The day Josh Wardle sold Wordle to the New York Times, in 2022, for more than a million dollars, should have been a moment of triumph. The game, which gives players six chances to guess a five-letter word, had unexpectedly become a global sensation, and Wardle had already begun to receive e-mails from puzzle designers seeking his input on their own ideas. “The underlying question was always, ‘How do I sell this to the New York Times?’ ” he told me. In their eyes, Wardle had achieved the ultimate success. Yet, as he fielded invitations from journalists, chat-show hosts, and podcast producers, Wardle felt only discombobulated, even borderline depressed.

The problem was not only the speed with which fame had upended his life but the fact that it had arrived uninvited. Wardle, who grew up on a farm in Wales, and who recently returned to the U.K. from Brooklyn, had built the game as a gift for his partner, Palak Shah: a bonding ritual they could share each evening. At the time, he didn’t consider himself a game designer, though play was deeply embedded in his professional life. While working as a product manager for Reddit, he had created a series of interactive projects for April Fools’ Day. One of them, from 2015, was the Button, a web page featuring a large button and a sixty-second countdown clock. Each time a visitor clicked the button, it reset the timer; if no one pressed it before the clock reached zero, the project would end. More than a million people participated before the countdown finally expired, three months later.

It was this sort of lightly mischievous design, combining play and social experiment, that led to Wordle. As the game slipped beyond Wardle’s living room, passing from friends to group chats to Twitter feeds around the world, many players started devotedly sharing their solve streaks—a dynamic not unlike the communal joy of the Button. For Wardle, though, the flood of attention felt overwhelming. “I’m not sure humans are built to handle going viral,” he said. He worried that others would copy and commercialize the game’s concept, whether he wanted them to or not; selling meant relinquishing control, along with responsibility for whatever followed. The day he sold the game was the last time he played it. He didn’t know what he would work on next, or how, exactly, he would pass his days.

Wardle found refuge in an unlikely place: the cryptic crossword, a kind of puzzle popular mostly in Britain, where it originated, in the nineteen-twenties. American, or “concise,” crosswords are typically exercises in trivia more so than wit. A conventional clue might read: “Got up.” If the solution line has four letters, two answers might fit—“rose” or “woke”—and only the crossing letters can settle the matter. The cryptic clue eliminates the ambiguity entirely. It might read: “Pairs of rowdy seagulls get up.” Here, “pairs of” is not a hint but an instruction: take pairs of letters from the following words—the “ro” of “rowdy,” the “se” of “seagulls”—and arrive at an answer, “rose,” that matches “get up.” The uncertainty collapses in a single, satisfying click. Chasing that click is the cryptic solver’s obsession.

Wardle had tried cryptic crosswords when he was younger, but found them to be impenetrable. “I didn’t know how to begin,” he told me. The rules could seem arcane, almost impossible to deduce. A clue containing the word “radio” could signal that “am” or “fm” belongs somewhere in the answer; “book” could imply “ot” or “nt” (Old and New Testament); “sailor” could require the letters “ab,” for “Able Bodied.” Anagrams were flagged by a bewildering range of indicators: “mixed,” “scrambled,” even “microwaved.” These codes often exhaust newcomers, who may feel as though they have arrived at the door of a private club, ignorant of its customs and wearing the wrong trousers. As P. G. Wodehouse wrote of the Times’s daily cryptic in 1945, “the humiliation of only being able to fill in about three words each day is too much for me.” Everyone needs a guide.

Wardle found his while listening to the podcast “Scriptnotes.” In one episode, the showrunner, Craig Mazin, of HBO’s “Chernobyl” and “The Last of Us,” enthused that “everybody should do cryptic crosswords,” and explained some of the form’s underlying logic. “It really stayed with me,” Wardle said. Conversance led to discernment. He became a devotee of the Toronto-based mathematics teacher and puzzle constructor Fraser Simpson, whose clues struck Wardle as contemporary and enviably taut, without the usual “fodder” used to make a clue legible. “Every word in the clue is either definition or wordplay; there are no connector words,” he said. “I was stunned when I saw that.”

