The Musings of Doctor Clue

It is generally considered unacceptable for a setter to arbitrarily remove a capital letter which is required by the cryptic reading of a clue. So ‘Take one around nice course’ for RUN is no good, since ‘nice’ has to be ‘Nice’ for the wordplay (R + UN) to work. So how about ‘Pass river, one in Nice’, also for RUN? Superficially, that looks ok. But a check in Chambers tells us that R is an abbreviation for ‘River’ (with a capital letter), but not ‘river’.

We could ask ourselves “Is the abbreviation ever used in real life to indicate the complete word without the initial capital?” In this instance (for ‘R=river’) the answer seems to me a definite “No” – the abbreviation is only seen in the names of specific rivers (typically on maps) as eg ‘R. Thames’, and is never used generically. So saying ‘river’ when you mean ‘River’ is really no different to saying ‘nice’ when you mean ‘Nice’, and presumably shouldn’t be allowed.

Apart from ‘river’, such a prohibition would rule out several old favourites lacking an initial capital: ‘street’, ‘road’ and ‘lake’ from the ‘geographical’ group, as well as ‘society’ (for ‘S’) and ‘king’ (for ‘R’, though not for ‘K’, which is  an abbreviation for ‘king’ in the context of chess or cards).

Knowing this, would I use, say, ‘river’ for R in a clue? Well, since I suspect that there isn’t a single solver out there who would bat an eyelid, I certainly wouldn’t reject a good clue purely to avoid it. I think it’s fair to say that some wordplay elements are so well established that it makes no sense to unilaterally boycott them, but this does show that even at a general level there is on occasion a lack of consistency when it comes to what is deemed allowable and what is not.

Like old soldiers, old words in the English language never die, their memory being preserved by dictionaries such as Chambers. The Big Red Book has a number of classifications for words which are not in common use, including ‘old’ and ‘rare’, but the largest groups consist of those categorized as ‘archaic’ or ‘obsolete’. There is, I think, general agreement that when such words appear in puzzles they need to be flagged by the setter, eg ‘old chestnut’ for the obsolete term FAVEL.

There seems to be a tendency among setters to treat archaisms and obsoletisms identically, but let’s consider what the two classifications mean:

• archaic. Describes words which are ‘not absolutely obsolete but no longer in general use’; they will typically have been common at some point in the past, but that could have been a long time ago. Examples would be immeritous for ‘undeserving’ and mouldwarp for a mole.
• obsolete. Words classified as obsolete may once have been common but are now completely out of use. Among such words are disembrangle (‘to free from dispute’) and ellops, a kind of sturgeon.

I recently saw the definition of a word shown by Chambers as ‘archaic’ being qualified in a clue by ‘dead’. This seems wrong to me – terms indicating total absence from today’s language are fine for obsolete words, but archaic ones are still hanging on in there, so although ‘once’ or ‘former’ is fine for an archaic word (or an archaic sense of a word), I don’t think ‘dead’ is valid. An ‘extinct sturgeon’ could be ELLOPS, but a MOULDWARP surely isn’t an ‘extinct mole’.

A question: what attribute was shared by the first crosswords published in The Times and The Telegraph, but not the first Guardian puzzle?

The answer is 180° rotational symmetry. It’s something that we take for granted, particularly in blocked puzzles, but why is it there, and would we care if it wasn’t?

I don’t think anyone is sure why it’s there, although it certainly simplifies grid construction (produce half the grid and the other half will look after itself) and checking, while the resulting layout is aesthetically pleasing. With the exception of the occasional barred puzzle, every crossword published in Britain will have at minimum 180° symmetry, with the odd one adding 90° symmetry on top of that. The main grid of every puzzle I’ve ever set has been symmetrical, and I’m a little put out when I see a barred puzzle where symmetry has been sacrificed to the demands of the theme. But why? The rules around unchecked letters (‘unches’) and disconnected sections of the grid (there were eight completely separate blocks in the first Telegraph puzzle, by the way) are there for reasons of fairness to the solver, but symmetry (except in skeleton or carte blanche puzzles, where the solver has to locate the blocks or bars themselves) has no effect of solvability.

I think the look of a puzzle is important, and I find a grid which is not symmetrical around a line or axis jarring, so my rating of an ‘offending’ puzzle will be downgraded accordingly. But I accept that this may be tantamount to criticizing Picasso’s artworks simply because they don’t look like paintings ought to. Views, anyone…?