Full Dark House (2004) - Christopher Fowler

 And to see my decree is obeyed as it should be;
To the shades I shall go in propria Personæ!

"Act 1 Finale," Orphée aux enfers, Jacques Offenbach

Tragedy has struck the London police force. The headquarters of the Peculiar Crimes Unit has been bombed and its leader, Detective Arthur Bryant, killed in the explosion. His best friend and partner, John May, takes it upon himself to find his killer, both for the sake of justice and to get some measure of closure after having his friend taken from him. He suspects that the bombing is connected to the first case they worked together, 60 years ago during the blitz. A dancer was horribly killed during the rehearsals for a new production of Offenbach’s Orphee aux enfers (one of the greatest comic operas of all time). More deaths follow among the actors, seemingly without any rhyme or reason. We follow the course of both of these investigations in Christopher Fowler’s Full Dark House.

I’ve seen authors start their series in many different ways, but this is the first time I’ve seen one begin by killing off one the main characters. Full Dark House is the first entry in the Bryant & May series (also known as the Peculiar Crimes Unit, or PCU, series). I’ve been wanting to read it for quite a while, since Fowler intended it in part as a homage to the Golden Age of detective fiction. Plus, a series centered around strange crimes with historical connections to London is just cool. I had actually intended to start with the second book in the series, The Water Room, which has a reputation as one of the strongest mysteries in the series, but since this one was in system at my library, I figured I might as well try reading a series in order for once. Was that a good idea? We shall see…

The book has a duel-narrative structure, alternating between the present and the 1940s. It’s used to good effect, drawing parallels between the two narratives and frequently using one of them to fill in gaps of information in the other. More time spent on the WWII murder plot, which was a wise choice for a couple of reasons. One is that the integration of the historical setting and the mystery is a central part of the series. The other is that, just as a general point, a mystery concerning baroque executed murders in the depths of an old theater is likely to be more interesting than the semi-procedural investigation of a bombing. (Which is not to say that the modern plot is uninteresting, it just couldn’t support the length accorded to the other plot.)

As for that other plot, it is suitably intriguing. Members of the production are being killed off in ways that reflect their roles (a dance has her feet cut off, the singer playing Jupiter is crushed by a large metal globe, etc.). There seems to be no way for anyone to have committed them as, although the theater is a labyrinthine building filled with odd nooks, crannies, and hiding places, the only door to the backstage is kept locked at all times and all the people known to have been there were accounted for. In addition, there have been sightings of a cloaked figure wearing a mask of comedy, who has potentially been stalking members of the cast. It seems likely that someone wants to see the production shut down, but as the government intends for it to boost wartime morale, it falls to Bryant and May to see that the killer is caught before that can happen.

Now, I really enjoyed this book, but the mystery plot wasn’t the reason why. Quite honestly, it had some serious problems. In the first place, it was not in any way fair. There’s no cluing to speak of and there’s a major Knox violation. It’s not unlikely that you’ll cotton on to the solution, the existence of gur phycevg jub jr'ir arire frra orsber is pretty obvious from the Phantom of the Opera parallels and another character’s ntbencubovn is hinted at in an unrigorous thematic way which is interesting but in no wise a clue. That said, none of these issues were the source of my problem with the plot. I knew going in that Full Dark House was not a fair-play mystery, but, as shall be seen, I liked other aspects of it so much that even that wasn’t going to seriously impact my enjoyment of it. No, my problem came from the way the false solution was handled. (Some may consider revealing even the existence of a false solution as something of a spoiler, but I disagree. When the detective gives their summation with something like a fifth of the book remaining, it’s a fair bet that things aren’t done yet. To discuss the issues with the way it’s handled, however, would involve going into details about the relation between the false and true solutions, details which, although they don’t spoil the culprit’s identity, would most certainly give too much away. Therefore, part of the discussion will be in rot-13. You can come back and read it after reading the book.)

