Crossword Help Forum
Forum Rules

In the Guardian cryptic 28,328, clue 23, Realm is supposed to be the definition for Egypt. I've seen categories before that were used to point to a specific instance of that category, but Realm feels like a broad category to me, that makes the definition too vague. This made me wonder about whether there's much consensus around what makes a good definition.

Hi Bigblind,

Definitions by example are usually clued with "maybe" or "perhaps" - I found this on Google:-

The definition in a cryptic clue is not always a synonym of the answer; it may be an example or sub-type of the answer. So, NOVELIST may be defined as "Dickens, perhaps", DOG as "setter, maybe".

It'll be interesting to hear what others think.

Anguish at heart of alien realm(5)
Anguish at heart of alien realm,maybe(5)
Anguish at heart of alien realm, perhaps(5)
Anguish at heart of alien realm?(5)

From the setter's point of view- perhaps and maybe- are a pain for the surface of the clue. I think a simple ? works as an indication for the solver to think outside the box but isn't that where we should be perhaps? Maybe!

I was dubious about this definition when doing the puzzle yesterday. I'm not sure that an indicator of a definition by example really helps. I'm no expert, but I don't think "realm" works as a definition of Egypt under any circumstances.

The EGYPT clue isn't really a definition by example, not according to standard convention. As Brendan correctly points out, Dickens and setter are DBEs of DOG and NOVELIST respectively, because Dickens and setter are just one example of a much larger group. Using "realm" to lead to the answer EGYPT is not a DBE, because realm is not an example of EGYPT. On the other hand, a clue which used "Egypt" as a definition to lead to the answer REALM would be a DBE.

(Whether modern Egypt really is a realm is another matter.)

I've noticed that The Times in particular has become less insistent on indicators like "maybe" for DBEs recently, in certain cases. There are reasonable grounds for this. For example, "Dachshund" to clue DOG is a DBE and technically this needs to be indicated in the clue. But when we see "dachshund" the first thing you think of is DOG, so do we really need the extra nudge?

A reductio ad absurdum of the DBE principle is something like

Zsa Zsa perhaps upset Garbo (5)

I mean, how many people called Zsa Zsa do we know of?

Correction: ...Dickens and setter are DBEs of NOVELIST and DOG respectively... Duh.

As loge says, 'realm' for EGYPT is not a definition by example (it is the reverse - a definition by generalization, you might say, as would be 'dog' for SETTER).

When assessing the validity of definitions, one has to remember that the term 'definition' in a crossword context is used specifically to distinguish a part of a clue which leads directly to the solution from a part which enables the solver to piece the solution together ('wordplay'); it does not have the same precise sense that it would in a lexicographic context. 'Indication' is probably a more accurate term (sometimes the definition is called the 'primary indication' and the wordplay the 'subsidiary indication').

That said, the 'indication' has to be fair. Setters often mistakenly assume a transitive property of word meanings, ie if a=b and b=c then a=c. 'Soft', for instance, can mean 'tender' or 'politically moderate', but 'tender' doesn't equate to 'politically moderate'.

Regarding Egypt, Chambers will tell you that a realm and a kingdom are similar things, and that 'kingdom' can be used to describe 'a region that was once a monarchical state'. Therefore Egypt is a 'realm'? I think not. 'Country' or 'nation' - fine.

Interesting point about the definition of "realm", Spoffy. Egypt had kings between 1922 and 1952 (or thereabouts) - and in ancient history of course - so "once a monarchical state" does possibly qualify it as a kingdom or realm. But by that token France could be called a kingdom/realm too, and it doesn't sit quite right, does it?

Agree with comments from Loge and Spoffy.
In addition, context can be important in determining if a definition indication is ‘fair’. For instance if the wordplay is complex or unusual then the definition indication should ideally be more straight forward, and vice versa. However, too many generalised definitions in the context of the entire puzzle can be frustrating, especially if such clues cross each other.

In addition to DBE, it might also be considered fair to include indicators for archaic words, American usage or unusual spelling variants etc.

The EGYPT clue would be very difficult in isolation but the fact that the comparatively straightforward HUDSON BAY clue gave the central Y made getting to Egypt a little easier. (I personally don’t think that Egypt can be described as a realm, perhaps 'ancient realm' would be ok but it wouldn’t fit into the word play. Also as GYP is a fairly unusual synonym for anguish the clue becomes less solvable).

A similar example of a general definition pointing to a specific country is:

Country with its capital in Czechoslovakia (6)

Here, however, because we have Oslo in the clue there is no ambiguity in getting to NORWAY.

There was an interesting discussion on Fifteen Squared recently about whether Stravinsky (in the wordplay) could lead to IGOR (as part of the answer) fairly without a DBE indicator. Clueing a forename by a surname (or vice-versa) adds an extra level of complication: is Stravinsky an example of people called Igor, or is Igor an example of people called Stravinsky?

I'd say that it's technically a DBE either way round, but in the example here, where Stravinsky is the indicator, what else could it lead to other than IGOR?