Kosman and Picciotto on their Nation puzzle, cryptic crosswords, wordplay and puzzles in general.
|Click here for Kosman and Picciotto’s tips on how to work their puzzles||116.49 KB|
[First, three links:
We get mail—not a lot, granted, but a steady stream of reactions to our work. Some complain the puzzles are too hard, some that they’re too easy (some recent examples are here). Some people even think that, like Goldilocks, we get it just right.
And then there are the outliers. Last week, one of us received, at home, a good old-fashioned envelope from the US Postal Service. A certain amount of effort had gone into keeping the mailing anonymous; it was addressed in a bland and undistinguished font, with no return address. Inside was a photocopy of the most recent Nation crossword, with every square correctly filled in. There was no signature, and no hint as to the sender’s identity.
At first glance, this mysterious missive seemed utterly unmotivated. It told us only that at least one person on the planet had been able to solve last week’s puzzle—a heartening development, in its way, but perhaps not especially earth-shaking.
Yet on further reflection, this veiled message did seem to get at something fundamental about the psychological dynamics of puzzling. A puzzle is a form of intellectual combat, a battle of wits between the constructor(s) and the solvers. It’s a friendly fight, to be sure, and like a WWF match it’s essentially fixed: the goal on both sides is for the solver to emerge victorious. But there is still a struggle being waged across the black and white squares of a crossword grid: we try to confound you with traps and snares, you try to evade them to arrive at your goal.
Seen in this light, the nature of our anonymous mailing becomes clearer. It was nothing less than a victory cry, an assertion of pride at having emerged triumphant from the week’s skirmish. Perhaps this was even the first time our correspondent had succeeded in completing the puzzle. In that case, the subtext might be: “You two must think you’re very clever. You have stumped me before, but what do you think of this?”
And to that we can only reply: Bravo, sir or madam. Well played. Next week, we’ll try to find new ways to deceive you.
Feel the need to crow, either anonymously or not? Please share your boasts below, along with comments, questions, kudos or complaints about the current puzzle or any previous puzzle.
In the lingo of the National Puzzlers’ League (NPL), a heteronym is a word or phrase that is spelled the same as another word or phrase, except perhaps for spaces. One example would be “to get her” and “together.” The original meaning of the word (which is still also in use) refers to words that are spelled exactly the same, but pronounced differently, such as “refuse” (verb) and “refuse” (noun).
This is of course quite relevant to cryptic crosswords. “Heteronym” could be a category of clues that are often classified differently. For example, here’s a normal double definition (from Puzzle 3198):
MOPED Vehicle was low (5)
This is definitely a heteronym, given the different pronunciations.
A less standard double definition (from Puzzle 3199):
LEGIT Acceptable run (5)
This is less standard, because LEG IT is not the same word as LEGIT. Some might consider it a “whole thing” charade.
The clues below are also charades of sorts, or you might think of them as jokey double definitions (respectively from Puzzles 3215, 3225, 3232, 3233, 3236, 3243):
MOCKING Monarch of comfy shoes? (That’s sarcastic) (7)
LACERATIONS Shoestring budgets resulting in tears (11)
GAS PEDAL “It makes the car go,” Unser said with difficulty (3,5)
AMPHITHEATER Band’s equipment collided with radiator in auditorium (12)
PREEN Groom ’em? (5)
BANJOIST Prohibit support for Pete Seeger, e.g. (8)
We use clues of this type quite often.
This one (from Puzzle 3215) was based on a classic heteronym discovered by NPL member Newrow (Edward Wolpow):
MOUNT ST. HELENS Prepares the telescope to find a volcano (5,2,6)
Heteronyms can be used in clues, but this is controversial among our fellow constructors. For example, one can use compound words in clues misleadingly: “redhead” could represent the letter R; “tailspin” could clue ALIT. We have no objection to this sort of wordplay—in fact, we have gone further, as in this example from Puzzle 3241:
PERSPIRE React to the heat with steeplechasing agent in retreat (8)
To understand the wordplay, you need to insert a space between “steeple” and “chasing.”
Our most unorthodox use of heteronymy was in this reversal clue for a Down word (from Puzzle 3237), where we suppressed the space between definition and wordplay:
ENID Eat uptown in Oklahoma? (4)
(The question mark was an attempt at warning the solver that something unusual was going on.) We only do this once in a blue moon, of course, but we enjoy it when we do!
