So how do all these genres add up? What’s the picture of crosswords today?
To answer that question, I’ve taken a small survey of 100 puzzles from this blog (August of last year). Just a few weeks ago, I was informed of a much larger survey by Kevin McCann, covering 2700 puzzles, which promptly disabused me of the cherished notion that I was the first to categorize crosswords on a really grand scale. McCann runs Cruciverb.com, which draws its survey data from more conservative sources than the Fiend.
McCann’s divisions and terminology sometimes disagree with mine, and I feel like this series already threatens to be too confusing. So I’ve reworked McCann’s categories into the ones in this series, as well as I can. Results are imperfect, but we’ll glean as much information as we can from them… and review the categories we’ve established at the same time.
McCann’s Data: Major Categories
Crossword Fiend Sampler: Major Categories
Themelesses might be underrepresented in McCann’s survey, which claims to be about “theme usage,” but includes four categories I identified as themeless. You might recall that we divided themelesses into four types: the artily-constructed Mieses, the vocabulary-driven seed patches, the homogenized puddings, and the wide-open, wildly-answered anything-goes.
McCann’s “Grid Design” category corresponds closely to the Mies, but his “stacked 15s” could be a Mies or seed patch, depending. Unlike my survey so far, he notes the “mini-theme,” two related entries in an otherwise themeless puzzle. And I’m putting his “other” in the seed patch category, where it may or may not belong… but it seems the most likely match. There were no puddings or anything-goeses in the blog survey, and there likely weren’t many in McCann’s data, as anything-goeses are barely extant and puddings rarely make the papers.
There’s an additional problem defining the Mies: the standards for puzzles keep changing as the software, and the use of software, improves. Once upon a time, two double stacks of 15s might qualify as an unusual, artsy grid, but today that’s more likely to register as a seed patch. For that reason if no other, the Mies will probably remain on the frontier, the seed patch at the core of the puzzle fans’ themelesses, and the pudding in the amateur ring. I continue to hope the anything-goes gains more traction among puzzles’ avant-garde, but we’ll just have to see.
Fact-finders make a respectable showing in both polls, coming in third and second, respectively. The commemorative (defined by McCann as both “tribute” and “holidays and special occasions”) naturally doesn’t see as much usage as the less time-dependent listmaker, corresponding roughly to the largest single category in McCann’s original survey, which he defines as “category types.” There’s no way to sift out the half-fact-based, half-wordplay-based Marian from McCann’s categorizations: some Marians might fit “category types” and others his “combined” division.
McCann’s Data: Subcategories
Crossword Fiend Sampler: Subcategories
The Fiend surveys several sources that supply nothing but themelesses. But surprisingly, themelesses aren’t even the top subcategory of crosswords in its survey– that distinction goes to the camouflage, which has developed in a wide variety of ways to hide everything from phonetic strings to anagrams to fact-themed groupings to rhythmicpatterns to accompaniers to monostrings. They’ll hide these all in one piece, or in merger, parenthetical, initial or blended configurations. Along with the similarly varied algebraics and freakshows, and the relatively straightforward punsters, they make up the wordplayer category, easily the most popular of the lot in either survey.
Algebraics further subdivide into the fairly self-explanatory additives, subtractives, substitutionals and cut-and-pastes. Freakshows split into pure accompaniers, pure anagrams, monoclues, straight-answer patterned-clues, pure etymologicals, pure cryptogrammatics, pure palindromics, theme letter banks, homogenous letter banks and chimeras.
Punsters split right into the Groucho and the Burns and Allen, but the former further splits into root-phrase, conversion and clue themes. McCann’s divisions don’t always observe the Groucho/Burns and Allen distinction, and a couple of them merge a camouflage and freakshow category (anagrams, accompaniers). But all in all, it maps rather well, and both surveys seem to be in agreement that the wordplayer in all its varieties is doing fine.
McCann’s Data: Subsubcategories
Crossword Fiend Sampler: Subsubcategories
Run-ons and tricksters make just minor appearances in each survey– 7% and 5% in McCann’s data, and 5% and 4% in the Fiend’s. Among the run-ons, McCann doesn’t distinguish between quotes and improvs, and I’m not entirely sure of myself when I map his “punchlines” category to narratives. Riddles make a small appearance in the McCann data, but not multi-parters, and neither of those are anywhere to be found in the Fiend sample.
Among tricksters, both surveys cite rebuses as one of the most popular subgenres, with the Fiend putting them neck-and-neck with the Moriartys (from a very limited sample) and McCann putting them clearly ahead, with Vector Threes in second place and Moriartys nowhere to be seen. McCann’s data also has nothing that maps to a single-shape or off-grid Vector Three, though, only a multi-shape one. Had I included the contest crossword in the Fiend survey it would’ve represented 4% of the total, but since every contest crossword is also part of another genre, I thought it best to leave them out. The McCann data doesn’t mention contests either. With only a few examples extant, the variable crossword makes neither survey. And forget the squeezeboxes and vowellesses.
While you’re at it, forget all the mutants. Cryptics, rows gardens, snake charmers, spirals, marching bands, petal pushers, stacked grids, cubic crosswords, acrostics, vocabs, fill-ins, diagramlesses, Blankouts, diagramless fill-ins, split decisions, crossnumbers, gryptics, word squares, double word squares, arrowwords, barred crosswords, barred and boxed crosswords, cipher crosswords, and experimental crosswords? These surveys don’t serve their kind. Most of these are kept alive in markets that the New York Times solver finds unusual– either foreign countries as in the cryptic and arrowword, big sections for the puzzle lover as in the acrostic and diagramless, or individual keepers of the fire as in the split decision and gryptic. Only the vocab puzzle has a wide American audience, and advanced solvers are inclined to discount it.
It seems fair to say that the American solver today likes a little extra cleverness in his puzzle– hence the greater popularity of the wordplayer, compared to the more straightforward fact-finder and nearly generic themeless– but not too much, hence the minority status of tricksters and mutants. The particular challenges of the run-on format make it as demanding of the solver as some of the more advanced forms, but it’s a rare one that can match them for cleverness. I suspect they’re only more popular than tricksters because they’re easier to build and their gimmick is easier to communicate without giving the whole game away.
Crosswords are at a crossroads. The traditional newspaper, long their greatest source of support, is in trouble itself, and slashing production budgets left and right. Will Shortz has preserved the New York Times crossword’s brand to the point where the paper invests significant resources into it, but it’s the exception that proves the rule. Independent crossword-makers have attracted devoted but modest followings.
Will the future of the field move more to the themeless or to the mutant side of the spectrum? More puzzles for all comers, or more specialized exotic beauties for the aficionado? Will crosswords be married to their current demographics or come to aggressively court a younger generation? Will they continue to encourage, and reward, knowledge of the current wide variety of trivia topics, or will they grow more specialized, more slanted?
I wish I had any answers, but hard experience has taught me that grand predictions go stale very quickly. All I can do is all that any of us on the creative side can do– continue to invent.
Next week, in our grand finale, we’ll look into the experimental crossword a little more closely– and discuss how to make the genres of tomorrow.