Ximenes On The Art Of The Crossword

---



Beginnings

This introductory chapter will not attempt to give a detailed historical account of the coming of crossword puzzles. Such an account would be alien to the purpose of this book, which is the ambitious one—perhaps too ambitious—of trying to arrive at a system of principles which can make the crossword more enjoyable and rewarding to solvers, whether they be among the millions of desultory solvers, content sometimes to fill in only part of a puzzle to pass an odd half-hour, or among the many thousands of real enthusiasts, determined to reach a full solution, whether of an easy or of a difficult puzzle. The main point I want to make is that the system will be just the same for all types of solver, though they may not all realise it: its first principle will be that the clue-writer’s aim must be never to leave the solver with a possible answer in his mind that he is afraid to write in because it doesn’t, for some reason, seem a fully satisfactory or certain answer, only to find, when he sees the solution, that it was the right answer. That aim will not be achieved unless the composer becomes an unwavering adherent of a principle laid down for clue-writers long ago by Afrit of the Listener: “I need not mean what I say, but I must say what I mean.” Of this much more anon, in its proper place, but our first paragraph cannot be complete without it.

Of very early crosswords I shall not say very much, simply because their clues were all definitions; and while I know there are still plenty of solvers who are satisfied to fill in a diagram by writing down the answers to definitions, the majority of solvers nowadays demand, I am sure, something more ingenious to pit their wits against, something which is provided by what is normally called the cryptic clue, whose coming we shall discuss in the next chapter. In the United States, the original source of this, as of so many other ingenious things, it is, I believe, curiously enough, otherwise: though the crossword puzzle there is over fifty years old, definition clues are still the normal thing. There must be, however, voices there crying in the wilderness, for I have some solvers there, as in many other parts of the world, of my Ximenes puzzles, and I know that many other British puzzles, with cryptic clues, are solved there too.

It was in 1913 that an American journalist named Arthur Wynne, who had earlier emigrated from Liverpool (so that we may at least claim that the inventor of crosswords was originally English) constructed several specimen puzzles and submitted them to the Editor of the New York Sunday World. He liked the idea and ran a crossword in his paper for about ten years before anyone else followed suit.

The next step came in 1923. In that year two young men lately down from Harvard, Mr Robert Simon and Mr Lincoln Schuster, compiled a book of crosswords in New York, containing fifty simple puzzles. Three-quarters of a million copies were sold within a few weeks of publication at about 5s. 6d. each, so that the venture was a very profitable one. Soon newspapers from New York to San Francisco began publishing puzzles daily. As far as Great Britain is concerned, there have been several claims and counter-claims on the subject of the crossword’s introduction. Some records are muddled or self-contradictory, others were lost in the last war; and whenever the files of papers or magazines have failed to survive their moves to different premises, the memories of present or former members of their staffs have become vague and historically untrustworthy. We have only included here what is based on reliable evidence. On 2 November 1924, about a year after the appearance of the American book, the Sunday Express published one of Mr Arthur Wynne’s puzzles, with certain alterations. This puzzle was one of a small batch offered by Mr Wynne to Mr C. W. Shepherd, a member of a syndicate known as “Newspaper Features”, who in turn sold half a dozen to the Sunday Express. As it happened, the first puzzle chosen for publication contained a word with an American spelling, and in order to eliminate it, Mr Shepherd was forced to make a drastic reconstruction of the diagram: so this puzzle must be called the joint work of Mr Wynne and Mr Shepherd. Solvers familiar with the more sophisticated puzzles of today, with their cryptic clues and their almost infinite variety of block-patterns or bar-arrangements, will no doubt be amused at this modest little effort, which was the first British crossword and startled and delighted the British public some forty years ago. It entitles Mr Wynne, who died in January 1945, to be regarded as not only the inventor of the crossword, but also as its introducer to this country.

Here it is:

Others were quick to adopt this new idea, and soon all the popular papers here had their own daily or weekly crosswords , and when, on 1 February 1930, The Times began to print a puzzle as a regular feature, the crossword’s future, as well as its respectability, was assured. In the first enthusiasm generated by the Wynne-type puzzles, solvers of all classes enjoyed the friendly rivalry of competing with each other in their speed of solution. Sir Josiah Stamp, a brilliant economist and a Director of the Bank of England, began the craze by claiming, in a letter to the Editor, to have solved a puzzle in fifty minutes. Sir Austen Chamberlain, at one time Foreign Secretary and Nobel Peace Prize winner, soon bettered this time with forty-one minutes, and others in turn announced gradually decreasing times, until one learned man asserted that he normally completed his daily crossword while his breakfast egg was boiling: he must surely have liked his eggs, but not his puzzles, hard.

Very soon after the establishment of these straightforward puzzles came the type of crossword which deliberately introduced the possibility of alternative answers to many of its clues, and which offered large money prizes for those solutions which agreed most nearly with those arrived at by a panel of judges. In the late 1920s — the peak period for competitions of this kind — the weekly value of prizes offered was between £20,000 and £25,000. With these crosswords, which are more or less sheer gambles, we have no concern in this book.

A further early development, still far removed from the modern crossword but at least a matter of skill rather than of luck, was the series of crosswords, each one harder than last, forming the basis of a competition with a large money prize, intended to go on until an ultimate winner emerged. Perhaps the most famous of these competitions, which older solvers of today may remember, appeared in the Daily News in 1925: the first prize was £5,000. The first puzzle was a very easy one: there were more than 100,000 entries (entry fee, 1s. per time). At this point Gilbert Frankau, the novelist, was called in as composer to complete the process of elimination. This he did, in four more rounds. The third of these demanded a knowledge of French, Spanish, German, Italian and higher mathematics: after this there were still over 300 competitors left. His fourth puzzle, of which I have been able to obtain a copy, was not wholly solved by anyone; the prize was shared by a syndicate of London solvers, who presumably had the fewest letters wrong. To show the almost incredible tortuosity that was resorted to, and to preserve what may be regarded as a historic work, this puzzle, cumbersome as it is, is appended. Perhaps there may still be gluttons for punishment who like to see how much of it they can solve.

It may be appropriate to add, with regard to competitions of this sort, that I don’t think the crossword lends itself to knockout competitions. The persistence and skill of the best solvers is such that only by unfairness can they, if many of them compete, be reduced to a single winner. Sheer difficulty in itself, unless it is difficulty of an ingenious sort, bringing legitimate satisfaction to the solver when it is conquered, is hardly a merit.

Before going on from this brief excursion into the early days of crosswords to consider their later development, it may be worth while to recount a war-time crossword incident, which gives an example of the unexpected problems that can discomfit the crossword composer. Shortly before D-Day in 1944, the late L. S. Dawe, who was the chief crossword-setter of the Daily Telegraph, was visited by representatives of MI5 and subjected to a rigorous examination. He had innocently used in some recent puzzles four words which happened to fit in his diagrams and which were, as everyday words, as familiar to his solvers as to himself. The offending words were mulberry, Pluto, Neptune, and overlord — each of them a highly confidential code-word connected with impending war operations. He was fortunately able to convince MI5 in the end that his use of these words was purely fortuitous, and that if the very first word with which he had started each of those diagrams had been different, probably not one of these code-words would have entered either his mind or his puzzles!

ACROSS

1.   Initialise a father.
202.   By adding hydrogen here, help the British Dyestuffs Corporation.
303.   Full of hops.
404.   Curtail a formula.
5.   Might describe the condition of a rustic wooer.
6.   A poetess.
7.   A make.
8.   The guns of Trafalgar.
9.   Curtail and decapitate an enormous oven.
10.   Reverse one part of a pulley block.
11.   One way out of the harem.
12.   Once decapitate and thrice curtail the race of a man whose skin might have been saved had he been able to exercise 11 across.
13.   Decapitate and curtail something which also answers to clue 44 across.
14.   Curtail and decapitate that which apparently did not fall.
15.   Twice curtail an exiguous meadow.
16.   Two code letters found on a certain brand of advertised cigarettes.
17.   Reverse either three-fourths or three-fifths of a neurosis.
18.   Add one and make a tragedy.
19.   Curtail and decapitate a diminutive.
20.   Her husband’s initials were J. R.
21.   Found under the table.
22.   Reverse two words which might describe the influence of several specimens on the judgment of a mining engineer.
23.   First reverse, then twice decapitate an idler.
24.   Look for this on the end of a blowpipe.
25.   Reverse one thorough paced scoundrel.
26.   Reverse something made of chestnuts.
27.   First reverse, then twice decapitate a city bossed by a bull.
28.   Might describe hawthorn.
29.   Curtail a Saracen’s stronghold.
30.   Curtail a word associated with mice.
31.   First reverse, then invert the Erst two letters of a knowledge invaluable to Mr. Royce.
32.   Ask your tailor about this.
33.   Found under expensive cigar boxes.
34.   Could also be initialised as u, Q, F, E, G, D.
35.   See house agents’ advertisements.
36.   Start an ode.
37.   Curtail one who would have been better shingled.
38.   The fore-runner of the land girl.
39.   Twice curtail that which put behind 72 across conveys social reform.
40.   By decapitation and the alteration of one letter, make print vocal.
41.   One of the lacertilia.
42.   Reverse the first half of a hill.
43.   A form of promotion.
44.   Dished up with vegetables.
45.   A female.
46.   Add a couple and make a row.
47.   To surround.
48.   Curtail a village. 4
49.   Set our fathers singing.
50.   Curtail one who goes veiled. 4
51.   Made men cry, “Bring out your dead”.
52.   A relative of Peeping Tom. 4
53.   One who entered on behalf of the House.
54.   Curtail a river.
55.   Might have been the first two words of a historic 4 South African cable.
56.   Remove two letters from contracts. 4
57. and 57a.   By combining these two make a lark.
58.   Decapitate a doubtful beautifier.
59.   Reverse a form of communication. . .
60.   Two-thirds of a military unit.
61.   Deduct an aromatic liquid from an aromatic plant.
62.   B.C. three thousand two hundred and sixty.
63.   Gives knights nightmare.
64.   Reverse projections.
65.   The last of the fairies.
66.   Curtail an inhabitant of the Gebbi.
67.   Singularise and reverse a reversal.
68.   First decapitate, then reverse a property vital to a play.
69.   Twice beheaded a fearsome monster.
70.   Nice nicknames (reversed) for a brace of peeresses.
71.   Once decapitate and three curtail a business which is also an art.
72.   See 39 across.
73.   A chorus.

DOWN

1.   Title of a poem.
40.   Seven score.
55.   Reverse that which makes men distrust legality.
2.   The novice oarsman’s friend.
3.   Halve a frisky lady.
4.   Reverse a process Biblically connected with a beard.
5.   An appellation of intimacy.
606.   Would apply to 21 across.
7.   Reverse a sign of decomposition.
38.   Reverse a name.
39.   A great handicap.
8.   The answer the doctor made to his nagging wife.
9.   Makes H. G. Wells see red.
10.   Reverse an inflammation.
11.   Reverse whom you curse.
12.   Thrice curtail a trade-mark.
13.   Reverse that which has as bad an effect as 9 down.
14.   Curtail a cereal.
15.   Cross.
16.   Reverse a revenue payer.
202.   Reverse a patient carrier.
203.   A racecourse.
59.   Depenultimise an African tribe.
204.   The first two letters of 303 across.
205.   Stops “fans” fanning.
206.   More trouble for H. G. Wells.
44.   Reverse a newspaper suggestion about 46 across.
207.   Half a fruit.
208.   Appeal to peelers.
209.   Leslie Henson does this.
210.   Reverse an obsolescent individual.
211.   Several Zevs.
212.   What the Yorkshirernan said when they offered him ginger beer.
213.   Reverse, then four times decapitate a war policy.
214.   The day to which no one looks forward.
215.   Reverse a flapper’s dream.
216.   Curtail that which tallies with 40 across. .
217.   Initialise and reverse a preserver (?) of morality.
57a.   A monopoly.
218.   Reverse a headman.
303.   Reverse a phrase which might be used by a Channel swimmer.
304.   Halve embellished.
305.   Stand a cup on its rim.
306.   Halve a cake.
307.   Bad in the lung, worse in the brain.
308.   Deduct first an American animal, then an English interjection from a barometrical term. Then reverse your result.
309.   Add a conveyance and make a spectacle.
310.   One of these rarely obtains 311 down.
311.   Rarely obtained by 310 down.
312.   Reverse a knight.
313.   Another Wells.
314.   Give this to your enemies.
316.   Curtail and reverse a provider of transport.
316.   Reverse the last two letters of 55 across.
317.   Reverse three primal liquids.
404.   First deduct the opposition from, then reverse a word whose original significance is precisely the opposite of opposition.
405.   Sets Chinamen gobbling.
406.   The wrong way to pluralise some Orientals.
37.   The gardener’s friend.
407.   Not the Southern’s, we hope.
408.   First reverse then decapitate, the forerunner of 55 down.
409.   Add the beginning and the end of 65 across to the ultimate of 1 across and the penultimate of 71 across.
410.   A reformer.
51.   Curtail a kicker.
411.   Was used for keening.
412.   Reverse a sign of utility,
47.   Initialise a story title.
413.   One of the things that is his.
414.   Reverse the first half of a Swinburnian town.
415.   Reverse and add a couple to 317 down.
415a.   Voyage, and learn this.
416.   Take the last four letters of 406 down and turn them into the past.
417.   Never wrote vers libre.
418.   Singularise, without authority, symbols of a jolly evening.
418a.   The typist’s curse
419.   The largest inhabitants of a South American Republic.

---



The Coming of the Cryptic Clue

We have seen that in the original American crosswords, and in their earliest successors in Britain, all the clues were definitions. But crosswords had not long been established in our papers before cryptic clues began to appear. What is a cryptic clue? In its widest and earliest sense, it can be described as a clue which is not a plain definition but can lead the solver to the answer by disguised, and more entertaining, means. Who was the benefactor of the crossword community who wrote the first cryptic clue?

After some research I doubt if it is possible at this length of time to make a definite pronouncement, but I think it safe to say that the first composer to use cryptic clues at all extensively was Torquemada, first in the Saturday Westminster, and soon afterwards when he started his famous Observer series in March 1926. Some appreciation of his work as a pioneer, and of his genius, will be found in the next chapter: it is enough here to say that whether he was the first to write a cryptic clue or not, such clues were making fitful appearances in many crosswords by 1927.

At first most of the deceptions were naturally very mild; an example from a 1927 Daily Telegraph puzzle is: “Normally grateful and comforting, but the wet variety spoils everything [7]”. Mild as the deception is, one can feel that the solver could get a little more kick out of arriving at BLANKET from this than from “Article of bed-clothing”.

Later chapters will be devoted to analysing the various types of the modern clue; here it may be of interest to do the same thing, less fully and inevitably less systematically, to the ancient variety. It will be convenient to use first the Times Crossword Puzzle No. 1, of February 1930. I don’t want to spoil the fun for those readers who may like to solve it, so I shall not refer to individual clues and their answers, but in order to exemplify ancient methods I have analysed and roughly classified the Across clues of this puzzle as follows:

1. Plain definitions: five. These might come straight from a dictionary.
2. Elusive definitions: seven. These, while they refer to nothing but the normal meaning of the word, define it in various roundabout ways: one of them I would venture to call inaccurate, since the answer is an adjective, whereas the clue could hardly suggest anything but either a verb or a noun.
3. Hints rather than definitions: six. These suggest the sort of word needed without actually defining it. Nowadays one might criticize such clues as too vague, but their composer, especially in those early days, might well defend them by saying “You call these clues hints: well, isn’t a hint a clue?”, and one could hardly deny that it is. Nevertheless, the tendency since those days has been away from such clues, and they do undoubtedly fail to satisfy one requirement of a good clue, namely that when the solver has thought of his answer he should be able to feel fairly sure that he is right.
4. Indications of the letters, not of the meaning, of the word: three. This type, like the rather vague hint at a meaning, was fairly common in the early days of the cryptic clue. That was natural enough at the time: an indication had been given — why give more than one? Here again, as with the vague hint, the modern tendency has been against clues that give no suggestion of the meaning of the word required, and I will say nothing till later except that I agree with that tendency. Even today such clues still appear in the most reputable circles; whether they deserve adverse criticism now may be thought a matter of opinion, but it would certainly have been harsh to criticize them in 1930.
5. Bald anagrams, indicated by the parenthetical “(anag.)”: two. This is really a specialized form of Type 4, just discussed: when the idea of using anagrams for clues was first lit upon, this form of presentation was universal. This type, too, is still sometimes used, though most composers prefer nowadays a less bald and more artistic approach to anagrams, of which more later.
6. Much more modern cryptic clues with two parts, one being a definition, the other a different kind of indication: four. It is very interesting to find these at such an early stage in the crossword’s development. Two of them include anagrams, disguised in the modern way, with a definition added: one of these, let us notice, is an indirect anagram, i.e. a synonym of the anagram is given in the clue, not the anagram itself. The fairness of this device will be discussed later, but we must note at once that it has been used by many composers throughout the crossword’s history.

To sum up, out of twenty-seven clues, four certainly, probably four or five others, and possibly a few more, might find favour in a reputable crossword today: at least half would probably fail nowadays to satisfy. But we mustn’t forget that only four years had passed since clues were all definitions. Progress, therefore, had been by no means slow.

Analysis of the Down clues shows slightly different results:

Types 3 and 5 disappear; there are six plain definitions, no less than sixteen elusive ones, three references to the letters of the word only, and six cryptic clues; but I don’t consider that these technical differences alter the main picture very much. Let us now examine some samples from later pre-war crosswords, with a view to observing developments. We will take first a rather later Times puzzle: how does it compare with No. 1?

1. The number of plain definitions has fallen from five to three.
2. There are nine elusive definitions instead of seven, so the total number of clues which we can include under the heading of definition remains the same.
3. The vague hints have gone altogether an interesting sign.
4. Indications of letters only: two instead of three, and one of these is of some interest, being an instance of what we now call a “hidden” clue, but a very antiquated form of it: “My dog wags his tail at policemen, but he won’t wag it at ordinary people (hidden) [8]”. Note that, as with anagrams of Type 5, the word “hidden” is actually given, there is no surprise. One merely has to search the clue for eight consecutive letters that form a word—not a difficult task, nor, I think, a very rewarding one. Yet a “hidden” clue in its more artistic modern form can, if not appearing too often, be very satisfying. More of this later.
5. Two bald anagrams, no change.
6. Twelve cryptic clues instead of four — a big advance. This, I think, is an indication of what was happening in crosswords everywhere, though the advance may have been quicker in The Times than elsewhere. There are, however, three of these cryptic clues with whose wording I personally should quarrel nowadays. They are as follows: “Caning me — and how the schoolmaster looks when he is” — MENACING. Here we have an anagram and a definition, with a near connection in sense between the two — the essence of an artistic cryptic clue.

One thing is lacking: there is nothing to show that “caning me” is an anagram. Don’t mistake me, I don’t want that dull little “(anag.)” back! But why not something that will give a hint of an anagram, something that will it in with the general sense and at the same time can be interpreted to mean that the letters are to be mixed? In this case the addition of “roughly” or “with abandon” would do. I shall not go fully into the principle which governs such matters yet; it is enough here to call attention to the point, and to mention incidentally that such anagram clues are still, even now, so common that a solver is apt to suspect any word, or combination of words, that has the same number of letters as the answer of indicating an anagram, even with no hint of letter-mixture: this I dislike.

“Coastal noises” — SOUNDS. This is cryptic, in that it is not a straight definition, but alludes with the adjective to a second meaning of the word. To my mind it does so inaccurately, so that if I thought of that answer I should hesitate to write it in. A sound is a narrow tract of water joining larger ones. It lies between coasts, but does that make it coastal? I doubt it. If I had to clue the word in that sort of way, I should prefer “Watery noises”.

“Suitable bonnets for ostlers, when they’ve finished work” — DUNSTABLES. This refers, quite entertainingly, to “done stables”, the generally discredited and corny pun has lived on in crosswords (where it is very helpful to the composer) and still seems to be tolerated if not overdone. To the wording used here I have the same sort of objection that I had to the anagram in No. 1. There is no hint thats pun is intended. Something like “… as ostlers say, we hear, when they’ve finished work” would serve , then the solver would know what he was looking for, which I think he has a right to do, as long as he reads the clue correctly.

But these are only three clues out of twelve: let me also quote, as instances to which no such objections can be raised, “Beverage that may produce a trance” — NECTAR. “Fruit that sounds like a bathing belle’s companion” — BEECH-NUT. “Form of remission obtained by providing a sailor with a key” — ABSOLUTION. And this crossword appeared over thirty years ago, and only about five years after cryptic clues were invented.

I said just now that the growth of cryptic clues may have been more rapid in the Times puzzle than elsewhere. This is borne out by an examination of some Everyman crosswords in the Observer, which appeared a few years later than the Times puzzle just analysed. I take two of these at random.

In each of them there are more plain definitions than in the Times puzzle — eight and seven, there are eleven elusive definitions in each, mere hints are rare — one in each, indications of letters only, one and three, bald anagrams, two in each, cryptic clues, nine in one, only five in the other; and in the second one there is one clue of a type that I haven’t mentioned yet, a plain quotation with a blank for the required word. My impression is that this type of clue was at first very rare except in Torquemada’s puzzles — he had his own special reason for using it, as I shall explain in the next chapter — but that it has, if anything, grown in prevalence since. I should like personally to see it die out, because I find its solution, whether easy or difficult, dull: I will give my reasons in due course.

I don’t want to give the impression that in writing this chapter I am trying cunningly to suggest that these were very inferior, early efforts, and that when I started composing crosswords I soon changed all that! Nothing could be further from the truth, and to prove it I will now mercilessly analyse my first effort in the Observer at an Everyman-type crossword, in 1941. It wasn’t called Everyman, because in war-time, after the first year or so, there was only room for one puzzle each Sunday: we had a Torquemada type one Sunday, an Everyman type the next, all of them under the traditional name of Torquemada. But the Everyman-type puzzles really were of the everyday kind, with blocks, not bars, in the diagram and with reasonably familiar words throughout, though they were supposed, for the sake of Torquemada addicts, to be a little more difficult than an everyday puzzle: nevertheless, they should be judged and analysed in the same way.

Now what do I find in this puzzle? Plain definitions: two; elusive definitions: no less than fourteen, one shameless hint, references to letters only: two (yes, I own up), cryptic clues: eleven, at least four of which I shouldn’t pass nowadays, and finally one quotation with a blank. An example of a cryptic clue of which I now strongly disapprove is “See someone in three times six and we turn back” — IN-TER-VI-EW. It makes nonsense: the reference to ter (Latin) is out of place here, “we turn back” does not mean “we turns back” — of that more later in its proper place. So I am not claiming to have done much in those days to forward the cause of progress.

What about the first post-war Everyman, the first puzzle of the present Everyman series, which I initiated? Definitions: two; elusive definitions: still, I’m sorry to say, ten, hints and references to letters only: none, bald anagrams: one, cryptic: fifteen, at least four of which I should now call unsatisfactory; and two quotations. “Is this terrier ever the limit?” — SKYE, is one of those I disapprove of. The query is hardly enough to hit at the pun.

Finally, lest these analyses should degenerate into a boring catalogue, we will leap to the present time, when the genuine cryptic clue in the majority of crosswords far outnumbers all other kinds of clue. Naturally composers vary in their methods: many do not subscribe to the principles I am going to lay down in later chapters, and no doubt they will continue to ignore them — why shouldn’t they? I shall give my reasons; but many may not think them cogent and may maintain that crosswords could easily become too stereotyped, and much can be said for that view. All I maintain is that for me these principles are right, and that, as long as I stick to them, I believe my clues to be fair; my successors in the Everyman crossword follow them, so in their puzzles you will find endless examples of clues that I approve of; and you will find very many examples elsewhere. If you found them always, everywhere, crosswords might well be too monotonous, and it is only my personal opinion that they would be fairer to the solver.