How do you feel about heteronyms? Any favorites? We’d love to hear from you. Please share below, along with comments, questions, kudos or complaints about the current puzzle or any previous puzzle.
[First, three links:
The other day we found ourselves in conversation with our old friend Fraser Simpson, the puzzler who sets the weekly Saturday cryptic in the Globe and Mail. In the midst of shop talk about principles of clueing and grid design, Fraser let loose with an offhand observation: “I do notice,” he said, “that you guys put a lot of three-letter entries in your grids.”
There wasn’t any malice behind it, but the accusation stung a bit. No one likes to see three-letter words in a cryptic crossword grid (regular crossword grids are a different story), and we’re no different. The reasons are multifold. For one thing, in a checkerboard grid every other letter in an entry is checked by the crossing word (see here for more on this topic). That’s fine in a longer word, but a three-letter word with the first and last letters filled in is almost pre-solved.
More importantly, it’s practically impossible to write an interesting clue for a three-letter word—there’s just not enough to work with. You can’t assemble a good charade or container clue, and anagrams and even reversals provide minimal variation.
Still, there’s a difference between forbidding something absolutely and trying to avoid it as much as possible—which has been our approach to three-letter entries. The most common reason to put them in a grid has been the constraints imposed by a group of (usually long) theme entries. And when we do have three-letter entries, we almost always clue them as part of a phrase with other entries, which alleviates the problem of finding interesting wordplay.
So the notion that we were lax about these matters seemed unfounded, and we sputtered an indignant protest: “We hardly ever use them, and every time we do there’s been a good reason!”
The truth, it turns out, lies somewhere in between. A check of our puzzles over the past year reveals that three-letter entries have appeared in ten puzzles, which is far more than we’d thought. But in every one of those cases, the three-letter entries were forced by a tough-to-fill thematic grid (e.g., #3207, #3224, or #3230), or used as part of a longer, multi-word entry (most notably in #3222, with its three-letter orgy TEA FOR TWO AND TWO FOR TEA), or both. And in some cases—as with ONO in the Beatles-themed #3210 or OIL and GAS in the Arab League-themed #3201—the three-letter entries were themselves thematic. This was never something we did lightly or inadvertently.
In other words, our approach to three-letter words is in keeping with our attitude to nearly every aspect of crossword construction: conscientious but not doctrinaire. If three-letter words help make a theme possible or lead to a clever clue, then let a few of them bloom!
What are your thoughts about three-letter words? Please share them below, along with comments, questions, kudos or complaints about the current puzzle or any previous puzzle.
We recently received a letter from a solver who stated: “The puzzles are WAY TOO HARD! I can never solve more than one or two of the clues. I have a suggestion which would increase MY enjoyment, at least: Do as the crossword puzzle editor of The New York Times does: vary the difficulty of the puzzles.”
Another correspondent wrote: “For the most part, the clues and answers pass muster. But too often, the answers are too convoluted and hard to reconcile with the clues.” He is not renewing his subscription.
Those comments came to us via the US Postal Service. The following three came by e-mail:
“You guys have some great, interesting and challenging clues and answers. Most satisfying is to work on puzzle while eating, say, for about 20 minutes at a stretch and get all that I can get, which might be five or six, and then do the same for about three more times in the week before the next mag comes. The combo of challenge and relaxation often works out very well. Best if I can get almost all, such as in 3231.”
“You have to slow down and keep at it. I start by going through the clues in order. I mark up the grid to show where multiple-word answers fall and fill in any answers I get as I go (not many), trying not to get frustrated if I go through, say, all the across clues without getting a single one. After the first pass I have a few answers. I look to see if the cross letters are any help. Then it’s time to go back though. I get a few more. I put it down and come back to it later. More answers. I keep doing this over a day or 2 et voilà! Done.”
“I really, really, really like your work. All of us were bereft by the death of the irreplaceable Frank Lewis, and I have no higher compliment than to say that you are worthy successors. I’m also happy to say that the past two puzzles were the very first of yours which I have succeeded in solving completely before the next issue came out, so I think I’m beginning to get on your wavelength—for better or for worse.”