Wardle wanted to make a game that could teach the rules of cryptics, just as Mazin had taught him. The idea was vague, germinal; the prospect of actually releasing a follow-up to Wordle felt “paralyzing,” he said, and for a time he worked as a consultant, helping others on game prototypes, which freed him creatively. Then he began to sense an appetite. Cryptics seemed to be experiencing a modest revival; the Australian YouTube channel Minute Cryptic, for example, had drawn tens of thousands of viewers to short videos that unpacked a single clue at a time. Wardle enlisted the support of Chris Dary and Matt Lee, two longtime collaborators from Reddit, and their new game, Parseword, is now available, with a title that came from Wardle’s partner, Shah.

Neither Dary nor Lee had been familiar with cryptics before development began, and Wardle believes this strengthened the game. They felt comfortable questioning conventions, pressing him on assumptions a veteran solver might take for granted. The result, Wardle hopes, is a way to introduce newcomers to the joys and agonies of the cryptic, whose curious but finally comforting logic had been, for him, a salve.

Others have tried this before. In 1968, the composer and lyricist Stephen Sondheim published an essay in New York explaining the rewards of the cryptic crossword. In his view, the form was superior to the American crossword, full of “cleverness, humor, even a pseudo-aphoristic grace.” Sondheim began publishing his own cryptics in the magazine, complete with rules, examples, and prizes (copies of “Chambers Twentieth Century Dictionary,” then priced at around five dollars). The experiment didn’t take. After forty-two puzzles, Sondheim retired the column, and cryptics remained—with occasional exceptions, including at this magazine—largely a British preoccupation.

The central premise of Parseword’s approach is to treat the cryptic not as a riddle to be intuited but as an equation to be solved. On “Scriptnotes,” Mazin had suggested that most cryptic clues contain a conceptual dividing line: on one side is the definition, on the other the wordplay. This led Wardle to consider how each clue might be broken into components; once one fragment was resolved, it might be substituted back into the whole, reducing the remaining complexity.

In Parseword, this process is pleasingly tactile. You can click on a word and see potential synonyms, as well as whether the word is a potential indicator. The game teaches you that in the clue “Taxi reduced fee,” for example, “reduced” is an indicator word, instructing you to shorten another word. Remove the “i” from “taxi” and you find the solution, which is confirmed by the definition: “fee.” Having learned the principle, a thoughtful player should be able to solve the next example: “Funk reduced joy.” Rather than asking newcomers to memorize a rulebook, Parseword makes the algebra visible, guiding players through a sequence of discrete steps. The mystery remains, but the path to clarity is no longer obscure.

Neither Wardle nor his collaborators started out as puzzle constructors, but they knew that cryptic creators are often as distinctive as prose writers. A friend introduced them to Emily Cox and Henry Rathvon, revered constructors who, in the course of fifty years, produced cryptics for the Wall Street Journal, The Atlantic, and The New Yorker. “They were incredibly generous,” Wardle said. “They have this bank of clues, and they said we could use them.”

In recent years, crosswords have faced a gentle reckoning, as younger constructors push to update the form’s conventions, moving away from fusty references, obscure Latinisms, and other arcane rules by which, for example, the appearance of the word “school,” in a cryptic crossword, invariably signals that “Eton” lurks somewhere in the answer. Cox and Rathvon appear to avoid these habits, and Wardle and his collaborators have worked carefully to further soften the learning curve, beginning with an elegant tutorial designed to introduce essential techniques: deleting letters, joining them, spotting anagrams and homophones. Each daily clue is supported by a system of hints and prompts, which, in time, the player will learn to rely on less. The game still demands patience and study, however, almost guaranteeing that it will never approach the ubiquity of Wardle’s previous creation.

Indeed, the contrast between Wordle and Parseword is stark. Wordle’s power rests in its elegant restraint. “Even people for whom English is their second language are able to play,” Wardle said. Cryptics occupy the opposite end of the word-game spectrum, and Parseword admits to its difficulty in a blunt subtitle: “A tricky wordplay game.” For Wardle, too, this project feels vastly different. “Releasing Parseword is happening more on my own terms, instead of happening to me,” he said. His ambition is modest and, for now, apparently noncommercial. He hopes, simply, to “reduce the problem space for new solvers,” a phrase that sounds as though it could be parsed for indicator, connector, and definition. ♦