The List

Word Definitions

A word definition is a rule for when a string of letters counts as a word, stated precisely enough that a solver can test it mechanically — and the hard part is never the common case, it is the boundary. Every crossword, word-search, anagram or Scrabble-style puzzle reduces to the same three decisions: what counts as a word, what counts as a match, and what counts as a valid answer. Get those in the right order and the puzzle is fair; get them in the wrong order and you end up with an argument at the table.

5 articles

Word Definitions

Word Definitions

Newest first
Defining words related to stirring up in crosswords
Word Definitions

Defining words related to stirring up in crosswords

To successfully solve a standard Sudoku puzzle, you must remember that the playing field is governed by three simultaneous constraints: every row, every column, and every one of the nine 3×3 subgrids…

16 SepOpen →
Finding the meaning of a pizzazz clue in puzzles
Word Definitions

Finding the meaning of a pizzazz clue in puzzles

If a "pizzazz" clue feels meaningless, it is almost always because you are interpreting it as an independent constraint when it must actually be mapped onto one of the existing structural rules…

11 SepOpen →
Fun synonyms for big fuss across different puzzles
Word Definitions

Fun synonyms for big fuss across different puzzles

If you are looking for synonyms for "big fuss" when describing a puzzle’s difficulty or complexity, I strongly recommend adopting a contextual approach rather than relying on simple synonym lists.

7 SepOpen →
Solving definitions related to hubbub and commotion
Word Definitions

Solving definitions related to hubbub and commotion

The core definition of a standard Sudoku puzzle requires that every row, every column, and each of the nine distinct 3×3 subgrids must contain all digits from 1 through 9 exactly once, with no…

4 SepOpen →
Solving common definitions like commotion in crosswords
Word Definitions

Solving common definitions like commotion in crosswords

When a definition like commotion pops up in a crossword clue, the modern approach emphasizes not just finding the direct synonym, but recognizing the specific type of wordplay requested…

29 AugOpen →

A word definition is a rule for when a string of letters counts as a word, stated precisely enough that a solver can test it mechanically — and the hard part is never the common case, it is the boundary. Every crossword, word-search, anagram or Scrabble-style puzzle reduces to the same three decisions: what counts as a word, what counts as a match, and what counts as a valid answer. Get those in the right order and the puzzle is fair; get them in the wrong order and you end up with an argument at the table.

What a dictionary actually promises, and what it doesn't

A dictionary entry is a record of usage, not a rule of law. That distinction matters the moment a definition leaves the page and becomes a puzzle constraint, because a solver needs a yes-or-no, and a dictionary gives you a history. The two things you must separate are lexical status (is this a word at all?) and puzzle status (does this puzzle accept it?). They overlap heavily and are almost never identical.

Lexical status has gradations. At the firm end sit headwords with their own entries. Further out sit inflections and derived forms — -ed, -ing, plurals — which many dictionaries fold into the base entry rather than listing separately. Further still sit compounds written open, hyphenated or solid, and further yet the productive formations that any speaker can build on demand. A dictionary is under no obligation to list every one of those, and none of them does. So "it's not in the dictionary" is a statement about the dictionary's editorial policy as much as about the word.

Then comes the question of which dictionary. British and American lexicography differ in coverage, in how quickly they admit new coinages, and in how they treat regional and dialect vocabulary. A word perfectly standard in one tradition can be absent from a major dictionary in the other. For puzzle work this is not pedantry: it is the single most common source of disputed answers, and it is why serious puzzle sets name their reference rather than gesturing at "the dictionary."

The limit of this section is real. No amount of careful definition fixes a puzzle whose answer key was built from an unstated list. If you cannot say which reference decided a word's status, you have not defined anything — you have deferred to whoever compiled the key.

Word lists, and why the reference you pick decides the argument

Competitive and published puzzles do not consult a dictionary at solve time; they consult a fixed word list agreed in advance, so that every player is working against the same boundary. That is the correct architecture. A dictionary is a moving target — revised, expanded, re-edited — and a word list is frozen. When you freeze it, you convert a debate about English into a lookup.

