MGWCC #760 — “Word for Word”

[photosurrealism]

Yes! Thanks. I was annoyingly close and am home now.

[Joe Ross]

We're in for it, this weekend. Week looms.

117. Joe Ross 12/27 - 20:19

[HunterX]

There's an element to this construction that I find quite impressive, but, at best, unnecessary, and at worst, potentially misleading. At least it was the latter for me for awhile.

But it's still very impressive!

Obviously, keep a lid on it until deadline passes tomorrow. But I'd love to hear what that was!

As it is after the noon hour and submissions are closed, I'll post my answer.

Firstly, the answer (as I submitted it) was CHAMPION. There were exactly 26 clues consisting of one word, whose answers were one word. Those 26 clues were pangrammatic, each starting with a different letter of the alphabet. The answers to those clues were also pangrammatic. (Only in the case of the 'P' clue did the answer start with the same letter.) 124A gave the answer MMXIII or the year 2013, commonly read as "Twenty thirteen." So if you arranged the clues in alphabetic order, and looked at the starting letters of the answers to those clues, those letters starting with the T clue (T being the 20th letter of the alphabet) and going backwards to the M clue (13th letter) spelled the meta answer.

My beef is that the answers did not need to be pangrammatic. It's amazing that they were. I can't imagine the brain energy it took to create two pangrammatic lists that matched up as 26 clues and answers for the grid, AND did so such that a stretch of 8 (in reverse order) spelled a word. That's a really impressive construction.

But the fact that the answers were also pangrammatic kept throwing me off. I felt that there HAD to be a reason for it. Are we supposed to take the 13th and 20th letters of the alphabet from the first list (clues) or second list (answers) or both successively? Do we arrange the list by the clues, find the answer words, then arrange them in alphabetical order and do something?

It was only when I finally saw the answer that I realized the answers didn't really need to be pangrammatic.

But great meta-puzzle, regardless!

[Laura M]

Nope. I got the letter mapping early but never did figure out what to do with it. I read the 124A clue so many times, trying to make it make sense. (I'm sure it's sour grapes but I'm a little underwhelmed with the actual solution!) Not sure I should even attempt week 5...

[woozy]

I think the answers needed to be pangrammatic in order for us to notice. 26 one word pangrammatic clues distributed evenly among over 130 clues total simply wouldn't register. But noting weird X words in both and answer and a clue is noticeable.

My issue is that there is nothing natural in my mind to reading Twenty Thirteen and interpreting it as a back listed range.

Then again I never even noticed there were 26 pangrammatic clues/words. I only noticed there were a few trios of SAHARAN = XERIC/HOT, XANTHOUS = YELLOW/TAN and with a stretch I could maybe stretch 2 or 3 more such as JOKE=GAG/KID and others. The lack of balance bugged me.

On the third hand though, After only having about a half hour on christmas eve to devote to this I was unmotivated to pick up a two day old puzzle come monday, had not a friendly muggle grabbed my by the throat and started throwing nudges at me.

[HoldThatThought]

Yes. I concur 100% with my friend Hunter's comment above.

Before I go any further (and I wish I didn't feel like I needed to do this every time) - I loved the puzzle, I love the challenge of doing Matt Gaffney's puzzles every week, but I don't love them all the same. My favorite author was Kurt Vonnegut, but his books aren't all equally brilliant, and I don't think that I would be insulting Mr. Vonnegut any time I dared to explain why I liked this book more than that book. Nor would it be fair for a heckler to say "If you don't like Deadeye Dick, you shouldn't be reading anything the guy wrote." We do that sometimes in crossword communities.

So - that aside - I need to agree with Hunter.

Successful solvers discovered something extremely curious - and therefore critical. Buried within a long set of clues were 26 one-word clues, and - ah the best part - their first letters formed a pangram. Can't be coincidence right? Here they are - our 26 key clues.

But then...there's a further creator's flourish - the answers to these 26 pangrammic clues were pangrammic, too! Surely, that must be important, as well.

I can't begin to explain how other people solve these puzzles, but I've been privileged to "meet" one or two fellow solvers who seem to approach these like I do. What is UNUSUAL about the grid or the clues. What's there that didn't have to be there? Is the grid asymmetric? Are there unusual words that seem forced, because a couple of letter changes elsewhere could have avoided them? Are there a lot of clues that reference numbers?