I will finish this chapter, as I started it, with a Times crossword, a typical one, I think, of today. I think it is fair to say that the Times crossword has achieved a sort of unwritten but generally acknowledged seniority among crosswords of the everyday type whose clues are mainly cryptic, and I have already stated my belief that cryptic clues grew quicker there than elsewhere. I always find there some clues that I like: I usually, too, find liberties taken which I should never take. There I will leave the matter and make this last analysis of a recent puzzle: Plain definitions: one; elusive definitions: five; mere hints: none; references to letters only: two, bald anagrams: none, all the rest, twenty-four are cryptic, and in ten of them, at the most, I would quarrel, perhaps carpingly here and there, with the wording, e.g. “Seamen who put rings on their fingernail” — DECK HANDS: I can’t reconcile myself to that “who”. “Livens our disgust” — REVULSION: here, as I noted in a former puzzle, we have an anagram of which no hint is given. At the same time let me commend the neatness of “Dead tired?” — OUT OF BREATH, and, especially, of “A person who uses a car has imbibed nothing, but this one has imbibed a lot” — CAROUSER. Both of these are perfect clues. Incidentally, there are no quotational clues in this puzzle. That is not, I believe, typical, more often than not one finds one or two.

---



Crossword Principles

Crossword puzzles nowadays, as acrostics did before them, aim to entertain two quite different publics. The first, and naturally by far the larger of these, wants, in the absence of a good book or interesting news in the paper, to make a railway journey pass quickly, to be diverted for half an hour in an armchair or over a solitary lunch in the heat of the day, or to relax in the evening with something which calls for thought of a not too exhausting kind. Often in the home, sometimes in the car, less often in the railway carriage, the effort at solution is a communal one. This public probably doesn’t mind if it doesn’t finish the puzzle, or even if it goes to sleep in the middle of it. Its members are not very critical: as long as they can solve a reasonable number of clues, they don’t mind very much how sound in small details those clues are.

But they do, rightly, make certain demands. They are not walking dictionaries or encyclopaedias, nor do they travel carrying such works or keep them beside their armchairs. They therefore expect the vocabulary of their crosswords to be that of the man in the street. Harassed composers in a hurry, when driven into an awkward corner, sometimes get out of it by using an unfamiliar word; but this is naughty of them: if they do it too often they deserve to lose their fans. Other demands which ordinary solvers make are these: they like some of the clues to be amusing: they like an element of surprise when the penny drops — “Oh, of course! I never thought of that: it’s a good one, that”, and above all they like, when they have thought of an answer, to feel absolutely sure that it’s the right answer. Though they are not, on the whole, irritable people — they don’t take crosswords seriously enough to be irritated by shortcomings in them — they can be irritated by this sort of sequence: “It looks as if x ought to be the answer, but I daren’t put it in; it doesn’t really satisfy the clue.” Next day, when the solution appears, “oh, it was x, after all; I wonder how the blighter would justify it.” But they seldom bother to write and ask him, and so the blighter goes on getting away with it. Nor do they in many cases realise just what was wrong with the clue. I intend in a later chapter, on cluemanship, to examine in detail what I think are the most important causes of unsoundness in clues. If the reader agrees with me, perhaps he’ll write that letter more often, and the blighter won’t go on getting away with it; or perhaps some of the blighters—I’ve been one myself often enough—may also agree with me, as I did long ago with my own mentor, Afrit of the Listener, and they won’t be blighters so often. Many of them already are blighters very seldom, or only in one or two particular ways.

The second, much smaller, public does its crossword not merely as a diversion but because it is really devoted to word puzzles and enjoys the challenge of a tough problem. Its members may sometimes solve a normal crossword for fun, though they won’t regard it as fun if they are held up beyond half an hour or so, because they will inevitably suspect (often justly) that the composer is doing the dirty on them. But what they really want is not a sound easy puzzle but a sound difficult puzzle. They don’t in the least mind being made to use dictionaries and, within reason and reach, other reference books; they are even prepared to find crosswords educative, and they en oy meeting extraordinary-looking importations from other languages, found in some dictionaries, such as sdrucciola, (triple, of rhyme, Italian), kibbutzim (Jewish communal settlements in Israel, Hebrew), and obsolete English words such as agraste (=aggraced, i.e. favored, Spenser). But, of course, to lead the way to such atrocities there must be, in addition to a definition, a subsidiary part of the clue alluding to the letters or parts of the word: then the ingenious solver can deduce what the word is without knowing it and merely look it up afterwards to make sure. Take “agraste” for instance: this is a particularly nasty specimen, because it isn’t even given in the only smallish dictionary which includes it (Chambers’s) under its own spelling but under “aggrace”, with two Gs, so that it’s hard to find. But suppose the clue is “Tucked in, getting outside fat in Paris, becoming well-favoured as of Old.” “Getting outside” makes it clear that the first part refers to parts of the word: “tucked in” must be a short word outside, or including, a French word meaning “fat”. The first can hardly be anything but “ate”: so we have AT _ _ _ _ E or A _ _ _ _ TE. What is your French like? If it’s as had as mine, you may not know that there are two words meaning “fat”—”gras”, a noun, or adjective, as in “foie gras”, and “gros”, an adjective: if you don’t, you probably have a dictionary or someone to ask. ATG is impossible even in Spenser: so it must be AGRASTE or AGROSTE. Now the definition was “well-favoured”, “favour” = “grace” will make you come down on the side of AGRASTE. And you needn’t even look it up in the dictionary: it must be right. Even if the French word is unattainable, there are crossing words which may supply enough letters to suggest to you an old spelling of something to do with “grace”.

I have written at length about this one clue, to show how a straightforward subsidiary indication can enable the solver to get at an unknown word—and, of course, knowing little French isn’t often a handicap. French isn’t widely used in my crossword clues or anyone else’s.

This second crossword public, it can fairly be said, was created by the late Edward Powys Mathers, widely known, loved by addicts, and feared by the uninitiated, as Torquemada. He started, when crosswords in Britain were very young, with a series called “Crosswords For Riper Years” in the Saturday Westminster. Soon afterwards began his famous series in the Observer. He combined a mind brilliantly agile in using letters and words with a vast and varied knowledge of literature, and he applied both to the composition of maddeningly ingenious crosswords which at first seemed quite impossible. Gradually one came to follow the workings of his mind, and the first complete solution of one of his puzzles was a legitimate cause for pride.

He was a pioneer in every way. Most crosswords, when he started, had nothing but definitions for clues—very dull work for the puzzle-minded. He determined to retain some definitions but to sprinkle his puzzles with clues which were much more elusive and amusing. He also determined to add to the difficulty by using obsolete and little-known words. He thought that in puzzles of the standard he intended to set a symmetrical diagram, like those most other composers were using, with blocks alternating with spaces for letters so that many letters were not checked by crossing words, would produce too much difficulty. So he (I am almost sure) invented the system of using bars between the squares, making an unsymmetrical diagram look less unsightly than one with unsymmetrically placed blocks. He disregarded symmetry: he wasn’t interested in the diagram as an artistic thing, and he didn’t always compose it himself: it was, I believe, often composed by his wife, who was provided by him with lists of words to be worked in. Such devices as reversed words, words split into two parts of the diagram, bits of words and two or three-letter abbreviations were all admitted, in order to include as many of his chosen words as possible and at the same time to reduce the number of the unchecked letters. This “bittiness” could be irritating, but one got used to it, and it was largely atoned for by the enthralling interest of the struggle with his mind.

His clues were of three sorts: (1) The verbally agile, the precursors of the modern cryptic clue. But they weren’t like anyone else’s cryptic clues: he had an indescribable wit of his own which distinguished them from the work of other authors. Many of them weren’t strictly sound by later standards; but one knew what to expect and very, very rarely felt that he had been unfair. This was his type of clue that I enjoyed, and this was Torquemada for me. (2) The literary. He loved quoting from his favourite authors, and often he would include three or four words from one source in the same puzzle. Many solvers found their greatest pleasure in these and loved being sent back to their shelves to recover a forgotten passage. I admit that I was not one of these: to me a direct rotational clue gives no pleasure. Either you know it, and there is no great thrill in writing the answer in; or more probably you don’t, and I am too lazy to enjoy research. But the words were fully, or nearly fully, checked, and such clues were never impossibly numerous. (3) The definition. Yes, he didn’t break away entirely from the existing norm. This happened sometimes, as he openly avowed in print, when he couldn’t think easily of anything amusing or tortuous and therefore decided that this was “no time for trifling”. Occasionally he did it deliberately as a double-cross, making the definition look as if it were a tortuous cryptic clue. And if anyone nowadays feels like calling this resort to a dull definition a weakness, let him remember that Torquemada was a pioneer, and also that he was a genius of a most original kind; rules of cluemanship are for ordinary mortals, not for such as he was.

But most of us are ordinary mortals, and I therefore maintain that we are unlikely to succeed, as Torquemada did, in building up a large and satisfied clientele of solvers, unless we follow certain principles and keep a firm check on ourselves lest we be unfair. And this applies as much to the everyday kind of crossword, for transient amusement only, as to the connoisseur’s type . I hope I am not being too egotistical in illustrating this idea by my own experience.

When I (with, at first, two partners in crime) succeeded Torquemada in the Observer in 1939, I began by trying to imitate my predecessor closely. Complaints from solvers were few, so perhaps my clues were not really so dreadful as they seem to me now and would, I am sure, seem to my present solvers. But it is more than possible that the absence of an outcry was due to the patience of solvers who realised that they couldn’t expect a new Torquemada to be built in a day. Nevertheless I feel sure that the close imitation was misguided: to succeed in crossword composition, an in literary work, one must be oneself. It took me a long time to discover that my particular crossword self was one that worked best with a framework of fixed principles within which to be original; and I should have taken even longer to discover lt, if it had not been for the arrival of one great influence, Afrit (the late Prebendary A. F. Ritchie) of the Listener.

He had been composing his crosswords for some years before I was first introduced to them by a solver of mine, Mr L. E. Myres, to whom I am eternally grateful: the dedication of this book to him is a small token of this gratitude. Afrit’s Listener puzzles were desperately difficult, too difficult, for some time, for me. One reason why they had to be difficult was that the Listener at that time was offering a prize for every correct solution received! But he composed, later, easy everyday crosswords too: there is a book of them called Armchair Crosswords, published by F. W. Warne & Co., now, alas, out of print. An example of this type (not from the book), as well as a tough Afrit, will be found among the puzzles at the end of this book. He asked me to write a foreword to Armchair Crosswords, and in it I expressed my conviction that the crossword world needed standards, and that Afrit was the man to set them.

In his introduction to that book he refers to an imaginary Book of the Crossword and summarises the principles he supposes to be set forth in it. If he had been a less busy man—he was a headmaster as well as prebendary of Wells Cathedral—or had lived longer, perhaps he would have written it; if he had, there would have been no room on anyone’s shelves for this book of mine. My chief excuse for writing it is that I want to enlarge on that necessarily brief introduction, especially on what he says about the art of writing clues, which is here the subject of Chapter V. It is many years since he wrote that introduction, and I have added further principles of my own, as well as deviating in some ways from his; but I want there to be no doubt about the extent of my debt to him.

---



Diagram Composition

In the early days of the crossword the diagram was so constructed that every letter contributed to two words, one across and one down. This meant, unless the composer was exceptionally ingenious, or lucky, or persistent and prepared to spend an immense amount of time, that there must be few long words and many short ones, and that he would only be able deliberately to choose a small proportion of his words; many would drive themselves into the diagram as being the only ones possible. Hence the spate of emus and gnus, of ores and odes, of ohms and ergs, that occurred in early crosswords. Even two-letter words and abbreviations were admitted, and the sun-god Ra achieved a new fame and became almost a household word.

This didn’t matter much as long as the clues remained mere definitions: one just read “flightless bird” and wrote in EMU, though it did tend to pall after a year or two. But as the cryptic clue gradually came in and more than a definition was sought, it became clear that the composer must have more choice. He must be able to use long, interesting words, or phrases (though these only came into common use later) and to avoid two-letter words altogether and to reduce three-letter words to a minimum. The result was that in many papers, quite early on, blocks were used to shut in some letters and save the composer from making them contribute to two words. At first this was only done in parts of the diagram: wide open spaces still appeared here lull there (see The Times Crossword No. l at the end of this book).

Torquemada, with his bars and lack of symmetry, still stuck to full, or nearly full, “checking”: this is the technical term for making a letter contribute to two words, letters contributing to only one word being called “unchecked”. The difficulty of Torquemada’s clues and the rarity of some of his words made him do this, and some straight-definition composers still do it today. But the use of many unchecked spaces grew in most everyday crosswords, until nowadays the normal thing is for roughly every other space to be unchecked.

Now this can be done in two ways; either the outside rows and columns can be free of blocks, with the even-numbered rows and columns blocked, thus:

or conversely the diagram can be constructed thus:

I very much prefer the first of these two methods, because it leads towards what I want to make the first principle of crossword diagram composition, that at least half the letters of every word should be checked. I consider that a five-letter word with only its second and fourth letters checked is unfair to the solver, though it is very common practice today. In some diagrams the two methods are combined; sometimes, also, small wide open spaces appear, as if to compensate for the lack of checking in other places. But do they compensate, especially when, as sometimes happens, the less-than-half-checked words have none too easy clues?

Another practice in diagram composition is the unchecking of two or even three consecutive letters, thus:

or thus:

The second arrangement, it is true, is generally used for long words or phrases (twelve to fifteen letters); but even so I think the composer is saving himself too much trouble at the solver’s expense. Many solvers are only seeking diversion and don’t mind if they fail to finish a puzzle; but there are many too who don’t like to be baffled unless they feel it’s their own fault. If, having filled in all the interlocking words, they are left with something like -A~E-, there are so many possibilities that even a fairish but somewhat elusive clue may fail to suggest the answer in reasonable time, and the solver has fair grounds for feeling dissatisfied. After all, interlocking, so that the solution of one clue will help towards solving another, is of the essence of the crossword; we have seen that there are reasons why it should not be complete, but it should be adequate.

There is another feature of diagram construction that we composers ought to study and often neglect. In very early crosswords the diagram was sometimes so arranged that the four quarters were completely cut off from each other: in effect, there were four small crosswords instead of one large one. This was soon felt to be unsatisfying, and one never sees it nowadays.

But one does meet diagrams still—I have used them myself, I admit —in which the quarters are very nearly isolated. A popular type is one like this (the north-west corner is illustrated) :

A solver may make a good start in this corner and complete it, perhaps, but be baffled by the second part of the ten-letter phrase at 4 down; then what does he find? He has no letter to help him to continue! After solving six clues, and part of another one, he may well feel entitled to some guides to continuation. Such a diagram has the advantage that it eliminates words of less than six letters. This frees the composer for once from short words that tend to become hackneyed, STIR and STOP and STEM and so on. But the disadvantage to the solver must outweigh this: the composer’s job is to entertain and to avoid irritating.

Nothing has been said yet about symmetry in diagrams. It is curious how obligatory in nearly all crosswords this has become.

Torquemada disregarded it, and in a few crosswords, where bars instead of blocks are used to separate words and to uncheck spaces, it is still sometimes disregarded, especially in very difficult puzzles with a complicated twist, where fairly full checking is considered necessary. But symmetry is the regular thing, certainly in diagrams where blocks are used. Is this entirely for the look of the thing? I think it is, very largely. Afrit argued that symmetry ensures that an even distribution of unchecked letters -Not more in one quarter than another-is maintained in the diagram. He thought, probably, that without its restraint a composer might give way easily to something like this:

North-west corner:

South-east corner:

We should then have a three-letter and a five-letter word with less than half their letters checked. But if he were a conscientious composer, he could still depart from symmetry without doing this, and he might be able to work in a word or phrase, lending itself to a sparkling clue, which symmetry would exclude. No, I think it is the look of the thing that does it; an unsymmetrical diagram would affect many people as a clashing color scheme does. It certainly does that to me, though very occasionally, in a puzzle involving special complications, like the inclusion of a large number of Fixed words, I have deviated slightly from it, and I believe most solvers didn’t notice it. At least hardly any said they did; perhaps they were too polite.

The technique of filling the diagram can best be shown by practical demonstrations, which follow later. But a few general observations may interest solvers who may have tried their hands, perhaps only occasionally and without much success: I have met many of these. And I might even manage to say something which has escaped actual practitioners.

First, the composer should be fairly ambitious but not too ambitious. Long words and phrases lead more readily to amusing clues than short ones, and the ideal 15 x 15-square blocked diagram should, I think, contain twenty-eight words rather than thirty-two or more, with four long words OI’ phrases, usually of twelve to fifteen letters, and two words, not three, in each of the other open rows and columns. There is no reason why all four long words or phrases should be of the same length, since the sort of symmetry which only makes the diagram look the same when turned upside down is just as legitimate as the sort which makes it look the same from all four aspects. Suppose you have thought of two which you can clue pleasantly, one of thirteen and one of fourteen letters. If they won’t fit into each other, put them at two edges of the diagram: if they will fit - and they will somewhere, more often than not - let them cross each other, and fill in the other necessary blocks for symmetry. If they have to go at the edges, you will have to consider first the thirteen-letter one: are its odd letters more helpful as beginnings or as endings of interlocking words? The odd letters are the vital ones: the even ones will be unchecked. For if you expose the even ones by putting a block in each corner of the diagram, you will have only six checked letters and seven unchecked, and this we have already decided to avoid as a rule. Letter-frequency at the beginnings and ends of words is a thing which the composer learns gradually: the following conclusions are based on experience, not on scientific statistics, but I feel pretty sure about most of them.

R, S, and T are friends at either end: so, on the whole, is G, especially in longish words, because of -ing. Most other consonants are more helpful at the beginning, except, emphatically, N, which is much more helpful at the end, and to a lesser extent, though definitely, I think, D and L. E is certainly our greatest friend at the end: S is a rival, if we are going to admit a lot of plurals or 3rd persons singular, but to do that is weakness. A is probably the best vowel at the beginning; the others are not bad for long words, thanks to compounds, ex-, in-, over-, un-, under-, etc., but are apt to involve one in tiresome repetition with short ones. A and O are tolerable at the end, I and U much too rare to to used there if we can help it. Y is far more helpful at the end.

And of course you hardly need to be told not to put J, Q, or V at the end, or X or Z at either end if you can help it.

I would sum up the whole matter thus: the only letters that are more helpful at the end than at the beginning are D, E, L, N and Y, while G, R, S, T and, perhaps, H and K are indifferent.

So, if your thirteen-letter word is BACK-SCRATCHER, put it at the top of the diagram or in the extreme left-hand column: all its odd letters are favourable for beginnings, B, C, and, perhaps, H are less favourable for endings. But if your word is NONRESISTANCE, it is five to one the other way (only the I is unfavourable); so put it at the bottom. Often the choice won’t be so obvious: if it isn’t, you must take a chance; but very often the scales will be tipped one way or the other.

Let us suppose your word was BACK-SCRATCHER and you have put it at the top, and that the extreme north-west corner of the diagram looks like this:

Let us also suppose that you wanted to use VERISIMILITUDE. It at once becomes obvious that you mustn’t use it in this puzzle. If you put it in the left-hand column, every word across on that side will start with a vowel, which is sure to be a nuisance; and you can’t put it in the right-hand column, which must begin with R.

So scrap your VERISIMILITUDE and keep it for a later puzzle.

If, when your four long spaces are settled, you have admitted a difficult ending such as U or I, continue your composing there and fix that word. And all the time, as you proceed, avoid putting blocks in places which will give you few possibilities to choose from, and put them where you will have many. Suppose there is an S somewhere in the middle: try to get a block immediately before it rather than give yourself -s to proceed from.

Whenever you put in a word, look to see if you are facing yourself with a very restricted choice or even an impossibility. If the choice somewhere is between _ H_ E and _ A_ E at the end of some word not yet Fixed, choose the latter. Only so will you avoid tiresome wholesale scrappings of whole quarters of the diagram.

You have to think of two things (at least) at once: word-possibility and word-clueability. The latter is desirable, but it may have to give way sometimes to the former, or you will never have a diagram to clue! And later you may find that an unlikely-looking word isn’t so difficult to think of a clue for after all.

If you are using the type of symmetry which only makes the diagram look the same when it is turned upside down (not if it is given a quarter-turn), it is important, when putting in blocks,

not to make two quarters much more difficult for you than the other two. The preceding diagram will make this clear.

Consideration of various points I have made may have led you to produce something like that. You may even have filled the difficult north-west corner with its many longish words: remember that the north-west corner, where you are working from beginnings, is the easiest. The south-west and north-east corners will be child’s play; but the south-east corner will probably be the very devil! Much better to close up the north-west and southeast corners a bit more with blocks, and include more longish words in the other two. You won’t like scrapping your beautiful north-west corner; but the odds are very long that you will have to do so in the end, and you’ll hate it much more in a couple of hours’ time! Why is the north-west corner the easiest to fill and the best place in which to start? Because one thinks forwards and not backwards: it is easier to proceed from beginnings than up to ends.

Rhyming dictionaries are a help in the latter process, but that isn’t enough to compensate. So as soon as you have settled your long spaces, and any specially difficult endings, like the I or U suggested above, go to the north~west. Go next to the north-east or south-west, whichever looks harder to settle: there you will at least be working half the time from beginnings. The south-east will be the hardest struggle, nearly every time. “Then why not start there and get it done?”, you ask. Because if you do, you can’t take advantage anywhere of working from beginnings, and the whole thing will take longer. This is the normal principle, but circumstances can make it wise to violate it. You will find that my colleague in Chapter IX and I myself in Chapter XII have both found particular reasons for working on the south-east corner before going to the north-west. But these are particular reasons, caused by apparent impending difficulties; they don’t affect the validity of the principle in cases where there are no such reasons.

Now for my final word on diagram composition, and the most vital. Don’t be obstinate. When things look black, scrap and start again! I only wish I always followed my own advice. So often you hit together a combination of words easy to clue, but the last one or two words in the corner just won’t come. You spend ages, and at last you have to give way. You change two or three words to duller ones, and the whole thing is finished in a few minutes. And remember that you may well be able to use those nice words that have been scrapped in your next puzzle. It simply doesn’t pay to go slogging on with the luck spitting at you; but I shall probably go on doing it from time to time!

---



Cluemanship

I hold very strong views on this subject, and it is these views that have produced the cacoethes scribendi (“itch for scribbling” — Chambers’s Dictionary — in case this cliché has escaped you) which has produced this book. As I have already said, it was Afrit who converted me to my present way of thinking, and many of the details are his too. I am hoping, provided that other crossword composers do me the honour of reading this book, to do a bit of proselytising on my own.

Why? Sheer arrogance? No, honestly not. It is because I believe the principles laid down in this chapter can, if followed, make crosswords more satisfying. I know that as they are, with their various styles and methods, they entertain millions. But the fact remains that solvers, whether they know why or not, are often left, through no fault of their own, wondering whether an answer is right when they have put it in; or, when they fail altogether, they don’t say when they see the answer next day or next week, “Oh yes! How stupid I was!” but ‘Well, I really don’t see how I could have got that” or (even worse) “I thought of that but I rejected it as impossible’. And I maintain that unless certain musts and must-nots are established, this is bound to happen. There are, I know, plenty of uncritical solvers, who are happy to guess and don’t mind whether a clue is sound or whether they are right or not. But my belief is that they are outnumbered by those who are occasionally dissatisfied. I don’t begin to claim perfection or anything like it: I have erred, especially before I met Afrit, like everyone else. But if one is determined to stick to certain principles, it surely follows that one errs less often and that solvers can trust one more.

What is a clue? It is, according to Chambers, “a thread that guides through a labyrinth: anything that points to the solution of a mystery”. It guides and it points: these are positive actions; so in crosswords its action, however superficially misleading, must be positive. However superficially misleading: I repeat these words because it is of the essence of the cryptic clue not to mean exactly what it appears to say.

This brings me to what is, I think, the whole essence of the clue. I mentioned in Chapter I Afrit’s summing up of the clue-writer’s duty. I mean, and I’m sure he meant, every bit of the heavy type here used: “I need not mean what I say, but I must say what I mean.” That in itself is a little cryptic to the uninitiated; I’m inclined to put it the wrong way round myself when I don’t think! So let me interpret. “I need not mean what I say” in the sense that my words may appear, if taken literally, to mean something quite different from the meaning I really intend. “Rending granite all to pieces’ suggests, if taken at its face value, some Hercules in action; I don’t really mean that at all. What I mean is ‘a word meaning ‘rending’ whose letters are those of ‘granite’ out of their normal order”. The point is that I must, in however veiled a way, actually say that.