Are you getting on our wavelength or canceling your subscription? Are the puzzles too hard? What is your approach to cracking them? We’d love to hear from you. Please share below, along with comments, questions, kudos or complaints about the current puzzle or any previous puzzle.
Fans of cryptic crosswords face a constant and grim economic law, which is that—in the United States, at least—demand always outstrips supply. Journalistic outlets for cryptics are few and far between, and although there are a handful of published collections available, many of us ran through those long ago.
So we were happy to discover, during last month’s annual convention of the National Puzzlers’ League, the arrival of a fine new compendium of no fewer than fifty variety cryptics. This self-published collection is the work of Roger Wolff, a former Microsoft programmer now pursuing a doctorate in computer science at Carnegie Mellon University, and they’re high-quality creations, with well-made grids, solid clues and entertaining gimmicks. Admittedly, the range of gimmicks is a little narrow—most of the puzzles involve altering a subset of entries in accordance with some thematic principle—but the number of variations Wolff comes up with on that basic idea is impressive, and he follows through deftly on each of them.
The book is recommended for beginning solvers and practiced hands alike. It’s available on Amazon or from the author at email@example.com. We caught up with him this week for a short phone chat.
How did you get started making cryptic crosswords?
I was introduced to them by my ninth grade English teacher, who I guess took a liking to me. I spent about ten years solving, and I had a great time doing that. Then at some point I got the idea to start making my own. The first cryptics I wrote were for the Microsoft Puzzle Hunt. I included those in the book, but I had to go back and rewrite every single clue. I really didn’t know what I was doing. But that was the first time the Puzzle Hunt included cryptics, so maybe no one knew any better.
So you just started writing puzzles and putting them in a drawer, like some crossword Emily Dickinson?
Occasionally I would try to get one published, but there’s no real outlet for them. I had one puzzle published in Games magazine. The puzzle had a baseball theme, and an editor rewrote every single clue to be about baseball. It was much improved, in fact.
How did you conceive of putting the book out yourself?
The whole time I was writing these, I kept telling myself, “When I get to 50, I’ll put them in a book.” And the technology just caught up with me. By the time I was ready, the possibilities for self-publishing were there.
Do you have any particular models for constructing cryptics?
My role models have been Emily Cox and Henry Rathvon. Over the years, they just stood out as the people who really know how to write variety puzzles that are fun to solve. They’re the best.
What plans do you have for a follow-up?
I’d like to go do another book of fifty, but this one took ten years and I tapped out all my creative ideas. My next project is a puzzle-a-month calendar. Writing a cryptic that is thematically appropriate to each month helps give a little structure to my thinking.
Do you have any suggestions for other published collections of cryptics? Please share them below, along with comments, questions, kudos or complaints about the last issue’s puzzle or any previous puzzle.
A few links before this week’s post:
• The current puzzle
• Our puzzle-solving guidelines
• A new Nation puzzle solver’s blog hosted by Braze, a Pennsylvania cryptic fan who is also a hockey referee (zebra). Braze will post his annotated solutions on the Monday following the appearance of the puzzle online. Fortunately, it seems like the hard-copy solvers (most of you, we think) can easily scroll back to a previous puzzle and avoid seeing solutions you don’t want to see.
• An interview with Henri on the Guardian (UK) crossword blog
As we mentioned in a previous post, our predecessor, Frank Lewis, occasionally wrote reversed clues, such as “Earth despair (6,5)” for BROKEN HEART. Here the definition is “despair” and the wordplay is, in a way, “earth.” This is the reverse of the usual cryptic crossword anagram, where “broken heart” would be used as wordplay for EARTH.
In the terminology of the National Puzzlers’ League, this sort of wordplay is a rebus. In a rebus, you start from a so-called rubric (“earth” in the above example), and figure out a way to read it that supplies the solution. Of course, some context is required, and in the case of a cryptic clue it is supplied by the definition.
Here are some examples of rebus clues we have used, with a brief explanation of each. First, some examples in the style of Frank Lewis:
From puzzle 3200:
ROCKING THE BOAT Ate both? Revolting! (7,3,4)
ROCKING is the anagram indicator and THE BOAT is the anagram fodder, yielding “Ate both.”
