Novel * Daniel H. Wilson * The Title is Pretty Self-Explanatory * 2011
Is it possible to be disappointed in a book that a) I had never heard of before picking up, and b) has a title like Robopocalypse? Like, I know I’m one to talk what with the name of the blog and all, but I’m also not selling anything to people. As you might surmise, this is a novel about robots overthrowing civilization. You know, like that episode of Futurama. Between the title, and the cover, and pretty much everything about this book apparent at a glance, I didn’t expect much. I was hoping for a pulpy, borderline silly romp. There are times when that’s what I got. Yet I think the overall issue I have with the novel is that there’s a good deal of ambition here that’s not matched by what’s actually on the page. There are lots of good, interesting ideas here. The execution of those ideas never really comes together, however. The story just kind of meanders, and the rest of the action is never quite over-the-top enough to compensate for the lack of coherent characterization or narrative. That sounds like a sick burn, but at least the book is breezy enough not to feel oppressive in its mediocrity.
Let’s roll back a little bit and set up what this novel actually is. Robopocalypse is a near-future sci-fi story about a self-aware artificial intelligence that infiltrates the world’s networks. Like self-aware A.I. are wont to do, Archos pretty much instantly judges humanity and finds us wanting. Civilization is a cancer, and we must be wiped out. The variation on this story in Robopocalypse is actually a neat subversion. Instead of going all Skynet and letting fly with the nukes, Archos is fascinated with life. All life. Therefore, it is suboptimal to simply blow everything up, or release plagues which could threaten other species. Civilization is the problem, so the solution Archos comes up with is to infect the machines rather than the humans. In this near-future version of our reality, robots continue to advance. All technology becomes “smart,” in which it communicates with other tech and world-wide networks. Once Archos is able to infiltrate and subvert that network, it can compromise the planet’s robot community and force them to turn on humanity.
Once that happens, civilization falls pretty quick. One thing Robopocalypse does well is underscore just how dependent we’ve become on technology, and how integrated into our social structures it’s become. Most of the damage done to civilization isn’t from automated war machines, it’s done by domestic robots and other common machines, such as smart cars and the like. Your phone starts lying to you, alerting you to emergency services and directing you to safe zones that turn out to be death camps. Once the malevolent super-intelligence takes over, the cities collapse almost immediately. Those who live in rural communities fare better, and the book’s protagonists center around an Indian tribe out in the middle of Oklahoma since they weren’t attacked right away. Because of course humanity fights back. Since Archos doesn’t go scorched earth on us, that gives humans the opportunity to do what we do best and adapt to the evil A.I.’s tactics.
The setup, then, is pretty cool. Not the most original thing in the world, but whatever, I’ve long held the belief that tropes and clichés are acceptable as long as they’re done well. And to Wilson’s credit, he tries to do something different with this particular malevolent A.I. story. Unfortunately, the narrative tricks used don’t work particularly well. Now, I’ve haven’t read World War Z (yet, it’s on my list), but I understand that it’s written in a sort of oral history style, in which the author writes from a variety of perspectives which illustrates a broad spectrum of an apocalyptic event. It seems to me that Robopocalypse was going for a similar effect, except with an apocalypse that could actually happen. The novel begins with the ending. The ragtag human resistance finally storms Archos’ underground hidey-hole in the wilds of Alaska and unplug it. Yay, humans! Once they pull the physical manifestation of Archos’ brain out of the tundra, the narrator character realizes that he can interface with it and cobble together the story of the war using various means, like surveillance cameras and phone transcripts and the like. The rest of the novel is laid out thusly, a series of snippets from various perspectives.
That’s actually a cool idea, you know? An apocalyptic event of any kind is rather beyond us. Humans have a hard time with large numbers of concrete things, which includes each other. Our brains aren’t equipped to process something like 10,000 people, so we have to abstract it. Once something is abstracted, scale becomes a difficult thing to capture. So using a structure like the one laid out in this novel could help convey scope and scale to the sheer size of the disaster. The first few chapters of the book unfold just like I expected, given the framing at the beginning. There’s an early chapter that’s a transcript between a cop and a victim of a robo-crime. A deranged domestic robot breaks into the shop where this kid is working and tries to kill him and his coworker. The chapter is just a window into a society at the very beginning of the coming apocalypse. There’s a chapter that’s the transcript of a hearing about a military officer who lost control of their peacekeeping robot. The first section of the novel shows a lot of promise in its scattershot depiction of a world slowly falling apart as the robots take over.
The downside of this approach is that you lose the ability to tell a proper story about a central group of characters. If you stick to the above structure, what you’re really doing is telling a series of short stories set in the same world. That is a totally acceptable way to do things, and in this instance I think Robopocalypse would have been a much stronger work if had stuck to that structure. That first section is by far the best, with the most vibrant ideas and the most compelling storytelling. The problem is that Wilson also wants to tell a longer story with familiar characters and tie everything into a continuous narrative, and it just doesn’t work. The book basically splits the difference between a series of framed short stories and a traditional novel, and both aspects suffer as a result. After that first section, when we start circling back to characters we’ve already met, the novelty of their situation and perspective is gone. Yet at the same time there simply isn’t enough time spent with them to flesh out their characters. The narrative is still fragmented, but the characters are the same, so now the structure is working against the story. By the end I’m simply not vested in anybody or what they’re doing, not only because I already know they win in the end, but because what they’re doing in the interim isn’t compelling.
I had next to no expectations when I picked up this book. I was looking for something quick and dirty, a nice pulpy in-between kind of book. Instead, for the first 100 pages or so, I was pleasantly surprised by it. Robopocalypse rudely adjusted my expectations, setting them much higher than when I first bought the book. As a result, I’m probably a little more down on the novel than I otherwise would be. It’s still a pulpy good time – there are some brutal action scenes in here that are thrilling and fun. However, this still isn’t the book I wanted. And to be fair, the book I want is the one promised in that first 100 pages. The following two-thirds of the story simply don’t keep that promise, which is a bummer.