And I have. “Rending”’ is a definition of TEARING: “all to pieces” does mean “smashed, so that the parts are out of their normal order’. So I have said what I mean and given the solver two separate true indications of TEARING. When he has thought of TEARING, he can’t possibly be afraid to write it in: humanly speaking it must be right.

An important point arises out of this clue with regard to punctuation. If I am really saying what I mean, oughtn’t I to put a dash or a colon between “rending” and “granite” to indicate the two parts, as a careful writer would? Now here I’m in a weak position, because I believe Afrit would have said “Yes, you ought.” It spoils the misleading sense, and the clue would have to be reworded; but I think he, as a purist, would have insisted, and I have declared my devotion to his principles.

But this matter of punctuation is one of very few in which I have decided to be, as he would have said, weak. This is how I defend my position.

Punctuation in ordinary writing is a guide telling the reader where and how long to pause. But clue-writing is not ordinary writing, because the clue-writer, instead of seeking to make his true meaning as clear as possible, is seeking to veil it, so that an enjoyable penny may drop when that meaning emerges. This seems to me to justify the convention, universally used, I think, by all composers nowadays, that in a two-part clue the parts may stand side by side without punctuation between them. But this does not justify the introduction of misleading punctuation, which makes nonsense of the true meaning of the clue, simply to help the misleading sense. If I omit punctuation, I can claim that I have said what I mean, though perhaps in telegraphese: if I put in false punctuation, I can’t honestly make that claim.

While we are on the subject of punctuation, what should a query or. an exclamation mark mean at the end of a clue? Composers vary enormously about this. My own practice has varied too. I now think that a query, unless it occurs simply because the clue, or part of it, is in the form of a question, should usually mean one of two things: either that the idea is a fanciful one, referring to something which might be true, but probably, or certainly, isn’t, or that a possible example is being referred to rather than an answer to a definition. Instances will make both possibilities clear. 1. “Extortionate gang — of tennis players?” — RACKETEERS. They aren’t really tennis players — hence the query; but the suggestion of “racket” is a help to solution. Or: “Memorial to physician — something precious?” — TOMBSTONE (to M.B. stone). This time the statement might be true — a stone might be precious — but it needn’t be. 2. “Insincere sympathy, swelling the Nile?” — CROCODILE TEARS. Without the query, this statement would imply that crocodiles are found nowhere but in the Nile and cry there. Or: “Sticking out for the potato insect?” — PROTUBERANT (pro-tuber-ant). Again, there are other tubers besides potatoes. The point here is that it is not true of words that because A=B, B=A. “Tuber” is a definition of “potato”; “potato” is not a definition of ‘“tuber”’ but merely an example. This is an important point which clue-writers sometimes forget.

As to exclamation marks, I am grateful to a solver who once wrote (none too politely) saying, in so many words, that I sprinkled my clues with them with no other purpose pain that of crying out “Aren’t I clever — isn’t that a good one?” I was irritated, as one is apt to be, at first; but on further thought I had to admit that he had got something. Now I try to use much more restraint in this matter and to use them only when I se am exclaiming or for a technical purpose, to call the solver’s attention to the fact that I’m doing something particularly outrageous, perhaps by deliberately misunderstanding the meaning of a word. Here is a bad example, a clue of mine printed not nearly long enough ago: ‘“Colonel, it’s enough to give one a pain!” — coLITis. That isn’t nearly exclamatory enough » justify it. A justifiable one? “I’m wet, but I’m only half ga-ga! — ONEGA (one ga!). That’s a difficult Ximenes clue, anyway: “I’m wet” is a rather vague suggestion of a lake, and the other part is, I think, outrageous enough to justify the exclamation mark. This clue would hardly do for an everyday puzzle, but it tickled me and I couldn’t resist it. I hope it tickled solvers.

A problem of a similar sort to that of punctuation is that of capital letters. May one use a capital, where it isn’t necessary, in order to deceive? May one abolish one, where it is, strictly speaking, needed, in order to deceive? My answer to the first question is: Yes, at a pinch; but try, if you can, to put the word first in the clue or after a full stop in the course of it. My answer to the second is: No! If you do abolish it, you aren’t saying what you mean. Illogical? I don’t think so. Capital letters are often obligatory; but they are also used in certain contexts, especially in nouns, in words that don’t normally need them. So the composer may claim that he is making his context such a context. But there are extreme cases in which he really mustn’t do it — in verbs, for example. I was recently faced with a horrible word to clue,

CLENCH. What could one do with those letters? I arrived eventually at the fact that ‘“c.’’ and ‘‘ch.” are both short for “chapter”; LEN at once suggests Sir Leonard Hutton, and to clench is to close (one’s fist). Close! There was the connexion — Yorkshire cricket. But I really couldn’t bring myself to write “Close”, with a capital, in the middle of a sentence and expect solvers to arrive at the fact that in my concealed meaning the verb “to close’? was intended. So in this case it had to be the first word, and I reached “Close, two short chapters about former Yorkshire captain.” But with a noun I think more latitude is justifiable.

I must add one little point here, really for Ximenes solvers, but it may amuse others. Entrants in my monthly competition occasionally seek to dodge this difficulty by writing the whole clue submitted in capital letters! All I need say is that they never get away with it.

To return to that word CLENCH. I said it was a horrible word to clue. Why? Because it has no anagram, and its letters can’t be fooled about with in any obvious way; nor can it be “hidden’’ (see later). Now the old-fashioned answer to this problem was a simple definition; and modern composers sometimes resort to that expedient still: many might write “Close [6]’ and leave it at that. I think that would be weak, if not unfair. I have just opened my Dictionary of Synonyms at ‘close’. It gives the following words of six letters: finish, clinch, closed, hidden, secret, stuffy, intent, stingy — and incidentally not ‘‘clench’’, though I can’t see why not. Any of those might be the answer.

I think the solver has a right to demand further guidance — remember the definition of a clue. So I insist, with very few exceptions, and those in cases where all these alternatives don’t exist, on a fuller clue, giving either references to at least two meanings of the word or a reference to one meaning plus some treatment of the letters of the word.

Now note that reference to the meaning of the word. When plain definitions began to give way to cryptic clues, composers, as I mentioned in Chapter II, often served up an indication of the letters of the word with no reference to its meaning. They do this far more seldom nowadays, but in some very reputable crosswords it is still occasionally done. One still occasionally meets something like ‘‘Son lent caption (anag.)” or “Fixed in a shaky pole” for CONSTANTINOPLE. I suppose, if we go back again to the definition of a clue, we can’t deny that these clues do guide the solver, and point to the solution, in their way.

But to my mind they do the job inadequately. I would go so far as to call their composers mean, inartistic, and even lazy! They’re mean because they withhold the one piece of information that it is surely natural for a person set to guess a word to demand; they don’t tell him whom or what he is trying to find. They’re inartistic because they haven’t presented a picture which catches one’s interest. And they’re lazy because they let themselves be satisfied with something that requires practically no effort.

CONSTANTINOPLE isn’t an obviously easy word to treat, but the second imaginary clue above hits on part of a perfectly adequate way of treating it. Start from ‘‘constant’’, and adapt its meaning to a famous city “standing firm”, perhaps. We’re left with INOPLE: “with no pile in ruins’’ isn’t brilliant, but it does make sense with “standing firm’. If you think ‘famous city’” is too easy a definition, use the double meaning of “capital’’ — words with more than one meaning are invaluable in clues — and say “Standing firm with no pile in ruins — that was capital.” Don’t put an exclamation mark at the end to pat yourself on the back. In any case this isn’t a very exciting clue, and it isn’t meant to be. It is simply meant to be adequate and it is sound.

Talking of laziness in clue-writing, there is another common type of clue which I think deserves that accusation, and that is the plain quotation with a blank for the required word and the author’s name in brackets at the end. I used to do this myself, I freely admit, falling in weakly with the practice of others.

Torquemada, whom I have described as a hero of mine, did this more than any other composer. But I have already shown that I think he was a law unto himself, and he did it, I’m quite sure, not through laziness but from a love of literature and a desire to make his solvers explore what he loved; and he did it in puzzles which were intended for solvers many of whom enjoyed the research involved and, what’s more, in puzzles with very full checking, so that those who, like me, don’t much like that sort of research could arrive at the required word by the aid of interlocking words. He didn’t do it, as some composers do, with every other letter, or even three out of five or four out of seven, unchecked. I have always hated meeting in a crossword a clue like “ ‘Oh lift me from the grass! I die! I faint! I ——!’ (Shelley)”, when all I can get from other sources is _ A_ L. Now three things are possible. Either I know it and write it straight in and get very little kick out of it; I know perfectly well that I was just lucky. Or I look in a dictionary of quotations and find it and write it in, and get no kick at all out of it plus a feeling that I’ve cheated; an everyday crossword is meant to be solved without books. Or I don’t find it or haven’t got the book handy, and I’m reduced to a guess between “fail” and ‘fall’, with “call” and even “wail” as possibilities. “Bawl’ is perhaps a little out of keeping; but I now see that Chambers gives “‘waul”’ or “wawl”, “to cry like a baby or a cat”: mightn’t Shelley have done that? Joking apart, I think that if these clues ceased, “they’d none of them be missed”. They can’t be called unfair when they’re chestnuts from well known poems like “I arise from dreams of thee” as this one that I imagined is; but one meets many that aren’t chestnuts at all, and memories do fade, as mine had in this case. I shouldn’t have been able to solve it without research.

Where did I get it? At random from a quotation dictionary.

Less direct literary references are fair game and amusing when additional help is given, but they are, I think, outrageously unfair without it. I came across one very recently; I dare not quote it for fear of retribution, but I have concocted what I think is an exact parallel. “Loan no public speaker wanted [4]”.

We will suppose that the solver can reach _ A_ S from interlocking words — that doesn’t help much, and remember that he might only have got one of those two letters and have turned to this clue for further guidance. He might think of the right context; I have purposely made it even more of a chestnut than the one in the actual clue I am thinking of. Stop reading and see if you can solve it. I won’t give the answer till the end of this chapter.

Before, in the next chapter, we have a look at the various legitimate devices that are used in what in this book are regarded as sound clues, there are some types of inaccuracy and, in my view, unfairness that I want to preach against. One is failure to indicate the right part of speech. This most often happens when the word is an adjective. I have seen clues such as “Can the station be altered? He sticks his toes in [9]”. The first half, though it isn’t syntactically accurate — we’ll discuss that in a moment — may well contain an anagram: the number of letters, as usual, is eloquent. If you’re quick at anagrams, you may see OBSTINATE at once; but the definition is impossible: the answer to it ought to be a masculine noun. I have lately met another clue where this violence was done to a verb. Again I’ll try to produce as near a parallel as I can. “Put the swot in the river: that will burst him.” Answer: EXPLODE. The merits of the first half don’t matter here; the point is that “that will burst him”’ is no indication of a verb meaning “‘burst”. The clue-writer hasn’t said what he means.

Now for syntax: go back to “Can the station be altered?” I believe many composers would pass it, and many, many solvers wouldn’t bother about it. But analyse it in detail, and you find that for the clue to be sound — to say what it means — those five words have got to convey the meaning “The letters of ‘station be’ are to have their order altered”, or, to put it interrogatively, ‘What emerges if ‘station be’ is altered?” Now the clue-writer has got to veil his meaning; but the veil mustn’t be so thick that the words can’t bear the meaning intended. “Can the station be be altered?” might pass syntactically with an understood “Yes”; but that, of course, would be intolerable. The clue- writer feels this and gets out of it by making “be” do double duty, which he has no right to do.

The best I can do with the idea is “Station be altered? I’m determined not to budge”’, the first half being an indignant query, the second now indicating an adjective, with “I” representing the answer describing itself, in the time-honoured manner of old-time acrostics, charades and other word-puzzles — which brings me to the vital syntax of pronouns.

Many clue-writers make no distinction between the pronoun “I” and the letter “I”. Both may legitimately be used in clues, but I can see no valid reason for the prevalent disregard of syntax in referring to the letter. The pronoun, mostly used as I said just now to represent the required word, is naturally followed in the present tense by “am”. But why should the letter be? It is not the first personal pronoun, to which alone “‘am”’ belongs. When we clue the word PLAIN by “I am in the plot, that’s clear’’, we really mean “I is in the plot’’, so we aren’t saying what we mean. It may be urged in defence that this is a common convention which deceives no one, so why worry? My answer is that a common convention involving inaccuracy is a bad one; and in this instance it cannot be claimed, as for instance in the case of omitted punctuation, which doesn’t involve anything like such blatant inaccuracy anyway, that this insistence on the letter of the law cramps the style of the writer. In nearly every case the trouble can be got over by saying “I must be", “I can be’’, etc., or by using ‘‘one’’. Often even worse violence is done to syntax by “You’ll see me in the plot’: how can “I’’, whether it is a pronoun or not, be defined by ‘“‘me’’? The same thing, of course, applies to “we’’, “us’’, “he’’, “him’’, etc., but they are needed far less often.

A few more instances of inaccurate types of wording may be helpful. Beginnings, middles and ends of words are apt to be treated very loosely. “Cylinder-head’’ may fairly indicate C, or ‘‘masthead’’ M; but why should ‘‘redhead”’ indicate R? It can’t mean ‘‘the top of red’’, and not even suggestions of cents and billiard-balls will convince me that it can. Middles should be exact middles: G is the middle of the night, ND is the heart of London, LOTH is the Heart of Midlothian; but IVE is not the centre, or hub, of the Universe. Both first and last letters are often too loosely indicated. I don’t like “Convivial occasion to separate with a happy ending” for PARTY; “a happy ending” doesn’t, to me, mean “the last letter of happy’’. One more point: how long can a beginning or an end, a head or a tail, be? It can obviously be one letter: in a compound word it can legitimately, I think, be the first or last member. But often it is used to cover two or three letters not forming a whole, or even for more than half the word. ‘‘There’s a horse in the stable with a lion’s tail” for STALLION is the sort of thing I mean: why should a lion have a tail three times as long as the rest of him?

Another form of looseness which I dislike is the use of ‘‘back” or “returns” or the like to point to the reversal of a down word or part of one. This, again, doesn’t say what it means. ‘Return of a star — nonsense!” is fine for an across word. (Note, incidentally, that we have here a justifiable exclamation mark: “rats!” is an exclamation.) But in down words the star must rise, not return, because that is what he, she or it actually does.

Two final hobby-horses, anagrams and “hidden” clues, and this chapter will be almost finished.

I feel three things very strongly about anagrams. First, they must be indicated, not by the dull “anag.” at the end, but by some reasonable sign that the letters are to be disordered. “The orchestra pulls a heavy weight” for CARTHORSE doesn’t say what it means: ‘Clumsy great creature puts the orchestra out of tune” does. This juxtaposition of anagram and definition, with no suggestion of mixture, is still a terribly common practice.

Once again, opponents may say that it is a convention: the number of letters is enough indication. To me it isn’t; I can’t see why a solver should be expected to look for an anagram when he is given no actual sign that it is there.

Secondly — and here, for once, I differ from Afrit — I hate what I call an indirect anagram. By that I mean “Tough form of monster” for HARDY (anagram of HYDRA). There may not be many monsters in five letters; but all the same I think the clue- writer is being mean and withholding information which the solver can reasonably demand. Why should he have to solve something before he can begin to use part of a clue? He has first to find “hydra” — and why shouldn’t it be “giant”? — and then use the anagrammatic information to help him think of “hardy”.

To make matters worse, this sort of thing is freely done with rare words in the connoisseur’s type of puzzle. As I said, I’m crossing swords with Afrit here; and I’m not forgetting how difficult his puzzles had to be. I never tried to convert him on this subject, because my own views then were so unformed, and I sometimes used indirect anagrams myself: I wonder if I could convert him if he were alive now? My real point is that the secondary part of the clue — other than the definition — is meant to help the solver. The indirect anagram, unless there are virtually no alternatives, hardly ever does. He only sees it after he has got his answer by other means.

Thirdly there shouldn’t be too many anagrams, or they will pall. Composers vary very much about this. Some use masses, some hardly any. I think, after long experience, that there is an ideal limit — three or four in a twenty-eight-word puzzle, never more than six in a thirty-six-word puzzle. Why do I suggest even as many as this, when there is a danger of satiety? Because I know many solvers rely on anagrams for a start; here, at least, there is something concrete — the letters are there, and the length pf the word, however skilfully the anagram is veiled, is apt to yield its secret. Part-anagrams, of course, are also common. These too should be used with restraint, more being allowable if there are few full anagrams among the clues. In the twenty-eight-word puzzle I would suggest two full anagrams plus three or four part-anagrams, or four full anagrams plus one or two part-anagrams as examples of a fair proportion.

What of “hidden” clues? These, unless the hiding is very skilfully done, are sure to be sitters for experienced solvers, especially if many are used. This is a pity for the lazy or rushed composer, for with short, dull words they are by far the easiest way out. But if he uses many of them, the solver will pounce on them every time, and the puzzle may become too easy. One per puzzle, or very occasionally two, is quite enough. I generally try to have a puzzle fairly often with none at all. Then there is a chance of surprise. We must remember that in this case the letters of the word are in front of the solver in their right order, and it isn’t easy to prevent the answer from sticking out a mile, unless … ! That brings me to my last point about “hidden” clues; it may be fair, but to my mind it is most inartistic, to have redundant words in the hiding-place, like this: ““This girl appears in black at every party she goes to [4]’’. I needn’t tell you the answer; but I hate those last four words of eyewash! Compare this with one of Afrit’s best: “Girl detected in imitating Lady Sneerwell [6]”. You won’t take long to find her, because in this context you know she’s hidden; but she might not catch your eye at once otherwise.

A final word on brevity. This is, I think, the last of the few points on which I differ from Afrit. His clues, especially in his ‘Armchair Crosswords’, tended to be rather long. I think a good short one beats a good long one every time. Perhaps I have been influenced by necessity during the war. Clues had to be short then, because space was precious. But even today I try to restrict as many clues as possible to the number of letters (and spaces) that will fit into one line of print in the Observer. I only give way to longer ones when I really like the idea I have hit upon and it just won’t go into one line. I’ll finish this chapter with one of my favourite short clues, not mine but a first prizewinner in one of my clue-writing competitions: Mr E. Gomersall’s “Cake with nuts on top” for MADCAP (a down word, justifying “on”). No long clue could be so attractive.

Answer to “Loan no public speaker wanted [4]”: EARS (‘‘Friends, Romans, countrymen … I am no orator as Brutus is.’’)

---



Types of Clue

We have now reached a stage at which it will be helpful to classify under their headings the chief devices which the composer of cryptic clues uses. The composer’s ideal, of course, would be a diagram in which every word, by the accident of its meanings or combination of letters, lent itself directly to the use of one of these devices. At the same time such an ideal diagram would consist of words which preserved a due proportion between the various devices to which they lent themselves. Such an ideal is in practice unattainable; the composer can, in the nature of things, only choose a certain number of his words, and some of those that he is forced to use to fill up each section of the diagram are sure to be intractable ones, which do not fall under any of the broad headings given in this chapter. These will therefore call for improvisation. In the end this may really be a good thing, though it gives the composer a lot of trouble; for the improvised clues which are eventually found to cover the intractable words will add variety to the whole puzzle and may even please the solver more, by their novelty, than the neatest of the normal types. Nevertheless, every sensible composer will still try to include in his diagram as many tractable words as he can: if he saves himself time in composing the diagram by always including, without regard for clueability, the first word which fits, he will find that the time saved, and probably more, is lost again when he comes to write his clues.

If one ignores minor variations, there are seven main headings under which clues to tractable words may be grouped:

1. Two or more meanings. 2. Reversals. 3. Charades. 4. Container and contents. 5. Puns. 6. Anagrams. 7. Hidden.

We will take these in order, but first it may be as well to point out that the clue-writer must never forget that his object is legitimately to deceive; so, if he is writing a clue to a word that lends itself readily to one of these seven kinds of treatment, he must do his utmost not to announce too obviously, by the form of the clue, the type to which that clue belongs. The really experienced solver, who knows the composer’s style, will often be impossible to deceive, however hard the composer tries; but every now and then an unfamiliar twist may do the trick, and the dropping of the penny, when it comes, will then give the solver that extra enjoyment which the composer aims to provide as often as he can.

I propose to use, as examples of each type, old Ximenes clues; but please — if you are not a Ximenean — do not be put off by this and think that it will all be too obscure! I shall choose, as far as I possibly can, clues to words in a normal person’s vocabulary, and clues, too, which do not demand in any of their parts specialised knowledge that can only be got from dictionaries or reference books. Very many of the examples given will not be so difficult as to be out of place in an everyday crossword. The answers to some of them will be held back till the end of the chapter, in case readers like to try to solve them from scratch.

1. Two or more meanings.

This is obviously the simplest form of cryptic clue: it departs less far than any other from the antiquated single straight definition or veiled definition. The composer will most often find it useful in dealing with short, common words of Anglo-Saxon origin; for these very often have a wide variety of meanings. His object will be to combine two or more of these meanings in a clue which embodies an artificial and misleading connexion in sense between them. The experienced solver will know, as usual, that what he reads doesn’t mean what it appears to mean; but he may be deceived for a while — for instance, he may not see at once whether only two meanings, or more, are being offered to him.

Let us take first two examples of the use of two meanings, “Fleeces, things often ordered by men of rank [6]”. This aims to suggest the noun ‘fleeces’, and the ordering of them from manufacturers by wealthy noblemen for use in luxurious wool-lined coats. But in fact it is the verb ‘‘fleeces” that is intended; ‘men of rank” are soldiers, not noblemen, who “order” the things referred to on parade. RIFLES is the answer: to fleece and to rifle are to plunder, and soldiers order arms.

“He went to hell — several of them, in fact [5]”. The solver will probably soon see that he wants the name of a person who went to hell (biblical?) and a word, which happens to be the same as that name, meaning hells, or hellish places. He may or may not quickly think of Dives (and Lazarus) and dives, disreputable resorts; if he doesn’t get it from scratch, no doubt a letter or two from crossing words will suggest it.

Now for three meanings. “What does a man in distress need? Bar that’s near [6]”. Notice that this doesn’t scream too loudly that three meanings are there: it seems at first sight to fall rather into two parts. But on further thought the acute solver may say to himself ‘Can there be a word that means ‘a near bar’?” Perhaps instead he wants a word that means in one sense ‘‘bar”’ and in another sense ‘near’. So the image of the neighbouring pub fades away: in time he thinks of “bar” =“except”, ‘‘near” = “stingy, economical” — and SAVING fits both of them, as well as being what a man in distress needs.

‘“That’s a bargain! — Swindled, proper [4]”. First note the exclamation mark; if I’m sticking to my resolution, the answer should be a genuine exclamation — or an outrage! This one really looks like a three-meaning clue, so try “swindled”. There can’t be many words so short with this meaning: DONE should suggest itself; and the done thing is the proper thing, and ‘‘Done!” satisfies the first part and justifies the exclamation mark.

Clues that include more than three meanings are rare: I use them chiefly in two ways. The first occurs when the addition of a fourth, and even a fifth, meaning seems to add something piquant to the picture presented by the clue, though it isn’t necessary or, strictly speaking, very helpful towards solution, e.g. ‘Pub charmer, serving spirit — 6s. 8d. to 10s. each [5]”. The first four words give three meanings of ANGEL — the name of a pub, a charming person, a spirit who serves; but the imaginary picture of a barmaid is, I think, improved by the suggestion of the high price of drinks, given by the fourth meaning, admittedly a dictionary one, of an old English coin with the stated value.

A glance at the dictionary would serve to make a solver sure he has got the right answer. The second purpose is to make, from the meanings of a word with many meanings, a large selection which, in telegraphese, can present the high spots of a dramatic career, thus: “Old-fashioned gamble — celebrate — arrest — custody — judge — defend successfully — continue as before [4]”. All these seven meanings will be found in the dictionary under HOLD.