From puzzle 3205:
INTEGER VARIABLE Non-fractional quantity in a program makes it green? (7,8)
VARIABLE is the anagram indicator and INTEGER is the anagram fodder, yielding “it green.”
From puzzle 3244:
LOOSE-LEAF Flea’s kind of binder (5-4)
The next three examples are based on the position of a specific letter in a word in the solution:
From puzzle 3198:
BEETHOVEN’S THIRD E is for “Eroica” (10,5)
From puzzle 3229:
MIDDLE OF NOWHERE H in an isolated location (6,2,7)
From puzzle 3237:
SECOND IN COMMAND O, deputy! (6,2,7)
And finally an idiosyncratic example, from puzzle 3222:
TEA FOR TWO AND TWO FOR TEA Song excerp2 from 19T5? (3,3,3,3,3,3,3)
This is a somewhat phonetic clue, where 2 was used for t, and T was used for 2 in the clue, to suggest the solution.
Rebus clues are very popular among our test solvers, and in fact, the first example above was suggested by a test solver (thanks, Yossi!). Obviously, we don’t often have the opportunity to use them, but we love to do it when we can.
If you have thoughts on this type of puzzle clue, please share them below, along with comments, questions, kudos or complaints about the current puzzle or any previous puzzle.
SPOILER ALERT: If you haven’t solved our previous puzzle yet, you might not want to read on.
One of the things we’ve been exploring during our first year at The Nation is how to add thematic content to the world of the black-square cryptic crossword. (Variety cryptics, of the kind that appear monthly in Harper’s and the Wall Street Journal, are themed by definition, and don’t need our help.) The themes have come in a variety of flavors, but one way to categorize them would be based on how pervasive a theme is in a given puzzle.
The simplest type of theme, which is where we began, entails a few distinctive entries while the rest of the puzzle remains standard. Our audition puzzle, for instance, included four related long phrases (BANANA REPUBLIC, ORANGE FREE STATE, JOHNNY APPLESEED and FRUIT OF THE LOOM), after which we just filled the grid as we could. That kind of theme continues to offer fertile opportunities, and is our most common go-to strategy.
Another sort of theme lives not in the diagram, but in the clues. For example, in Puzzle #3201, which was published during the Arab spring, every one of the twenty-two countries in the Arab League appears in one of the clues. In Puzzle #3231, which appeared in the “Occupy” special issue, many clues included variations on the word: occupational, preoccupy, unoccupied and so on.
In the past month, though, we’ve turned up the pressure on ourselves by adopting what we might call “full-compliance” themes—that is, ones in which every aspect of the puzzle needs to fit the theme. The first such was Puzzle #3242, for The Nation’s special issue about Amazon. We began with the less ambitious idea of getting a few river names into the clues, but we wound up fitting in so many that we decided to go all the way and put a river into every clue. (And believe us, the last few we wrote were a bit of a stretch.)
Last issue’s puzzle, in which the grid included no vowels other than E, was even more challenging to construct. The idea came from a wordplay chestnut, the fact that Ellen Degeneres’s name is a slew of E’s. (Years ago, Joshua created a little entertainment for a National Puzzlers’ League gathering, based on a dozen such E-only notables as Renee Zellweger, Helen Keller, Peewee Reese and Bennett Cerf. It was aptly titled “Twelve Celebs.”) With ELLEN DEGENERES and EVERGREEN TREES symmetrically placed, it was then a matter of filling in the rest of the grid with words that also fit the constraint.
In theory, we could have gone even further. One early plan—or rather dream—was to match the E-only grid with clues that complemented it by excluding any use of the letter E (in the manner of George Perec’s great 1969 novel La Disparition, translated into English by Gilbert Adair as A Void). But we quickly abandoned that, for two reasons. One was that a grid full of E’s together with clues lacking E’s would have meant that we couldn’t do any anagram clues, which are the life’s blood of cryptic crosswords. The second was, Are you freaking kidding? However, we don’t rule out an E-less set of clues for some future puzzle!
Do you have any thoughts on these or other puzzle themes? Please share them below, along with comments, questions, kudos or complaints about last issue’s puzzle or any previous puzzle.
