The goal is to test reasoning power, not knowledge. “What’s great about it is, you don’t need a Ph.D. in math,” says Todd Geldon, who played for the B team in Kraljevica (and who does, as it happens, have a Ph.D. in math). “It’s very accessible in terms of what you’re looking at, what the rules are, what the goals are. But actually doing it is very difficult.”

In fact, some pure-logic snobs look down on sudoku, whose rules are always the same. At the WPC, the rules of the game change from problem to problem. Each one requires a whole different set of solving strategies–the term of art is heuristics–which players often must devise on the fly. You’re not just figuring out how to fix something; you’re inventing the tools as you go. “What we’re really measuring in this competition is who is the best at developing these heuristics rapidly, essentially as you’re solving the second puzzle you’ve ever seen of that style,” says Nick Baxter, 55, a veteran competitor who now acts as captain of the U.S. team. “How quickly you grok the subtleties and the interesting interactions between the givens and the constraints, and how you get solutions.”

For example, one problem in the 2012 championship consisted of a table-size grid representing Kraljevica’s harbor, on which teams had to place as many ships of three different sizes as they could in a manner that satisfied 17 separate conditions. Every boat had to touch the shore, no boat could touch any other boat, and so on. The puzzle rather starkly revealed the gulf between the Platonic world of logic and our fallen reality: picture a ballroom full of geniuses bent over stylized maps, carefully placing toy boats, one by one, in precise configurations, while in the windows the sun goes down on the actual Kraljevica harbor, standing idle and empty of any ships whatsoever–an incorrectly solved square on the global economic grid.

The Art of the Puzzle

To connoisseurs, a puzzle is like a poem, except that whereas poems deal in feelings and ideas, puzzles traffic in pure logical epiphanies. A well-constructed puzzle leads you down a particular chain of reasoning to a specific elegant or counterintuitive cognitive leap, and as you make it, you know at that moment that you’re making precisely the same leap the puzzle designer did. There’s a moment of silent intellectual communion, of shared aha!

Snyder got a taste of that aha! feeling when he was 14. “I was reading an account of the third world championship,” he says. “They had a puzzle where the organizer was walking to the dinner area, and he was wearing a very odd shirt with various patterns and shapes. And he drops a plate and makes a commotion, and everybody’s supposed to look at him and see the shirt. An hour later they were given a round where they had to reconstruct what he was wearing. And I thought, That sounds cool. I would like to do this.”

But professional puzzling has never acquired the same public profile as, say, poker, or even poetry, and one reason is that at that level the puzzles are really hard. Unearthly hard–so hard that it’s hard to even explain how hard they are. At the WPC there are point bonuses just for finishing all the puzzles in a single round, but not many players claim them. In 2012 one round featured 21 puzzles in which each of the answers was linked to the others, and no instructions were given for any of them; before you could solve the puzzles, you had to deduce what the rules were.

Another reason you don’t see much puzzling on ESPN is that the rapid development of heuristics is not a very telegenic activity. In Kraljevica the rounds were conducted with about as much fanfare as a practice-SAT session. An announcer said go, and a large TV at the front of the room started ticking off seconds. Chairs creaked. Someone puffed and swept away eraser fluff and started over. Someone else slapped his forehead, V8-ad-style. It’s a serious test of nerve, speed and intelligence, but it’s not a spectator sport. The action is internal, and the triumphs and disasters come out only after the fact. “You’ll hear some very weird conversations during the breaks,” says Baxter. “‘Oh, I cooked that one’ or ‘I broke that puzzle,’ which means that you went down and made a logical deduction that wasn’t valid, and everything felt good till two-thirds of the way through, then you hit a contradiction and had to erase the entire puzzle–something like that. It’s an adventure, every puzzle.”

The finals of the 2012 WPC demonstrated both the grand intellectual drama of high-level puzzle solving and the difficulty of translating it into anything resembling popular entertainment. After two days and 13 rounds of competition, the top eight individual scorers were pitted against one another in a three-round playoff. The U.S. wound up with two solvers in the final eight: Snyder was in second place, and Mebane was in fourth. Japan also had two finalists; the Netherlands, Poland and Hungary had one each. Germany had one too: Voigt.

In an effort to heighten the drama, the Croatian organizers set up the finals as a series of head-to-head matches, with the players working on 4-by-4-ft. easels, in permanent marker, so the crowd could follow the action. Not everybody was thrilled about the format. “When’s the last time you solved a puzzle on a four-foot-square piece of paper?” Baxter asks. “Standing up? Not being able to erase?”