The first is obsolete, but most of the others, either as verbs or nouns, are familiar. Here there is little attempt at deception — pretty clearly all seven bits are meanings; but the solver may be held up by not being able to see the wood for the trees. This last is obviously a device to be used only very occasionally.

Finally try these three; solutions will be found at the end to the chapter.

(1) “Irritation and worry? Eat out [4]”.

(2) “V.I.P. at fête is for broaching the booze [6]”.

(3) “There’s very little room outstanding at Oxford [5]”.

2. Reversals.

Opportunities for these are limited: not many words, when turned back to front, will form another word or words without more ado. But when this chance does arise, a neat clue is usually easy to find. One point, which has already been mentioned, is here vital: across words reversed go back, down words reversed go up, and we shall be guilty of inaccuracy if we disregard these facts in the interests of misleading sense, as some composers often do. There is always at hand one device open to the composer who is particularly pleased with a reversal clue, say, to a down word which involves making it go back, not up. The whole diagram can be turned inside out, so that the across words go down and vice versa. No difficulties arise: try it and see. It may be a nuisance, but it is worth while if the clue is a really good one and cannot be made sound by other means.

Here are two examples with explanations, and one for you to try to solve from scratch:

‘Heater with electric valve reflex — one just out’ [9 — an across word]. “Reflex” is clearly appropriate to the reversal of an across word: it would hardly suit a down word. An Etna is a heater, and tube can mean an electric valve; hence we get etnatubed reversed into DEBUTANTE, one who is “just out”, a nice change from the picture of the modern gadget.

“Loved one with glamour rises, splendidly attired” [7 — a down word]. This is an easy one for the hardened solver. ‘‘Rises”’ in a down clue always screams reversal, and “glamour” at once suggests that the reversal will end with “it” or “S.A.”: try TI_ _ _ _ _  or AS_ _ _ _ _ . TIARAED, dear-a-it reversed, should not be long in coming.

Now try

(4) “An ideal wife needs to get a meal to go round” [4 — an across word] — or “… needs to get a meal brought up” would do, if it were a down word.

3. Charades.

You all know this old-fashioned kind of riddle: ‘my first is …, my second is …, etc., my whole is…”. No doubt you acted in them as children. The essential thing was a word which would fall into convenient complete parts, and these are fairly common and are very useful material for cryptic clues.

As will be seen later, words not wholly appropriate for this treatment can be so used if one part is reversed, is mutilated in some simple way, or has its letters jumbled, or if an abbreviation is used for it — each of these devices must be clearly indicated in the clue.

We will take straightforward examples first, then a few more complex ones.

"Remains precisely how he is [5]”. Here is simplicity itself but simplicity disguised, perhaps, enough to deceive for a time, The solver might not at once think of “remains” as a noun, though he should always try to include every possibility. Again the idea of “he is” merely representing “he’s” might not: strike him at once because of its sheer artlessness. However, the answer is ASHES (as he’s).

“Voting against, told to go and vote for [7]”. A rather more interesting one, but not very difficult. Remember that in a charade clue the whole may come either first, as above, or last, as here, though in the old-fashioned riddle and game it always came last. “Voting against’? Think of pro and con: to tell someone to go is to send: therefore, CONSENT.

“Alas, the best of lawns do this [10]”. This is the subtlest kind of charade clue (not often possible) where the parts, in this case three words, form a coherent phrase: this makes the discovery of the type of clue being used harder for the solver. To cut a long story short, the answer is LACKADAISY (lack a daisy).

Before turning to the end of the chapter, try these two.

(5) “One in flames managed to make a landing [4]”.

(6) ”Puts hat on back to front — often clicks [9]”.

Note that the second of these is a good deceiver, because “back to front” at once suggests a reversal; but as the clue is given you in this section, you know it is a straightforward charade.

We will now proceed to complex charade clues, where one or more parts must be doctored in some way; the indication of the whole, the definition, can never be doctored thus, of course.

“What was Gertrude over Hamlet’s upbringing? She must have been crazy’ [6 — a down word]. ‘“Upbringing”’ points to reversal: Hamlet will probably be the Dane, reversed. _ _ ENAD, and G. was his mother: hence MAENAD, a crazed female worshipper of Bacchus.

“Spoil, or knock about? Mother, the latter’s wrong [8]”. Ma again? Yes. And ‘wrong” is a common sign of an anagram, or, more strictly; a jumbling of letters: a proper anagram is itself a perfect word, but for our purpose here a mere jumble serves just as well. So we want MA _ _ _ _ _ _ , the last six letters being those of “latter”: MALTREAT. Note the use of a double definition, harmless — perhaps even helpful — to the solver, used to complete the imaginary picture presented by the clue.

“To make one giggle in morning service [5]”. Here we have the use of an abbreviation for one part of the charade: AMUSE (a.m. — use) — a very simple one. Such a common abbreviation need hardly be indicated: rarer ones I usually hint at by “short”, etc.

‘“Armadillo-like tailless cat has burrowing habits [8]”. Zoologically speaking, this is evidently nonsense of a wild kind; seek rather an adjective meaning “like an armadillo” followed by a cat with the tail (last letter) removed, the whole being a burrowing animal. PLATYPUS (platy-pus-s), it must be. Finally a charade clue with both parts mutilated — not very neat, but legitimate:

“Bombers: half last Russian order nearly completed [6]”.

The first word fairly obviously indicates the whole: the first part must surely be la or st. If you remember the last war, you should think of STUKAS (st-ukas-e). Not heard of ukase? Bad luck! I said I would avoid dictionary words as far as possible, and I don’t think I have included many.

One for you to try, a not very difficult one, of the complex type, i.e. with one part doctored.

(7) ‘‘After the war, ruined, scrape, devoid of love [10]”.

4. Container and Contents.

Such clues resemble charades in having wholes and parts, but their parts are outside and inside instead of side by side. Opportunities for this treatment are probably more frequent for the composer than those for charades. These clues too can be simple or complex.

“I’ll be kept in by the beak for talking too much [5]”. A very easy one: NOISE (no-I-se). Notice “I’ll be”: “I am” would be criminally inaccurate, as we have already seen, because the letter I is third person. |

“With hardly a stitch? Do let me in unobserved [8]”. This resembles in subtlety the clue above to LACKADAISY, in that the contents form a self-subsisting (though very short) phrase. “Do let me” is equivalent to “can’t I?”: the container is sly, unobserved: hence SCANTILY.

Now for a clue in which the container is a two-word phrase, the contents a short word:

“Passing round the low-down (not compromising) [12]”.

Quite often you may do well to suspect an apparently unimportant parenthesis of being vital and, as in this case, giving you the definition. “Round” (unless it means the letter O, as it often does) may well point to a clue of the type we are discussing.

Then a word or phrase meaning passing will surround gen, the lowdown. The result is INTRANSIGENT (in transit).

Two for you to try.

(8) “Habits that are touching in cows [8]”. (Don’t forget that “cows” can be a verb.)

(9) “Parking solutions? ‘A tale told by an idiot, full of sound and fury’” [7]. (Allow for a little poetic licence in the use of the tag from Macbeth.)

Now for the complex form of this type. "Two cheats had up — an hour, kept in — or this?” [8 — a down word]. It looks as if the contents will be ‘hr.”, an abbreviation: two words meaning “cheat”, each reversed, will contain it. The: definition, then, can only be "or this?”, which surely must mean an alternative form of punishment to being kept in. It shouldn’t be too hard to find BIRCH-ROD (crib, hr., do). What about the hyphen? Yes, in Everyman and in most puzzles it would be indicated, but as Ximenes I withhold such help, though I do say ‘(2 words)” when that is applicable.

“Bridges, for example, long overdue: a top-class lot of engineers to be called in [8]”.

Abbreviations again. ‘“Top-class” must be AI (i.e. A1) or U (cf. non-U): “lot of engineers’? can only be R.E. As the general picture suggests engineering, get right away from it and remember that Bridges was poet LAUREATE (l-a-U-R.E.-ate). There are rather a lot of parts this time, and it is a somewhat clumsy clue as a result; but it illustrates possibilities.

Now try these two.

(10) “The Reds are 6 up in busy exchanges [8 — a down word]”.

(11) ‘“I’m very abstemious about hock — only a suspicion [7]”.

Before I leave this type, I want to point out that “over” is by some composers used for “round” It ought not to be, because it doesn’t really mean that.

5. Puns.

This type depends usually on what are technically called homonyms, words of different meaning having the same sound as each other. It can become monotonous, as puns in conversation undoubtedly can, and must not be overdone; but it is useful to the composer and, when skilfully used, can even amuse the solver, I believe.

A very simple example is: “German sounds in need of a regular brushing up [4]”. HERR (hair). This has, perhaps, little merit except the possible deception in the word ‘‘up”, suggesting a reversal, if it is a down word.

A rather more difficult one: “One of us is seen at Chiswick — or more, we hear [5]. Think of the boat-race and Chiswick Eyot: the answer is EYOTS (eights).

One involving two parts (and not actual homonyms): “Madame’s wear, we hear, is to be Scots-tweedy [7]”. “Madame’s” is probably just “her”: her suit is to be HIRSUTE.

I leave you to solve one which I prefer to any of these:

(12) “Sounds like the old grass to weed, and I’m tired! [7]”.

Don’t ignore the exclamation mark, which is a useful hint.

This type, I repeat, should not appear too often: for one thing, there are very few ways of indicating it to the solver, not nearly so many as there are with other types, such as the next that we come to.

6. Anagrams.

I am particularly fussy about these, and I am more inclined to be critical about their use by others than about that of any other type of clue. I have already emphasised how strongly I feel that they should be indicated by some word or phrase that can suggest that mixture of letters is to take place, that restraint should be exercised in the number used, and that they should not be indirect anagrams, which only give a synonym of the anagram instead of the anagram itself. Those who use this last type of anagram seem to me to disregard the fact that the subsidiary part of the clue, which is added to the definition, is meant to help, not to be a further problem in itself.

The anagram clues I like best are of two sorts — those in which there is a really appropriate connexion between the anagram and the definition, and those which come under my heading “& lit.”, which will be explained in Chapter VIII. There follow a few examples of the first sort.

“Halo Iris (mixed) for handsome heads (scented) [8]”. This sounds like something out of a nurseryman’s catalogue, but “mixed”, the most obvious of all indications, though appropriate here, shows the way. HAIR-OILS is the answer.

“The seating’s untidy: get a broom [7]”. Here “untidy” is the operative word, and you will soon find GENISTA.

“Port’s apt to make you steer it erratically [7]” Not your car, but TRIESTE.

“Le Spice is very naughty: cut out heavenly body [7]”. No pin-up here — merely the letters of ECLIPSE.

Finally a long one, involving the invention of what, I am sure, is a non-existent disease, but it completes the picture:

“Cerebro-acne’s rampant — a spreading growth [12]”. It might take you a little time here to find ARBORESCENCE.

A word here about long whole-word anagrams may not be out of place. Composers should beware of them, as they are apt to become hackneyed, e.g. carthorse and orchestra, antagonist and stagnation. I am conscious of having erred in this way myself, not, I hope, too often. The longest one I ever concocted for a Ximenes puzzle involved making an adjective plural — a bit of a cheat: ‘“This kind of poppy is a variety of the schizothecals [13]”’.

ESCHSCHOLTZIA. The adjective schizothecal was in the edition of Chambers that we were using at the time; now it has disappeared. Its meaning, of course, doesn’t matter.

Before we leave anagrams, try these.

(13) “Exertion makes us alter our belt [7]”.

(14) “Director of a buffet with questionable merits [6]”.

7. Hidden.

I will repeat here my caution to composers that this type is hopelessly obvious if it is at all overdone. Restraint 1s needed far more than with anagrams, Secondly, I hope I have converted composers who have read Chapter V to my belief that redundant words in the hiding-place — mere window-dressing — are inartistic, if not unfair; no word other than an article should occur in the hiding-place that does not contain at least one letter of the required word. a I won’t give many examples, because the type is so familiar:

“Give power to some of the policemen ~ a blessing [6]”.

ENABLE.

“Where they wear saris in thin diamanté [5]”. INDIA.

And finally one for you, possibly more subtle than most.

(15) ‘‘An item in fuel is somewhat fluctuating supply [6]”.

Answers to clues unsolved in the text above:

1. FRET.
2. OPENER.
3. ORIEL (it can mean this as well as a window).
4. ENID (Idylls of the King).
5. ALIT (a-lit).
6. TURNSTILE (turns-tile).
7. HEARTWHOLE (heartw-hole)
8. APPARELS (re in appals).
9. GARAGES (rage in gas).
10. MARXISTS (six reversed in marts).
11. THOUGHT (t.t. about hough).
12. HEIGH-HO! (hay, hoe).
13. TROUBLE.
14. SMITER (that kind of buffet).
15. LISSOM (súpply, not supplý).

---



Improvised Clues

What does the setter do with words which do not conveniently allow any of the treatments described in the last chapter? In the old days he often used to fall back on a mere definition. That won’t do for us. Our principle is that every clue must have some entertainment value, and this can very rarely be achieved unless a clue has two parts. So we must improvise. Sometimes it may be a long time before a good idea arrives, but it is always possible that an inspiration may produce a really good clue. In this chapter my aim is to give some examples of devices which, though they are not used as often as those classified in the last chapter, may from time to time be useful. I shall finish these off with some wild extravagances, the sort to amuse if they don’t appear too often.

1. Heads and Tails.

This is not an uncommon device, and there are traces of its use in the previous chapter; but I think it is used rarely enough for this to be the best place for it. It is the type which indicates the required word or some part of it by alluding to a word which must lose its first ar last letter, thus: “Wail — South African beheaded not long ago” — ULULATE (Z-ulu, late). “Hard workers have a limit: one exam unfinished” — TERMITES (term, i, tes-t). A certain caution is needed here, lest this type should get out of hand and become questionably fair. Some composers, as I mentioned above, allow heads and tails to consist of more than one letter, sometimes even of half the letters of a word. I question the soundness of this, though, as I said, I would make an exception in favour of a hyphened word. If, say, two letters are referred to, I think this should be made clear, by writing “heads” or in some such way. A further point is that while head and tail may be applied to either across or down words, and so may some other similar indications such as beginning and end, top and bottom are only appropriate to down words. This, I think, is only logical.

Here is a straightforward example for you to solve.

(1) “Besetting sin cut short, doing household chore [10]”.

2. Peculiarities of speech.

These include local accents, lisping, stuttering and even the effects of the common cold. Take first the cockney. DIME is not a very promising word: it has only one meaning, though one can perhaps use a reference to a dime novel, the American equivalent of the penny dreadful or — more probably nowadays — shilling shocker, But what about “Arry’s pronunciation of dame”? That may lead to “The figure of a shocker — ’Arry says there’s nothing like ’er in South Pacific.” That song “There is nothing like a dame” is a bit out of date now, but everyone knew it at the time when this clue appeared, and probably many remember it still. Then the Welshman’s habit of softening B into P may serve its turn, so that for PLIGHTER we may offer “The Welsh beggar has taken the pledge.” You may notice that this word would provide light in per; but that doesn’t, to me at least, suggest much of interest. Or we may turn to the north for TARTUFFE: the north country pronunciation of tough suggests “I’m a hypocrite to denigrate a Liverpool rocker? Sounds like it.” Notice that the hint of a pun, “sounds like it”, is essential for accuracy. Again, the dropping of the G in the ending -ing may be useful. Try for yourself:

(2) “Goin’ for a ride, in short, in minimum attire [6]”, bearing in mind that ‘i’ is short for in. Local or poetical you say ? Not necessarily: the dictionary doesn’t attribute it to any special usage.

Lisping is the sort of device that mustn’t be overdone and may be too obvious, but we’ll have one example: “A chap like Macbeth’s quite rethponthible” — THANE. Stuttering can be, perhaps, a little more entertaining, if, again, it isn’t overdone: “I was thought magical, but I’m fer-futile’ —- VERVAIN (a magical plant).

Finally, with a cold one tends to make Ms into Bs and Ns into Ds: hence (for you to solve):

(3) “Effect of cold on one undressed? One who’s overdressed [4]”.

3. Words treated as parts of other words.

This method can often be applied to words unhelpful in themselves, but it needs care: it can easily become much too difficult, and it is probably best to restrict it to cases where the longer word is a reasonably familiar one, thus: “After an outing, terrier wants a drink [5]”.

Don’t forget that “wants” can mean “lacks” as well as “desires”; so you should look for a word, probably a past participle of a verb, meaning “after an outing”, and for a kind of terrier from which the letters spelling a drink have been removed: AIRED (Aired-ale). Again, a rather more difficult one this time: “A lifter causes swearing (four letters first) [5]”. The solver has to understand the parenthesis to be conditional, i.e. to mean “if four letters are put in front of it”; tricky, perhaps, but just fair, I think. Don’t try naughty four-letter words: these don’t occur in crosswords. Try to think of a lifting device, which, with four extra letters in front of it, suggests swearing or an oath. The answer is DAVIT (affi-davit). Here is an easy one for you to try.

(4) “Trade without an effort feeds many mouths [5]”.

4. Initial or final letters.

When the setter is in real desperation with a dull or intractable word, he may try a clue which indicates its letters as the initial or final letters of other words. Unless rarely, and reasonably skilfully, used, this method is apt to be transparently obvious, and I hardly ever admit it to a puzzle. I used it once for ESME, a name which can very easily be indicated — by a ‘hidden” clue, but I had done that before, and I wanted something new. So I tried “Name of the last of the Legends Barham wrote.” I doubt if many solvers were green enough to turn to their Ingoldsby, but I like, wickedly, to think that a few did. The device is less obvious if it is only used for part of the required. word, as in the following.

(5) “Mocking bird with tips of tail blue or grey [8]”.

5. Foreign Languages.

A word whose letters are unsympathetic to the setter as long as he thinks in English may, perhaps, be more helpful in some other language. But he has no right to assume that his solvers are linguists, even if he is one himself, and he had better restrict himself to very elementary references, of this kind. Mine are usually to French, the only modern language of which I have any knowledge; they are necessarily elementary and carefully checked with a dictionary — I know my limitations.

This is the sort of thing: “English writer, but understood by all Frenchmen” — MAIS. The best that can be said for it is that it has a little more point than the almost inevitable “Mother is … or “A reverse for Thailand”. A slightly more venturesome sally is the following: “Very critical — would be bolder about Vichy? [10]”. Various signs here should point the way: “would be… about” suggests that the inclusion in the required word of the letters of a word satisfying “Vichy?” would produce “bolder”.

Then note the question mark: that probably means that Vichy is an example of a generic word — water — eau, as it’s French.

These considerations may lead to MORDACIOUS (mor-e au- dacious).

Here is an easier one for you to try.

(6) “What’s French? That’s French: it’s all French [10]”.

The knowledge of French required is most elementary.

6. Literary references, etc.

I have already expressed my dislike of purely quotational clues; but I have no such dislike of making part of a clue have a literary connexion or, on a lower plane, a connexion with some familiar phrase. Even if this proves to be elusive to a solver, he has the other part of the clue to fall back on. In the first example quoted, the solver has two indications as well as the quotational one: “Poet whose ‘sculptor is paid’ in advance [5]. The quotation, though in the Oxford Book of English Verse, is not a chestnut, and there is no reason why you should know it; but you don’t need to. PRIOR means “in advance” and is also the name of a poet; so that must be right. The quotation (which I used because it fitted so neatly with “in advance”) comes from Matthew Prior’s “For My Own Monument”.

The next example is a biblical reference, familiar to everyone, once they see the point — which is somewhat disguised. “Procedures followed by Romans [4]” — ACTS: no need to explain that one. Again, a disguised reference to a common mild expletive; “Fed up about being at home, like my aunt? [7]”. “At home” is usually “in”; answer: SAINTED (sa-in-ted, and ‘My sainted aunt!”).

Finally I offer you a reference to a poem familiar to very many people.

(7) “They make tongues wag: you can believe what one of them tells you more than twice [7]”.

7. Outsides.

This is an unusual device, which can take several forms. I quote two, which I have used in order to deal with intractable words. “Cavalrymen disheartened in Normandy [4]”.

This is a converse of a “hidden” clue. All you have to do is to discard the heart of cavalrymen — and notice that it must be an exact heart, leaving an equal number of letters at each end — and you get CAEN. You have only to spot the idea to get the answer, but it is an unusual idea and may cause trouble for a little while.

A simple anagram would have been possible, but I couldn’t make one interesting.

The second example uses this idea in a different way: “Put me in the bloomin’ ward — you’ll see what I am [4]”. I could do nothing with GAWK, with the simple meaning of a clumsy, awkward person, till I saw it as part of “blooming awkward”!

8. Various.

Finally we come to unclassified extravagances. These may hit the setter, if he is lucky, when he has been staring for some time at a word which suggests nothing of one of the more usual kinds. For instance, BLAST is a pretty dull word, until one notices B last: that suggested to me “Withering effect, invariably found in rhubarb”, an unusual form of charade, but at least a fair one.

Sometimes one may reasonably allude to an imaginary word that might exist, thus: “A great big swimmer, like a barrel? [5]”.

Note the question mark. In this case it suggests that the last three words are questionable, because there is no such adjective as TUNNY, derived from tun; but there might be. This clue would be definitely unfair without the question mark.

One may occasionally ask the solver to make a logical deduction in a slightly unusual way: “One who will be (Part II) married? [5]”. This might suggest that the person concerned is (Part I) unmarried but likely to be married later; and so he is, if he is a PARTI. But he needn’t ever get married; hence the question mark.

Normally the artistic setter likes his clues to make misleading sense. Occasionally a good idea may strike him, which simply refuses to submit to such “sensification”: then he may, just for once, have recourse to a puzzling sort of nonsense. I had deliberately put AUTOMOBILE into a diagram, intending, as one who has suffered occasionally (being unmechanical) from motor-mowers, to have fun with the pun on “ought to mow” followed by “bile”. After more than half an hour trying to make sense of this, I relapsed instead into this nonsense, which, I was later told, proved not unamusing when the penny dropped: ‘Mowers which are this sound of more than half: if not, the rest follows’. In other words, mowers which are automobile ought to mow (sound of six letters — more than half); if not, bile follows. Here is a simple one of the same kind for you to try.

(8) “When the clouds inside outside, inside it! [6]”.

Usually these improvised clues arise, as I have said, from the setter’s difficulties: they may also be used because he wants to avoid using for a very common word a method that has been used all too often before. Take the familiar VETO, which has appeared so often with an anagram of vote or in “hidden” form.

Wanting a change, I tried a rather unusual use of a split blank for the word required, thus: “Lea— — act? On the contrary!” (lea-ve-to act).

My next quoted clue I include with very strong reservations, because I regard its method as clumsy and definitely to be avoided, unless there is some strong compensating merit. Others, I think, use it much more readily than I do. This is the imperfect anagram, an anagram in which some letter or letters must be added or subtracted. The essence of an anagram is, I think, its completeness. This usage spoils it. However, here goes: “Clumsy tailors are, almost, of their work [9]”. The answer is SARTORIAL: “almost”’ indicates that the final E of “are” must be disregarded.

I’m not proud of it, but I passed it because I couldn’t resist the appropriateness of working tailors into an anagram.

I will finish with two extreme outrages of mine, and then add two much less extreme ones for you to toy with. Faced with the dull word ERRATA (err at a? — rat in era? — neither was promising), I thought of the idea of indicating misprints by including some in my clue. A literal is another name for a misprint, so “literals listed in books” is a reasonable definition. I therefore dared to print “Liberals lifted, in boots — there are three there!”’ At least I offered an exclamation mark to indicate the outrage, and no one protested. Secondly, when unable to think of anything very bright for DIAGONAL, I said to myself, “Why not refer in the clue to one of the long diagonals of this diagram? Does either of them lend itself?”? One of them read AOMGCROELIOS. At least this ended with two words, one of them reversed: Roe and soil.

A.O. and M.G.C. both happen to be military abbreviations. So again I was rash and offered “e.g. here: Army Order, Machine Gun Corps, imaginary defendant, rising ground! [8]”. (Richard Roe and John Doe are familiar to lawyers, at least.) At first it might seem that this clue could only be used when the rest of the puzzle had been solved; but this is not really so. A solver might think of the idea that “here” meant “in this diagram’; “eg suggests that there are more than one; and a few letters towards DIAGONAL, combined with these ideas, might make the penny drop. Anyway, here, too, there were no protests! But let me emphasize that I only do this sort of thing very occasionally.

