Murasaki Shikibu, Death Note, and A Study in Scarlet

            Detective fiction is super neat. Sherlock Holmes remains a household name for a reason—a well-written mystery holds nearly universal appeal, and while he was not the first fictional detective, Sherlock is the symbol of the entire genre. Reading or watching the great detective reason his way through cleverly-plotted cases is, quite simply, really fun.
            As detective fiction has developed, however, writers have taken to using its techniques and tropes for purposes beyond pure entertainment value. I’ve written about this in the context of other works--Kara no Kyoukai and Danganronpa V3 in particular—but a while back I came across a particularly interesting representation of how the genre has grown over time.
            Madoi Ban is a Japanese mystery writer who also works as a member of Fate/Grand Order’s main writing team. He recently wrote a mystery-themed event called Analysis of the Perplexing Meihousou. The event was presented as a frame story, with Murasaki Shikibu—author of The Tale of Genji—writing and directing a movie, assisted by Sherlock Holmes’s arch-rival James Moriarty. Because it’s a Fate work, all of the actors are historical or mythological figures, ranging from the composer Antonio Salieri to the Egyptian pharaoh Ozymandias. It’s a fun time.
            The story’s setup places the filmmaking process on a strict deadline, and the event jumps back and forth between the “in-frame” story that’s being filmed and the “out-of-frame” events surrounding the filming itself. Early on, Murasaki collapses while on-camera in what was intended to be a small director cameo role. With the writer and director down for the count, the actors (and assistant director Moriarty) are forced to improvise, and the “in-frame” story develops into a murder mystery centered on Murasaki’s character’s death while the “out-of-frame” story becomes a secondary mystery centered on her collapse and the “intended” course of her hastily-written story.
            This would be neat even on its own. Two separate mysteries playing out side-by-side, one fictional within the universe and the other real? Super cool, super unique, very on-brand for a modern mystery writer like Madoi. The story is (much to its credit) a level more complex than this, though. With Murasaki out of commission and the filming on a tight schedule, the main cast calls in the assistance of three experts—Hans Christian Andersen and William Shakespeare for writing advice, and Sherlock Holmes himself for the solving of the mystery. Hans and Shakespeare refuse to finish Murasaki’s story on principle (saying, essentially, that one author should not impose his thematic ideas onto another author’s work unless the second author expressly asks for it), but they do offer guidance on good writing to the crew and actors who frantically try to improvise through the story Murasaki might have intended.
            Hans in particular carries a special significance within Nasu Kinoko’s universe, as he serves as something like an author insert. His character has little basis in the historical Hans Christian Andersen but rather is drawn from Nasu’s own quirks, and when he appears it is often in the de-facto role of speaking for the author. Madoi uses Shakespeare similarly, as his own voice when he decides to speak about writing more generally within his work in FGO.
            As such, the advice that comes from Hans and from Shakespeare ends up being something resembling the philosophy Nasu and Madoi espouse when writing mysteries. Nasu is not a “mystery writer” in the traditional sense, but his writing owes a great debt to the techniques of detective fiction (and particularly to Edogawa Ranpo and Ayatsuji Yukito), while mysteries are truly Madoi’s specialty. There are a lot of neat quips about writing throughout the event (one of which I’ll return to in a bit) but for now suffice it to say that Moriarty and the actors take their advice—with some tips about deduction from Sherlock—and start forming their own ideas for how the in-frame mystery should end.
            Bumbling through the first few scenes, the actors end up establishing a set of facts on which the murder must be based. Character backgrounds are revealed, a murder weapon is discovered, a second murder occurs, and so on. None of this in-frame action is the result of any out-of-frame plan for how the mystery should end, and it seems for a time that the story is headed to a dissatisfying and chaotic conclusion. Surely a mystery not planned in advance cannot resolve satisfactorily, yes?