When The Nation announced an opening for cryptic crossword constructors, each of us had been solving cryptic crosswords for many years. Like many of the experienced solvers we know, we had largely lost interest in standard American black-square cryptics, such as the ones found in Games magazine and the occasional Sunday New York Times. Those puzzles have a limited audience: beginners find them impenetrable, but for more experienced solvers they are, frankly, rather boring. Most experienced solvers we know sooner or later migrated to so-called variety cryptics, such as the ones in Harper’s or The Enigma, or to British puzzles.
This state of affairs made us wonder whether applying for the job at The Nation made any sense. Would we want to create puzzles of a kind that we ourselves do not enjoy solving? But further reflection convinced us it would be worth a try. One of the things that swayed us was the example of Frank Lewis, who had a broad and loyal following that extended over many decades. In other words, he had achieved something unique among US constructors of black-square cryptics. What was his secret? Could it be the fact that his style was well outside the US cryptic mainstream?
The conventional wisdom among US cryptic constructors—and many of their fans—was that Lewis’s style was too freewheeling. For example, he was not very vigilant about matching the parts of speech in definition and answer. His diagrams often included consecutive unchecked letters, which is generally considered unfair. The two senses in his double definitions were often very close to each other, leading some solvers to coin the phrase “1.01 def” to describe those clues. He often resorted to non-cryptic punny definitions (what the British call “cryptic definitions”). Finally, his cryptic repertoire was very limited—he overused the same few anagram indicators, as well as Roman numerals and cardinal points.
The result was that many of his solvers spent time and energy excoriating him. And yet they kept coming back for more. Why? There are probably many different answers to this question, but here are some features of his puzzles that we enjoyed:
• He often used very long entries, including phrases and entire sentences, distributing them in multiple location in the diagram.
• He usually had lovely, humorous, and fresh insights into how to reinterpret those long entries, which led him to create entertaining clues.
• He drew from almost every field of knowledge, and sometimes used interesting unfamiliar words.
• He sometimes reversed the clue-answer relationship, for example by using a clue like “Earth despair (6,5)” for BROKEN HEART. The anagram indicator and fodder were in the answer, and the result was in the clue! Likewise, he sometimes used (duly indicated) homophones of words in the clue.
• He occasionally made references to The Nation’s politics.
• His clues were varied in length and style, and he sometimes completely surprised the solver with an unexpected and idiosyncratic twist to the usual routine.
Our plan was to take some of Frank’s whimsy and inventiveness and join it with the structural rigor we were accustomed to. As members of the National Puzzlers’ League and veterans of many kinds of puzzles, we are familiar with a wide range of approaches to wordplay. Perhaps we could combine all those ingredients to create a type of black-square puzzle that would be fun for us to construct, and for you to solve?
So we ended up applying for the job, and here we are. We have thoroughly enjoyed the experience so far—although given our advanced ages, we will not be able to match Frank’s prodigious sixty-year run.
PS: A Hard Nut to Crack
When constructing last week’s puzzle, we rejected this clue for PECAN:
California imprisoned nut (5)
(CA in PEN = California imprisoned.) We thought it was too difficult. What do you think? Please share your thoughts below, where you can also post comments about Frank Lewis, and questions, kudos, or complaints about last week’s puzzle or any previous puzzle.
The other day we heard the behavioral economist Dan Ariely talking on the radio about his new book, The Honest Truth About Dishonesty. One of its themes is that people’s dishonesty can vary across a wide spectrum of behavior, depending on the circumstance. That in turn got us thinking about cheating in solving crosswords. Is there even such a thing? And if so, what are the parameters?
At a first approximation, of course, cheating on a crossword is like cheating at solitaire—unless you’re enrolled in a competition, it’s a contradiction in terms. Solving a puzzle is something each of us does for our own enjoyment, in our own way. Any technique I use to solve a crossword is no business of yours, and vice versa.
But in practical terms, crossword solvers do tend to follow certain guidelines. In the extreme case, if you were to simply look up the answer to each clue in the next issue, no one would realistically say that you had “solved” it—but more importantly, it’s hard to see how you would have derived any pleasure from the exercise. If a puzzle is worth doing at all, it’s worth doing while observing certain constraints.
What’s interesting is how those self-imposed constraints vary from person to person—and how changing the parameters just a little reveals the psychology underlying the process. For instance, many solvers have no compunction about asking a friend, spouse or office-mate for an answer, yet would never ask Google (or Siri) the same question. There seems to be something about sharing the solving process with a fellow, and equally fallible, human being, that feels different from consulting the cyborg mind.