But it worked, up to a point. The mood in the room was electric. In order to give the top finishers from the first two days an advantage, they were allowed to pick their puzzle types; this led to a considerable amount of gamesmanship, with players sizing up their own and others’ strengths and weaknesses. When play began, Huang and the other Americans provided color commentary, calling out the names of the puzzles in stage whispers as they came up: Dubai Skyscrapers, Magnets, Battleships, Pipes, X-Kakuro, Tapa, High Winds, Rural Tourism. Voigt–a big guy with a rabbinical beard and a receding hairline, as if the heat of his frontal lobes had burned the hair away from his forehead–wobbled early, nearly losing his first match with the Pole. Snyder and Mebane, the defending champ, bulled through their first rounds in straight sets. When a contestant finished, he (the participants in the playoffs were all male) stepped away from the puzzle and raised a hand, whereupon a Croatian official would check his work for a tense minute, then turn to the audience and say, curtly, “Iz korrekt.”

The most exciting moment came in the semifinals when Mebane faced off against Voigt. Both looked invincible. Mebane stood in front of his easel with one hip cocked and a dead-eyed gunslinger’s stare. They traded off victories–holding serve, effectively–but in the third and last puzzle, a complex sudoku variant called Andy’s Sudoku, there was drama. Voigt took an early lead, his marker flickering over the grid, filling it with numbers, but then he froze, stared and lunged for a red marker, with which he started making laborious corrections.

Mebane raised his arm first, but after a long, tense pause, the judges waved him over. Iz not korrekt! He had to spend a minute in the penalty box before he could go back and fix it. Meanwhile, Voigt finished his corrections and raised his hand. But the judges ruled his answer wrong too. The room exploded with chatter, and the stage flooded with vexed puzzlers. A rumor raced through the crowd: The puzzle itself was broken! It would be tossed out as invalid! That had happened before.

But no–the problem was just reading Voigt’s tangled corrections, which eventually ran to three colors. The judges reversed their verdict, and the solution was upheld. For a second, Mebane appeared stunned, until you realized that that was just his customary blank facial expression. Then he grinned and shook Voigt’s hand. The two titans, Snyder and Voigt, would meet in the finals.

Which were, maybe inevitably, an anti-climax. Both parties would later agree that the key factor was solving the meta-puzzle of who got the puzzle types they were best at. For the first three rounds, Voigt and Snyder traded wins, but the fourth puzzle, which Snyder had picked, was his downfall. He made an unlucky guess, and it was exacerbated by the large format, which made it hard to take in the whole puzzle at once. While he was getting back on track, Voigt churned through the puzzle ahead of him.

And that was the win. The Croatian master of ceremonies muttered something deep, Slavic and totally unintelligible into the microphone. Voigt had taken back the title. He was, for a record eighth time, the world puzzle champion. There was no cash prize, though later that day he would receive a medal and a hideous trophy. The whale had eluded Ahab again.

Snyder was gracious and upbeat in defeat. In fact, in three days of white-hot intellectual competition, I never saw a single unsportsmanlike word or gesture or even facial expression. Ultimately the U.S. would finish third in the team standings, behind Germany and Japan, its lowest team finish in a decade. But the A team took two of the top three individual spots; moreover, an MIT freshman named Anderson Wang, who barely made it to Kraljevica straight from his midterms, placed a respectable 42nd overall. There is hope for the future.

Since then, Snyder has left his biotech job, following a change in direction at ImmuMetrix. “At the WPC, the problems always have one solution and people generally agree with each other when they’ve found it,” he writes in an e-mail. “In life, things are never as clean.” He’s launched his own company, Grandmaster Puzzles, which produces gloriously elegant, handcrafted puzzles, in contrast to the computer-generated puzzles that fill dozens of sudoku rags. Meanwhile, the World Puzzle Federation, which oversees the WPC, has started a Sudoku Grand Prix: eight tournaments, hosted by eight countries, all played online. The 2013 WPC has been set for next October in Beijing, and the rumor is that the sponsor, Beijing Media Network, is going to pump a lot of money into it. Maybe the Chinese will crack the TV problem and Team USA will be big in Asia.

Not that Team USA’s members seem to care, particularly. What they care about is solving. No more than 10 minutes after the finals ended, Huang broke out a pack of cards, and the U.S. puzzlers started in on a four-handed Chinese card game called Tichu. It was 1:30 in the afternoon; they had time to kill before 3:00, which was the opening bell of an online sudoku tournament run out of Japan. First prize was a T-shirt (which Mebane would claim). In a way, elite puzzlers are like extreme big-wave surfers, endlessly chasing that pure logical high, the correct chain of reasoning that will lead them to the ultimate aha. But they’re just like the rest of us too. All they really want are questions that have answers.

More Must-Reads from TIME

Contact us at letters@time.com