The basic problem here is that the false solution is more interesting and more satisfying than the true one. But while that’s an accurate description as far as it goes, it is also overly reductive. The false solution is prompted by Oelnag'f ernyvmngvba gung gur rnpu bs gur xvyyvatf jnf pnevrq bhg va n jnl gung fhttrfgf bar bs gur Zhfrf bs Terrx zlgubybtl. Ur pbaarpgf guvf jvgu n ehzbe nobhg gur zna jub svanaprq gur cebqhpgvba, bar fnlvat gung uvf zbgure envfrq uvz va gur eryvtvba bs pynffvpny Terrpr, cynpvat uvz haqre gur cebgrpgvba bs gur Zhfrf. Oelnag gurbevmrf gung, ol pbzzvgvat zheqref cngrearq nsgre rnpu bs gur Zhfrf, va n gurngevpny frggvat, ur vf pbzzvgvat fnpevyntr ntnvafg gurz, qvfnibjvat gurz naq fubjvat gung ur pna fgnaq nf n frys-znqr zna jvgubhg gurve cebgrpgvba. Shegurezber, vg furqf yvtug ba jul ur jnf vagrerfgrq va svanapvat gur cebqhpgvba. Becurr nhk rasref gnxrf n fnglevpny naq veerirenag crefcrpgvir ba pynffvpny zlgubybtl, naq jbhyq guhf cebivqr gur cresrpg onpxqebc sbe fhpu n cyna. When this theory is introduced, we, and the other characters, can see that it has major flaws. In fact, it has to be, since it serves a narrative purpose outside of the mystery. However, despite its holes, this proposed solution has a number of excellent points. It has that element of grandeur that is possessed by many of the genre’s greatest works, it is deeply rooted in specific aspects of the characters and setting, and, most importantly, it gives meaning to the preceding events. I don’t mean that it just explains them, although it certainly does (and in a mystery that sort of meaning is absolutely necessary). Perhaps it would be better to say that it gives the events not just meaning, but also significance. There is a satisfying pattern inherent in that solution, which, were it correct, would also shed light on the characters involved.

The real solution, on the other hand, does none of these things. In it, gur pbeerfcbaqrapr orgjrra gur zheqref naq gur Zhfrf vf fgvyy qryvorengr, ohg bayl nf n jnl gb senzr gur svanapvre va beqre gb sbepr gur cebqhpgvba gb pybfr. And that’s not exactly the most interesting solution. Sure, the zlgubybtvpny cnenyyryf are striking, but without a reason to exist they just feel like window dressing. (Not to mention that it was an unreliable and bizarrely convoluted way for the culprit to achieve their goal. If the case had been investigated by someone without Bryant’s interesting methodology, it couldn’t achieve what it was meant to. It’s like in those Ellery Queen novels where the criminal counts on Ellery’s taste for complicated explanations, only without the culprit having any reason to think that there would be a “great detective” to manipulate…) But leaving that rather large flaw aside, the bigger problem is that the solution has been evacuated of everything that gave it significance. All of the thematic resonance and context from the characters that gave it “meaning” is missing. Vg jnf whfg n trarevp jnl gb senzr fbzrbar. Jub jnf senzrq be ubj gurl jrer senzrq orpbzrf na veeryrinag. On its own the true solution would have been tolerable, but following a more interesting and meaningful false solution makes it disappointingly anticlimactic.

Even without the benefit of a satisfyingly resolved mystery, however, Full Dark House has much to recommend it. First off, as I alluded to in my discussion of the plot, the characterization in the novel is superb. The dual narrative allows for a moving portrayal of the friendship between Bryant and May, alternating as it does between their first getting to know each other and May’s grieving for his friend. The rest of the cast is equally well drawn. In particular, the performers in the opera may not be quite as fully realized, but they are well sketched, with several subtle touches. Furthermore, the writing is extremely good throughout. The dialogue and description were both a joy to read. The tone was mostly comic, but at times it switched seamlessly to the menacing or the sincere. That is not an easy thing to do, but Fowler made it look effortless. Finally, excellent use is made of the historical setting, both as regards the plot and in general. Fowler thoroughly researched London during the blitz, and it shows, not in long blocks of exposition, but in all sorts of interesting historical details worked into the narrative. These make reading the novel a very immersive experience. I had a very hard time putting it down when I was reading it.

So, despite some notable flaws, I wholeheartedly recommend Full Dark House. While the mystery is weak, it succeeds admirably as an entertainment and as a novel. Having finished it, I now look forward to reading The Water Room all the more. The promise of this kind of writing and a satisfying mystery is a very exciting one indeed.

The Transvection Machine (1971) – Ed Hoch

There was a little girl,
Who had a little curl,
Right in the middle of her forehead.
When she was good,
She was very good indeed,
But when she was bad she was horrid.