Now here are your two, neither of them anything like as weird as the last two.

(9) “Last appearance of Garbo G.? Wrong surname [8]”.

(10) “Likely to produce a deep dent in a luggage-carrier [6]”.

Before giving the answers to the improvised clues I have offered to you to solve, I will refer, for the sake of completeness, to a further device that I use occasionally, namely the coupling of two clues, with dots at the end of one and the beginning of the next. I don’t care to use it unless there is really a strong point of connexion. I deprecate the use of it, which one sometimes meets, simply for its own sake. When there is real point, its use may give pleasure to the solver by its variation from normal types. The rarity of my use of it makes it difficult to find an example which doesn’t involve at least one “dictionary” word; but this will show what I mean:

“Engineer breaks into quarters — ones for women [6] …” followed by “… Hooked one of those and tucked in [6]’. The answers are HAREMS (hams, R.E.) and HAMATE (ham, ate) — you may not know this word, which means hooked. The point lies in the two references to “ham”, and in the misleading reference of “those” in the second clue to “quarters”, not to “women”, which is unexpected but possible, and therefore legitimate. But such opportunities arise rarely, and the temptation to drag in these couplings pointlessly should, I think, be resisted.

Answers to clues unsolved in the text above:

1. ENVIRONING (env-y, ironing)
2. BIKINI (bikin’, i’).
3. DUDE (nude).
4. INDUS (indus-try).
5. RAILLERY (rail, l-e-r-y).
6. MARTINIQUE (Martini, que).
7. BELLMEN (“What I tell you three times is true”, Carroll, Hunting of the Snark).
8. BROLLY (roll by).
9. SWAN-SONG (Swanson, Gloria).
10. GRAVID (a V in grid: def. “likely to produce”).

---

“& Lit.” Clues

I have kept till last the type of clue which is, on the whole, my favourite, and I am giving it a short chapter to itself. One reason is that its ramifications cut across many other types. Another is that it calls for a good deal of explanation, since the name I have given it for the sake of brevity is not self-explanatory and has from time to time caused misunderstanding among my solver-competitors. I hope here to make it really clear, both to them and to the uninitiated.

The term “& lit.” is short for “This clue both indicates the letters or parts of the required word, in one of the ways already explained in this book, and can also be read, in toto, literally, as an indication of the meaning of the whole word, whether as a straight or as a veiled definition.” So the solver can read the whole clue in two quite different ways, first as an indication of the letters or parts, and secondly as an indication of the whole word.

Every word of the clue does double duty.

This type of clue seems to me to have a neatness and an appropriateness which are entirely its own. Opportunities for it only arise occasionally, most often, perhaps, when a longish word yields an appropriate anagram, thus: “I don’t exactly get more dim — I last, if I’m this [12]”. “I don’t exactly get” suggests an anagram — I get these letters, but I don’t get them exactly, i.e. correctly, arranged; and the next four words contain twelve letters, the required number. So you can read the whole clue first as meaning “If I am what I am, I get the letters of ‘more dim I last’ arranged in a different order.”? You can also read the whole clue as a very full definition of the required word: “If I (and in this case I may be anyone) am this, my memory does not become dimmer, but lasts.” IMMORTALISED precisely answers both ways of reading the clue. (I don’t often use that vague “I”, but here I think the merits of the clue compensate for this irregularity.)

Two more examples of the “anagram & lit.” clue follow, the second, because of its brevity, a particular favourite of mine. “Can make you uncommonly sore with flickering tip [7]”. Here, for the first reading, the parts of the anagram are separated, each with its own indicating word, “uncommonly” and “flickering”.

Hence we want an anagram of “sore” plus “tip”. The answer is RIPOSTE, whose meaning is also described by the whole clue. “Origin of the great Red [8]”. Clearly we want a word which can be the origin, i.e. yield the letters, of great Red. In the literal sense we want the origin, perhaps the birthplace, of a famous Socialist. Aneurin Bevan came from TREDEGAR. The composer, of course, doesn’t often have the luck to be presented with a chance like this.

Here is an artlessly simple clue of this kind for you to solve.

(1) “This is when models that have lost their shape are employed [6]”.

And here is a slightly more complicated one.

(2) “One of three maids O.K. for his issue? Yes [6]”.

Let us now take examples of three other types, “reversed & lit.”, “charade, or container and contents, & lit.”, “hidden & lit.”.

Opportunities for “reversed & lit.” are rare. The next clue quoted is, I think, a pleasantly neat one, in spite of its unpleasant subject. “This back may show the result of strokes of the cat in profusion [8]”. This refers to an across word, so for our first reading we readily look for a reversal. There appear to be two parts, “the result of strokes of the cat” being contained in “profusion”. As often, it will be wise to reject the picture suggested by the clue, in this case the gruesome one of flogging, and to try quite a different result of strokes of the cat, namely purr! This must be reversed into rrup, and included in the reversal of a word meaning profusion, or a large quantity or number.

Probably “rrup” will in any case suggest LARRUPED, purr reversed inside deal reversed; and for our second reading we revert to the gruesome picture and substitute “larruped” for “this” as an epithet of “back”.

Next, a “charade & lit.” clue. “If you want to pass nothing as sound, try this [4]”. Here, to cut a long story short, is VETO again: to vet nought is to pass nothing as sound, and if you try a veto, you will get the same effect.

Similarly, we can use “container and contents & lit.” “The place to make one robust without culture [6]”. Remember, for the first reading, that ‘without’? can mean “outside”. This may point to the place to make one robust as the container and culture (almost certainly art) as the contents. A spa is supposed to improve one’s health; hence SPARTA — and the whole clue describes the traditional nature of that town in antiquity.

Try this one.

(3) “The ultimate of turpitude in Lent [5]”.

Chances for “hidden & lit”. are sure to present themselves from time to time with short words: I will repeat here that the composer must remember how easy hidden clues become if the solver is on the look-out for them; so he must not use them often. “Show the outcome of self-pride ignored [5]”. Terribly obvious, when you know what you are looking for, but note the “& lit.” effect: the letters of DEIGN are the outcome, and the whole clue gives you a possible definition. Here is another that won’t detain you long.

(4) “All you’ll get out of a misbegotten nuisance [5]”.

Incidentally, I am all too conscious of having used various forms of this clue much more than once; it seems to me the only neat kind of clue for this particularly intractable word.

Now that I have fully explained and illustrated the “& lit.” clue in its perfect form, I hope it will not be confusing if I briefly describe an offshoot of it to which I apply, for convenience, the same name. In this offshoot type, what I have called the second way of reading the clue remains the same, in that the whole clue gives a definition of the required word, possibly a veiled one. But the first way of reading it differs from that of the perfect “& lit.” clue; for we find not only an indication of an anagram, charade, etc., but a brief definition plus that indication. There are thus two definitions offered, a brief one in the first reading and a fuller one when we read the clue again as a whole. An example should make this clear: “Tower where one might see a man like Col. Fairfax show impatience [9]”. I hope you know your Yeomen of the Guard; the Colonel was immured in the BEAUCHAMP Tower. You will see that the first reading of this clue gives “tower” as a brief definition, too vague by itself, with the further indication that you will find beau (the colonel was handsome) champ in it. The second reading gives a much fuller definition of the Beauchamp Tower.

Try this similar type.

(5) “What a bishop may have had before getting a crook [7]”.

I am not quite so fond of this offshoot type as of the perfect type; but it is equally sound and often available when the perfect type won’t quite work.

There are two more things to be said about “& lit.” clues.

They involve a certain danger that does not seem to arise quite so much with other types; this danger is that they are apt, if the composer isn’t careful, ‘to waste their sweetness on the desert air”. It sometimes happens that the second reading, giving a full definition, points too easily and obviously to the answer: then, especially if the first reading, giving an anagram or reference to the word’s parts, is rather obscure or involved, the solver will leap straight to the answer and not even realise that he has an “& lit.” clue at all. This makes the composer’s effort a waste of time and detracts from the value of the clue. Ideally, one or the other reading should suggest the answer, without making the solver quite certain he is right; then the other reading should dawn on him and add the certainty that makes a clue fully satisfying.

Finally, I wish I knew who first produced a clue of this sort.

I can’t remember Torquemada using one, though he may have.

I should like to think that Afrit was the originator. He was most certainly a master of the sentence that can be read in more than one way, and he very well may have been the first to write one.

He certainly did use the method, long before I did; but whether he was the pioneer I cannot say.

Answers to clues unsolved in the text above:

1. SELDOM.
2. MIKADO (anagram of maid, O.K., and reference to Yum-Yum marrying Nanki-Poo).
3. FEAST (e in fast).
4. ENNUI.
5. PREBEND (first six words form a short definition; prebend also indicated; the whole also a full definition).

---

Composing an Everyman: The Diagram and Words

How do you begin to compose a crossword? Do you think of the clues first and then put the words in to fit? Is your diagram complete with blocks before you start? Do you use a lot of reference books?

Such questions are always asked whenever someone hears, with some feeling of awe, that you actually compose crosswords for a newspaper. There are, of course, many technical things which one learns with experience; but in this chapter I shall try to show that even a beginner, with a bit of help, can soon learn the principles of composition and avoid most of the snags.

I shall therefore ask you to come with me through all the stages of composition of a typical Everyman diagram, with the warning that each puzzle presents its own problems which cannot be adequately covered at one attempt.

Before we begin, we must know what we are trying to do. We want a diagram 15 squares by 15 in size, containing a symmetrical arrangement of blocks forming a pattern. By the term “symmetrical’ we mean that the design of the blocks — the black squares — will look exactly the same if the diagram is turned upside down.

The best arrangement of blocks, and the fairest for the solver, is one in which every word used will have its alternate letters exposed so as to be used by another word or words crossing them.

It follows from this that every other letter will be hidden between blocks and used only by the word or phrase to which it belongs.

The exposed letters are called checked letters, and the hidden ones unchecked letters, which for the sake of brevity we shall call “unches”.

Everyman puzzles are intended to be solvable by a person of average education on a park bench, in bed, on a train journey, and so on, without the use of a dictionary. The composer must therefore try to include in his diagram only such words or phrases as he can reasonably expect his solvers to know, including a few place-names or names of well-known people. This is an ideal which difficulties in composition may sometimes do their best to sabotage; nevertheless, we must force ourselves to shun the more recondite words.

A symmetrical pattern and well-known words, then, are our aim. What else? If we start by blocking in alternate squares on alternate lines, we shall find that we have eight 15-letter rows going across and eight 15-letter columns going down, and these must be split up to make the words workable. The most useful plan is to have two long words or phrases (of 12, 13, 14 or 15 letters each) running somewhere across the diagram, and similarly two running down it, with the rest split up into two words per row or column. This will produce fourteen answers across and fourteen down.

We are now ready to begin. We have our diagram with alternate blocks all ready, and for the purpose of easier explanation as we proceed, we shall number each square at the top from 1 to 15 and. also each square at the side similarly.

It is vital for the composer to fit in the long words before the short ones. For example, if we simply started at the top left-hand corner and began to build from it, we should soon find that we were forced to make our long words or phrases fit into impossible combinations of letters, and we should have to begin again. So we think of a familiar phrase, preferably one which looks suitable for a neat clue, the details of which we shall work — out later.

Let’s try LETTER OF THE LAW (a subpoena?). Where shall we put it? It has 14 letters, so it may go into any row, beginning at Column 1 or Column 2. If we begin it in Column 2, the checked letters will be E, T, R, F, H, L, W, of which F, H, and W might prove awkward for fitting words or phrases running down the puzzle. But if begun in Column 1, the same phrase would leave exposed the letters L, T, E, O, T, E, A, all of which are much more accommodating for intersecting words.

Which row shall we use? With such useful letters we could put it in Row 15 for the word-endings of down words, but the O might cause trouble later. Let’s try the middle, where none of the checked letters should cause any difficulty at all: say, Row 7. We block in the empty square in Column 15, and to preserve the symmetry we immediately block in the corresponding space in Row 9, Column 1. This means that the first down word in Column 1 must finish with L or L—: which is not difficult. Our diagram is now started.

Figure 12

We are going to leave the corresponding phrase for Row 9 alone for the moment. Why? Because if we put one in now, we shall have to find our two Down phrases with two letters of each already fixed, and it’s obviously easier to work with only one letter fixed.

So we glance down Columns 3, 5, 7, 9, 11, and 13 of Fig. 12, and. see whether any phrase which we’ve previously jotted down for future use will fit anywhere. What about FISH OUT OF WATER, which cries out for a nice clue about the rescue of a small boy playing near a pond? It will fit in Column 3. But wait: the awkward letter F will then be the second letter of our Row 9 phrase — not promising, if the corresponding phrase in Column 13 is awkward too.

Well, then, Column 9? That puts the F right in the middle of a yet-to-be-found phrase. Too restrictive, this; we’d better postpone the child’s accident until next week’s puzzle, and we’ll make sure that both the F’s are nice harmless unches in that one.

Let’s try again — we mustn’t let the occasional obstacle get us down. Down? DOWN IN THE MOUTH — there must be a good clue here about a careless young shaver! That will go in Column 9, with a block at the end, a corresponding block at the beginning of Column 7, and only that W in Row 3 to keep an eye on. In fact, the safest thing to do is to put a block in front of it (Column 8) and then it will have to begin a word, which is easy. So in goes the block at Row 3, Column 8, and its partner at Row 13, Column 8.

We must now find a fourteen-letter phrase for Row 9, and another for Column 7 to complete the symmetry. We’d better do Column 7 first. Why? Because we’ve got to remember that our second letter will end the word in Row 3, and our twelfth is the last letter of the word in Row 13; and if we’ve got another letter to worry about as well as the O of oF (Row 7), it’s going to be tricky. Let’s try E as the second letter. After some trial and error with fourteen-letter phrases we hit upon SENT TO COVENTRY, with a useful T to finish the first word in Row 13.

Now, are there any snags? Yes, two. The second word in Row 1 begins D: there aren’t too many words with D as the second letter, and you’ve probably used most of the obvious ones in previous puzzles. Also, the first word in Row 15 ends with Y—, which isn’t at all promising. An inspiration! Let’s put a block in front of the D and balance it with one after the Y, and we’ve killed two nuisances on one pair of chopping-blocks.

Figure 13

We are now left with the problem of finding the final phrase for Row 9, with the letters O-E now fixed in the middle. OPEN? LOVER? OVER? MAN OVERBOARD! (Chairman of directors?) — no, not enough letters: a pity, but we must use that another time. ONE’S? That’s better — there must be plenty of phrases with ONE’S in the middle. SAVES ONE’S BACON seems a nice ambiguous idea for a clue — let’s put it in and take stock of the diagram so far.

Figure 14

It’s now time for us to study the symmetry, because we’ve almost reached the stage at which we can tackle the composition one quarter at a time. We must take care, however, not to arrange the remaining blocks in such a way that each quarter is cut off from the rest of the diagram, since the solver is entitled to some help from cross-checked letters from the moment when he has solved your first clue. It isn’t fair to make him start again from scratch in each quarter.

It’s easier to build up words from first or second letters than from final or penultimate ones, which are more restrictive and impossible to look up in the dictionary, though it is possible to look them up in a rhyming dictionary; you should provide yourself with one of these. Even so, whenever possible we work from the beginnings of words. Unfortunately, this is not always possible, as you will see as we get on with the composition of this diagram.

We look first at the south-east corner, since that is the most awkward part of the diagram to compose. (This, briefly, is because the second across word in Row 15 and the second down word in Column 15 must use their checked letters as the final letters of intersecting words; so we must be careful to end our south-east corner words with versatile letters, such as D, E, N, R, S, T, with possibly C, G or L for such endings as —IC, —ING or —AL.) And experience will show that if our block arrangement works in the south-east corner, it will almost certainly work elsewhere.

A glance at Fig. 14, Row 11, shows us that we can’t put in a block at Column 8, because that would give us a fairly solid column of blocks almost dividing the puzzle into left and right halves.

If we put it in Column 10, we’ll have to find a word ending in E-O, an unhappy prospect.

If a block goes in Column 4 or 12, we shall have a three-letter word in the row, which we try to avoid.

This leaves Column 5 or 6. Column 6 is all right in Row 11, with a word beginning E—O (EVOLUTION? EMOLUMENT?) — but hold on! The corresponding block in Row 5, Column 10, would give a nine-letter word ending in T-I, and we usually limit our Italian composers in crosswords to VERDI!

We have no choice — the block must go in Row 11 at Column 5, with its mate in Row 5 at Column 11.

These two new blocks have simultaneously helped us, by breaking up Columns 5 and 11 into two words each. Are they all right? Column 5 seems fair enough, ending in E-E- (SQUANDERED? Keep it in mind by pencilling it in for the moment).

Column 11 is less easy, beginning with -E-B, but not impossible.

REMBRANDTS? The plural is not very good, and the S might be tiresome as a second letter. MEMBERSHIP, with possible play on the subject of MEMBER’S HIP (politician’s joint?) seems more promising, so in it goes, and we see what trouble it has landed us in.

Figure 15

The second word in Row 13 is the only one with obvious difficulties, so if that doesn’t work we must think again. TEHERAN fits — there surely can’t be any other word here — so put it in.

Now let’s get rid of the second word in Column 15, with one eye trained on Row 11. If the latter were DEMOCRATIC, that would give -N-C-N–, Is there such a word? Yes, ANACONDA! That’s a bit of luck, with only the final A to torment us. What word will fit in Row 15, and still leave a possible word in Column 13? SPARTA, and CATERER. That settles it: our final blocks must go above the C of CATERER and below the T in Column 3 to balance i.

Now our diagram is fully blocked-in, with each quarter deriving help from words or phrases begun elsewhere in the diagram, and the tricky south-east corner is complete.

Figure 16

We now notice that our two end words, ANACONDA and SPARTA, have used only three of the most versatile letters, the R of SPARTA and the two Ns of ANACONDA, and we have had to fall back on our second line of defence with a C, as well as a P and an A, which we hadn’t expected to use.

If it hadn’t come off, we should have had to scrap this corner and begin again, since it usually pays in the long run to be ruthless with yourself at this stage, rather than waste endless time with an impossible combination of letters.

We are now ready to go on with our composition, beginning with the north-west quarter, which already contains a few letters.

Following the principle of disposing of the longer words first, we must find something to fit Row 5.

Let’s look at it. Our tentative SQUANDERED in Column 5 will give ----N-T-I-. No single word springs to mind, so what about a phrase? The N suggests AND. And what? AND TAIL? Yes, TOP AND TAIL is familiar, so now we have Row 5 and Column 5 both filled in, and we’re nearly there.

Look at Column 3. If we put another letter in it from a word crossing it, we’re going to find the ending P-T difficult, so let’s deal with it before that happens. HOT-SPOT? That H is asking for trouble. PITAPAT looks better. Now all we need is a word in Column 1, whose first letter will help to complete Row 1, and whose third letter will give us Row 3. This is a matter of trial and error again. TELL-TALE? No good for Rows 1 and 3. PRETTILY? Still no good. DAINTILY! With that, we can have INTRUDE in Row 3, and DEPOSE in Row 1. And that, ladies and gentlemen, conquers the north-west frontier.

The north-east and south-west corners can now be tackled, in either order. In the north-east corner the only word to be wary about is the one in Column 13, ending with A. This is by no means as difficult an ending as I, O or U, though naturally far less common than E; but it will become harder if other letters are allowed to appear in the word haphazardly. So let’s think of something to put in and see whether Rows 1 and 3 are prepared to co-operate.

Lots of place-names will fit, no doubt, but since we ve already been sent to Coventry and driven to visit Teheran and Sparta we’ve done enough travelling for one puzzle.

Figure 17

Any words beginning with A and ending with A? ANTENNA might do — it could lead to a clue using ant, the composer’s favourite worker, and the reverse of ANNE, his favourite girl: and it doesn’t make the four-letter word in Row 5 impossible, so let’s have it, unless it proves a deceiver later.

Now Row 1: let’s work through the alphabet. D-A-A–? Nothing obvious there. D-B-A–? Same again. D-C-A-:? DICTATE, with a nice E at the end — put it in.

Row 3. Go through the alphabet again if you like, but WHISTLE leaps to the mind, with another useful E to finish.

Column 11. CHIP? CLIP? CHIN? An embarrassment of riches! Please yourself which you choose — it will be the one that you can clue best or the one that you haven’t used lately.

Column 15. EVENTS? ELECTS? EXERTS? let’s try to avoid the plural and third person singular if we can. ExEMPT — that’s more like it.

Finally, Row 5. SNAP or SNIP — put them both in, and make up your mind which of them to use tomorrow.

Three up and one to go!

Figure 18

So our ANTENNA proved a useful feeler after all! If it hadn’t worked, we should have tried another word. By now we have got used to finding that our first thoughts don’t always click.

We are left with one corner — the south-west — to fill in, and a glance at Fig. 16 will show that three letters only are almost irrevocably fixed, exactly as in the north-east corner which we have just done.

There are lots of words which will fit into these spaces, and it might be fun at this final stage, if we all had a go at this corner independently. You know the rules. We are going to try to avoid proper names, plurals and the third person singular of verbs too, if we can; and we are going to use familiar words.

When you are thinking of words to fit Rows 13 and 15, keep one eye riveted on the four-letter word in Column 5. Ready? Off we go!

My own version will be found in the completed puzzle (Figure 19), but don’t look at it until you’ve finished yours.

There remains one more important task before we can put the diagram aside: to fill in the clue numbers. Beginners might be grateful for precise instructions here, and the more experienced reader will perhaps excuse what may seem a statement of the obvious.

Begin at the top left-hand square of the diagram and work to the right of Row 1, putting in successive numbers at every square which begins a word or phrase, whether it goes across or down. Continue throughout the diagram, one row at a time, being careful not to miss any down words. You will find that puzzles of this type are nearly always numbered up to 25, 26 or 27.

Now look at Fig. 19 and compare its south-west corner with your own. The chances are that most of you have got completely different words from mine, which shows what variety you can get in crosswords.

The diagram is finished, but no one can yet say whether it will be a good or a bad puzzle. Given a diagram with reasonable words in it, what will make or mar the puzzle will be the standard of fairness and the entertainment value of your clues.

Figure 19

---

Composing an Everyman: The Clues

An Everyman puzzle differs from a Ximenes chiefly in its standard of difficulty. As has been shown, its diagram is constructed from familiar words and phrases, whereas solving a Ximenes often requires a dictionary’s help; but exactly the same principles of clueing are common to both types. Furthermore, although we are attempting to compose a fairly easy rather than a stiff puzzle, we must try to make sure that solvers can trust us, in small details as well as in large.

If we now work our way through the puzzle that we completed in the previous chapter, and examine the make-up of each word and phrase, we shall perhaps be able to show a fair number of different ways of tackling them. We can’t hope to cover every possibility in one puzzle of twenty-eight clues, but this set of clues ought to give a general picture of what we consider to be sound, accurate and, above all, fair clueing practice.

We could start at 1 Across and continue in a methodical way right through the puzzle; but I prefer to dispose of the four long words or phrases first. Why? Because often phrases call for a slightly different clueing technique from that required for shorter words. There may be something in a phrase which will enable us to give it an unexpected twist, without using a reference to its letters as such. For instance, if we were trying to clue DILATED PUPILS, we could picture Billy Bunter and Co. after a large tea-party, as well as the real meaning; or with the phrase GET INTO HOT WATER, the idea of somebody having a bath leaps to the mind. Let’s see if our four phrases can produce some similar situations.

When we put 14 Across (LETTER OF THE LAW) in our diagram, we thought of a subpoena. Let’s work on it. It’s something that you stick to if you’re scrupulous, but in a fanciful way it could be thought of as a letter which the law, or lawyers generally, might send to people, such as a summons or a subpoena. It isn’t really, so we mustn’t say it is. But we can suggest it as a possible way of reading the phrase by the use of a question-mark, thus: “Subpoena? Scrupulous people stick to it [6, 2, 3, 3]”.