When something out of the ordinary is found - the difficult next question is "Well, is that random coincidence, or does it mean something?" Is it a pathway to the solution or is it random noise?" Anyone who has ever worked in any field that involves data analysis knows this challenge all too well. "Is this data blip meaningful, or just chance noise, to be ignored?"

A lot of times, there's no sure way to tell. An experienced analyst goes with the gut - an experienced meta solver does that, too.

If Matt Gaffney makes both the clues and their associated grid entries pangrammic, that must mean something!! (Said my gut)

So...I (and the crack team of brilliant folks I discussed this with) spent a very long time trying to figure out what that meant.

Surely the answer wasn't a panagram - a 26-letter word? A phrase of 26 words? A shorter phrase like "The five boxing wizards jump quickly?"

No. The pangrammic answers were irrelevant. The creator's artistic flourish. The solution had nothing to do with pangrams, and the entries could have been anything, so long as the alphabetical ordering of the clue-entry pairs - by clue word - lined up associated entries that spelled NOIPMAHC in positions 13 through 20.

A whole lot of other time was invested in reading puzzle grids from 2013, but that was my own misinterpretation of the solving clue - that's always fair game, and what makes these later month puzzles so challenging and fun. That was an unforced error. What wasn't unforced was the lily gilding pangrammic answers - which was too unlikely to be coincidence - and therefore, surely, part of the solution. But not.

That all being said - another great and ultimately satisfying puzzle, with a hiccup that spoiled the eventual celebration.

My opinion and a whole lot of patience won't get me from Podunk to Palookaville on Southwest Airlines.

[pgw]

No. The pangrammic answers were irrelevant. The creator's artistic flourish. The solution had nothing to do with pangrams, and the entries could have been anything, so long as the alphabetical ordering of the clue-entry pairs - by clue word - lined up associated entries that spelled NOIPMAHC in positions 13 through 20.

I guess it is technically true that the answer words didn't also have to cover the whole alphabet, but I think it was pretty key to the solving process and therefore definitely not irrelevant. The fact that there are 26 one-word clues (each with a one-word answer) is a pretty strong hint to wonder if they make a cipher. You check them and they do, so you *know* this has to be part of the solution. And a natural thing to do is to write down the cipher in alphabetical order. Once you do that, if you're like me you stare at it for two days without finding the answer and then you randomly glance at it one afternoon and you see a backwards word and you're like "oh, there it is" (and then the "T=20--M=13" thing confirms it). Not easy ... but it would have been much harder without the mid-solve "you're on the right track" confirmation provided by the fact that the 26 clue/answer pairs make a substitution cipher, and the resulting motivation to write the cipher down.

pgw

[BrennerTJ]

That was one well-camouflaged Champion...

[KayW]

I got this thanks to some help from @AaronT and the Last Call Zoom Crew (is that a rock band?) after two days of staring at the pangram "ciphers" and all Matt's 2013 puzzles.

This meta reminds me of something Matt called a "front door" meta. He describes it in the write up for MGWCC #700. In Matt's words:

Name comes from playing hide-and-go-seek in the house, where the searcher starts outside on the front porch. Hiding behind the front door is a risky move, but with possible big payoff: if the searcher doesn’t find you right away, then they’re in for a long look around an empty house before they circle back and find you.

And yep. BOTH TIMES. I went for a long look around an empty house before circling back to find his fiendishly clever answer right there behind the front door. Fooled me twice, Mr. Gaffney. Maybe one day I'll learn.

A VERY clever meta.

[ky-mike]

We're in for it, this weekend. Week looms.

117. Joe Ross 12/27 - 20:19

From Crossword Fiend - Matt Comments "... 147 right answers of which just 27 were solo solves.

I didn’t intend for the final step to be so difficult although I must admit that Consigliere warned it could be. I just put 2013 in Roman numerals because it’d have been awkward to use Arabic numerals in the grid. I thought the upside- down CHAMPION would stand out much more than it did — figured many would get it just from a quick scan of those 26, but it was much more hidden than it appeared.

That was obviously Week 5 so we’ll just aim for some easier fun this Friday."

[Abide]

If you don’t use the pangramming on both sides, how would you utilize 2013? Have ten one-word clues that lead to the words in grid that can be anagrammed to CHAMPION? That would be a pretty loose concept and deserving of critical response.

This instead was a thing of beauty. It was almost impossible to interpret the 2013 correctly, but 27 did. And a large number of us had written CHAMPION backwards on our worksheet, hiding in plain sight.

