Listen With Others

Are you sitting comfortably? Then we’ll begin

3966 – Mercury's Whereabouts by Dysart – Setter's Blog

Posted by Listen With Others on 15 Feb 2008

The idea for this puzzle came to me shortly after I’d compiled my first puzzle, Deal. The novel has always been a favourite of mine, and the Joseph Losey film, starring Julie Christie and Alan Bates, scripted by Harold Pinter, is a delightful adaptation. The title invited cryptic treatment so I set about compiling a list of approximate synonyms for “go” and constructing a grid; at that stage my only worry was that I could waste a lot of time for nothing if the theme had already been used before in The Listener.
From the outset I decided to make it tougher than Deal, though I didn’t intend it to be as tough as it eventually turned out to be. While I was working on it, Gos’s Witness Protection was published, with a tough set of clues and jumbled entries that were moved in the grid, and I think this suggested a standard of difficulty for my puzzle.

I felt that the best route to the novel was through the memorable opening words of the prologue, “The past is a foreign country”, and this dictated the choice of 24 across or down entries. The quotation is fairly familiar and is in the compact version of ODQ, conveniently indexed under “past”, “foreign” and “country”. This meant that solvers struggling after solving only a selection of across clues might have enough to recognize one of these words and check in ODQ.
My first grid was symmetrical. Initially I experimented with words that could become new words with the insertion of synonyms for “go”, but this proved to be too severe a constraint and I quickly abandoned the idea, but it did give me “whereabouts”, thus supplying the title of the puzzle. With greater flexibility in the words to be used and their precise placement in a selection of across and down entries, I soon had a usable grid. However, after writing a complete set of clues I became conscious of what I regarded as a flaw: some of the synonyms for “go” that I’d used were verbs, but the definite article in the novel’s title, cryptically interpreted, really demanded words that could function as nouns. I therefore scrapped everything and started again. Now the task became far harder as the choice of suitable words was far more limited. With the other constraints of a central block of 12 cells in which LEO was sandwiched between MARIAN and TED, and the need for exactly 24 across clues, I found it impossible to produce a symmetrical grid.
I decided to place the “goes” in the down entries to avoid overloading across clues/entries with abnormalities; it was at this point that I decided to limit myself to 13 words and 13 down clues. Even with the luxury of an asymmetric grid it was still not easy, and I had to experiment a lot, shifting the “goes” around. One of the most problematic features of the final grid for solvers is the number of unchecked letters in PLEONASTE in the central column. Other experimental grids I produced all exhibited this problem somewhere or other. In the end I decided to risk it with the editors, writing a clue that provided all the necessary letters in the reference to 7 across and the letters of EAST, but I fully understand any solvers who found solving the clue a frustrating experience, especially if they hadn’t solved 7 across or didn’t make the connection. I tried to compensate elsewhere by making the checking as generous as I could. Twelve of the across entries have no unchecked letters, and most of the 8-letter or 9-letter answers to down clues have no more than 2 unchecked letters.
Initially the down clues were presented normally and it was my intention that solvers should enter synonyms for “go” wherever they would fit. However, that presented some possible ambiguities where there were unchecked letters; I could have given solvers an anagram of unchecked letters, but it struck me as more consistent thematically to use the extra words in the clues. At the time I placed the words wherever they best fitted the surface of the clue. Now, in hindsight, I think I should have tried to ensure that all the extra words were placed within the clues (i.e. not at the beginning or end) so they would be “between” in the clues as well as in the grid. As it is, in three cases the synonym for “go” starts the clue. It’s not something I noticed until shortly before publication date.
Some of the extra words were tricky to disguise in the clues, and this may have led to some oblique wordplay or definitions. Where possible I wanted the meaning of the word in the clue to be unrelated to its thematic meaning. Thinking of the household cleaner when I incorporated ‘vim’ in the clue for ‘deer’, I saw that the latter was just the first five letters of DETERGENT minus T; the old chestnut of “does perhaps” for ‘deer’ suited the surface nicely, so I eschewed originality and used that as the definition. ‘Pep’ was another word that was difficult to incorporate into a clue without sticking out like the proverbial sore thumb. Once I turned my attention to “spell” I couldn’t resist using “spellbinding” for a container clue; I thought the editors might object, so I added a preamble warning and offered two alternative clues in which ‘spell’ was a separate word; the preamble warning was enough to satisfy them.
Once I’d written the clues it occurred to me that an extra element to the puzzle could be the presentation of the down clues in a sequence matching the occurrence of GO words in each column, left to right. My intention wasn’t really to add difficulty for it’s own sake, but to create a penny-dropping moment when solvers realized what was going on. I’m not sure this was successful since it would appear that for many the penny dropped very late in the solving process, or didn’t drop at all. This is probably the one feature that adds most to the difficulty of the puzzle and had I realized how tough it was going to be I might have dropped the idea, but after it had been test-solved without too much difficulty (by a very accomplished solver, it must be said) I decided to submit it in that form, adding a note to the editors at the end of my solution saying that if they felt the disordered downs made the puzzle too hard I would not raise any objections to a normal sequence. Since that offer was not taken up I assume the editors were happy to have a tough puzzle to schedule.
The only significant feature of the puzzle that the editors changed was the highlighting requirement (a change that I was entirely happy with). In my original submission solvers were required to highlight LEO only; the intention was to make solvers look for a justification for choosing between the vertical LEO in the centre and the horizontal one in the north-east corner, but both the test-solver and the first editor highlighted the central LEO (presumably on the grounds that he is the central character) without noticing the other characters. Incidentally, the existence of MARION as well as MARIAN was coincidental, not a deliberate trap. If it had been a trap I don’t suppose anyone would have fallen for the wrong spelling at the highlighting stage, since MARION and LEO share the O, so it would have been neither elegant nor justified.

One Response to “3966 – Mercury's Whereabouts by Dysart – Setter's Blog”

  1. obviously like your web site but you have to test the spelling on quite a few of your posts. Several of them are rife with spelling issues and I to find it very bothersome to inform the truth then again I will definitely come again again.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: