Nine hours later, with a blazing headache and a sore throat from cursing the setter (apologies to whoever it was!) I filled in the last entry. Well, I thought, this was clearly a one-off. Nobody’s going to tackle puzzles that hard every week, are they?

Move forward a few weeks to when I next bought the Saturday Times (I didn’t buy the paper regularly in those days). There, in all its glory, was another Listener crossword, something about bridge this time… which of course I attempted and again ended up with puffy eyes, a serious headache and the sense that this couldn’t go on. In those days I didn’t have a lot of free time or much of a social life and didn’t want to waste what little I did have in favour of spending a ridiculous amount of time on a single crossword. So, to avoid temptation, I made a point of not buying the Times on a Saturday.

I returned to the Listener two or three years later and completed it relatively quickly. It was probably an easy one, and of course my solving skills had improved considerably. From then on I have been a devotee of this puzzle and am happy to say that I can quite often complete it in two or three hours – helped of course by modern computer aids like wordfinders, anagram solvers and the Internet. I attempt the Listener near enough every week, but there are a few puzzles I don't bother with. These are:

Mathematical puzzles.

Four puzzles a year are mathematical. That means four weeks off a year for me. Now, I’m perfectly numerate and enjoyed maths at school, and certainly not one of those bores who show their ignorance by proudly boasting “I’m hopeless at maths, me.” The Listener maths puzzles, as required by the Notes for Setters, should not and do not require mathematical knowledge beyond GCSE level so there’s no advanced calculus or trigonometry or anything like that. My objection to these puzzles is that although the maths itself is relatively simple, solving these puzzles is usually a very hard slog requiring a lot of trial and error. If you’re an advanced computer programmer or, obviously, a mathematics professor, you can probably develop processes to solve these puzzles more quickly. For the rest of us though, what happens when we are confronted by a clue such as “Square, whose root is prime (5)”? You have to work out all the squares that have five digits, then test all their square roots to see if they are primes. The opportunity for missing one out (which will always be the right answer) is enormous and the whole process is a heck of a slog for which, frankly, I haven’t the time or the patience. In addition, these puzzles often have long, off-putting preambles (see below) and perhaps most important of all, they are not crosswords! However good a theme may be, some fun should be had solving the clues and there is no way that “Not a cube” – a clue which appeared in one mathematical puzzle – has any of the wit, ingenuity or enjoyment of the best efforts by Elgin, Chalicea, Phi et al. I appreciate that these puzzles are popular with some so would never suggest that they shouldn’t be included, but they are not for me.

Playfair puzzles.

The Playfair code square, historically used by the military, has been used in several Listener puzzles, usually to the tune of one a year. The rules are long and complex and rather than take up space here, this link explains the principle very well. The main problem I have with Playfair is that, in many cases, it is a fiddly and unnecessary obstacle tacked on to the end of a puzzle just to make it harder, perhaps to disguise a weak theme. I see little point in getting 95% of a puzzle done, then spending three times as long again trying to find the Playfair keyword just so that the last six answers can be entered in code. What’s more, the Listener Setters’ rules dictate that the keyword should not be able to be guessed from any thematic material in the puzzle. I wouldn’t mind so much if having got Julius Caesar, Cymbeline and Hamlet in the grid I was able to make an educated guess that Shakespeare is the keyword. But I don’t want to waste time just to discover that, having got all the other thematic material, the word needed to code the last few answers is something irrelevant like PLEONASTIC. In the past Playfair puzzles always meant another week off for me, although thanks to the advent of online code-breakers I am more likely to attempt them these days if the rest of the puzzle looks interesting.

Convoluted Preambles.

Imagine something like this:

“All answers are to be treated as Rosicrucians, Visigoths or Boy Scouts. The relationship between them can be determined by encoding half the letters in the grid (to be determined by the solver) with a code also to be determined by the solver. This gives instructions in an Ugro-Finnic language which may be interpreted in seven different ways. On discovering the relationship a Lorentzian transformation is required to locate five virtual elements in the grid...”