[Bird Lives]

I too am in debt to Aaron T on the Zoom call. As he was stumbling around to come up with a way to nudge the rest of us without giving it away, I was looking at the Excel spreadsheet I had made so that I could sort alphabetically either by clue or by entry. Then I noticed a crucial column that Excel had provided -- the numbers for the rows. I looked at Row 20 and scanned up to Row 13. Solved.

[HunterX]

If you don’t use the pangramming on both sides, how would you utilize 2013? Have ten one-word clues that lead to the words in grid that can be anagrammed to CHAMPION? That would be a pretty loose concept and deserving of critical response.

This instead was a thing of beauty. It was almost impossible to interpret the 2013 correctly, but 27 did. And a large number of us had written CHAMPION backwards on our worksheet, hiding in plain sight.

I agree that the puzzle was a thing of beauty. I would say, however, that you use the 2013 on the clues' initial letters, but not on the answers' initial letters. The fact that there were 26 one-word clues, with their initial letters comprising all the letters of the alphabet, begged the conclusion that they needed to be put in order. And the clue at 124A told you that you needed to use 2013 somehow (or possibly MMXIII). There are only so many ways to that, and the fact that you have every letter of the alphabet represented strongly suggested you somehow use the position of the letters in the alphabet (or possibly in the list of clues). The way it was constructed, you can substitute other C, H, A, M, P, etc. words in as the answers for the A, B, C, D clues and the meta answer would still be CHAMPION. The answer to "Ardor" could be a word starting with any of the letters of CHAMPION without changing the answer of the meta.

I do agree with @pgw that the fact that the answers were also pangrammatic strongly suggests a cipher. I didn't pick up on that aspect. I just saw the CHAMPION in the stretch between the 20th letter and the 13th letter and knew it was the answer. So I don't think the answers had to be pangrammatic.

[HoldThatThought]

If you don’t use the pangramming on both sides, how would you utilize 2013? Have ten one-word clues that lead to the words in grid that can be anagrammed to CHAMPION?

I am utterly confused by this comment.

The identifying quality of the 26 key clues was simply that the 26 were both the only single-word clues AND the 26 provided a pangrammic set. The additional fact that these 26 clues cued 26 pangrammic "answers" did not additionally differentiate the 26. It may have confirmed that these clues were "special", but the identifying characteristic was "one word clues that were discovered to form a pangrammic set".

Once those 26 were identified, the actual "answer" to those clues was now constrained only by the requirement that the M clue had an N answer, the N clue an O answer, the O clue an I answer, etc.

Admittedly, the selected answers were fine answers; certainly Moniker clues Name and Nightmare is a reasonable enough clue for Ordeal, but an intentional effort was made to insure, not just that M paired with N, N paired with O, O paired with I, but that the entire set of answers was pangrammic. That may be elegant, but it was not necessary to the solve, and the "look how both the clues and their answers are pangrammic sets" created a reasonable expectation that the solution relied, somehow, on a pangrammic quality.

I don't understand how "they both had to be pangrams for this meta-chanism to work" is derived. Please explain.

[Abide]

Please explain.

I'll try. What you are proposing is a 15x15 with normal number of clues, of which 10 would be one-word. (With only ten entries Matt would have no need to do a large grid, maybe a 17x17 at most.) Most puzzles of that size have 8-15 one word clues already, so for many it would be harder to get a toehold. I guess some people would pick up that the one-word clues go from N to T, but if the clues are scattered that would still be hard to see. For those who picked up on a N to T progression, it would be a simple step to anagramming the corresponding answers, which makes the 2013 step somewhat superfluous. And the whole puzzle would have been a bit boring.

The pangramming aspect set the table for the second 2013 step [i.e. T is the 20th letter in the 26 letter alphabet], which Matt admits he misjudged the difficulty on.

Rephasing the helper clue as "Year the crossword puzzle celebrated its100th anniversary; or, where to find the meta answer, going from Point A to B" would have helped. I can see the frustration getting to the second step, but taking it out on the first step is misguided (IMO).

[HunterX]

Please explain.

I'll try. What you are proposing is a 15x15 with normal number of clues, of which 10 would be one-word. (With only ten entries Matt would have no need to do a large grid, maybe a 17x17 at most.) Most puzzles of that size have 8-15 one word clues already, so for many it would be harder to get a toehold. I guess some people would pick up that the one-word clues go from N to T, but if the clues are scattered that would still be hard to see. For those who picked up on a N to T progression, it would be a simple step to anagramming the corresponding answers, which makes the 2013 step somewhat superfluous. And the whole puzzle would have been a bit boring.

