The Origin Story
Arthur Wynne and the World That Almost Missed the Crossword
It was a Sunday two days before Christmas, 1913. The New York World's "Fun" supplement editor needed something to fill space. Arthur Wynne, a 44-year-old journalist born in Liverpool, England, reached back to a childhood memory: a Victorian word game called "magic squares" his grandfather had shown him, where words read the same horizontally and vertically in a simple grid.
What Wynne created was something new: a diamond-shaped grid with no black squares, numbered clues in two directions, and blank spaces to fill in. He called it a "word-cross" — the hyphen later migrated to produce "crossword." The puzzle ran on page 4, between a quiz about the Pilgrims and a cartoon about Christmas shopping. No one thought much of it.
Except readers loved it. Letters poured in asking for more. The puzzle became a weekly feature. Other newspapers noticed. By 1924, when Simon & Schuster released the first crossword puzzle book (complete with a pencil attached, acknowledging that people would make mistakes), it sold 100,000 copies and launched a national craze. The New York Times — famously snobbish about the form — declared crosswords "a waste of time" and refused to publish them for eighteen more years.
Wynne's Curious Legacy
Arthur Wynne never sought credit or celebrity for his invention. He continued working as a journalist and puzzle editor, treating the crossword as one assignment among many. When interviewed in his later years, he seemed mildly bemused by the cultural phenomenon he'd accidentally created. He died in 1945, by which point crossword puzzles were appearing in virtually every American newspaper and had spread internationally. His original diamond grid — unusually, without any black squares — remains in the Library of Congress and is reproduced in countless crossword history books.
The Architect of the Modern Form
Margaret Farrar: The Woman Who Made Crosswords Respectable
When the New York Times finally reversed its position on crosswords in 1942 — spurred partly by the argument that puzzles gave Americans something useful to do during blackout periods in World War II — they chose Margaret Farrar as their first editor. It was a choice that would define the crossword for generations.
Farrar had been editing crosswords for Simon & Schuster since 1924, essentially functioning as the form's first professional gatekeeper. She arrived at the Times with deeply held aesthetic convictions: a crossword should be fair, elegant, and genuinely entertaining. Over her 27-year tenure, she established or codified virtually every convention still used in American crosswords today.
The 180-degree rotational symmetry requirement — meaning the black-square pattern looks the same upside down — was her insistence. "It makes the grid beautiful," she said. The rule that every letter must be checked by both an Across and a Down answer eliminated the frustrating "unchecked" letters of early puzzles. The ban on repeated words in a single grid. The preference for theme entries of equal or symmetrical length. The graduated difficulty across the week, from Monday's accessibility to Saturday's fiendishness. These weren't inevitable features of the crossword; they were Farrar's deliberate choices becoming standards.
She also championed themed puzzles — grids organized around a central unifying idea — as the highest expression of the art form. A good theme, in Farrar's view, should be surprising, internally consistent, and elegant: the solver should finish the puzzle, understand the theme, and feel a flash of genuine delight.
Deep History
The Editorial Succession That Built the Modern Crossword
One way to understand the evolution of the American crossword is through the succession of its editors — each of whom imposed their own aesthetic on the dominant form.
1942
Margaret Farrar takes the NYT helm
Establishes symmetry, checking, and theme as core requirements. The crossword becomes a respectable intellectual pursuit.
1969
Will Weng succeeds Farrar
Weng, a former news editor, maintained Farrar's standards while occasionally loosening vocabulary restrictions. A transitional figure.
1977
Eugene T. Maleska era begins
Classical scholar and poet. Made NYT crosswords the hardest in print. Accepted obscure vocabulary that would make later editors wince. Loathed pop culture references. Revered by dedicated solvers.
1993
Will Shortz appointed — modern era begins
Immediately introduces contemporary pop culture, film, television, and music. Removes many obscure Latin terms. The crossword gets younger and broader in its cultural range.
2014
NYT Crossword goes digital-first
App release accelerates constructor diversity efforts. New voices from younger, more demographically diverse constructors begin appearing with greater frequency.
Merl Reagle: The Comedian's Constructor
Merl Reagle occupies a special category among puzzle constructors: he wasn't just building puzzles, he was performing comedy in grid form. His Sunday puzzles routinely featured elaborate punning themes that required you to groan and laugh simultaneously — a reaction he called "the Reagle reflex." A 1994 puzzle built around spoonerisms of famous names, a grid where every theme entry was a food that had been "wrongly cooked," a puzzle where every answer contained a hidden expletive that made the title "Profanity" both accurate and wildly misleading. He appeared in the documentary Wordplay (2006), which introduced many casual solvers to the concept that puzzle construction was a genuine art form with genuine artists.
The Craft
How Professional Puzzle Constructors Actually Work
Most crossword solvers have a vague mental image of puzzle construction as something close to magic — a mysterious process where the right words somehow materialize in the right places. The reality is part art and part engineering, and understanding it enriches your solving experience considerably.
Professional constructors typically begin with theme entries — the long answers (usually 7-15 letters) that embody the puzzle's central idea. A constructor might brainstorm dozens of candidate theme entries before selecting the three or four that work best together: the same type of wordplay, the same length, the best combination of freshness and accessibility. This conceptual phase is where the puzzle's personality is established.