            This setup leads to one of the most creative concepts I’ve seen in a mystery: specifically, as the in-frame story approaches its conclusion, the out-of-frame actors convene to determine how the in-frame mystery should end. We then get five alternative theories, with the out-of-frame characters reasoning their ways to different resolutions of the mystery, all consistent with the (ostensibly) arbitrarily-produced facts. These sections read like chains of logical deduction—like a detective explaining a murder and pointing out the true culprit—but none of the solutions are inherently correct or true. They are five entirely different, entirely satisfying ways to resolve the same setup.
            This section of the story calls to mind the multiple endings of the movie Clue. Take away the frame story setup and it functions much the same: multiple solutions to the same mystery. Meihousou’s justification for its multiple endings is, of course, somewhat more thorough, but the audience experience is more-or-less the same. You effectively get the “payoff” for several mysteries despite only needing the “setup” for one. It’s economical if nothing else.
            Where this takes a step beyond being just a tighter Clue is in the progression of Meihousou’s alternate in-frame endings. The proposed endings themselves mirror the development of detective fiction as a whole. The first is mostly straightforward, filled with twists and surprises but mostly focused on the direct plot and the killer’s motive. The second builds on the first, taking much the same style but adding stronger character interaction, emotional weight, and thematic purpose. The third sheds much of the actual “mystery” and instead uses the trappings of a mystery to tell an otherwise unrelated story in the vein of more “serious” literature. The fourth returns to the mystery-plot focus but violates the expected rules of detective fiction in an extremely cool way. The fifth breaks from traditional mystery storytelling further by presenting the out-of-frame mystery as the true mystery and proposing that the in-frame mystery is just a red herring to divert attention from the actual plot—the solution involves including footage of the filming process itself in the final film.
            In these five stories we see the transition from standard plot-centric detective story (as in Doyle’s original Sherlock Holmes stories), to “non-mysteries” that subvert the genre towards a larger goal (as in the works of Edogawa Ranpo and Nasu Kinoko), to meta-mysteries that break the established rules of the genre to great effect (as is characteristic of Ayatsuji Yukito and of Madoi himself).
            Meihousou is a fantastically entertaining mystery—a collection of mysteries, really—but even beyond that it’s a lesson in the structure and history of detective fiction, and it’s hugely effective in that regard. It’s also one of the few Fate/Grand Order events to have been published in traditional book form, for good reason.

Reasoning Backwards

            One of the more interesting observations to come out of Meihousou’s advice on the writing of mysteries comes in the form of a recommendation to plan backwards. As one of the characters struggles to produce a satisfying conclusion to the in-frame story, he receives some advice from Sherlock and the author duo. The advice, in essence, amounts to this:
            It is basically impossible to deduce things in the manner Sherlock does, and even more so when producing a work of fiction that (by necessity) does not have a true answer already waiting to be found. A writer is not a detective, and as such does not come at a mystery’s solution by reasoning forward. Instead, when writing a mystery, it is helpful to think first of an interesting twist or solution and then to develop a set of clues and deductions that lead to that intended solution.
            In other words, to write a mystery, decide on the answer and then create the clues that lead to that answer, somewhat like designing a puzzle. This process allows a writer to create the illusion of cleverness while in fact being in control of the situation from the beginning. A writer doesn’t need to deduce a killer’s identity—the author needs only to create tools that might lead his or her capable detective to deduce things the author already knows to be true in a way that strikes the reader as plausible.
            Perhaps this bit of advice is obvious—of course the author knows everything, right?—but it’s nonetheless a potential sticking point for someone new to mystery writing (as is the hapless Sir Tristan in Meihousou). An inexperienced writer may have an intriguing setup or interesting mystery in mind, but no clue how to go about solving it. Writing backwards solves the problem.