-Henry Wadsworth Longfellow

In the far flung feature of the 21st century, trouble is brewing for the united states. It’s beset by anti-technology terrorists, growing tensions with the Russo-Chinese alliance (as the cold war is apparently still going), there’s unrest on its moon colony, and Vander DeFoe, the cabinet secretary of extra-terrestrial defense, has died during a routine appendectomy. That last is problematic for two reasons. One, all routine surgeries are performed by medical robots, and it’s unheard of for anyone to die during a robotic operation. Second, and more importantly, DeFoe is the inventor of the transvection machine, which allows teleportation over great distances. It was first demonstrated by instantaneously transporting a woman from the US to India. Though still experimental, the hope was that it would allow for rapid transportation between Earth and the moon, giving the US an advantage in colonization. (This possibility is why DeFoe was appointed to the cabinet.) Without him, research on the device will likely stall out. It falls to “computer cop” Carl Crader to investigate his death. It seems impossible that it could have been murder, since the cassette tape (yes, I know…) on which the program for the operation was stored was both untampered with and located in a secure, central national medical facility. The tape's data was transmitted via telephone wire to the robot at the time of the surgery, precluding any interference. But if it somehow was murder, there’s certainly no shortage of suspects: DeFoe’s estranged wife, the ex-partner he cut out from their invention, or the aforementioned terrorists. Was it murder, or wasn’t it? And if it was, how could it have been committed? These are the central questions in Ed Hoch’s The Transvection Machine.

I started this book with high hopes. There are essentially no reviews of this book, so all I knew about it came from the blurb, but the outline made it seem like a hybrid mystery, and, as you can tell from past reviews, I’m a big fan of those. And it was by Ed Hoch, whose short stories were consistently good and who wrote some of the acknowledged masterpieces of the impossible crime sub-genre. So surely, I thought, this would be an excellent, or at least good, book. You no doubt already know how accurate that thought was. Anytime anyone starts out saying they had high hopes for something, it’s practically a sure thing that those hopes were disappointed. And this is a sterling confirmation of that trend. Bluntly, this book was truly awful. And not in the so-bad-it’s-good way either. No, it was awful in the agonizingly painful way.

As I am quite candid about the plot being my favorite aspect of mystery fiction, and as it was the plot which first attracted me to this, it makes sense to start my discussion there. As a preliminary, I should point out that this is most emphatically not a hybrid mystery. I had hopes that it would be an overlooked example, but it was not to be. As for the plot as it is (rather than as it might have been), the thing that strikes you about it as you read is how remarkably unfocused it is. For over half the book the character go back and forth (and back and forth and back and forth and back and forth…) trying to decide if it’s murder or not (while investigating in ways that only make sense if they already know that it is a murder). Now, there’s quite a lot of potential in that ambiguity, but unfortunately none of it is utilized here. It is blatantly obvious to the reader that, yes, it was murder. So why does it take the characters so long to figure it out? Well, midway through they address that very question. It seems that murders are so rare (despite the rampant violent terrorism?) that autopsies just aren’t routinely carried out anymore. This comes up when Crader, who, as the “most powerful law enforcement official since J. Edgar Hoover,” should presumably know this, asks why there hasn’t been an autopsy. And after finding this out he...doesn’t order one. Why not? It’s unaddressed in the story, but I can think of a couple of reasons. If an autopsy were performed, it would lead to the instant complete solution of the mystery. And if it were solved that soon, this would be a short story rather than a novel. It’s almost too obvious to point out, but this is terrible plotting. If your mystery story depends on the characters overlooking the obvious in order to stretch out the plot, something is wrong. And if that plot is so transparent that simply performing an autopsy would clear it up, something is very wrong. And if you, as the writer, absolutely can’t have an autopsy performed, maybe don’t set your book in a future with highly advanced medical science.

And speaking of things too obvious to point out, even after the characters definitively realize that a murder took place (over halfway through the book, mind), it still takes a frustratingly long time for them to solve it. And make no mistake, the who, how, and why are very easy to figure out. You could get close just by guessing, and with even a slight bit of effort will probably arrive at the solution, if it doesn’t just jump out at you altogether. That’s not to say that it’s well clued, as the cluing is rather indifferent. It’s simply that the plot is just that transparent. (And that the misdirection is literally just Crader saying rneyl va gur obbx gung ur qbrfa'g frr ubj gur phycevg pbhyq unir qbar vg, naq yngre ernyvmvat ubj ur pbhyq unir jvgubhg zhpu be nal arj rivqrapr.) In my last review I said that Philo Vance should be ashamed of himself for taking so long to identify the culprit, but compared to Carl Crader he’s truly a criminological genius.

