I usually just talk about fiction in terms of how it can be reflected in games, but today I'm going to talk about the actual fiction itself. It should be no great surprise that I'm a fan of capers, and by extension, a certain sort of mystery and thriller. Plot, story and character are all important, of course, but these stories add an extra layer of puzzle onto the fiction. Over and above the interest in the story, there's an intellectual challenge to it. Can you, as the reader, stay a step ahead of what's going on? Can you solve the puzzle? Can you figure out the twist?
Writing these stories walks a fine line, and I do not envy the authors who try it.They need to strike a balance of showing the reader enough to allow her to try to puzzle things out, but no so much that it gives it all away. To make this possible there are a number of rules, tropes and traditions that allow the puzzler-author to get away with more things, specifically to allow for information to be obscured.
Now, this can be tricky. If the author withholds a piece of information that one of the characters have, it can utterly ruin the puzzle. You see this a lot in bad (but often popular) mysteries where the reveal revolves around something the detective hasn't mentioned to the narrator (because, hey, why write well when you can use unreliable narrator as a bulletproof shield?) until that very moment. That's a cheat because it means the reader could never have figured it out.
But at the same time, you don't want to show everything all at once. If the brains behind a caper is making preparations for a twist, we don't necessarily need to see what those preparations are, just that preparations are being made. This is often the difference between a good and bad flashback. A good flashback is built on a stub of an earlier scene - a bad flashback comes out of nowhere in an oh-by-the-way fashion.
If you read a book for the puzzle, you buy into the rules and expect the author to do the same, and when he does, that's a real problem.
It is with this in mind that I both love and deeply hate The Lies of Locke Lamora. This is a shame, because it's exactly the kind of book I SHOULD enjoy, and it's full of things I still remember fondly. If it had been merely an adventure novel, I would probably have enjoyed it a great deal. But it's a puzzler with both mystery and caper elements, and in that regard it falls down very hard.
Lynch cheats in three very specific ways in the book, and i use the word cheat very deliberately. In each case he uses his authorial power to break the rules in order to keep readers in the dark. This is an admirable goal, but it's rather like scoring a touchdown in chess - it's not very impressive if you don't tell the other party what game you're playing.
So, first, we have the identity of the Gray King. This is set up very nicely, and there are any number of people it potentially could be, all of whom have some good reason (like being dead) why it couldn't possibly be them. This is a classic mystery setup, and in time you discover the lie and it reveals who's alibi doesn't hold up. It's set up so well that it ends up feeling like a double cop-out when the reveal is "Ha Ha, it's this guy that got mentioned on page 23 who has no connection at all to the story so far!" (and, in fact, it feels like a triple cop out if you view it as a "Ha Ha, I as an author am deliberately foiling your expectations!"). It's a cheat designed to make sure that everyone's guesses are wrong.
Second, we have the Bondsmagi. To put it bluntly, he's a walking Deus ex Machina, and that's _terrible_. Introducing magic into genre fiction is always hard, but it's super hard for puzzlers because it has the potential to undercut any and all logic. This is the reason, for example, it's so hard to do a locked room mystery in a D&D setting - there are so many ways to get in and out of a locked room that it's hardly a mystery at all. If you're going to add magic to your puzzlers, you need to have rules that you stick by (because nothing's cheesier than using violations of your own rules as resolution). If you want to bend or break those rules then you need to make sure the limit is not the rules but a character's understanding (*cough*Harry Dresdean*cough*). Better still, you don't want magic to ever be the _answer_ to a mystery unless it's also part of the question. Bottom line, the Bondsmagi is an unending font of cheating.
Last, LLL performs the ultimate cop-out of any caper. In an attempt to make the protagonist clever, Lynch often takes the shortcut of making everyone else stupid. The best and most obvious example of this is the fact that the virtually all-powerful and for-hire magi are an absolute blind spot for every single person in the city, including the spymaster who I can't help suspect reminded lynch of an NPC he disliked. Yes, they're crazy expensive, but we're talking about a rich trading city here. even if they can't keep one on retainer, it stretches credulity that no one else has thought of using one of these guys before (especially when you can apparently use one to steal huge amounts of money, which certainly seems to offset the cost).
Now, as I say all that, bear in mind that I think the book is full of fantastic, clever, well-written scenes. Many of these scenes are so good that I still waffle on my opinion of the book despite how much my problems with the main plot grate at me. But taken as a whole, it's a great book about how to run small cons, and a terrible book about how to run a caper.
That said, enough people have said good things about the sequel that I've downloaded the audio book and put it in my queue. I'm willing to give it a fair shot, if only because Jean is awesome, and I'm totally willing to read the adventures of Jean and his chatty sidekick.