            While Madoi expands on the idea, the advice actually echoes something from Sherlock’s debut work, Sir Arthur Conan Doyle’s A Study in Scarlet. In explaining his deductive process in this original work, Sherlock talks about reasoning backward—looking at a conclusion and deducing the steps that must have led to that conclusion. In the context in which he means it, Sherlock refers to the process of coming upon a crime scene and gradually piecing together the events that must have led to its happening. This is standard for murder mysteries, of course, though in light of Madoi’s advice in Meihousou it takes on a slightly different significance.
            Not only was A Study in Scarlet (along with Doyle’s later writing) likely written using the process Madoi outlines, it is presented to the reader in exactly that fashion. Sherlock almost never presents a series of facts and deductions leading to a conclusion—he does the reverse, presenting the reader (and nominally Watson) with a conclusion, and only after great prodding explaining the observations and deductions that led him there. His deductions are presented backwards, just as they must have been written. Doyle tells us “the culprit came by cab” and then only afterward gives the reasoning why Sherlock would know that to be true. Other mysteries do this as well, but it’s most readily apparent in Doyle’s works as Watson frequently draws attention to the practice.
            On the surface, the reversed order seems to be primarily a way to characterize Sherlock as brilliant, eccentric, and more than a little impatient. He’s the clever student in a math class who doesn’t like to show work because the answer is “too obvious.” It works well, and Doyle’s Sherlock is, if nothing else, fairly uniform in his personality.
            Beyond that, though, the flipped presentation serves to do that at which detective fiction is most adept: tricking the reader. I don’t mean this in the sense I might in the context of more modern works—the flipped reasoning doesn’t lead the reader to mistaken conclusions—but it does create an illusion of brilliance where said brilliance doesn’t truly exist.
            Placing conclusions before deductions leads the reader to accept leaps in logic he or she might otherwise question. An example of this comes very early in A Study in Scarlet, where Sherlock identifies a passing man as a retired marine sergeant. Without the foreknowledge that he is correct, his logic seems shaky. Sherlock assumes he is a seaman because of a tattoo of an anchor—a reasonable guess, sure, but not a foolproof one. He assumes the man is from the military due to his haircut—a somewhat stronger deduction. Finally, he concludes the man must be a sergeant because of… his age and the way he carries himself? This last piece in particular is a strange logical leap that would be difficult to accept without the certainty that Sherlock is, in fact, right.
            And yet, when reading the piece front to back, nothing about the exchange jumps out as odd. Sherlock makes his claim, which Watson questions, asking the man to verify Sherlock’s intuition. When the man confirms Sherlock is correct, Sherlock is able to present what would otherwise come across as strange guesses as superhuman insight. Watson’s response, impressed by the reasoning, leads the reader to accept it as presented. It’s quite clever.
            Essentially, the backwards reasoning means Doyle does not need to produce chains of deduction that would actually make someone a brilliant detective, as the reader’s assumptions that Sherlock is correct fill in the gaps. We don’t question Sherlock because, well, he’s Sherlock. He might as well be a superhero (a character type with which he in fact shares many qualities, right down to his surpassing physical strength).
            There is an element of suspension of disbelief involved here as well. Even after noticing the holes in Sherlock’s logic, readers are able to enjoy watching him work because the writing is structured so well. You have to actually stop and dig into the logic for it to ring false, and to do so is to do the one thing you should never do—question Sherlock Holmes. The direct steps of Doyle’s logical chains don’t matter nearly so much as the fact Sherlock is the one producing them, and the writing moves so quickly it’s quite easy to just accept that Sherlock is right and move on.

L is not Sherlock

            Roundabout as it seems, all this background brings me to the main point of this post: Death Note.
            It’s related, I promise.
            Death Note, for the unfamiliar, is essentially a semi-supernatural crime drama. There is one single supernatural component with clearly-defined rules—the death note itself—and everything else adheres to the rules of a “normal” crime drama. The criticisms I’m about to point at Death Note can be written off by saying I’m applying mystery logic to a work that is at its core a thriller and not a mystery, and while that’s a fair point, Death Note owes enough of a debt to detective fiction that I think it’s worth acknowledging the ways it misuses the science of deduction.