Grid construction comes next. The constructor places theme entries in the grid (typically symmetrically, per Farrar's convention), then fills in the remaining squares using a combination of construction software and their own vocabulary knowledge. Software like Crossword Compiler, Across Lite, and Black Ink can generate fill automatically, but most professionals treat auto-fill as a starting suggestion rather than a final answer — automated fill tends toward boring, overused words ("crosswordese") that experienced constructors actively avoid.
The cluing phase is where voice and personality emerge. The same entry — say, CRANE — could be clued as the bird, the construction equipment, the filmmaker (David Crane, co-creator of Friends), the poet (Hart Crane), or the action of stretching your neck. Expert clue-writing selects the angle that provides appropriate difficulty for the puzzle's target day, creates elegant misdirection (suggesting one interpretation while intending another), and occasionally delights with genuine wit. Maleska might have clued CRANE as "Whooping bird"; Shortz might clue it "Selfie-taking action, perhaps."
After construction, the puzzle goes to the editor — who may suggest or require changes to fill (terms that have dated, entries that duplicate others in recent puzzles, clues that are factually incorrect or unintentionally offensive). At the NYT, accepted puzzles are test-solved by multiple solvers before publication to calibrate difficulty. The typical time from submission to publication at the NYT is three to eighteen months.
The New Wave
Democratizing Construction: Who's Building Puzzles Now
For most of the crossword's history, construction was a genteel, largely amateur pursuit dominated by a relatively narrow demographic: educated, older, and predominantly white. The constructors whose work appeared in the NYT reflected the readership the Times imagined: people who knew their Gilbert and Sullivan, could identify obscure Roman emperors, and had strong opinions about proper cluing for EIRE.
The digital era has changed this significantly. Online communities like XWord Info, the Crossword Fiend blog, and dedicated subreddits have made construction knowledge far more accessible. Software tools have lowered the technical barrier. And editors — especially at the NYT under Shortz, and increasingly at newer venues like the LA Times, USA Today (where Erik Agard edited from 2021), and the independent puzzle sites that have proliferated since 2020 — have actively sought more diverse voices.
The results are visible in the puzzles themselves: wider cultural references, fresher vocabulary, themes that reflect contemporary experience rather than mid-century assumptions, and grids where the entries feel like they were chosen by someone who actually uses language in 2025. The tension between honoring the form's tradition and evolving it for new audiences is one of the most interesting conversations happening in the puzzle world right now.
The Amateur Constructor Community
One of the most heartening aspects of contemporary puzzle construction is the thriving amateur community. Constructors who will never appear in the NYT create excellent puzzles for independent publications, personal blogs, and community venues like the American Values Club Crossword (AVCX), the Lollapuzzoola tournament, and Inkubator (which focuses exclusively on female and non-binary constructors). Many professional NYT constructors began in these spaces. The barrier between amateur and professional is genuinely permeable — which is itself a form of the puzzle community's ongoing appeal.
Listener Questions
Q: Why do some constructors' names feel so familiar? I keep seeing the same people in the NYT.
A: There are probably 200–300 constructors who appear in the NYT with any regularity, but certain names appear much more frequently because they've built prolific output and strong editorial relationships. Prolific constructors like David Steinberg, Paolo Pasco, and Robyn Weintraub submit and have accepted multiple puzzles per year. Because the Times publishes 365 puzzles annually, even a constructor with ten published puzzles per year represents only about 3% of the output. Many readers develop personal favorites — a particular constructor's voice (wit, difficulty, theme style) becomes recognizable and anticipated.
Q: Is there a difference between constructors who use software and those who fill grids by hand?
A: Virtually all modern professionals use software for at least some phases. The distinction that matters more is how much constructors rely on automated fill versus curating it manually. Software can fill a 15x15 grid in seconds, but the results often include stale or ugly words. The most respected constructors use software as a tool but make extensive manual choices, sometimes rejecting an automatically-generated grid entirely and rebuilding it from scratch to avoid specific entries they find unacceptable. The creative decisions remain entirely human; software is more like a very fast reference library than a creative partner.
Q: Do constructors solve other people's crosswords? Or does knowing how they're made ruin the fun?
A: Most constructors are enthusiastic solvers — it's typically how they got into construction in the first place. Knowing the mechanics doesn't ruin the fun any more than knowing how a novel is structured ruins reading fiction. If anything, constructors often have a heightened appreciation for the clever moments: a particularly elegant bit of misdirection, an unusually fresh piece of fill, a theme they wish they'd thought of. There's also a professional dimension — studying published puzzles is how constructors develop their craft and stay current with what editors are accepting. It's research that happens to be enjoyable.
Q: Are there famous puzzle constructors outside the crossword world?
A: Absolutely. In the logic puzzle world, Nikoli (the Japanese puzzle company) has several legendary designers, including the creators who developed and refined Sudoku, Kakuro, and Slitherlink. Sam Loyd (1841–1911) was arguably the most famous American puzzle inventor before the crossword era — he created thousands of mechanical and mathematical puzzles, including the famous 15-puzzle and the "Get Off the Earth" illusion puzzle. Henry Dudeney in England was his contemporary rival and arguably superior mathematician, creating puzzles that still appear in puzzle books today. In the riddle and logic puzzle tradition, Martin Gardner's 25-year Mathematical Games column in Scientific American introduced millions of readers to mathematical puzzles and their designers.