The solver is thus given two shots at the same target from two different angles, in accordance with good clueing principles. “Is it a subpoena that we’re trying to find?’ is one hint at the answer, and “Scrupulous people stick to it” is a straightforward statement. Whichever part of the clue strikes the solver first, the other part should settle the matter. If it does, he knows he’s right. If it doesn’t, either he or the clue hasn’t done the job properly.

16 Across: SAVES ONE’S BACON. Again, there are two ideas to play with here: the real meaning, “escapes loss”, and also some picture of a man hoarding his supply of bacon. If he makes do with eggs for a few consecutive breakfasts, he saves his bacon for later. Here’s an idea, then: “Escapes loss — by breakfasting only on eggs? [5, 4, 5]”.

This should be a real help. The definition, although accurate, may not get the solver very far — the answer might be lots of things. But the subsidiary reference to eggs and breakfast ought to set him thinking about bacon too, and once he starts chewing on that, the answer shouldn’t be long in coming.

8 Down: SENT TO COVENTRY. Who or what is sent there? A social outcast, metaphorically, and orders for cars, literally. Let’s put both ideas together: “Socially excluded, as orders for cars may be [4, 2, 8]”.

A different clue-writer, whose thoughts turned to white horses rather than horse-power, might have tackled this same phrase with Lady Godiva in mind and produced something like this: “Despatched to report on Godiva’s ride? Socially excluded [4, 2, 8]”.

4 Down: DOWN IN THE MOUTH — and our clumsy young shaver! The phrase means “low-spirited’”, but we must do something about the other meaning of down, that is, soft hair.

Whenever a word has two different meanings, the obvious thing is to get both in the clue somehow. Here’s a suggestion: “What clumsy young shaver’s got? Low-spirited [4, 2, 3, 5]. | Does that cover everything? “Shaver” suggests a man shaving; clumsy” gives a hint that he’s making a messy job of it; and “young” makes it clear that he’s removing down rather than bristle — yes, it’s all there, and we can’t in fairness omit any of these pointers. Note the question-mark again — we haven’t said he gets down in his mouth, but rather asked if that’s a picture which the phrase could bring to mind.

We’ve managed, then, to clue our four long phrases by playing with the ideas of their words, not with their actual letters, and that’s what we set out to do when we tackled them out of their order. Now we can go ahead in sequence, without jumping about all over the place. Let’s write down the clue-numbers in their correct order on our sheet of paper — fourteen across, fourteen down — put in the four we’ve just done, and check each new one against the diagram as we deal with it.

Just a word or two before we get going in real earnest with the rest of the puzzle. We are hoping to write clues which appear to make some sense, but also by various means to send the solver on a temporary wild-goose chase before he returns with the answer. We want to puzzle, but not indefinitely, so here goes.

1 Across: DEPOSE. It means to oust, or remove from a high position. Let’s examine the word’s make-up. DE (=of, French) and POSE (=attitude) — no obvious connexion with the word’s meaning here. We could, quite accurately, make the clue “Remove from high position of French attitude”. We could, but of course we shan’t, because the whole thing reads like nonsense. What about SOP, written backwards, in DEE? Hopeless. Does it make any interesting anagrams? SEED OP, E. DOPES (in Chinese opium dens — but what’s that got to do with “to oust”?), so DEEP, SPEEDO (the wonder petrol?), O SPEED, or zero SPEED — wait a minute! Zero speed when driving out of a high position — this is more promising, except for “high position’. “Elevated station” — that’s better! A little more thought, to mislead the solver into visualizing car-driving instead of despot-driving, and then: “To drive out of elevated station, change from zero speed [6]”.

Have we kept to our principles? We’ve got a definition in one part, and a subsidiary clue in the other part in the form of an anagram, suggested by the words “change from”; we’ve created a picture of a car leaving a station; and we’ve told the solver to jumble the letters of O SPEED.

4 Across: DICTATE. This sounds like Dick Tate — is there, or was there, a famous one? The England bowler is the only one who comes readily to mind, but he wasn’t Dick, so let’s forget it.

There’s no obvious way of separating the parts of this word, and we don’t want to use an anagram again so soon, so let’s try hiding it in some other words, as we tend to do for one clue in a puzzle. To dictate is to command, and the word EDICT is related to it and might help us to make a sensible-sounding clue: “Feature of an edict — a temporary command [7]”.

‘Feature of” suggests that the answer is a noticeable part of what follows, which is “an eDICT — A TEmporary”, and the definition “command” completes the clue. It will be seen that the definition part may come at either the beginning or the end of a clue.

9 Across: INTRUDE. We have almost two words side by side here: “INT and RUDE, so let’s see if we can devise some connexion. A hint is a tip, and rude means vulgar. Now how can we suggest that HINT has dropped its aitch? Obviously by finding a synonym, or synonymous phrase, that also has an aitch that we can drop. What about “a helpful tip’? If we drop the aitch, it will become “a ’elpful” — not so good. “A tip that’s ’elpful” is better. We now want a definition of the complete answer; “gatecrash” sounds all right. So the full clue will be: “Gate-crash — a tip that’s ’elpful and vulgar [7]”.

10 Across: WHISTLE. What are we going to be in this clue — referees or wolves? It all depends on what the word’s letters suggest. It is rather an awkward combination of letters, but we can see that there is the familiar abbreviation st. (for street) inside the word WHILE, or conversely, WHILE about st. One does not normally expect to see referees whistling around a street, but that’s precisely where the wolves congregate to show their appreciation of transient beauty! It will be an easy clue to solve, but it seems too pleasant a thought to discard, so let’s have: “Act like a wolf, while standing about the street [7]”.

The definition is “act like a wolf”, which colloquially suggests to “whistle”; and the word WHILE is standing, or positioned, about the abbreviation of street — so all the solver has to do is put one part round the other.

11 Across: TOP AND TAIL. Another phrase, and we try to think of two different ways of looking at it. The trouble here is that we can’t get away from the same idea of top and bottom in the usual senses, so we’d better abandon that tack and try something else. If it weren’t for that T in TAIL, we could have caught a PANDA in the TOIL — a pity. It’s no use, it’s not being a bit helpful, so let’s think about another anagram.

It’s always nice to work out an anagram which also has some connexion with the answer whenever this is possible — which isn’t all that often; but let’s see. To top and tail is to cut both ends off, especially in the vegetable kingdom. Vegetable? We suddenly notice that the letters of A PLANT are all in TOP AND TAIL — let’s remove them and see what’s left: T-O-D-I. This can become I DOT — no sense there — or DO IT, making A PLANT DO IT. We must indicate the anagram somehow, and after various attempts we finish with: “To cut both ends off a plant, do it haphazardly [3, 3, 4]”.

Note that the solver sometimes, as here, has to supply his own punctuation, and is required to pause after “to cut both ends — off”. He is then given “A PLANT, DO IT haphazardly’, which tells him to work out an anagram.

Before we leave this clue, we must carefully check each letter of the anagram against the answer in case we’ve made a mistake (it’s easily done if one tries it on sight only), and also put a pencilled mark at the side of the clue (to be erased later) to remind us that we’ve now used up two of the four full anagrams, the maximum that we usually allow ourselves.

12 Across: SNAP or SNIP. Reverse of PANS or reverse of PINS — which one is likely to clue better? There’s not much in it, but it occurs to us that we have got the word SNIPPET at 23 Across, so let’s not repeat the snip idea but play instead at SNAP. “Sudden cold spell — vessels returning [4]”.

A snap is a sudden cold spell — there’s the definition; and the solver is further helped by being told that a word meaning “vessels” is returning, or is written backwards. Since this is an across clue, we must be careful to say that PANS should be written backwards, whereas if it had been a down clue, we’d have had to say something like “vessels held up”, to show that the letters go upwards.

20 Across: ONUS. “Responsibility — at our expense [4]”. Two different ways of looking at the word have been used here. In addition to the definition (responsibility), we can think of it as the two separate words on and vs. If you’re lucky enough to hear some people offering you some refreshment with the words “It’s on us’, they mean, of course, ‘It’s at our expense’. So the words ON US, taken out of their full sentence, mean “at our expense”. )

21 Across: DEMOCRATIC. There are no obvious words side by side here, or any fitting of one word inside another, so this may be the time to use up the third of our anagrams. The letters can be rearranged to make I MET ACCORD, and the answer means “having popular government”. Put them together, then, as follows: “Having popular government, I met accord, somehow [10]”.

The clue’s final word suggests that the letters before it have t be arranged somehow, in other words, in a different order.

23 Across: SNIPPET. The most simple approach here is to treat the word as an order to snip one’s pet, or to give one’s dog a trim. How can we connect that idea with the word’s meaning? We must also be careful not to say definitely that only a dog is involved, because a pet might be any of a great number of creatures, so let’s suggest it by the use of a question-mark, thus: “Give doggie a trim? ‘That’s a bit of news [7]”.

The last-minute thought of changing “dog” to “doggie” ought to help the solver to think of a favourite domestic animal rather than other kinds, like greyhounds or hunting-dogs, which one is less likely to think of as “doggies”.

24 Across: TEHERAN. It is most important here, where we have the name of a place which is capable of different spellings, all equally acceptable, to take special care to give the solver a clue which will lead unmistakably to the spelling in our diagram and to no other. It makes HERA in TEN, or in a reversed NET, but that won’t be much help to solvers who aren’t familiar with the Greek goddess. There’s also HER+A in TEN - is that any good? HER is quite familiar as “the lady”; a is “one”; and we’d better leave TEN unchanged, or it will begin to look too hard. “The lady has one in ten” doesn’t make much sense, but “The lady attracts (i.e. draws towards her) one in ten” does. We mustn’t forget to indicate quite clearly where this place is, which is Persia. “The lady attracts one in ten in Persia [7]”.

It’s not an easy clue, but on the other hand most solvers, if asked to name a city in Persia, would probably think first of TEHERAN; and a quick check on the subsidiary clue should bring TE-HER-A-N and full satisfaction.

25 Across: LEGALLY. LEG ( a limb, or member of the body) and ALLY (partner) seem the best bet, giving something like “A member should have a partner, according to law”. But we notice that the answer to 13 Down is MEMBERSHIP, and we’re likely to need the “limb” idea there, so let’s think a bit more.

Something with a partner, but not LEG. What else is there? There’s L+8.G.:1, for instance, with a partner. L is a common abbreviation for “pound”, “lake”, or — yes, this might do — ‘Liberal’. Here we are: ‘Liberal, for instance, must have partner, according to law [7]”.

26 Across: SPARTA. We’re going to talk about a famous place, so let’s see if this time we can make a clue that fits the idea of a place.completely. It’s made up of PART (a region) inside the letters S.A., which can stand for South America. Sparta was well-known as the breeding-ground of hardy people, so it might make a nice misleading picture, as follows: ‘Region in S. America, whence hardy folk came [6]”.

Sparta isn’t, of course, a region in S. America, though it is a place whence hardy folk came. We’ve said something which isn’t true as a whole, but we’ve said precisely what we mean in the clue, namely that PART inside S.A. is where hardy folk came from.

Now it’s the turn of the Down clues, which we’ll tackle in numerical order as we did the Acrosses.

1 Down: DAINTILY. Here we have a word meaning “delicately”, consisting of DAILY (a charwoman, or char, familiarly) around Int. We’ve already used INT in the clue to 9 Across, so let’s think of some other approach to these three letters. They can be shuffled to make TIN. Now what would a char do with a tin? Probably throw it into the dust-bin when tidying up. That suggests “messy tin”, the word “messy” indicating to the solver that the letters of TIN are in a mess, that is, jumbled. Finally, how can we show that ‘messy tin” is within the word for a char? Better, perhaps, the other way round, with the char gripping, or holding, the tin. The picture is now complete: “Char holds messy tin delicately [8]”. 2

Only part of this word has been treated as an anagram, but we still don’t want to overdo this kind of clue, so we make a note at the side of the clue that a part-anagram has been used, to remind us to try not to do this more than once in a puzzle which looks like having four complete anagrams as well — but at a pinch, not more than twice.

2 Down: PITAPAT. We’ve got I TAP in PAT, A PAT upwards in PIT, neither of which is exactly inspiring; or there’s the straightforward PIT+A+PAT — let’s work on that. A pit is a mine. A Pat, colloquially, is an Irishman. ‘Mine an Irishman”, however, suggests blowing the poor chap up — that won’t do! How can we say “mine” is followed by “an Irishman” so as to make some sense? “Mine’s an Irishman” is nice and misleading, because the solver can’t help reading it at first sight as “mine is…’ whereas really it means “mine has … .”, because PIT can be said to have (next to it) A PAT.

There’s still the definition to work in. It means “palpitatingly”, but that won’t fit the general sense. The adverbial phrase “with palpitations’ is better, as follows: “Mine’s an Irishman, with palpitations [7]”.

That’s rather an unenviable boast for a girl to make about her current boy-friend, but it’s better than being on the shelf!

3 Down: SQUANDERED. This is an awkward-looking word. It makes SQUARED around an anagram of END, but let’s try to manage as long as possible without any more part-anagrams. In any case, that treatment doesn’t suggest any mental picture.

The word means “spent lavishly’, and the latter part of it is the latter part of PLUNDERED, which has some vague connexion — can we do something with that? squa is part of SQUALIDLY, and the words “squalidly plundered” make sense.

We must now indicate to the solver that the interior of these two words is missing — what about “gutless’”? That’s it: “Squalidly plundered — gutless — spent lavishly [10]”.

This clue is really the opposite idea to that used at 4 Across, where DICTATE was hidden in its correct order of letters inside the wording. Here the actual letters of the answer are given, also without any change of order and without synonyms, but they are to be found at the extremes of two consecutive words from which an internal part, or “guts”, must be removed.

5 Down: CHIN, CHIP or CLIP. Which one shall we clue? CHIN — a double one (CHIN-CHIN!) will lead to cheers! That’s a thought, but what about its definition? Leave it for a moment.

CHIP — on one’s shoulder, or in a greasy newspaper? Nothing obvious there at first sight. CLIP — on the ear? C + LIP? Hang on — we might have a bit of fun here. This clue, numbered 5 Down, comes straight after 4 Down (DOWN IN THE MOUTH), with its careless young shaver. Every now and again we find, accidentally, a superficial connexion between two successive answers which is too tempting to miss, and this may be a good opportunity. If a careless young shaver gets down in his mouth, a cheeky young shaver may get a clip on the ear, or round the cheek. Now we’ve just spotted that CLIP is made up of c. (a common abbreviation in history books for “circa”, meaning “about”) and LIP, which can mean impudence, or cheek. Let’s try it now: “What cheeky young shaver may get — about cheek [4]”.

Not every solver will appreciate this kind of lark, or even approve of it, but it’s irresistible! If the clue is read right through, it should lead to the straightforward answer, “He’ll get a CLIP about his cheek”. Then if it’s read over again, this time in two parts, we have cx1P for the answer to “What cheeky young shaver may get’, and we also have a subsidiary clue, “about cheek”; which should lead to c. — LIP.

6 Down: ANTENNA. We have here our old friend the ant, as well as the reverse of ANNE. Both parts are easy for regular solvers, so how can we make it a little more teasing? ANTENNA is a TV aerial, also easy, but if we make “aerial” an adjective instead of a noun it may hold up the solution for a moment or two and make the solver’s brain-cells earn their keep. We must also remember that this is a down clue, so the reverse of ANNE must use some word like “up”, not “back”. The following should make good sense: “Aerial worker takes girl up [7]”.

Once again the solver is required to pause after “aerial”, and then to make ANT take the reverse of ANNE.

7 Down: EXEMPT. The immediate idea that strikes one is that there’s EX, which means “former”, “late’, or “lately”, and EMPT, which is nearly the word EMPTY, meaning “vacant”; and the whole word can mean “released”. “Released, formerly vacant, almost” is very weak. What about “unoccupied” instead of “vacant”? We can paint a sad picture of a redundant worker getting the sack, thus: “Released, being lately almost unoccupied [6]”.

13 Down: MEMBERSHIP. The people who make up a club form its membership, and the word consists of MEMBERS (of a club, or Parliament, or the body) and H1P (a joint). It’s no use playing with the same idea of “members’” in the clue as in the answer, so let’s consider the members of the body, such as legs or arms. “Arms, joint” is an idea, especially with allied forces in mind, and the word “club” suggests the Nuclear Club, which is on the same wavelength. But just a moment — we can’t clue MEMBERS definitely as “arms”, because they might be other limbs besides. This is where the useful word “perhaps” comes in, as follows: “Arms, perhaps, joint, for all who are in the Club [10]”.

If we check this, we see that “arms, perhaps’ gives MEMBERS, ‘joint’ gives HIP, and the full answer means “all who are in — that is, belong to — the Club”.

15 Down: ANACONDA. ANA+CON+DA — no, these aren’t familiar words, though they’re all in the dictionary, so let’s try something else. It’s the reverse of AD-NO-CAN-A— “Notice no vessel, one turned up’? — what nonsense! We’ve still got one complete anagram up our sleeves — perhaps this is the time to use it. It makes A CAD ANON. Who will become a cad anon?

Someone who isn’t one now; but where’s the connexion with the real answer? An anaconda is a constrictor; it grips, it squeezes, it embraces — ah! Someone who’s doing some embracing may misbehave — given the chance! Let’s give him his chance here and put the words of the clue in his own mouth: “I’ve a strong embrace, and may become a cad anon [8]”.

Notice that the required word is addressing the solver direct, saying, “I, the answer, have a strong embrace, and may become (if my letters are rearranged) A CAD ANON”. Make a note too that we’ve used up our self-imposed ration of four full anagrams.

17 Down: AMUSING. “Capable of causing laughter’ — in a word, “laughable” or “funny” (since we must try to be economical of space whenever we can). The word itself has quite a number of reasonable possibilities this time. There’s AM+USING, or the reverse of MA+ USING, AM “U”+SING, US in A MING, A MUG about SIN, or A+MUSING. How many of them can be regarded as “funny”? Let’s take them in turn: ‘Am employing what’s funny”, (am-using); “Mother turns up, employing what’s funny” (rev. of ma+using); “Am top-drawer, having to warble — it’s laughable’, (am “U”-sing); “It’s funny, placing us in a Chinese dynasty”? (a M-us-ing); “A duffer about sin — it’s laughable” (a mu-sin-g); “The article needs pondering — it’s funny” (a-musing).

It’s now purely a matter of personal preference as to which one we use. However, since we must make a choice, the one which seems to present the neatest and simplest picture is: “A duffer about sin — it’s laughable [7]”.

18 Down: CATERER. What connexion can we make between “a person who provides food or entertainment” and the letters of the answer? CAT and ER-ER doesn’t make any sense. Does CARER about TE? “One who cares about an unfinished tea’ — that’s better, but let’s wait before deciding on it. There’s car about TERE, which suggests a CAR wrapped around a broken TREE — can that be improved? We can perhaps provide a rather gruesome entertainment by making a car crash and wrap itself round a tree, and after some thought about the wording, produce: “Car around tree, having crashed, can provide entertainment [7]”.

Thus the solver is told that CAR, around TREE that has crashed, or changed its shape, may become CA-TERE-R, who is someone who can provide entertainment. And we’ve had to employ a second part-anagram — this must be our last one in this puzzle.

19 Down: DORSAL. A dorsal fin is one which is found on, or at, the back of a fish. Now what can we put at the back? Reverse of ROD + SAL? What about “Staff turns up with Sarah, at the back”? Very dull. There’s also D (a penny) +0R+SAL. Luckily Penny is a girl’s name (which goes sensibly with sa), and if we put it as the first word in the clue, the capital letter will read naturally. This gives us: “Penny or Sarah should go in, at the back [6]”.

22 Down: OPAL. Even with so short a word, several ideas are suggested by the letters. There’s OP. (short for a musical work) and AL (nearly ALL, or a familiar name for Albert, etc.). Connect this with a gem found in a necklace or ring, and you could have “Make short work with Albert in the ring’. The weakness there is that one makes short work of, not with, so let’s abandon it.

The reverse of LAP + O doesn’t look very promising; nor does PA in the reverse of LO. What about the straightforward 0 + PAL? “Love a friend, one in the ring?” That’s not bad, but it’s not an easily imagined situation. The same O—PAL could represent a person addressing his friend with the words “O pal!”, and it’s quite easy to picture a boxer’s friend shouting advice to him from outside the ring. Here we are, then: “Address to a friend, one that’s in the ring? [4]”.

A final reminder that, since an opal might be found in a ring, a brooch, a bracelet, etc., we can’t say definitely that it is in the ring, but we can suggest the possibility — hence the question-mark.

Now that the clues are complete, we hope that we have managed to provide a puzzle that is both interesting and challenging to the average solver, and that we have introduced a reasonable variety of types of clue. The attempt to cover as many different aspects of “cluemanship’” as possible may have given some of the clues an element of stiffness which a typical Everyman would avoid. But since in this case all the answers have been given and the clues explained, perhaps not too many solvers will complain about the difficulty!

There can, happily, be no final word on clueing methods, because one can never fully forecast what new trains of thought the sight or sound of a given word or phrase will set in motion.

Nor can one anticipate the extent to which the English language can be manipulated with ingenuity and freshness. Nevertheless it is hoped that enough diversity has been provided here to point the way, and to give a fairly revealing insight into the workings of what one clue-writer likes to think of as his mind.

---



Composing a Ximenes, the bars

If you are like many people I have talked to, you probably think that the task now set you is beyond not only you but anyone except a slightly deranged specialist. I hope to show you that it isn’t. It is harder than the Everyman task in some ways: for example, in settling the diagram you will need more discretion, and you will have to think of more things at the same time. If you play golf, you know how hard it is to concentrate on looking at the ball, slow back, not swaying, not pressing, bracing your left side, following through, and so on, all at once. You may play best when you only think about one of them. That won’t work in composing this kind of crossword. What you have to do is much more like what you had to do at school, if you ever suffered from having to write Latin verse. Such sufferers have to think of scansion, grammar, syntax, word-order, intelligibility, and the meaning of the English, all at once. And if, like me, you are the sort of lunatic who enjoys trying to do that, you will certainly enjoy doing this. Is the converse also true? I hope not, because in other ways this task is easier than the Everyman task: for example, you never have to say to yourself “That word won’t do: someone may not have heard of it.” Within the limits of Chambers’s Dictionary, no holds are barred. And talking of bars, you will find that bars are, on the whole, more flexible than blocks.

Provide yourself first with a diagram, entirely blank and 12 squares by 12. Number the rows and columns as Everyman told you to, outside the diagram.

We will suppose that you have decided to include the Greek philosophers ANAXIMENES (for obvious reasons) and ANAXIMANDER (to go with him), SDRUCCIOLA (solvers like extraordinary-looking words) and TENACES (six of which will go up a card-sharper’s sleeve in their clue). Their positions in the diagram must be tentatively settled first.

The three long words are all unsuitable for the rows and columns at the extreme edges. They have too many vowels to be helpful in Row 1 or Col. 1 and too many difficult ending letters for Row 12 or Col. 12, e.g. I, U, X (and even A and O are none too common).

ANAXIMENES and ANAXIMANDER can cross each other almost anywhere, so put ANAXIMENES in Row 2, with two unches after him, and ANAXIMANDER in Col. 2, with one unch under him. Why? Remember the principle that it is easier to work away from beginnings than up to endings.

That will also fix SDRUCCIOLA: since the diagram must be symmetrical, put it in Row 11, with two unches before it. (An unch here is made with a bar instead of with a block. This economizes space — there will be 144 letters in your 12 by 12 diagram, nearly as many as in an average 15 by 15 diagram with blocks.)

Don’t at this point try to think of another nice eleven-letter word for Col. 11. If you do, and put it in, it will cramp your style by making you work backwards from the east side. SDRUCCIOLA is already going to do this at the bottom, but you can at least avoid taking this risk twice.

Now for TENACES. Try Row 1 and Col. 1 first. In Row 1 the first A of ANAXIMANDER drives it to the right: when it gets there, it produces TM, EE, NN, AE and CS as beginnings of words. Of these only NN is impossible: TMESIS and CSARDAS, for instance, do exist; but things will certainly become very difficult because your choice will be so limited.