Some, but fortunately not many, puzzles contain preambles that make as much sense to me as the nonsense I’ve written above. I find my eyes glaze over before I’ve got to the end and my response is “Sorry mate, no can do. I’m off to the pub.” Often these are the puzzles which attract the most praise on crossword forums which suggests (a) I’m too stupid to appreciate the really good Listeners (b) some people like to show off how clever they are or (c) a mixture of the two. I have no doubt that these puzzles are superbly crafted, and that if one puts in enough time they can be very satisfying to solve. The fact is, I’m not willing to spend twenty or so hours struggling with a puzzle that, by the looks of the preamble, is probably beyond me anyway. Again, I add that as these puzzles are popular with some solvers I have no problem with their inclusion – I don’t subscribe to the current trend of banning things that a significant number of people enjoy just because those things are not to my taste.

That still leaves me with well over 90% of Listener puzzles to enjoy. These days I finish almost all of the ones I attempt, but of course I am defeated every now and then, usually by puzzles in which the tail wags the dog. By this I mean that it takes thirty minutes to fill the grid, then another four hours trying to work out the last step. I am not keen on this type of puzzle, especially if this last step turns out to be rather tacked-on and inconsequential. I’m reluctant to spend longer on the endgame than it took to fill the grid, and in the past that would mean giving up and waiting for the solution to be published. Nowadays there are sites where people trade answers and hints, and if I really don’t think I'm going to reach the end unaided I will resort to them.

Some people disapprove of “cheat” sites, but I solve these puzzles entirely for pleasure and even if I still resided within the UK I would not bother sending in completed puzzles. I am not interested in the prize or in puzzle completion statistics. The reason most people want their statistics recorded is that they are hoping to get an all-correct run in any given year. The Listener statistician does an excellent job recording solvers’ success, but for the reasons described in the previous section I am never going to get a 100% record and I see little point in wasting his time. Therefore I see the occasional use of “cheat” sites as no different from looking at the published solution, since in both cases I concede that the puzzle defeated me and make no claim to have completed it successfully.

I am conscious that much of the foregoing is rather negative in tone. I certainly don’t want to come across like some of the contributors to certain crossword blogs who never seem to enjoy anything. I still get great satisfaction when I complete a puzzle, easy or hard, and there have been many in the last few years that have prompted a “Wow!” from me. Rather than name all of these, I am going to give an outline of what, in my view, makes for a good Listener crossword.

Title. The title is the first thing to catch the solver’s attention and an interesting one whets the appetite. Usually the title gives some cryptic indication of the theme, but sometimes the title can fascinate in its own right. A title like “Work by a Famous Poet” is dull, even if the puzzle is not, but one like “We Interrupt this Programme...”, the title of an excellent puzzle by Phi, immediately aroused my curiosity as there were so many possibilities.

Preamble. As I have already said, lengthy and convoluted preambles are off-putting. The preamble doesn’t have to give too much away, and may be a bit mysterious at first, but it should at least give the impression that it will make sense at some point when solving the puzzle. Clarity and conciseness are a must.

Clues. The best Listener setters write clues which are fun to solve. I have come across some Listener puzzles where solving the clues is a chore. In a sense the clues are the legwork needed to get to the theme but it’s so much more enjoyable when the clues are interesting. Some setters play safe (as I did with my own Listener puzzles) and to avoid rejections, produce clues which would pass muster with the most pedantically Ximenean critic but which are, as a consequence, rather boring. Listener clues tend to be hard, and so they should be, but I soon lose patience when they are peppered with obscure Spenserian, Shakespearean and Scottish words.

Theme. I am constantly amazed by the rich variety of themes that appear in the Listener series. I’m happy with any theme so long as its implementation doesn’t involve a final step which requires unreasonable leaps of faith. In the best puzzles, the theme becomes apparent as you solve the clues. It could be a quotation or set of instructions revealed by extra/misprinted letters in the clues, or it may become obvious from the way the answers fit into the grid itself. By the time the grid is complete, the solver should be fully aware of what the theme is, and if there is more to do, it should be reasonably obvious. One of the best examples of this was a puzzle some years ago (I forget the setter, I’m afraid) which, on completion, required the solver to erase all answers from the grid. That’s the kind of thing that makes a puzzle special for me – not just because it got me wondering how many people sent in blank grids every week after that on the off chance the theme had been repeated!

