The phrase “Late Queen Problem” is itself problematic. As I said above, the debate was triggered by the 1995 essay “Early Queen Theory” (shoki kuīn ron) by mystery novelist and critic Norizuki Rintarō, published in the Japanese journal Contemporary Thought. It focused primarily on the novels The Greek Coffin Mystery, The Tragedy of Y, and The Siamese Twins Mystery, all written by Dannay and Lee, in the years 1932 and 1933.

The terms “Problem” and “Theory” in these titles are also confusing. In English, problem generally signifies a difficulty or challenge. If so, what’s the problem? While some have argued that Norizuki’s article proved that fair play in detective fiction is impossible, that view is far from universal. The word mondai (問題) can refer to a question (such as in a quiz), a societal or political issue, a controversy, or a difficulty. But it may also refer to a discussion or debate. Similarly, ron (論) is usually translated as “theory,” but can also mean “dispute,” “discourse,” or “treatise.” In my view, the “Late Queen Problem” may contain some controversial ideas, but it doesn’t constitute a “problem” per se, and while Norizuki’s essay offers a variety of fresh ideas and theories, it doesn’t constitute a unified group of propositions in the same sense that we use the term in English (such as in “Einstein’s General Theory of Relativity”). Nevertheless, for this article we’ll stick with the translations “Early Queen Theory” for Norizuki’s initial article, and “Late Queen Problem” for the wider response it generated.

Another problematic term is Honkaku, a concept which is central to this discussion but doesn’t easily translate into English. Honkaku is understood as meaning “authentic” or “traditional,” but as used by Norizuki and others in the context of the “Late Queen Problem,” Honkaku specifically refers to detective fiction that adheres to the rules of fair play (in which the author provides all the relevant clues needed to solve the mystery before the detective reveals the solution).

The figure of “Ellery Queen” presents an interesting problem that, as we’ll see, is very relevant to this discussion. “Ellery Queen” was the pen name used by Frederic Dannay and Manfred Lee from 1929 to 1971. But prior to 1936, their identity was unknown to the public. The author of the novels was only known as “Ellery Queen,” which is also the name of the hero-detective character in the Queen books. Since this article only concerns books written prior to 1936, I will generally refer to the author of the Queen books as “Ellery Queen” or simply as “Queen” (rather than “Dannay and Lee”). The fictional detective-hero Ellery, son of Inspector Richard Queen, will be referred to as “Ellery” or occasionally, in the Japanese fashion, as “Detective Ellery.”

A few important caveats: Norizuki and others have taken some liberty in applying philosophical principles to the works of Queen. Specifically, Norizuki uses Russell’s “Type Theory,” Gödel’s “Incompleteness Theorems,” and the concept of metalanguage. Much of this is based on the work of Kojin Karatani as well as from Douglas R. Hofstadter’s Gödel, Escher, Bach. Whether or not these theories were intended to analyze detective fiction, I leave up to the reader. I’m not versed enough in theoretical math or logic to pass judgement, but I found the ideas compelling, and they’ve enriched my appreciation of the works of Ellery Queen.

Several of the interpretations question the very idea of Fair Play. I take this with a grain of salt, particularly because several of those who have made such claims have gone on to successful careers writing detective stories based on Fair Play plotting. Rather than marking the death of traditional crime stories, the “Late Queen Problem” seems to have triggered a renaissance with various writers experimenting with and expanding on the plot-devices of Queen.

Norizuki’s article shook the Japanese mystery world. Debates, discussions, and disputations occurred on university campuses, in coffee shops, and online. A year later, Kasai Kiyoshi addressed the issue and suggested that this “Late Queen Problem” marked the beginning of the “Third Wave of Honkaku Detective Fiction.”

Form-Fitting Fiction

Willard Huntington Wright (who also wrote as S.S. Van Dine) was the first to formalize detective stories. An art critic immersed in the avant-garde world, Wright described detective fiction as “a complicated and extended puzzle cast in fictional form.” He went on to say that detective fiction “has set up its own standards, drawn up its own rules, adhered to its own heritages, advanced along its own narrow-gauge track, and created its own form and technic.”

The Barber Paradox Mystery and the Invention of Queen

When Dannay and Lee set out to create their own detective, they drew on Van Dine’s popular “Philo Vance” series, and ultimately improved on the model. “He influenced us because he made so much money,” Dannay once admitted, but added, “the kind of thing he did appealed to us in those days. It was complex, logical, deductive, almost entirely intellectual.” Biographer Francis M. Nevins wrote, “In short, no major mystery writer anywhere was more influenced by S.S. Van Dine than was Queen.”

S.S. Van Dine puts a twist in his storytelling by having the adventures of Philo Vance narrated by Philo Vance’s friend, coincidentally named “S.S. Van Dine.” By using his own pen name as the identity of the first-person narrator, Van Dine was free to interact with the reader, providing observations and frequent footnotes, from within the novel.