What about Col. 1? This time TENACES must sink to the bottom because of the first A of ANAXIMENES, and it will give you TM, EA, NN, AD and CE — better, but still not promising.

Row 12% It.can go on either side. The left gives SN, DA, RC, UE and CS as endings — hopeless. The right gives UT, CE, CN, IA, OC, LE and AS — much better: only CN is impossible and only OC is difficult. Put it in tentatively, with a bar between C and N, a corresponding bar above the I of ANAXIMENES, a bar to the left of TENACES and a corresponding bar as before, after the seventh square of Row 1.

You have now reached Fig. 21 (I hope). Have a break — have some elevenses.

The next thing is to complete the diagram with bars, remembering, as you do it, all the instructions you were given by Everyman and some others which didn’t apply there.

1. Number of unches in a word. Every word should have at least one unch, so that it won’t solve itself without being able to say to the solver “You must look at my clue before you finish me”.

Sometimes it will solve itself even with unches, when only one

word will fit, e.g. P-T-TO. No one can help that, but you can prevent BEAN from solving itself by making it -EAN, B-AN, BE-N, or BEA-. On the other hand, anything but a low proportion of unches is unfair in a crossword containing unfamiliar words, as this one will. A good proportion is:

• 4 and 5 letters, 1 unch.
• 6 and 7 letters, 1 or 2 unches.
• 8 and 9 letters, 2 or 3 unches.
• 10, 11 and 12 letters, 3 or 4 unches.

There will then be between 40 and 70 unches in all: 52 or 54 is ideal. Three-letter words are rarely used: when they are, it is perhaps fairest to leave them fully checked and let them solve themselves if they like.

2. Wide open spaces. These are very dangerous. You want to be able to choose words, not to be reduced more often than is necessary to using the only possible word. You will give yourself most freedom by distributing bars fairly evenly over the diagram, and by not letting two or more longish words run side by side over much of their courses.

3. Number of words in a 12 by 12 diagram with bars. The normal number is thirty-six words. This means that six rows and six columns will each contain only one word: the other twelve will each contain two words. Four of the twelve one-word rows and columns will contain your four long words: the other eight will be reduced in length by bars to fit shorter words.

Now we can proceed to the bars: they are now, perhaps, open, so pause for a pint.

Begin by unchecking difficult-looking letters in ANAXIMENES. You need three unches in Row 2, so that he will have the normal ration for a ten-letter word. Where shall they go?

Choose the three letters of ANAXIMENES which are likely to be most tiresome as second letters of down words. These will be consonants: X, M and S are the worst, N being common as a second letter. But you can’t uncheck any of them: if you do, you will have two consecutive unches, and this is not allowed. So you must uncheck three helpful letters and leave open all the unhelpful ones, which is absurd. So the bar above I in Col. 5 must go, and TENACES must be moved, since you must preserve symmetry, and the CN ending at the bottom of Col. 8 is impossible. Expunge TENACES: where shall it go? It may as well use a letter that is already in position, so try one of the Es of ANAXIMENES, the first one for choice. You want to keep away from the eastern edge for as long as possible. Put it into Col. 7, a bar below it, and a corresponding bar below the fifth letter of Col. 6.

Now back to the top two rows. Put bars above and below X and M: this is possible now that you have got rid of the bar above I.

There is already a bar under the third square below M; so join up by putting in two more horizontal bars — very good exercise.

Before you can fix the third unch, you must settle where the break between words is to come in Row 1. There are only three possibilities — after four squares, after six squares and after eight squares. (If you put the break after five or seven squares, either the five-letter word will have two unches or the seven-letter word will have three: try, and see). With the break after six squares, the first six-letter word will have its fourth and sixth letters unchecked, and therefore the corresponding word in Row 12 will have its first and third letters unchecked: this is not actually forbidden, but it is rather unkind to the solver, as you probably know from your own experience. With the break after eight squares, the eight-letter word will run its whole course side by side with ANAXIMENES — uncomfortable, since you will have to find a word of which five letters must fit those of another word. So put a bar after four letters of Row 1, and its fellow after eight letters of Row 12. There are now no apparent snags.

You are left with a choice between the second N and the second EK for the third unch. There is not much to choose, though E is a vowel, because of the commonness of N as a second letter: you will soon learn that N is a friend everywhere except at the beginning of a word, where it is the most hostile common letter.

So spread your bars as widely as possible and choose the second E: put four bars experimentally at the top of Col. 9. I needn’t tell you what to do next. Then you may as well make the bars at the top of Col. 4 up to four, and again I needn’t tell you …

Finally, don’t forget the bottom of Col. 7.

You have now reached Fig. 22; check to see that you really have.

Congratulations on a good morning’s work. Let’s have lunch.

It’s half past two: wake up now and get on. Count the horizontal bars. There are 28. Now we said on p. 107 that the ideal total number of unches was 52 or 54, and you will need at least one each in Rows 6 and 7. That will mean 30 unches in the acrosses: you will therefore try to have not more than 24 in the downs. How many are there already? Only 4: so decide now on some more, starting on ANAXIMANDER, who should have three.

Uncheck the awkward-looking X? If so, you will either introduce another pair of ten-letter words — difficult — or a nine-letter word with its first and third letters unchecked — to be avoided if possible. Why not two words in that row? Because there are three unches in it already, and the proportions will be bad. Then try M and N, the only consonants left except D, which is to be avoided, you know now why.

Now continue eastwards. Col. 3 will need at least two unches, so you must put another bar each in Row 6 and Row 8. The four-letter word in Col. 4 will need one unch: choose the second letter rather than the fourth, because it is kinder to the solver to leave both ends open whenever you conveniently can. Continue with two more vertical bars. When you complete the symmetry, you will see why you can’t go any further.

Now count: twenty-two unches in the downs, which may well lead to an ideal total. We decided above that there must be at least one unch in Row 6: can we fix it yet? Nearly: the only possible places are in Col. 8 and Col. 10. Any preference? Yes: if you put it in Col. 10, no more unches will be possible, and you will be left with a nine-letter word in Row 5 with only one unch.

So put it in Row 8, and you can then uncheck the square above it.

Finally you must divide Rows 3 and 4 and Cols. 1 and 3 into two words each. Starting in Row 3, observe that the eighth square is so far destined to hold the third letter of a fully checked four-letter word: it is always most essential to uncheck one letter rather than the fourth, because it is kinder to the solver to leave both ends open whenever you conveniently can. Continue with two more vertical bars. When you complete the symmetry, you will see why you can’t go any further.

Now count: twenty-two unches in the downs, which may well lead to an ideal total. We decided above that there must be at least one unch in Row 6: can we fix it yet? Nearly: the only possible places are in Col. 8 and Col. 10. Any preference? Yes: if you put it in Col. 10, no more unches will be possible, and you will be left with a nine-letter word in Row 5 with only one unch.

So put it in Row 8, and you can then uncheck the square above it.

Finally you must divide Rows 3 and 4 and Cols. 1 and 3 into two words each. Starting in Row 3, observe that the eighth square is so far destined to hold the third letter of a fully checked four-letter word: it is always most essential to uncheck one letter of a word of that sinister length! Do so, and Row 3 will consist of a seven-letter and a four-letter word, both suitably unchecked, and the one unch belonging to Col. 8.

Row 4 offers several possibilities: as it’s nearly tea-time, divide it, say, into 4 and 8, and change it later if necessary.

Col. 1? The bar can’t go higher than below the sixth square: if it does, you will have a fully checked word at the top. Nor can it go lower: you see why. So divide this column into 6 and 6.

Col. 3? It can be 6 and 6 or 7 and 5: try 7 and 5, and change it later if necessary.

The diagram is now complete; any snags? No, and there are fifty-four unches in all, as ordered by the doctor.

Tea is ready, and no more work, mind, till tomorrow morning! Your brain is in a whirl, and you won’t think straight if you go on; and if you do any work after supper, you won’t go to sleep till about 3 a.m.

---



Composing a Ximenes, the letters

In bed last night you suddenly remembered that when SDRUCCIOLA caught your eye in the dictionary, you noticed that there was another word next to it beginning with SD. So this morning you look it up to see if it would be fun to work it in. You find

there are several — ’SDEATH, SDEIGNE, SDEIN, SDAINE, SDAYN, SDEIGNFULL(Y). And there’s already a second letter D in Row 9. Eight letters: put in SDEIGNES. What about putting SDEIGNFULLY in Col. 11? Naughty, and rash; but put it there as long as you are prepared to scrap it at the least provocation. Does it cause any obvious awkwardness? Perhaps the F in Row 8 is the most dangerous feature: use the rhyming dictionary. It opens, naturally enough, at the -Es: try -FE. VOUCHSAFE and HOUSEWIFE. You notice that VOUCHSAFE produces -HS— in Col. 8: so put in HOUSEWIFE.

Here you ask “Can’t we stop talking about Rows and Columns now, and put in the numbers?” No, because you will keep rubbing one out when you alter a letter. “Not if I write them in ink.” What if you want to move a bar? Besides, you wouldn’t save anything: “Col. 8” is as short as “22 dn.” “I see.”

Col. 6 must come next: --UN-U-. Use Chambers: -ful, -um, -us? Wait: you think there’s a name BLUNDUN. Can you remember who or what it is? No. Then try S-UN-U-, which has a possible air about it. SPUN-OUT; but you would prefer a consonant for the fifth letter, because it starts an across word. Back to B, etc., etc. … till you find ROUND-UP: it would be the fourteenth consonant you tried, with T the only other likely one left! But remember that you got HOUSEWIFE quickly: luck evens out when you’re composing crosswords, just as it does at bridge: there are good runs as well as bad.

Col. 8? Is there a word RESECT? Yes. Row 10? You probably don’t know there’s a word DREVILL, but you will come to it early, as it begins with D: very lucky, and don’t you forget it when you get stuck later on ~ you haven’t got stuck at all yet.

The south-east corner now looks like this:

Dots represent the missing letters of the six words still needed, which won’t be seriously affected by any other words. The south-east corner is apt to be sticky, because you so often have to work from endings and not from beginnings; but this one really doesn’t look formidable.

So leave it till later and move on to the north-west corner, which is usually the easiest.

Here you must first find a nine letter word for Row 5: -I----C–: B looks bad for Col. 1, so start in Chambers at C (unless you can think of a word quickly? No, you can’t: nor can I). No Cs strike you: go on to D. Pause at the comic-looking DIPHYCERCAL, and enter it in the back of Chambers, where you’re going to keep a list for future puzzles. Soon you come to DISCALCED: I might have to ban that, because I think I’ve used it lately — I’ll go and see. . . . No, it’s all right: I must have met it when solving: put it in.

The final D now sends you back to Fig. 24, where D–W now crosses E—G-. You had better finish your “easy corner” before going any further.

Row 6: ENGAGE? No: it gives DG-W, which might be a place in Czechoslovakia, but I doubt it. ERIDGE and Belvedere? Or has that got two Rs? Atlas: lunatic! It’s Erith, of course. Is ERIDGE anything? Only the end of Coleridge. (Actually I discovered later that there is a small place called Eridge Green in Sussex; but DI-W is impossible anyhow). EATAGE? No: it gives you an attack of D.T.s. ENRAGE! Idiotic not to think of it before! Put it in. You now find there is DROW, as well as DRAW and DREW, which are dull. And look! It’s a back word going down: that should be easy to clue. You might want to make it go across instead of down: if so, you merely turn the diagram inside out when you’ve finished it, so that the acrosses become downs and vice versa.

Now “the island valley of AVILION” (pavilion with the roof off) stares you in the face. It hasn’t got two Ls has it? No! Sure? Yes! Better look, all the same. … Right.

SNYE in Row 12? No such word. I thought there was: perhaps it’s in the old edition: look and see. … No: I was thinking of SNY, which is there. ONYX? Then -E-LAX in Col. 12: No. ONYM? BEDLAM in Col. 12. Nearly there, but look at -LUD! ‘Sblud! Wasn’t there a left-hander in the O.T. called Elud? Ehud, you ass! But Elud does sound possible: look at Cruden’s Concordance.

A useless search except that you noted down Keilah for future use! Then try Philips’ Record Atlas, the Everyman Encyclopaedia Atlas, and the gazetteer at the end of the old edition of Webster: still nothing, except Petlad for future use!

In final desperation, try the main body of the old Webster, which can make almost any short pronounceable collection of letters into a word. At last! PLUD, “a puddle, pool. (Obs. except dial.)” in a footnote. You would try very hard to avoid that in a competition puzzle, but we’ll suppose that this one isn’t, and give PLUD a very easy clue,

Well, that will teach me to talk about easy south-east corners! It was my fault, not yours: I ought to have known better. Still, that hold-up took only forty minutes, and I have had many much longer ones.

Back to the north-west corner. On the way, C–H in Col. 4, catches your eye, crossing -A—O in Row 7: you had better settle that. CASTRO? No, Cath isn’t given on p. 1350, though one would have said it was commoner than such names as Berenice and Clothilda — that clot, Hilda? Get on, and don’t waste time! CASSIO. Leave C-SH till you write the clues.

Col. 3. Is there a word PARESIS? Yes.

Row 4. OXER? Yes, unless I’ve used it lately. … No: put it in.

Col. 1 next: -A-OD-. PARODY? No: RAR—N looks hopeless in Row 3. If you can’t think of another quickly, try AXE- instead of OXER. SALADS, or SALADE, spring to mind, and isn’t there a word LARDOON? Yes. Put your very old friend Colonel SAPT in at the top. When did I last use him? . . . July 1960: high time he had another outing.

That completes the north-west corner, and you have reached

Good heavens! It’s after 1 o’clock, and we feel hungry and thirsty. That must be because we got engrossed in letters and forgot all about bars. Well, at least we’ll stop for lunch now.

3 o’clock. I was hoping you would wake up soon. I’ve just been checking the proportion of rare to common words in the diagram so far. I make it fourteen rare, ten common. I always aim at fifty-fifty, or rather eighteen-eighteen, so you mustn’t have many more rare ones.

Now, if you’re really awake, why don’t you carry on composing by yourself, and see if you can race me? We’re sure both to put the same word into Col. 5 to complete -IO-A-: how shall we continue from there? Don’t bother with dictionaries unless you get stuck: let the two or three rare words we want come out of your own stock.

4 o’clock. Finished yet? If not, I’ll try to help. I see you’ve got VIOLAS: I was sure you would. But you’ve got VATICANS in Row 1: surely there’s only one? Try my VATICIDE instead.

You can leave your INNS in Col. 8 (much nicer than INKS). I don’t much like your LEASURES in Row 4: has that got a genuine plural? But I couldn’t find anything else either, except LEASOWES with an impossible W: and perhaps Spenser, in those spacious days, had several leisures on end. Anyhow we’ll risk it, and I doubt if anyone will complain.

In Col. 10 you’ve got ASARA, in Nigeria, and I’ve got ISARI, in Greece. No one has ever heard of either, but they’ve all got atlases — if they want to know which I use most, there was a hint on p. 115 — and you can make the clue easy — I’ll get a SARI, IRAS and I — yes. And that’s the lot — no: wait! You’ve got SOSSLE in Col. 12, and it won’t fit my VATICIDE, and, what’s more, I don’t believe it exists! Sossled — that’s what you are. The old edition? … You’re quite right: there it is, under SOSS.

I’ve got ENISLE there — dull and hackneyed in this sort of puzzle, but I couldn’t find anything else to fit. Incidentally you might have had SISSIE — no, I believe Chambers only gives SISSY and CISSY, but I expect he gives them all under “The More Common English Christian Names” — goodness knows what the less common ones are. Finally CADI or KADI for Row 3.

We’ve nearly finished, but there’s still the south-west corner. Last lap.

Get the rhyming dictionary for Row 12. ENTREPOT, “a bonded warehouse”.

Col. 1. CASTLE. Leave -E-ST, TE-- and G-RE till you write the clues. At last we really have finished: the last lap took a minute.

Let’s make a final check of the proportion of rare to common words. We find that when we choose the last six words (-ADI must be rare) we can easily equalize them.

This, at last, is the time to put the numbers in: do it carefully — it is easy to make slips (as the writer actually did in preparing Fig. 26. When a job has become a routine one, it’s very hard to concentrate).

How long have we taken? Between eight and nine hours. We took as long yesterday over the bars as we have today over the words. But that was because there was so much explaining to do yesterday. Next time, unless you are unlucky, you will be much quicker. (In actual fact, when composing this puzzle and scribbling down his thoughts as he went along — which at least doubles the time taken — the writer took about six hours.)

---



Composing a Ximenes, the clues

Cluemanship (see Chapter V), even more than composition, calls for freshness: when one is tired, ideas just won’t come. So after breakfast, and half an hour in an armchair digesting it, your letters and the news, collect plenty of paper, Chambers, the dictionary of synonyms and the diagram, and prepare to start, let’s say at 9.15. Don’t forget what you learnt in Chapter V.

Apart from the rules of soundness given there, remember especially (1) Brevity — keep as many clues as possible down to fifty letters and spaces, so that they’ll need only one line of print. (2) Not more than six anagrams; so don’t use them up too soon — you may want very badly to use one or two towards the end. (3) Easyish clues, as a rule, for rare words, not-too-easy clues for common words.

Start with 1 across and go straight ahead: don’t mind if at any point you take a long time to think of something pleasing. If you give up and go on to the next one when you’re stuck, you will be left with a maddening accumulation of difficulties at the end, and your total time will probably be longer. So now for it: begin with our old friend the Colonel.

SAPT. The last clue I used for him was “Constable of fiction: brooded about a Prince.” (sa-P.-t, & lit. - Do you remember what “& lit.” means? If not, see Chapter VIII.) New subsidiary idea: S.A.-P.T., or pt. S. Africa? S. America? No connexion. It? Always possible. It jerks? It has little point? It calls for exercise? It takes training? Try the chapter titles. Nothing in The Prisoner of Zenda; but what about “A King Up His Sleeve” in Rupert of Hentzau? “Colonel who had a King up his sleeve: it takes training.” A bit long? Yes, fifty-five letters and spaces: “Colonel with King” saves the necessary six. (That one took longer than we expected.)

VATICIDE. Killer or killing of prophets, plus the idea of killing (which we can call knocking back) drinks (vat). “Who killest the prophets” — wasn’t that Jerusalem? Look it up in Cruden’s Concordance: yes, it was. “Jerusalem’s crime — knocking back a lot of drinks?” isn’t bad: the query is essential, to show that it doesn’t really mean that but is a fanciful idea. Difficult for those who have forgotten their bibles? Well, they shouldn’t have.

ANAXIMENES. This should be easy to clue. “Ana” means personal anecdotes, table-talk, etc, The philosopher believed that everything was made of air: it seems reasonable to call him “airy”. So we arrive at “Airy philosopher had a lot of chat with Cardinal.” Easy for classics and hard for everyone else? Perhaps, but we don’t often offend in this way.

LARDOON. A strip of bacon used for stuffing. Possibilities: ‘ard in loon (which has various meanings, including “tart”), Lar-Doon (or do on). “A strip of bacon: that’s ’arsh in a tart”? Not too bad, but we might do better. What about rasher? But perhaps it isn’t quite the same thing. . . . Nothing seems to work really nicely. . . . “We haven’t had an anagram yet’, you say.

No: let’s try: old roan — or an old — Orlando! Why didn’t we think of that before? Obviously we must have Orlando Furioso; and Bacon, the writer. “Bacon’s version of Orlando Furioso” — the best so far, I think, and it was your suggestion that produced it. Nearly a quarter of an hour on one clue, but perhaps worth it.

Three quarters of an hour for four clues: slow — hope the next ones will come more quickly.

CADI or KADI. Magistrate — beak. Sound of caddy? But the correct pronunciation doesn’t sound quite like caddy, and anyway we must indicate the first letter, which is unchecked: a clue referring to sound won’t do. “I’ll be seen with a bounder before the beak.” To see the point the solver will have to pause after “before”: punctuation is really needed between the indication of cad-I and the definition. This omission is allowed, but the clue would be more attractive without it. What about “This beak is bounder No. 1”? Better, but perhaps too easy: cad = bounder is very obvious. “And probably you’ve used it before”, you say?

Very likely: let’s look. Yes, and not long ago. What else is a ‘synonym of “cad”? Dictionary of synonyms — no help: the wretched book doesn’t give it! Any modern slang words? Heel: that’s better. “Magistrate known as ‘Heel Number One’.”

AXES. The obvious treatment is a and sex reversed. “Symbol of Fascists, a retrograde type of person” — or perhaps better, using the verbal meaning of axe, “Treats as redundant a …”

LEASURES. We must make that questionable plural clear. Lures about sea mixed, or about curtailed modern leisure (ease), which avoids the mixture: “rough sea” is a chestnut which would leap to the solver’s eye too quickly to give much of a kick. Lures. Attractions? Seductions? Why not? “Old-fashioned holidays rest almost entirely in seductions.” But do they? We should be able to do better. “Old-fashioned holidays — curtailed satisfaction in seductions.” About ten letters and spaces too many. “Dated” will help; and try “reduced facility”. We might do better than “reduced”: is there room for ‘restricted’? Yes. Then let it be ‘“Dated holidays — restricted facility in seductions.”

DISCALCED. Bare-footed (as a penance). Dis = Hell? I’ve used that rather often lately. Anagram? A bit soon after LARDOON, and two Ds plus two Cs are tiresome. Disc-laced, a part-anagram, is more promising, with the connexion between bare-footed and laced. Can we connect disc with this? Look it up. Not very hopeful, but to disc is to harrow: can’t we have Harrow boys not lacing their shoes properly and suffering the punishment to fit the crime? Yes — “Harrow: untidily laced. Going barefoot as penance.” (Note that “Harrow” should be first word, to give it a misleading capital letter: it could hardly indicate a verb fairly with a capital anywhere else.)

ENRAGE. Anger leaps to the eye, or a green. Another anagram so soon? Well, it is on the other side of the diagram. “A green that isn’t true: it’s sure to.’” Rather corny. Definitions? Nettle, gall suggest themselves among those in the dictionary of synonyms. What about “A green, knotted gall’? More misleading (it’s a common word, justifying this) and less corny. 10.45; we’ye been rather slow so far. What about taking a bit of exercise, and continuing after lunch and a rest? Let’s play a few holes of golf: there should be time before lunch.

3 p.m. Pleasantly tired in body and refreshed in mind, we return to consider:

CASSIO. You remember that he was involved in a street-brawl? Get your Othello to make sure. We’ll have a head-line clue. C.O. at the ends: ass I in the middle. His name was Michael, and he was a Lieutenant. “Lt. Mike involved in street brawl — ‘Clot ! I got caught by Colonel’.” Too long, but the idea demands it. Also a rather thin indication of who it is. Let’s add another detail, as long as it doesn’t run to three lines of print, Desdemona interceded for him; so we add “ — General’s wife interceded”, Anyone who knows Othello should get that.

HOUSEWIFE. With “sew” in the middle, this screams for an “& lit.” clue. “How to sew if. . .”? “She’s got, we hear, the way to sew. …”. If E? Not easy to finish it, Try again, Hou®-sew-I-Fe (Fe = iron). “I have most of the time to stitch — then I iron.” That’s nice, and perfectly sound.

SDEIGNES. Obsolete word meaning “scorns”, “Looks down old nose” perhaps. (Remember that obsoleteness must be indicated, as here by “old”). “Old scouts” (scout, vb.) is also possible. Subsidiary part? It looks like another anagram — no other possibility. “Digs seen” seems the likeliest — yes. “Looks down old nose — digs seen to be very untidy.”

Quicker since the break: three in twenty minutes.

PLUD. This must be an easy one, as solvers who haven’t access to Webster can’t verify its existence. It means a pool: p. = soft is a hoary chestnut, and lud! can be indicated clearly by gad! “Soft — gad! Here be ye olde worlde poole”. No one can miss that.

TE-- we leave till later.

DREVILL. Obsolete word meaning a foul person, d-rev,-ill: “back” for “rev.” (=reversed) will mislead them legitimately, Dill is used in pickling. So: “The dirty old man is back in a pickle’s embrace.” That might raise a smile when the penny drops.