There have been so many different themes in the history of the Listener that it’s impossible to categorise them all. It is possible though to identify the four most common types of theme. The most prevalent is probably that in which answers undergo some sort of modification before entry into the grid. This is usually in line with an instruction or quotation which reveals itself as you solve the clues. The instruction may be the first letters of extra words in the clues or a quotation around the perimeter of the grid, to name two possibilities. Thus, if the instruction turns out to be ALL ROADS LEAD TO ROME, you may have to replace answers or parts of answers which contain a synonym for “road” with the word “Rome”, e.g. the entry for the answer STREETWISE would be ROMEWISE.

The second common type is grid-based modifications. For example, it may appear at first glance that the answers don’t fit properly in the available squares. There may be too many or not enough squares. Say you have the answers CHINWAG and APHID but there are only 5 and 3 squares available respectively. It may be that, following a hint from the title or somewhere else in the puzzle, the solver has to deduce that any letter sequences which correspond to Greek letters should be replaced by the appropriate Greek character, i.e. ΧNWAG and AΦD.

Another common theme is what I would call the wordsearch. Usually this occurs when the puzzle is intended to celebrate a particular person. For example you might have special clues which lead to PRAGUE, LINZ, JUPITER, HAFFNER etc and the solver is required to locate and highlight MOZART (the composer of these symphonies) in the final grid. Finally, there is the pictorial representation puzzle, where the final stage requires shading certain letters to portray, perhaps, the St George flag to coincide with the puzzle appearing on 23^rd April. (That one would probably fall foul of the PC brigade these days!)

I enjoy pretty much any theme, whether I’m familiar with the subject or not, provided – as stated before – its implementation doesn’t involve obscure or irrelevant (i.e. unfair) last steps when the grid is complete. Obviously I’ll warm more to a theme if the subject’s one I am interested in – classical music for example – but I keep an open mind. It is a credit to the setters of this puzzle, present and past, that the Listener can be relied on to cover such a diverse range of themes and subjects.

As I state in the introduction to this site, I have had a couple of Listeners published. The first was called Europe’s Ports and the second Surprise. I can’t reproduce them here, because they are now the property of News International and I don’t want threatening letters from their lawyers. If you subscribe to Times Online Crossword Club you can find them there (12^th April 2003 and 16^th June 2007). They received a moderately enthusiastic reception and I would be the first to admit that they are certainly not the best Listener puzzles ever published, though I was quite pleased with them at the time and the fact that they were accepted shows they were of the required standard. I had another puzzle rejected, rightly – the theme was weak and its implementation poor, as a result of me trying to force a puzzle out. The fact is, I’m not particularly good at coming up with innovative themes, or ways of implementing them. I regard myself primarily as a cluesmith; I don’t claim to be good at many things, but these days I feel that when I’m on form I can write clues as well as anybody. My puzzles for the FT and Independent have been very enthusiastically received, and it is in that direction that my strengths lie. I’m more comfortable getting the grid filled and then trying to come up with original, witty clues to entertain the solver than I am thinking up themes or constructing clever grids. I have no plans to write any more Listeners at present, but if a brilliant idea strikes me as I’m enjoying a Pilsner and watching Prague go by, then who knows?

This final section is aimed at those who are interested in having a go at solving Listener puzzles but are a little anxious as to whether they would have any success. To start with, I would say that by far the main factor in determining this is clue solving ability. Take that out of the equation and every Listener solver has pretty much an equal chance, since we all face a new, unknown challenge each week. Sure, experience does count for something: you get to know that some setters tend towards certain themes, and you get to recognise certain tricks setters use to hide information in the clues, grid or title. Intelligence also counts too – someone who thinks that reality shows provide intellectual stimulation is unlikely to polish the Listener off before breakfast on the day it appears in the paper. But it is the ability to solve clues that is paramount, and the good news is that the clues for the hardest Listener generally rely on exactly the same techniques as they do in any ordinary broadsheet crossword. There may be extra complications, such as misprinted definitions or extra words, but the techniques for interpreting the wordplay in order to get the definition are no different. There is nothing new to learn.