The ethics of looking things up arises much more often in standard crosswords than in cryptics, because cryptic clues are less overt and more deceptive about what they’re asking for. If you see “actress Witherspoon” in a clue, it’s probably asking for “Reese,” but you can’t be sure—and in any case there may be more steps to take before reaching a final answer. So there’s a certain amount of unavoidable brainwork that can’t be circumvented by cheating.
What you can do, with both standard and cryptic puzzles, is use external sources to confirm an answer that you’ve gotten but aren’t sure of. This seems to be the most common, and most widely acceptable, practice that might be construed as “cheating.” All of us have holes in our knowledge, and we all find ourselves at one point or another thinking, “Everything in this clue suggests that such-and-such is the name of a pop singer”—or Impressionist painter, or chemical compound or African capital—“that I’ve never heard of.” And very few of us would refrain from checking that in a dictionary or an on-line search.
What about the situation where you know the answer is A?R?O?L, and is some sort of rotor blade, or perhaps a monk, but no actual word comes to mind? (This may have happened to you in last week’s puzzle.) You also know that Kosman and Picciotto are not shy about throwing in an occasional word you’ve never heard of. Well, fear not: there are electronic aids that will help you fill in the blanks. See, for example, OneLook Dictionary Search, or the National Puzzlers’ League site or the Franklin Crossword Solver devices or (for the iPhone) the Crossword Help app.
Oops. Have we been abetting cheating by providing these suggestions? What are your thoughts about cheating when solving a crossword? Do you have favorite electronic helpers? Are there some practices you think are never OK? Please share your thoughts below, where you can also post comments, questions, kudos or complaints about last week’s puzzle or any previous puzzle.
Henri: When you’re solving a cryptic, it can be fun to argue with the constructor in absentia. In our weekend breakfast solving group, for instance, we used to have ongoing arguments with Frank Lewis, The Nation’s cryptic constructor, year after year, decade after decade. In our imagination, we composed a letter that kept getting longer and longer, listing our objections to his clues. This was all in fun, and as much as we complained, we still had fun solving his puzzles.
Joshua: I’ve written those letters in my head myself, more than once. But I never would have considered sending one. What would be the point? Solving puzzles you don’t enjoy, or that go against your personal aesthetics, would be a bore. If the puzzles are that bad, your time is better spent doing something else.
Henri: That’s true. When I pick up my solver’s pencil, I try to put away my editor’s pen. Of course, it’s also fun to solve a puzzle where the clues are so consistently elegant that there is nothing to argue with. The obvious example is Emily Cox and Henry Rathvon’s variety cryptics, which appear monthly in the Wall Street Journal.
Joshua: Different solvers have different preferences, and even the same solver may have different preferences depending on context and mood. For example, while I like clear boundaries in a variety cryptic crossword, I also enjoy wide-open challenges in a puzzle hunt, such as DASH, where you don’t even know a priori what the rules are, and are not given instructions.
Henri: Another contrast is between a situation where I know I can solve the puzzle, as opposed to one where that is in doubt for a long time. In the first situation, I proceed methodically from A to Z and it’s just a matter of time before I’m done. In the second, I may solve a few clues and hit a wall. Then I might put the puzzle aside, and come back to it a few hours later or the next day, and solve a few more. I’m not always in the mood for this sort of challenge, but that kind of puzzle is more satisfying once it is finally conquered. As Piet Hein, the Danish designer of geometric puzzles, put it:
Problems worthy of attack
prove their worth by fighting back.
Joshua: In fact, the same idea applies to a single clue. I love coming across a surprisingly offbeat clue. If I solve a clue on sight, it is often not as pleasurable an experience as cracking a clue where it takes a while for the penny to drop.
Henri: Inevitably, our own attitudes and experiences as solvers must influence our choices as constructors of the Nation puzzle. We try to write clues we’d like to solve!
Hopefully they are fun for you also. How do you relate to puzzles in general? to The Nation’s puzzle in particular? Please share your thoughts below, where you can also post comments, questions, kudos or complaints about last week’s puzzle or any previous puzzle.