            Most of Death Note focuses on the interplay between its villain-protagonist, Light Yagami, and its hero-antagonist, L. The story begins with Light finding a journal that allows him to cause people to die by simply writing their names, and he uses the journal to undertake a sort of vigilante justice, killing criminals who he views as not being punished effectively by the law. This gets the attention of the police, who eventually call in the eccentric, brilliant detective L to identify and capture the strange killer.
            The excitement of Death Note comes in the form of the back-and-forth between the two, as L tries to close in on Light and Light outsmarts him again and again. L is nearly always almost able to catch Light, which forces Light to take more and more extreme measures to avoid capture. It’s pretty entertaining if you don’t think about it too hard..
            The core issue I have with Death Note lies with the deductive prowess exhibited by both L and Light as they try to predict and outsmart one another. On multiple occasions, both characters—but especially L—make assumptions about the behavior of the other and then act on those assumptions. Light’s are usually pretty well-justified, as he has ways of gathering information about the individuals targeting him, but L’s tend to be a bit… out there.
            As in the case of Sherlock’s deductions, L’s assumptions are nearly always correct. Unfortunately, L’s reasoning rings false far more often than Sherlock’s does. L seems at times to operate off of random guesses that strain suspension of disbelief, and the only reason they don’t come across as stranger is because the viewer already knows them to be true. As we see the criminal’s actions directly, L can simply assert things that we already know to be true and his insights seem brilliant rather than random because we know in advance that they are correct.
            This is a misuse of the backwards reasoning that characterizes Sherlock Holmes.
            Sherlock’s backwards reasoning works for two primary reasons: first, the reader and the narrator both believe Sherlock to be nearly infallible and thus trust him implicitly even when his direct explanations don’t quite hold up, and second, the reader does not already know the answers to the mystery when Sherlock presents his deductions.
            These seem somewhat contradictory. The first points to the deductions working because we know them to be true, while the second implies they work because we don’t know them to be true. What gives?
            The answer has to do with the source of our information. Sherlock works as a detective because he is both a reliable source of information (the former point) and our only source of information (the latter). We do not inherently know that he is correct, but we are adequately prepared to believe he is correct.
            L falls flat on both counts. As we are more-or-less omniscient when reading or watching Death Note, we know what L does not know, and so we know that he is not omniscient and is occasionally incorrect. We know that L is not a completely reliable source of information (nor is he meant to be), and he is also not our primary source of information.
            This isn’t inherently problematic, and were Death Note less mystery-esque in structure L’s weakness as a detective character would not be a weakness with the work overall (and might in fact be a strength). Unfortunately, Death Note wants to present L as a Sherlock-esque figure. It wants the illusion of brilliance Doyle achieves, but it lacks the foundation required to enable it.
            The result is that L’s deduction sequences feel hollow. When he makes a brilliant “deduction” that leads him close to catching Light, if seems unfair rather than impressive. His hunches are correct far too often without the character buildup or the plot-driven smoke-and-mirrors needed to cover it. He feels like he has the analytical equivalent of plot armor—all it would take is one or two wrong guesses to send him off the trail and leave Light free to do his thing unmolested, but L seems inexplicably immune to major errors (for the most part). His reasoning is weak and his assumptions often take the form of pretty wild logical leaps, but he gets away with it because we know he’s right.
            Again, maybe this isn’t a problem. Maybe a thriller’s detective can afford a flimsier foundation than a mystery’s detective’s. I think it makes the overall work feel quite contrived, though. Suspension of disbelief carries me through Sherlock’s trains of thought without issue, but it runs out somewhere before L’s.
            Death Note wants to be a competition between great minds, a battle between Sherlock and Moriarty that follows both and consequently leaves its outcome uncertain until the very end. It’s a noble attempt, and Light in particular is pretty well-executed, but L rides on his plot handicap a bit too much, and I find the end product stretches believability.