What distinguishes Listener clues from the Telegraph or Times is that they make use of a far wider range of vocabulary and abbreviations. Anything lexicographically justified by Chambers is acceptable so long as it isn’t offensive. The words or meanings of words used as answers are often unfamiliar. For example, in a daily puzzle’s clues a word like PORT will be indicated as a harbour, the left side, an opening or a fortified wine. These are the first four meanings listed in Chambers. In a Listener clue, PORT may be indicated as to carry, a borough, a bagpipe composition or a suitcase (the following four meanings in Chambers). Likewise indications for the letter R in a broadsheet are usually river, right or runs, whereas in a Listener R may be indicated by 80, 80000, Rector or rule. Some setters make more use of obscurities than others. In my opinion the best setters avoid overuse as they know it can make the solving process frustrating and even boring.

How do you know if you’re up to the challenge? Obviously the best way is to get hold of a puzzle, either in the Saturday Times or from the Times Crossword club if you’re a subscriber, and have a go. As a rule of thumb, I would say that if you regularly finish the Times puzzle in 15 minutes or less, you are certainly made of stern enough stuff to tackle the Listener. I use the Times as a yardstick as its puzzles are of consistent difficulty and fairness – whereas the Guardian is wildly inconsistent with some puzzles which are very easy, some very hard and some which take “libertarian” clueing to ridiculously unfair extremes.

A gentler route to the Listener is to try the Sunday Times Mephisto or plain Azed puzzles. These use similar vocabulary to the Listener but do not have the added complication of a theme. This will get you used to the style of advanced, barred cryptics. If you do well on these you’re certainly ready to have a go at the Listener. Likewise there are some puzzles similar to the Listener but usually easier – the Sunday Telegraph Enigmatic Variations series is a good example of this. The Inquisitor puzzle, formerly in the Independent and now in the i newspaper on Saturday, is also a thematic, barred puzzle, but be aware that while some of these puzzles are quite easy, others are every bit as hard as the Listener.

If you are a good solver but, after attempting the Listener for the first time, you find that after three hours you have one possible answer and aren’t sure where to put it in the grid, don’t despair. It may be that you’ve chosen to start with the hardest puzzle of the year. Try again the following week. Listener puzzles aren’t graded for difficulty, so until you get to know the setters you don’t know what you’re in for, unless the puzzle has the sort of convoluted preamble I’ve written about above. There are some easy puzzles each year, and I think this is intentional in order to attract new solvers. There are a few die-hard Listener solvers who grumble whenever an easy puzzle appears, as if they think that the Listener Crossword should belong only to a select few. Some elitism is unavoidable with a puzzle at this level of course, but unless there are a few entry-level puzzles each year the number of solvers will slowly die out until it is not worth publishing it at all.

Is the Listener crossword the best puzzle there is? It’s certainly the most consistently difficult, and composing a good Listener puzzle is probably the highest achievement any setter can aspire to. But to say that a Listener puzzle is “better” than a Times or Telegraph puzzle is probably a bit like saying that Wagner’s Ring is “better” than, say, a Schubert song. In comparison to the sheer complexity and epic scale of Wagner’s masterpiece, not to mention the huge effort involved in staging it, the Schubert may come across as relatively lightweight, but I happen to enjoy both. So it is with the crosswords – a Listener requires great skill to think up, construct, clue and of course solve, whereas a daily cryptic requires far less, but in their own ways both can provide equal pleasure. It’s not a matter of comparisons then, but I thoroughly recommend trying the Listener crossword to any keen and competent solver. I have expressed my likes and dislikes about this type of puzzle – yours of course may be different or even the opposite. If so, good! Variety is an excellent thing. It only remains for me to thank the Listener setters for providing so much rich entertainment in the past, present and, one hopes, the future – may you never run out of ideas!