The plot has other structural problems as well. It feels like parts from four or five books were chopped up and shuffled together. There are sections that feel like the sci-fi mystery we would expect, ones that feel like a political thriller, and ones that feel like a quasi-hard boiled story. There’s an extended sequence of someone escaping from prison on the moon (which could possibly have worked on its own, but which doesn’t at all fit in this novel) and another of Crader being kidnapped by the techno-terrorists on a tropical island. That last is especially egregious, going nowhere and serving only as padding. And since this wasn’t exactly a long novel to begin with, it stands out. The Transvection Machine feels like a confused patchwork of story elements that just don’t fit. 

But of course, plot isn’t everything. Good writing and characterization can save a book with a poor plot. Sadly, neither of those are in attendance here. The writing is very, very bland. It’s serviceable and nothing more. As for the characterization, it is, if you can believe it, even worse than the plot. At best the characters are cardboard cutouts, names on the page with no distinguishing features whatsoever (Crader himself is a notable example of this). At worst, they go from indistinguishable to profoundly unsympathetic. The best example of this is Earl Jazine, the creepy Archie Goodwin to Crader’s incompetent Nero Wolfe, who, despite apparently being in a relationship with Crader’s secretary, chases after every female character he comes across. He strings along the nurse who was handling DeFoe’s operation and sleeps with DeFoe’s widow, who he was sent to question and who he actively suspects of being his killer. And this is one of the protagonists!

And then there’s Euler Frost, an anti-machine terr...(you know, I’m just going them neo-ludites from now on. It’s simpler to type.) a neo-ludite who escaped incarceration on the moon, returned to Earth, and is a suspect in the murder. He (and his whole plot line) exist only to voice Hoch’s complaints about the encroachment of technology into modern life. Not, mind you, that any of its supposedly deleterious effects are ever actually shown. But some of the characters complain about it, so it must be bad. This plot element culminates in what may be the biggest misfire in the book, something which blows even Jazine’s skeevyness out of the water. (Spoilers for the finale of the book, which has nothing to do with the mystery.) Gur arb-yhqvgrf, nf gur ortvaavat bs gurve eroryvba ntnvafg grpuabybtl, oybj hc gur cerivbhfyl zragvbarq prageny zrqvpny yvoenel naq pbzchgre flfgrz, juvpu unaqyrf gur bccrengvbaf sbe nyy bs gur Havgrq Fgngrf. N terng qrny vf znqr bs ubj gurl znxr fher abg gb xvyy nal bs gur fgnss, ohg vg'f zragvbarq rneyvre va gur obbx ubj ener fhetrbaf ner orpnhfr bs gur cerinyrapr naq fnsrgl bs pbzchgrevmrq zrqvpvar. Fb gurl pbaqrza hagbyq zvyyvbaf gb ceriragnoyr fhssrevat naq qrngu orpnhfr "gur znpuvarf ner whfg fb rivy." Naq abg bayl ner gurl gerngrq flzcngurgvpnyyl, Penqre rira fnlf ur jbhyq wbva gurz vs ur jrer lbhatre! These characters are all horrible people!

Ultimately, The Transvection Machine is something of an enigma. Hoch was a consistently good writer of short stories, and was by all accounts a very kind man. So what prompted him to write this, a novel with an embarrassingly transparent plot and main characters who are who are significantly worse than the actual murderer, is baffling. I have heard interesting things about the other, equally obscure, book in the series, The Frankenstein Factory, so I’ll probably read it at some point. But as for this, I have no hesitation in calling it the worst mystery I’ve ever read. It is equal parts infuriating and insulting to the reader’s intelligence. If I wind up doing some kind of end of year list, this will most assuredly by named as one of the lowlights of my  year. In one respect, at least, I’m glad I read this. As no other mystery blog has reviewed it, there’s no one warning people away from seeking out this book. Beware! Learn from my experience, so that my suffering may not be for naught. Avoid it as you would the cozy mystery section in a bookstore. For like that benighted place, The Transvection Machine has nothing to offer fans of clever mysteries.