Dannay and Lee took this a few steps further. They used their pen name – “Ellery Queen” – for their hero detective as well as the third-person narrator of the novels. A close reading of the books’ forewords (by a fictional J.J. McC) and in the “Challenge to the Reader” makes this fact clear. The following, for example, is from the “Challenge” in The Dutch Shoe Mystery (emphasis added):

Translated into English, it means something like: “The set of all sets can only be a member of itself if it is not a member of itself.”

A popular analogy is the “Barber Paradox”:

Assuming the barber is shaved, he’s in the paradoxical situation of having to shave himself, while not being able to shave himself according to the rules.

Russell offered a solution to his paradox called “Type Theory” which eliminates the vicious circles of paradox by requiring that sets be ordered into a hierarchy of types. Using the analogy of the Barber Paradox, the barber is of a different “type.” He exists on a different hierarchical level than the rest of the townspeople, so the rule “who shaves all and only those who do not shave themselves” doesn’t apply to him.

A detective story is like the village in the Barber Paradox. The narrator of a detective story exists on a higher level than the characters within the story. The author and reader of the story exist on an even higher level. When a character from one hierarchical level trespasses into another, paradox occurs. Norizuki refers to these intrusions as “Logical Type Confusion” or “Logical Type Transgression.”

In Golden Age detective fiction, the most famous “Type transgression” occurs in a particular novel by Agatha Christie in which the narrator turns out to be the killer. (I don’t need to name the novel. If you don’t know what I’m referring to then you haven’t read enough Christie).

(As an interesting sidenote, Van Dine and Queen had different opinions about that particular Christie novel. Van Dine called the plot “hardly a legitimate device of the detective-story writer” and in his “Twenty Rules” referred to it as “bald trickery, on par with offering some one a bright penny for a five-dollar gold piece. It’s false pretenses.” Queen, on the other hand, reviewing this book in the first issue of Mystery League, said that the book was “Just short of excellent. Deserves classic rating on basis of its surprise solution.” Regarding the fair play of the book, Queen gave it a score of 10 out of 10.)

The Work can be seen nested within the Meta-level:

Note that on the left side of the triangle, both the Reader (outside of the work) and Detective (within the work) are engaged in the parallel tasks of solving the mystery. On the right side of the triangle, as though on the opposite side of the chessboard, Author and Villain are in opposition to Reader and Detective respectively.

When a character trespasses from one hierarchical level to another, Norizuki calls it “logical Type confusion” or “logical Type transgression.” We’ve already seen an example of this in Queen’s work when the Author and Detective, as well as the Narrator, are all the same person.

Norizuki refers to the structure of Greek Coffin as an “infinite meta-ladder,” suggesting the possibility that after a series of incorrect solutions, what if the fourth solution is also incorrect. In his book Ellery Queen Theory, Iiki Yusan points out that The Greek Coffin Mystery is ostensibly (and explicitly stated in the actual book’s foreword) Ellery’s own memoir of a case early in his career. What if, a year or two after the publication of the book, new evidence was to surface proving the innocence of Suspect #4 and that a fifth suspect set the entire thing up?

We know, of course, that Greek Coffin is a work of fiction, and that the entire story is contained within its covers. But for Norizuki, Kasai, and others, the story suggests an interesting application of Gödel’s “Incompleteness Theorems.” Hofstadter (Gödel, Escher, Bach pg. 17) paraphrases Gödel’s theorem: “All consistent axiomatic formulations of number theory include undecidable propositions.” It was Kojin Karatani, beginning in 1983, who suggested that Gödel’s Theorems have applications beyond mathematics.

Much of the debates and discussions of the “Late Queen Problem” seem to focus on Gödel’s Incompleteness Theorems to the point that many seem to equate the two. I find this a little troublesome, first, because this mathematical square peg wasn’t intended to be forced into the round hole of literature (6), and second, because it overshadows so many other fascinating aspects of Norizuki’s “Early Queen Theory” and “On the Masterpieces of 1932”as well as the entire debate called the “Late Queen Problem.”

The most compelling features of the “Late Queen Problem” for me are the ways Norizuki and others explored the logic of detective fiction, and the use of Russell’s Type Theory to map the participants of a story – both within the text and outside the text – in order to understand the paradoxical plot-twists that made Queen such an important mystery writer.

I learned a lot during the six months I spent at the bottom of the “Queen Problem” rabbit hole. It has expanded my understanding of the Japanese idea of honkaku mysteries and has deepened my appreciation of Ellery Queen. Early in my research, I felt I needed to take a closer look at Queen. So over a period of a few months, I sequentially reread the first fourteen Queen novels (7) and was enriched by that experience. If I had more time and space, I’d further explore the ramifications of the “Late Queen Problem” on interpreting The Egyptian Cross Mystery, The Siamese Twin Mystery, and The Tragedy of Y. There is much more to say, but for now I beg your pardon as I remain humbly yours, incomplete.