The pangramming aspect set the table for the second 2013 step [i.e. T is the 20th letter in the 26 letter alphabet], which Matt admits he misjudged the difficulty on.

Rephasing the helper clue as "Year the crossword puzzle celebrated its100th anniversary; or, where to find the meta answer, going from Point A to B" would have helped. I can see the frustration getting to the second step, but taking it out on the first step is misguided (IMO).

There seems to be a misunderstanding here. Neither I nor @HoldThatThought were suggesting that the CLUES (sorry for the all caps emphasis, it works better than italics) not be the full 26 with all letters of the alphabet. It was that the ANSWERS to those clues didn't need to contain all the letters of the alphabet. I agree that having only 10 one-word clues that went from M to T would have been significantly more difficult to recognize, if that mechanism would even be recognizable at all. I might even consider that an unfair mechanism.

Having all 26 letters represented in 26 one-word clues was necessary, in my opinion. Having them all represented in the answers was not.

Does that clarify our point?

[Al Sisti]

I did see the pangrammatic clues/entries (and hence had also had written the answer in my notebook without noticing it), but I also spent too much time comparing Matt's clues to Wynne's 1913 puzzle:

Matt's "saharan" [HOT] and "dry" [XERIC] --> Wynne's SERE;
Matt's "Jungle creature" [APE] --> Wynne's LION
Matt's "Have a ___ for news" [NOSE] --> Wynne's "Part of your head" [FACE]
Matt's "Youth" [KID] --> Wynne's "A boy" [LAD]
Matt's "Chesapeake Bay catch" [CRAB] --> Wynne's "Found on the seashore"
Matt's "Just" --> Wynn's "Such and nothing more" [MERE]

When nothing made sense (and given the inconsistencies of Matt's clues being answers to Wynne's entries and vice versa), I almost sent in "Wynner/Winner" as my hail Mary. And only now can I see how close that was to the answer.

[HoldThatThought]

Wait wait. I'm not talking about making any structural changes to the grid. Envision the exact same grid, or, more accurately, a very similar, super sized grid, with lots and lots of fill, much like this one.

Imagine, also that there are 26, and just 26 clues of one word; in fact, there's no reason these 26 clues in my alternate world grid can't be exactly the same as the ones used in the puzzle we're discussing. Or they can be different, if you prefer, as long as the full set of 26 is pangrammic.

Imagine, however, that the answer set was not pangrammic; of course, that means that some of the answers were different than the ones Gaffney used.

Solvers would still have thought it potentially useful to make a two-column list of clue/answer pairs, and, further, that it might be helpful to alphabetize this list (after all, pangrams, right?)

My point is simply that NOIPMAHC could have appeared consecutively in the reordered list of answers (reordered by alphabetizing the clue words), without any need for the answer set to be pangrammic. The MMXIII clue would still have directed solver's attention to list positions 20 through 13, and the answer still would have been CHAMPION.

The only "modification" needed to assure this was to be sure to make the T Clue's answer a C word, let's say, for the sake of argument, that the T clue had been "Talk", and the answer "Chat". (And so on and so forth for the other 7 critical clues, as well)

This arrangement could be achieved without there being a Gag/Joke pair, or a Dry/Xeric pair, both of which are irrelevant to the solution, and exist only to make the full answer set pangrammic.

[Bird Lives]

Neither I nor @HoldThatThought were suggesting that the CLUES ... not be the full 26 with all letters of the alphabet. It was that the ANSWERS to those clues didn't need to contain all the letters of the alphabet.

The pangram of entries was not necessary, but I thought the symmetry made for a certain elegance. Yes it suggested a simple substitution cipher, and I wasted a lot of time in that rabbit hole. But not only did P=P, but whether you went from clue to entry or the reverse, some very frequent letters (A, E, T) coded very infrequent letters like Z.

[Abide]

So if what you're saying is 26 one word clues, but only 10 one-word answers that spell CHAMPION. that still could be accomplished in a normal grid or 17x17. And your letter string would yield a lot of random letters except for CHAMPION. [kinda like what happened in the original].But you wouldn't want to use many of the CHAMPION letters outside of 13 to 20. And you wouldn't want to spell out any other words. And you might have to change the title, since there are now a lot of single words in the grid instead of a defined set.

That would have been much easier for Matt. But you still have to pick out CHAMPION backwards, which as we know is not intuitive. So he had two choices, and took the superhuman route. I can't see a nit for that.