When Clues Feel Like Cheat Codes
I've spent years filling out New York Times Sunday puzzles and a few smaller indie ones too. Most of the time, you get a straightforward clue, you get a satisfying "yes," and you move on. But every now and then, you hit a clue that makes you put the pen down and stare at the grid like it just insulted your mother. That's what happened to me with "Perpetrating" on a particular Sunday. The clue was just "Perpetrating." Eight letters. The crossing letters gave me C-O-M-M-I-T-T-I-N-G. At first glance, that should be fine. Commencing, carrying out, perpetrating — they're all synonyms. But here's the thing that bugs me: "Perpetrating" doesn't mean "committing" in any normal sentence you'd actually hear. You don't say "he was perpetrating a crime" at a dinner party. It's textbook crosswordese — a word chosen because it fits the crossing letters, not because it's a natural synonym in modern usage. I've noticed this pattern across constructors who prioritize fill over clue elegance. The clue uses an archaic or overly formal synonym that nobody actually uses, while the answer is a perfectly ordinary word. That's the reverse of what most good clues do. Normally, you'd see a fancy word as the answer and a plain-English clue. Flipping it makes the solver feel like they're being tested on vocabulary trivia rather than solving a puzzle. It's a cheap trick, and constructors know it.
The real problem isn't that the clue has one correct answer. It's that "perpetrating" and "committing" share only a narrow sliver of synonym overlap, and that overlap lives entirely in crossword databases. In the real world, perpetrate carries a connotation of doing something harmful or illegal with a deliberate, almost theatrical quality that commit doesn't necessarily carry. A jury perpetrates nothing. A con artist perpetrates schemes. Commit works for crimes, for promises, for suicide. The semantic range doesn't actually match. I learned this the hard way when I spent twenty minutes on a similar clue in a USA Today puzzle — "Doing, in a legal sense" with the answer PERPETRATING — and almost filled in COMMITTING before a crossing forced me to reconsider. It was a bad clue. The intended answer was defensible but only barely. Here's what most solvers don't realize: clue fairness depends heavily on the publication. A New York Times clue is held to a higher standard than a regional paper's Wednesday afternoon fill. The constructor standards vary. Some outlets run clue reviews. Others don't. When you're dealing with smaller constructors or self-published puzzles, the "is this even fair?" question becomes much harder to answer because there's no editorial gatekeeper catching the loose synonymy. If you want to check whether a clue like this is genuinely fair, my workflow is simple. I look up the clue in the Crossword Tracker or the NYT clue database and see how often that particular word pairing has appeared. If "Perpetrating" = "COMMITTING" shows up in fewer than five instances across all major puzzles, it's likely a stretch that the constructor pulled from a thesaurus rather than from actual usage. That's a red flag. I also cross-reference with Merriam-Webster or Oxford. If the dictionary lists "commit" as a synonym for "perpetrate," that's one thing. If it doesn't, the clue is basically a guess dressed up as a definition.
Another thing people miss: the difficulty curve within a single puzzle matters more than any individual clue. A Monday "Perpetrating" clue is acceptable because the puzzle is supposed to be easy and the crossings will guide you. A Friday or Saturday version of the same clue is where constructors should be held accountable. By the middle of the week, solvers are expected to handle harder vocabulary and trickier wordplay. When a constructor lazily drops a thesaurus synonym on a Thursday, it feels like a betrayal of the contract between writer and solver. The workaround I use when I suspect a clue is unfair is to look at the theme or the surrounding clues for context. Sometimes the answer makes sense within the puzzle's broader theme even if it doesn't make sense in isolation. Other times, it's just a bad fill word and there's no saving it. In those cases, I mark it, move on, and come back later. Filling in the crossing answers from other clues often reveals whether the intended word actually fits or whether the clue is forcing a wrong answer into a good grid position. I've also started keeping a personal spreadsheet of clues that felt unfair. Not for publication or anything dramatic, just for my own reference. After a few dozen entries, a pattern emerged. The constructors who consistently push loose synonyms tend to be newer to puzzle design or they're working under tight word-count constraints that force them to prioritize grid fill over clue integrity. Experienced constructors know that a good clue is a conversation, not a gotcha. When someone writes "Perpetrating" as a straightforward definition clue for COMMITTING, they're not having a conversation. They're checking a box.
Get the Full Details
For anyone who wants to dig into this more, the American Crossword Puzzle Tournament has published guidelines on clue fairness over the years. They're not rules, but they reflect a community standard that most serious constructors follow. The basic principle is simple: a clue should lead to one unambiguous answer that a reasonably informed solver could arrive at through legitimate deduction, not through lucky guessing or obscure vocabulary knowledge. "Perpetrating" for COMMITTING fails that test on a medium-difficulty puzzle because the synonym relationship is too narrow and too crossword-specific. I don't think constructors should be expected to write perfect clues every time. Puzzles are hard to make, and some misfires are inevitable. But when the same type of lazy clue shows up repeatedly, it's worth calling out. The crossword community is small enough that constructors pay attention to feedback. A well-reasoned complaint about a specific clue gets more response than a generic "your puzzles are bad" post. If you want to push for better clues, be specific about what made the clue unfair and suggest a better alternative. That's how the standards improve.