The trade-off is that a frozen list is arbitrary at its edges in ways a dictionary is not. Lists built for word games optimise for being checkable, which means they tend to include short obscure forms and exclude longer transparent ones. Lists built from general dictionaries optimise for being recognisable, which means the reverse. Neither is wrong; they answer different questions. If your puzzle's pleasure comes from finding a clever ordinary word, a recognisability-first list serves you. If it comes from the hunt through unlikely letter combinations, a checkability-first list does.

Two details decide most disputes. First, whether the list includes inflected forms as separate entries or expects you to derive them — this changes the answer to "is -s allowed" from puzzle to puzzle, and solvers who assume one convention will lose points under the other. Second, whether it includes proper nouns. Most word puzzles exclude them, but the exclusion needs stating, because a capitalised name is a perfectly good string of letters and the solver has no way to know your policy except that you told them.

Say the reference and the version. A list is a document with a date; a puzzle built on last year's list and solved against this year's will produce exactly one or two disagreements, which is enough to poison a competition. The cost of naming your source is a line of rules text. The cost of not naming it is every disputed answer for the life of the puzzle.

Matching rules: exact strings, case, accents and spelling variants

Once you know what a word is, you must decide what counts as the same word, and this is where most puzzle software quietly makes choices for you. The decision order is: case, then diacritics, then spelling variant, then inflection. Each step down that list widens what the puzzle accepts, and each widening is a design choice, not a default you can leave implicit.

Case. Folding everything to one case makes a word-search or anagram grid behave predictably. It also means a grid cannot distinguish a name from a common noun, so if you fold case you have effectively admitted proper nouns whether you meant to or not.

Diacritics. A loanword printed with its accent in the dictionary and typed without it on the answer line is the same word to a human and a different string to a machine. Either strip accents on both sides of the comparison or require them on both — but pick one, and say so, because a solver who loses a mark to a missing accent will not accept "the software decided."

Spelling variants. British and American forms of the same word, and the legitimate variant spellings within a single tradition, are the greyest area. Treating them as interchangeable is generous and occasionally wrong; treating them as distinct is strict and occasionally petty. The honest move is to state the convention for the specific puzzle rather than assert a general principle, because there isn't one.

Inflection. Whether run and running are one answer or two is not a fact about English. It is a rule about your answer key, and it must be fixed before the key is built, not after a solver submits the form you didn't want.

The limit: every one of these rules is a tolerance setting, and tolerances that are generous in one place tend to be exploited in another. Loosen case and you admit names; loosen variants and you admit spellings half your solvers have never seen. There is no setting that is both maximally fair and maximally strict. Choose, publish, hold.

Definitions as clues: the difference between a gloss and a riddle

A clue is not a definition, even when it is built from one. A dictionary gloss tells you what a word means; a cryptic or crossword clue tells you what a word means and then hides that fact inside wordplay. Confusing the two produces clues that are technically accurate and completely unsolvable, because the solver is given the meaning without the mechanism.

The mechanism is the contract. In a well-made clue, every word is accounted for: part of the surface reading is the definition, and the rest is an instruction — a container, a reversal, an anagram indicator, a hidden-word signal. The definition component is usually a direct synonym or a tight phrase, and its quality is judged by whether it could stand as a dictionary gloss for the answer without the rest of the clue. If it couldn't, the definition is loose, and loose definitions are where clues become guesswork.

Wordplay, by contrast, is judged by whether it is fair rather than whether it is true. A hidden-word clue that genuinely conceals the answer across a word boundary is fair even if the surface sentence is nonsense. An anagram whose indicator does not signal anagramming is unfair even if the answer is right. This is why clue-writing is a craft separate from lexicography: the lexicographer's job ends when the meaning is captured, and the setter's job begins there.