SDRUCCIOLA. An adjective meaning triple, of rhyme, We had already thought of aloic curds reversed. “Aloic” is a horridly formed word: does it exist? No. Nevertheless this looks the only likely idea. It will have to be a long clue with subtraction of “et”? from “aloetic”. “Aloetic and old rejected coagulum turns …? Now what? “into e.g. … .” followed by a triple rhyme, if possible. “Coagulum” suggests “cheese”: rhymes? “Squeeze” is appropriate, so why not “Brie’s cheese — squeeze”? “But it’s an adjective, and ‘e.g. Brie’s cheese’ suggests a noun”, you sagely remark. Well done! That’s a snag which often leads to unsound clues. Instead of “into e.g.” we’ll put “like”. Then it leads soundly to an adjective.

ENTREPOT. Its most convenient meaning is port: this will go with toper reversed. A nice, easy one to clue — the quickest yet: “After 10, sozzled, drunk returning: have a port.” “But drunk is an adjective, and toper is a noun!” you protest. Drunk is a noun too, you ass! ‘Sorry, teacher!”

ONYM. What does it mean? … . Nasty shock — not in the dictionary! Surely it was in an old edition? . . . Yes, that’s a mercy: technical name of a species. This will need a footnote, to show that it isn’t in Chambers except for older editions. “On my back” is obvious: is it so obvious that it will double-cross them? Probably not: so we’ll put “Species of animal that gets on this chap’s back”, I being this chap.

We may as well finish the acrosses now by going back to 29, TE–. This and 34 down must both be easy words, to keep the balance. So think of a convenient TEA- word (A will fit into 24 down), preferably with two or three meanings. TEAR is the obvious one: “There’s a drop in rent.” Corny? Well, I don’t remember meeting it before, though it has almost certainly been used. ‘Where does ‘in’ come in?” you ask. Oh, in = consisting in is very commonly used, though I admit I don’t admire it much.

That finishes the acrosses — and it’s only just after four o’clock. The second half came more quickly than the first. “Shan’t we start the downs? SALADE looks an easy word to clue”. No! Tea! And not much more work today, anyway, if we want the clues to be any good.

In fact, we did no more: it’s now the next morning, and we resume full of hope. First, we check the number of anagrams in the acrosses: three, and one partial one, so there mustn’t be more than three in the downs. And we haven’t had a “hidden” clue yet: you should remember that we decided in Chapter V that one of these was enough: they become far too easy if the solver is looking out for them all the time. A puzzle with none of them every now and then is wise. Now for the downs.

SALADE. You said yesterday that it looked an easy one: are you sure? S.A.-lade, S.-a-lad-EH., sa-lad-e (Scots sae = so) are all possible. S.A.= it has been overworked, but, thanks to the commonness of pronouns, it may not stick out: the solver won’t look at every “it” in the clues with that idea in mind. Try the meanings of lade. Apart from the obvious ones, it can mean empty, and a salade was a helmet: what’s under someone’s hat (his brain) might be empty, and this is where “it” is so useful. “Old-fashioned headgear: what’s under it is empty.” That didn’t take long: it’s always pleasant when the first idea works,

ANAXIMANDER. We must not suggest any connexion with Anaximenes. That would make it absurdly easy for classics, and the connexion will give an extra kick when it appears, whichever of the two the solver gets first. “’Im and ‘er” looks promising: ’Arry and ’Arriet, perhaps. An ax! We’ve used the verbal sense of axe, but it can be spelt either way and may well be used again. To axe (or ax) is to sack: “One has to sack ‘Arry and ’Arriet” makes good sense: Anaximander’s views must give us the reason for dismissal. Now he followed Thales, who was all in favour of water: Anaximander, on the other hand, believed in a variety of sources for the world. So they could be sacked for preferring other things to water. But we shall have to say “he”, not “they”. Put “and ’Arriet” in brackets, therefore, to give ‘Arry more prominence, and continue “he preferred absolutely anything to water.” Long, but there’s a lot to get in.

PARESIS. A sort of paralysis. There is an old clue of mine available from back in 1954 — “Trim the girl for not working properly.” No one will remember it, but it could be brought up to date a bit by introducing baby for sis, and let’s make trim an adjective and say “Trim baby stops one functioning properly”, a slightly better definition.

VIOLAS. There is a clue to the singular available, but that was used barely a year ago — too recent. It ran thus: “She wore the breeches at the palace — a big fiddle” (ref. Twelfth Night). She isn’t any use this time, because there was only one of her: better try for a completely new idea. “Is oval” is an anagram, and violins, etc., are pear-shaped: there’s our connexion. “They’re pearshaped: otherwise shape is oval” (shape being, of course, an imperative verb). Easy? Yes, but we want some easy ones. (That’s four anagrams, remember.)

TENACES. Ah! We have this clue already in our heads. “Pairs of high cards, six of them being up a sharper’s sleeve?” Long, but it would be hard to shorten, and this is only the third clue that will need two lines. (Much quicker than yesterday so far.)

IN-S. Inns and inks aren’t easy to make interesting: what about the Spanish name Ines, given in that wonderful list of “The More Common English Christian Names” in Chambers. Wasn’t there a famous one? Look up the name in Brewer’s Reader’s Handbook. We find she was the mother of Don Juan, and wanted to make her son a model of all virtues. “Wanted her child to become a model?” is pleasantly misleading (towards a daughter) and at the same time fair. Now what about the subsidiary? After much cogitation, we decide to look up words ending in -iness in the rhyming dictionary, with a view to using a word that contains her. “Dreaminess” can be connected with her dreams about her son’s future: “Had dreams about her in reverie’ (dream-Ines-s) continues to mislead towards a girl. Alter the first part to “Could her child become a model?”, and we have a perfectly fair but difficult clue. It was time for a difficult one, after two very straightforward ones. (That, incidentally, has taken twenty-five minutes, almost as long as the other five put together: that’s how it goes.)

ISARI. This calls for an easy clue, being a little known place in Greece, and it should be easy for us too. Either sari or Iras: Cleopatra was at the time topical, so “Lady’s maid of ancient Queen, I, up in Greece.” That should leap to the eye, and there will really be no need to search gazetteers.

ENISLE. This dull word again, one of the sort that keep pushing themselves in because no others like them are available. Can’t we use an old clue? After looking at several, we decide that the best one is an anagram: “Senile, decrepit: put in a home, perhaps” (it means isolate). Very obvious — not proud of it, but it’ll do.

SDEIGNFULLY. This one took ages, and we cursed ourselves for deliberately putting the word in, One more anagram was possible, but nothing remotely connected with disdain emerged. (Perhaps readers will find a good one.) At long last we treated S as Spenser’s head (it’s a Spenserian word) and wrote “See my poet’s head descend, quite haughtily” (quite = fully), Not very interesting, but at least it’s sound,

Half the downs done — let’s have some elevenses.

After that we got on with a rush: an interval is nearly always helpful.

C-SH. Cash and cosh have been used fairly often: try cush. Two meanings, billiards cushion and thigh-armour. Then obviously “Green elastic lining for fortifying an old behind.” A smile, perhaps, for the solver; it was time for one.

DROW. Scotch mist. Look up “Word”, Message is possible: “Message comes up — Scotch mist descending.” Not quite our original idea, but that often happens.

ROUND-UP. Look up the parts. “All over” and “rising” would do, if we could reverse them and say ‘Rising all over — wholesale arrests.” The reversal is difficult, but it’s syntactically possible: a slope up is a rising slope; and the difficulty is compensated for by an easy definition of a familiar word.

AVILION. §avilion, our original idea, will surely work here. “Have a roofless tent: it never rains there” (ref. Passing of Arthur — “Where falls not hail, or rain, or any snow”),

CASTLE. Two not too obvious meanings: a tower on an elephant’s back and a warship. “Warship carried on an elephant’s back.” (So that’s the origin of the Elephant and Castle — one lives and learns.)

RESECT. Dull word — to cut away part of. Secret anagram will be too easy, probably. Sec = dry in ret = dampen or rot. Dry rot? “Remove part of dry rot outside”? Too jerky: two punctuation marks left out. So reverse the order: “Dry rot outside: remove unsound part.” Only one punctuation mark to be supplied now — that, at least, is better.

BEDLAM. “Place for cuckoo — nest to line” (line = beat)? No, I was wrong — can’t find line = beat. Look up bed for another idea. Bottom. Therefore: “Uproar: that means a bottom to thrash.”

No “hidden” clue yet. We mentioned above that a puzzle without one is a good idea sometimes, but to make this set of clues complete, we’ll try for one in one of the last two. |

REAST. To become rancid; also, to cure with smoke (Scot.) “Cure haddies in fire a st …”? No: can’t finish it. “To cure a steak? Look there — Scots method.” Easy, as “hidden” clues always are, but the definition providing the hiding-place gives a new twist.

G-RE. Gore has been used lately. GARE = miserly (Scot.). Aberdonian? French station — tipping porters? That doesn’t seem to work. Chemin-de-fer is another idea with a money connexion. “Miserly Aberdonian limit at chemin-de-fer?” Note the query, necessary because a station needn’t be a terminus.

Finished, and not even lunch-time: about two and a quarter hours for the downs, in spite of one or two bad hold-ups. Total time for clues about six hours — longer than usual, perhaps, because here and there we took rather special trouble over this one and weren’t content with what seemed merely adequate. The result when the puzzle appeared — No. 774 in November 1963 — was that it proved quite a bit harder than the average. That was because clues that come quickly are usually less tortuous than those that need a lot of cerebration, and we had several of the latter, e.g. Sapt, leasures, Ines, sdeignfully. But this doesn’t apply to lardoon: in that clue the obvious idea was absurdly slow in coming. Another reason for the difficulty was that the preceding puzzle had been rather an easy one, and I therefore didn’t simplify any of the clues when I was reading the proof, as I usually do when preceding puzzles have proved to be hard. In spite of all this, I don’t think this set of clues is far off an average specimen.

I hope that both here and in the chapters on composition I have given a coherent and intelligible picture of my methods, and also that I have incidentally exploded to some extent the belief that the solution of such puzzles is only for geniuses, and the composition of them a work of superhuman and devilish ingenuity.

Both are just matters of acquired technique (knack, if you like) plus thinking cruciverbally. So why not have a try yourself?

---



Specialized Crosswords

I don’t intend to write at length on this subject, because I know its fans, though enthusiastic, are a small minority; but the book would be incomplete without some mention of it. It was Torquemada who started such crosswords, and in his Torquemada Puzzle Book he divides those he used into four classes:

1. Examinations, i.e. puzzles in which all or some of the clues refer to a specified subject or author.

2. Inserted clue puzzles, where the clue numbers are inserted in a narrative which the answers complete.

3. Hidden clue puzzles, where the clues consist of single words or consecutive words in pieces of prose or verse, and the solver must find them by trial and error.

4. Missing clue puzzles, where certain words have no clues, but the solver is told that they fall under some particular heading.

He occasionally used other types, one of which happens to be my favourite specialized Torquemada puzzle, and I have chosen it to appear in the Appendix of this book. It might be called a distant relation of the fourth class. All these four types have been extensively used by later composers, especially in the Listener, most of whose crosswords have always been specialized. Hundreds of new types have been invented by its many composers, of which I shall mention here those which have fascinated me most.

Two were among the many creations of Afrit, “Printer’s Devilry”, nowadays the favourite specialized type of many Ximenes solvers, and ‘Playfair’, in which a few answers are concealed by the use of the Playfair code. This last type is a great favourite with some, but there are others who can’t cope with it: I therefore produce one once a year only. “Printer’s Devilry” I have chosen for the Appendix as an example of my specialized puzzles: its nature is fully explained there,

One type that I always used to enjoy in the Listener was the “Theme”, in which some of the solution could only be reached when an unnamed theme had been discovered by deduction.

This type was, I think, invented by Proton (Mr A. McIntyre), and I use something like it myself occasionally under the title “Spot the Theme’. This type especially well fulfils my chief demand from a specialized crossword, that its solution should involve the dropping of an interesting penny at some stage: the setter’s main difficulty is to see that it drops neither too quickly nor too slowly for the solver’s satisfaction.

Another type, which I have never been brave enough to imitate, though the herculean task of solving it was always fascinating, was the “Knight’s Move” of Cocos (Mr D. H. 8. Cox).

In this the down words made knight’s moves, as in chess, instead of going straight downwards: this made things very complicated, but, with sound clues, as they always were, it was a most satisfying struggle.

Of my own inventions I will mention three: “Misprints”, where a single letter is misleadingly misprinted either in each clue or in half the clues and half the answers; “Right and Left”, where with one exception (to give a start) each clue is really two clues, side by side, leading to two answers of the same length — the diagram is divided into two similar halves, and the solver must decide into which side each answer will fit; and thirdly “Theme and Variations’, which belongs really to Torquemada’s fourth type, clues not being given to a small number of themewords, e.g. the names of the Cinque Ports, nor to their pairs of variations, i.e. words connected in some way with them. This, again, is the “penny-dropping” type of puzzle, which I like best.

There are some types that I dislike solving myself and therefore don’t use. Foremost among these is Torquemada’s second class, the Narrative: in solving these I have always found irritating the search through the narrative for the number of a word that I think I may be able to get from letters already fitted into the diagram. More patient solvers probably don’t mind this process, but it is not for me. I am none too fond of the Examination type: it depends, for me, too much on knowledge and research rather than on ingenuity. Torquemada’s third class I use in a somewhat altered form, which I believe to have been invented by an experienced solver of mine, Mr T. W. Melluish, who used it, I think, for the first time in his classical crosswords which used to appear in Greece and Rome. In this form of that class a letter-mixture of the word required is hidden as well as a definition: this makes it easier for the solver to get a start. It is a weakness of this type that they tend to be very hard to start but too easy to finish, once one has got a start: the last quarter of the answers may tumble out in a rush. But they are great fun to compose, giving opportunities for fatuous humour of the nonsensical kind that I like; and I know they are popular with some solvers, so I serve one up about once a year.

The example of Afrit’s specialized crosswords that I have chosen for the Appendix comes under none of these headings.

By fiendish ingenuity he produced, as those of you who like a tough struggle will find, a puzzle in which every clue has two completely legitimate answers, covered by the whole of its length, so that the crossword is in effect two separate crosswords.

I once had a try at this, and it eventually appeared under the title “Double Entendre”, after I had derived many ideas from consultation by post with Afrit himself. Solvers asked for more, but it took such an unconscionable time to compose that I doubt if they will ever get it. I used the same idea for one quarter of a diagram, to provide a leg-pull, on the last occasion when April 1st fell on a Sunday. The pairs of answers all fitted into each other with the exception of one space, which made one set of answers obligatory for solving the puzzle. The leg-pull succeeded more violently than I expected, and there were only about forty correct solutions. What shall I do when April 1st is again a Sunday? Something dastardly, if I’m spared, but not the same as last time!

---



Looking Back

The previous chapter brought to an end my views on crossword puzzles — strong views in some cases, with which I can’t expect all solvers, let alone all composers, to agree. I am well aware that I am arbitrary — perhaps even fanatical, which is worse — about some details; and no doubt there will be many who think that crosswords are a transient form of amusement, not justifying the expenditure of so much powder and shot. I can but plead that to me crosswords are an art, in much the same way as I know chess problems to be an art to the chess problemist. I tried some years ago to penetrate into the chess problemist’s world, only to find that I hadn’t got the right sort of mental equipment: the solution (rather slow) of two-move problems was as far as I could get: my attempts at composition, though for a time painstaking, never passed the puerile stage. But I did get far enough, helped a good deal by the late Brian Harley of the Observer, to understand the standards and the outlook of chess problemists, and to see how much small things, such as economy of the force used, mattered to them. I think that, indirectly, my short-lived venture into a field too difficult for me had some influence on my attitude to crosswords, so that I have come to find that a badly worded or inadequate clue really hurts: my own early efforts, when I look at them again nowadays, do just that, quite as much as any clues written by others have ever done.

The process which has brought me to this state has, of course, been a very gradual one, with my first acquaintance with Afrit as the outstanding landmark. I first met crosswords before they appeared in Britain. An uncle in Montreal sent me, when I was at Cambridge, a book of American puzzles, and the interlocking words at once fascinated me. It wasn’t long before I began trying to fit them together myself: soon after crosswords arrived over here I won my first prize for composing a crossword in a local Leicester paper. The clues were all still definitions: it was not, I think, till Torquemada burst upon the crossword world that it occurred to me that they could be anything else.

I remember well thinking at first, after putting aside a puzzle of his with only three or four words filled in, that this man was quite beyond me. But he fascinated me, and I persevered, till after two or three years I became a regular solver and finished the puzzle more often than not.

Imitation rather than originality has always, I think, been my line: long before I even dreamt of succeeding him, I used to make up what I flattered myself were Torquemada-ish puzzles for my friends to try. I solved other puzzles too; but Torquemada was my hero — I seldom took the others seriously enough to mind if I couldn’t finish, whereas I really did try hard every week to finish Torquemada. I hardly ever remember feeling, when I had failed, that he had been unfair: | may have said at times “Well, that really was a bit steep”, but no more, And this in spite of the fact that his standard of difficulty varied far more than is usual nowadays: sometimes hundreds and hundreds solved him, but on at least two occasions, I believe, only one lady did so. Through all those years his puzzle was a weekly treat to look forward to, and his untimely death in 1939 was a real shock to many thousands.

Having already tried my hand at imitating Torquemada (however feebly), I made up after his death a puzzle containing, unclued, the words of a quatrain paying tribute to his skill. (This type of puzzle, with unclued words forming a quatrain, was one he had often used.) I sent it up to the Observer and they published it. This emboldened me to ask if a successor was to be appointed and to apply for the job. Eventually three of us were appointed, contributing, in turns, what was now called “Torquemada Crossword” instead of “Crossword by Torquemada”, Soon afterwards came the war and, after a year or so, reductions in the sizes of papers; so there had to be one crossword in the Observer instead of two. One week the crossword was of the Torquemada type, next week of the Everyman type. This lasted till after the end of the war when, in the summer of 1946, Everyman, with its wider appeal, became weekly, the harder puzzle having to remain fortnightly till paper had at last become plentiful enough for its weekly appearance near the end of 1952.

It was in 1945 that I was given complete responsibility for both Observer crosswords. For the harder puzzle I had already, in 1943, adopted the pseudonym “Ximenes”, the name of the Cardinal who succeeded the original Torquemada as head of the Spanish Inquisition. Now a new series started, both for Ximenes and for Everyman, each being numbered afresh from No. 1. The style of the Everyman puzzle changed very gradually towards its present style; for Ximenes I introduced at once some definite changes. I was under Afrit’s influence by now, and I had become convinced that he was right to include, even in difficult puzzles with the barred type of diagram, far more unchecked letters than had been used in the “Torquemada Style’ crosswords: this title, by the way, was dropped when the new series started. I also, from this point onwards, made my diagrams symmetrical: hitherto I had followed Torquemada in disregarding symmetry. The biggest innovation in 1945 was the competition with prizes offered for the best clues sent in for a specified word, given in the paper a definition only instead of a cryptic clue. This competition, at first fortnightly, later to become monthly, has lasted ever since and has lost nothing in popularity: if anything the number of entries tends to be higher nowadays than it was at first, and their quality has advanced enormously.

The result of this innovation which I value most is the number of friends it has made for me. Very early on competitors began to write little notes with their entries, and correspondence with the regular ones gradually started. After about a dozen competitions someone suggested that a slip should be sent round to those who applied for it, containing in full the clues of all prizewinners and of those highly commended (there was only room for the first prizewinner’s clue in the paper). This idea at once became popular.

I soon took to adding comments on each competition (very brief at first, longer later). In these I began to circularize among competitors my growing views on the principles of cluemanship.

Nowadays something like three-quarters of the competitors send for slips. In a recent competition there were over 700 entries: the record entry was just under 900: it has rarely fallen below 200.

All the time I have been getting to know many of my solvers better and better, some only postally, others in the flesh, largely as a result of the opportunities of meeting them provided by the Ximenes Dinners, which have been a great delight both to my wife and to myself. The first celebrated No. 100: others have accompanied No. 200, No. 500 and No, 750, Another bright idea, which shows that a bond exists among Ximeneans, is the Ximenes tie for men and scarf for ladies, which were started over two years ago and have sold well.

It may be worth while to mention here the number of different people who have won prizes and commendations in the competitions. Naturally we have our experts, whose names crop up quite often in the lists; but we are by no means a closed shop.

As I write, over 400 different people have won prizes, and over 1,400 have been at one time or another highly commended.

How many people solve Ximenes? I have always wished there were some means of finding out, for the competition entries are certainly no guide. I have myself met or heard of hundreds of solvers who don’t fancy their cluemanship (or their luck in the weeks when there is a draw for prizes) enough to enter, and I feel safe in guessing that there are a good many thousands who solve, regularly or irregularly.

This chapter has, perhaps, had too much Ximenes in it and not enough Everyman: that I can’t help, for this type of puzzle has always been my first love. Nevertheless, I kept the Everyman series going from 1945 till the autumn of 1963, when I handed over that section to two successors, Miss D. W. Taylor and my collaborator in this book, Mr A. Robins, who are as scrupulous as I ever was about cluemanship, if not more so, Good luck to them both, and to all solvers of both Observer crosswords, and to all composers and solvers of all the other crosswords too.

---



Appendix: Specimen Crosswords

Ten specimen crosswords are here presented so that readers may compare techniques, ancient and modern. Nos. I-IV and No.

VII should be soluble by all average solvers. The other five will suit better those who like a tough struggle. A few notes about them follow.

I. This period piece, the first Times crossword, has been discussed already in Chapter II.

II. This is a fairly early Everyman puzzle of mine, which appeared in 1948. I have chosen it not for its merits but so that readers may compare its technique with that of Nºs. III and IV, modern Everyman puzzles by my two successors who, with many admirable additions proceeding from their own original minds, have based their methods largely on the principles I was using when I handed over. Readers may like also to compare Nº. II with the criticisms I made of my Everyman No. 1 in Chapter II, and to criticize it for themselves in the light of what I have written there.

V. This one I have chosen as a typical Torquemada puzzle, dating from the height of his career in 1932. It contains, perhaps, even more of his brilliant sallies of wit than most of his puzzles, and rather fewer literary references than usual.

VI. I have already said that this is the specialized Torquemada puzzle that I remember with most affection.

VII. I hope average solvers will not be put off by the blank diagram feature, which is not nearly so formidable as it may at first appear. The title “Armchair” shows that the puzzle exemplifies Afrit at his mildest, but at the same time it gives a clear idea of the perfect soundness and clarity of his clues.

VIII. Here, on the other hand, we have one of his difficult masterpieces, a wonderful achievement. I hope there will be many to appreciate it.

IX. I now reveal to Ximenes solvers that in December 1963 I tried a mild experiment on them. I presented, as No. 778, the diagram and most of the words of Ximenes No. 1 (1945). I altered a few words to get rid of the word then set for competitors to write a clue to: I thought it would ring a bell among old solvers.

I modernized any clues that I didn’t still approve of, and wondered if anyone would write to say he, or she, had spotted what I had done. As was to be expected, no one did. Now, so that changes in technique may be noted, I have presented the diagram to be filled in exactly as it was in 1945, with no changes; and I have presented two sets of clues: the original ones, unaltered, and those of 1963, plus new ones for the words I changed, for the reason stated above, in 1963. In the second set I have used the clue which won the first prize in the first competition, to replace the definition given in the original version: its author was Mr J. R. Tilley. You will see that I have not altered all the clues: those I have left would, I think, pass muster nowadays. I hope the comparison may be interesting in the cases where I have made changes.

X. Finally I have included, from 1959, a “Printer’s Devilry”, which, as I said before, is undoubtedly the favourite specialized type among my solvers. Dare I hope that new converts may be found to wrestle with it? If so, they will find a Ohambers’s Dictionary helpful for verifying unfamiliar words. I may add that over half the words in the puzzle are in any average person’s vocabulary. Here, as in the revised clues to IX, one clue, that to 30 across, is the first prizewinner in the competition associated with this puzzle: its author was Mr R. N. Haygarth.

Solutions

Frankau's puzzle from Chapter 1.

I

II

III

IV

V

VI

VII

VIII

IX

X

---