Growing up on the cryptic crosswords I found in Games magazine and The New York Times, I'm used to a very rigid rule for clues: Every clue comprises two parts, one of which directly (if opaquely) clues the word as in a non-cryptic crossword and the other of which does so cryptically. (These may be separated by "is" or "to yield" or some similar copula. And so-called &lit clues are an exception.) The cryptic half, in turn, has some fairly rigid rules: If it uses an anagram, it must indicate as much (thus, e.g., "turned handle" for HANDEL, never just "handle" for HANDEL); and if it hides a word within another, it must indicate as much (thus, e.g., "in the best" for THEBES, never just "the best" for THEBES).

But I've seen cryptics — mostly, I think, from Britain — whose clues don't follow these rules. They seem to allow anagrams and words hidden within other words even without signaling as much, and they even seem to allow cryptic-half-only clues, without a non-cryptic component of the clue. What rules, if any, do such crosswords follow for clue composition?

  • $\begingroup$ Cryptics in the NYT? $\endgroup$ – Kevin May 15 '14 at 1:22
  • $\begingroup$ Also, I've seen a lot of British cryptics with anagrams and embedded words, but all I've seen indicate both the anagram and the non-cryptic answer. Do you have a particular example? $\endgroup$ – Kevin May 15 '14 at 1:24
  • $\begingroup$ Are you sure there were no signals? The signals aren't always obvious. For example, here's a list of over 700 anagram indicators. $\endgroup$ – Gilles 'SO- stop being evil' May 15 '14 at 2:30
  • $\begingroup$ @Kevin, re the NYT: Sunday, beneath the main crossword, IIRC. It's been some years, though; maybe I'm mistaken. $\endgroup$ – msh210 May 15 '14 at 2:33
  • 1
    $\begingroup$ @Kevin see my new answer here puzzling.stackexchange.com/a/26221/4050 for some examples of British clues without 'straight' definitions. $\endgroup$ – A E Feb 7 '16 at 20:05

Broadly speaking, there are two styles of clues in British crosswords.

Ximenean clues (named after the pseudonym of Derrick Somerset Macnutt) adhere to a fairly strict recipe of [precise definition] + [cryptic indication], and are probably the sort of clues that American solvers would be most familiar with. For example:

I feel wretched getting no Latin to penetrate (5)

IMBUE — (I'M BLUE, without L for Latin)

Ximenes also favoured another type of clue called Printer's Devilry (PD), where the clue is a phrase from which a word has been removed, and then the letters rearranged to disguise the fact, e.g.:

She lay in the sunny meadow, chew in gab, utterly (5)

CUPID — (chewing a butterCUP IDly)

However, PD clues are generally not mixed with other types of clue.

Libertarian clues break the rules in various ways. For example, they will use less precise language in the definition or cryptic parts as long as the clue remains sufficiently unambiguous. Solving them typically requires a leap of the imagination. The late lamented Araucaria (who sadly passed away last year) was one of the best setters of this type. Some examples:

A forbidding place (7,4)

Auction Room

Of of of of of of of of of of (10)


Yogdaws (3, 5, 2, 10, 4)

God moves in mysterious ways

By their very nature, Libertarian clues are difficult to define in terms of a set of rules. But they all make perfect sense once you know what the answer is.

| improve this answer | |
  • 2
    $\begingroup$ It's also common for libertarian crosswords to carry the byline of the composer, so you have some idea of what to expect. Anonymous crosswords (e.g. in the Times) are more likely to be Ximenean. $\endgroup$ – Rupert Morrish Sep 12 '17 at 2:58

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.