What’s a seven-letter word for an action causing a general public outcry? The answer (“scandal”) is on the minds of many in the crossword community this weekend, as FiveThirtyEight reported a substantial case of puzzle plagiarism last night.
Using a database of tens of thousands of crossword puzzles created by Saul Pwanson, cruciverbalist (crossword puzzle maker) Ben Tausig discovered in late February that Universal Uclick—a company that distributes the syndicated Universal Crossword to numerous newspapers and websites—re-ran a 2004 crossword he wrote for them in 2008 under a pseudonymous byline. Pwanson then wrote a script to uncover any puzzles in the database that were at least 25 percent similar; in doing so, he revealed that Universal Crossword editor Timothy Parker may have plagiarized dozens of other cruciverbalists’ work and recycled even more of his own puzzles under pseudonyms. The discovery has sparked outrage across the crossword community, and for good reason: constructing a crossword puzzle can be as difficult as solving one.
Had a xword in USA Today 2004. They tweaked it in 2008 w/pseudonym. Then ran it AGAIN in 2015. Seedy business indeed pic.twitter.com/MwIe70YIs6— Tausig Ben (@datageneral) February 25, 2016
Crossword puzzle construction is governed by established rules set by organizations like Simon & Schuster and the New York Times. Sam Bellotto, Jr. who developed the crossword construction software Crossdown, explains the standard Simon & Schuster rules on the Crossdown website. There’s a standard set of grid sizes—15 x 15 is the most common, with 21 x 21 grids used in Sunday editions—and grids should be diagonally symmetrical. Words must have at least three letters and each square should be “keyed”—that is, each square must be part of both an Across and Down answer. The New York Times explains that cruciverbalists appreciate the constraints, as constructors must be clever in their attempts to work within the set boundaries. Puzzles usually have a set theme (for example, “Criminal Acts”) that helps guide word selection, and modern cruciverbalists have software available to help suggest words to fit specific spaces.
It’s the practice of “theming” that has led to trouble for Mr. Parker. 65 of Parker’s puzzles were found to replicate themes previously used in The New York Times, among others. In FiveThirtyEight’s report, writer Oliver Roeder describes these cases as “shady” duplications by Parker. Here’s one example:
On Jan. 8, 2001, The New York Times published a puzzle with three long theme answers (“DRIVEUPTHEWALL,” “GETONONESNERVES” and “RUBTHEWRONGWAY”) that all had clues containing the world “exasperate.” On June 4, 2010, USA Today published a puzzle with the same theme and the same theme answers in the same order, with the same placement and clues that all included the word “exasperate.”
The idea of one of the widest-syndicated puzzle distributors lifting other cruciverbalists’ hard work is understandably upsetting to those in the community. New York Times puzzle editor Will Shortz told FiveThirtyEight, “To me, it’s an obvious case of plagiarism. It’s unethical, and I would never publish a person who plagiarizes another person’s work.”
This is not the first time the issue has come up.
In a 2009 Slate article, Matt Gaffney used an incident where he unknowingly replicated a crossword theme previously used by Mike Shenk to tease out the crossword community's guidelines on plagiarism. Unlike the situation with Parker, Gaffney had merely used a few of the same themes by coincidence—the clues were entirely different—but pointed out that he had not copied, or even seen Shenk's crossword. Plagiarism is a big deal, he explains:
Just as in journalism or literature, plagiarism is reviled in the crossword world. The community of Americans who write crosswords for major publications is tiny, perhaps 300 people, and most of us know one another. Someone who ripped off themes from other writers would not be able to get away with it for long.
In 2013, a user of public question-and-answer site Quora asked if crossword clues and answers were proprietary, and if legal issues could arise from republishing them. Cruciverbalists and copyright experts weighed in, stating that while citing a puzzle for a database (such as Pwanson’s) would likely be acceptable, wholesale reproduction of puzzles without attribution was certainly copyright infringement.
So far, Parker denies the allegations of plagiarism, and Ben Tausig is concerned that USA Today and other outlets that use Parker’s work will ignore the situation until it blows over. Therefore, he is tirelessly unearthing evidence that Parker has previously been confronted about the issue, sharing his findings via Twitter. Unfortunately, lawyers have told FiveThirtyEight that even outright plagiarism like the example provided above may be in a legal grey area. A vaguely-defined “substantial” amount of clues and answers would need to be reappropriated to qualify as actionable infringement. In the meanwhile, Tausig and other cruciverbalists continue to search for answers to this puzzling case.
Update, 3/11: We changed the wording of a reference to a 2009 Slate article because it incorrectly suggested an apology for an accidental duplication of theme entries.