What this costs: tight clues take longer to write than loose ones, and a set built on loose definitions will still be solvable by a strong solver who simply guesses the setter's habits. That is the failure mode to watch for — when solvers start solving the setter rather than the clue, the definitions have stopped doing work.

Ambiguity, homographs and the words that break the rules

The words that cause trouble are rarely obscure. They are the ordinary ones with more than one sense, more than one pronunciation, or more than one spelling — and puzzles force you to choose a sense before the solver has any way to know which. A definition that is accurate for one sense and misleading for another is a defect even though nothing in it is false.

Homographs are the sharpest case. Two words spelled identically but unrelated in origin will share an entry or split across two, depending on the dictionary's house style, and either treatment is defensible. For a puzzle, what matters is that the clue's definition component pins down a single sense. If the answer could be reached through two different meanings and only one is clued, the other is a coincidence, not a solution path.

There is a broader category worth naming: strings that behave like words without being listed as such. Abbreviations, symbols read aloud, initialisms, and the productive compounds that any speaker can form but no dictionary bothers to record. Each puzzle must rule on these explicitly, because the general principle "it's a word if people use it" is true and useless at the answer key.

The limit here is that over-specification has a cost too. A rules page that tries to anticipate every ambiguity becomes longer than the puzzle and no more decisive, because the truly awkward cases are the ones nobody thought of. The practical compromise is to rule firmly on the categories — proper nouns, inflections, variants, abbreviations — and accept that a handful of individual words will be argued about. That residue is unavoidable; the categories are not.

Where word definitions meet number and letter grids

Grid puzzles make the definition's job sharper, because the grid enforces constraints the solver can check independently of meaning. The clearest example is Sudoku: a logic-based number-placement puzzle played on a 9×9 grid subdivided into nine 3×3 boxes, where the objective is to fill all 81 cells so that every row, every column, and every 3×3 box contains the digits 1 through 9 exactly once, with no repeats. That is a definition doing real work — it specifies the playing field (81 cells arranged in 9 rows and 9 columns, divided by thicker lines into nine 3×3 subgrids), the symbol set (digits 1 through 9), and the constraint on each row, each column and each box.

Compare that to a word puzzle. Sudoku's rules are complete: a filled grid either satisfies them or it doesn't, and no appeal to usage is required. A word puzzle's rules are never complete in the same way, because "counts as a word" cannot be settled by inspection. This is the structural difference between the two families, and it explains why Sudoku disputes are about solving logic while word-puzzle disputes are about vocabulary.

It also explains why Sudoku travels so well. The definition is short, the constraint is total, and the puzzle needs no reference list at all. The digits carry no meaning; only their distinctness matters. Swap them for nine letters or nine colours and the puzzle is unchanged — which is precisely the property a word puzzle can never have, since the meaning of a word is the whole point of using it.

What the number-puzzle literature does not establish is worth stating plainly. Research on number puzzles and cognition in older adults has assessed 14 cognitive measures in adults aged 50 to 93, which tells you the studies were broad in their measurement — and broad measurement is exactly the kind that struggles to isolate a single cause. Nothing in that design licenses the claim that a specific puzzle habit produces a specific cognitive benefit. Treat any such claim as unsupported.

The practical consequence for anyone building or choosing puzzles is a priority order. First, make the constraint checkable: state the grid, the symbol set, and the repetition rule so a solver can verify an answer without asking you. Second, make the vocabulary reference explicit, with its version. Third, decide the tolerance rules — case, accents, variants, inflections — and publish them alongside the puzzle rather than in a forum thread afterwards. Fourth, write definitions that pin a single sense. Everything after that is craft, and craft is where puzzles earn their reputation.

Get the order wrong and the failure is predictable: a solver produces a defensible answer, you produce a defensible objection, and neither of you can point to a rule that settles it. Get it right and the argument never happens — which is the whole value of a definition, and the reason it is worth more attention than it usually gets. The definition is not the boring part before the puzzle. It is the part that makes the puzzle a puzzle rather than a disagreement with a grid drawn around it.