There has been much discussion about the relative merits and de-merits of this puzzle. Whilst recognising many of the admirable attributes of this puzzle, I still think it’s pretty poor overall and I would like to have the opportunity to explain why.
Back in the 1960s, the difficult (worth solving) puzzles were by Ximenes, who wrote a book called The Art of the Crossword which became the bible for setters of that period. In those days, puzzles were mostly about solving clues; the theme-based endgame had yet to be invented. Derrick MacNutt (Ximenes) had much to say about what constituted an acceptable clue and what not. As I recall it all boiled down to two simple rules:
1) The setter does not have to say what he means, but he should mean what he says.
2) However hard a clue is, once the the solver has found the correct solution it MUST be 100% clear to him that he / she is correct.
These days we have The Listener with its theme based endgame. If Derrick was still alive, I’m sure that he would demand the same two rules for the endgame.
Let’s look at a couple of examples of recent Listeners to see what this means in practice.
In September 2012, we were treated to Listener 4207 – Loco by Ferret. The rubric was open to a number of interpretations, but when the solver found the correct one he was rewarded with the BR Logo. It is an example of a self-confirming endgame completely obeying Rule 2. The result is a fulfilling puzzle for the solver, many appreciative thanks and hardly any criticism at all. In fact, only a self-confirming endgame can ever truly obey Rule 2.
Now let’s look at Listener 4222 – 27 by Mango. It starts off rather badly by revealing the entire theme at the clue solving stage. And I do mean entire. Not just the general theme, but the exact phase of the theme which is the subject of the endgame. From this point, no self-confirming endgame is possible short of telling the solver precisely what to do. But we have to have an endgame and it can’t be made too easy, so let’s concoct something vaguely (hardly at all in parts) related to theme and then let’s not really tell the solver what to do. The result is a puzzle for which several of us can produce at least 4 different grids, each of which fulfils the demands of the rubric. An endgame which does not lead to a unique solution is utterly pointless. For the solver there’s no real PDM in this endgame – it’s as rewarding as sex without climax.
It is quite possible to study the rubric and have complete understanding of every last nuance of the theme without necessarily producing the correct grid. The only question that the setters appear to be asking of the solver in the endgame is “Are you thinking what we’re thinking?” Well the answer is, “Who knows? Maybe, maybe not.”
It’s a question that didn’t do Michael Howard much good, and it’s one I hope to never see again in a Listener Crossword, though it’s unavoidable to some degree in any non-self-confirming endgame.
I can assure those that think this puzzle is OK that the questions are not coming from those who haven’t studied the rubric. On the contrary, I think we may have studied it too much. You may well have found the required final grid, but just how confident are you really? If Ladbrokes took bets on the Listener would you gamble £20 on your solution being correct? Then let me ask you this. “Would you bet your house on it?” If the answer is negative, then the endgame does not comply with Rule 2.
Well, I’ve had my say now and I’ve submitted my last Listener Crossword. Does this mean I’m throwing the toys out of the pram? Definitely not. I’m going to try my hand at setting. Will I ever do a Listener again? Probably, when the weather is awful and I have some free time, but I won’t submit it. In the meantime, so that we don’t all waste too much time on future puzzles, would the first person to post on this forum each week kindly tell everybody else whether the endgame is self-confirming or not? Thanks.