Halting State
Warning: Spoilers Ahead
Halting State for my money, is Charlie Stross’s best science fiction novel. Not his most fun novel - that award collectively goes to the slightly-borked-alternative-reality Merchant Princes series that Paul talks about. Nor his most wildly inventive novel (which is surely Accelerando). But it’s the novel where fun and speculation come together most successfully. It works both as an entertaining read and as a fascinating discussion of an encroaching low-level singularity. It’s one of the best pieces of sociological-political extrapolation that I’ve ever read.
But the book is sneaky - the sociological speculation tiptoes up to you very, very quietly before it pounces. It begins like a genre novel, albeit not a genre science-fiction novel. Instead, it steals its opening chops from Christopher Brookmyre, a writer of satirical thrillers who doesn’t get nearly as much attention as he deserves on this side of the Atlantic (US readers should imagine a Scottish Carl Hiaasen with a keen interest in weapons and consumer technology). Brookmyre’s books regularly make the bestseller list in the UK - he has succeeded (as most SF writers haven’t) in marketing his books to the MMORPG generation (one of his thrillers is built around references to the classic FPS game Quake). The set-up at the beginning of the novel - where a frantic executive tries to explain to an Edinburgh cop how a team of orcs with a dragon have carried off an online bank-heist - is classic Brookmyre (so much so that I suspect it is a deliberate homage). But Stross is much more ambitious than Brookmyre. Rather than throwing political jab-punches at authority, Stross is interested in figuring out what happens to society when political authority evanesces away.
As noted, the novel is sneaky. Much of its argument (if I’m right) is oblique. It begins with what seems to be a targeted recruiting email dense with personal details about its intended target, but which is most likely automatically generated spam. But as the novel continues, you find out that the person to whom the spam is addressed is himself non-existent. Perhaps he’s a fake identity who was created by the security services for one of the book’s main characters to slot into. Or perhaps he’s simply based on that character. Or he’s a cutout for British intelligence. Or perhaps something else entirely. As with many aspects of the novel’s underlying plot, this is left a little ambiguous. The novel concludes with another email which appears to be standard Nigerian 419 spam - an offer to transfer large amounts of money from an account in a Lagos bank to the account of the major villain of the book. Despite appearances, the email is very plausibly legitimate.
These emails present, in miniature, one of the major themes of the book. In a world of open communications, there is no good way to distinguish fake identities from real ones, or false flag messages from legitimate communications. This hopelessly entangles both the good and the nasty characters of Halting State in a web of miscommunications.
Hence the plot is a morass of confusions about people’s real identities and motivations. The characters themselves speculate about the big picture, sometimes plausibly, sometimes not so, but nobody sees it entire. Halting State emphasizes that everything is subjective by presenting the story in the second person singular - “You backtrack, trying to work out what confused her” and so on - like a very complicated old-style text based computer adventure (maybe one of the old Infocom or Level 9 games). But unlike one of those old games, there isn’t any underlying plot that has been created by the game’s author, where you go through the locations, solve the puzzles and win the prizes. Instead, the game is generated by the characters themselves, none of whom fully understand what the others are doing.
Thus, for example, the main plotline - World of Warcraft meets The Producers. Start with a company - Hayek Associates,1which purportedly provides financial back-end services to the online gaming industry, but which in fact has various shady links with British intelligence. Add a psychopathic CEO who doesn’t realise that his company is a front, and is trying (with the help of his sidekick) to drive his company bankrupt so he can profit from exotic futures contracts that he has placed on its demise. Then have the CEO selling the company’s copies of the national backbone’s one time pads via an anonymized blacknet to a crowd whom he probably thinks are Russian mafiya, but in fact are Chinese government hackers. Add a fictitious employee (‘whose’ apartment hosts a node for the blacknet), a gaming clan of dedicated griefers working with the Chinese state in some vaguely defined cooperative relationship, a low level member of aforementioned gaming clan who decides to cop some money on the side by organizing an online bank raid (with Orcs – hence the opening scene), a confused effort by the boss’s sidekick-in-crime (who doesn’t fully understand what is going on) to inform the police, an even more confused attempt by some shadowy EU intelligence body to mount a raid on the company, and the merry-go-round is in full swing.
But this isn’t simply Feydeau without the sex. There’s a quite serious and interesting underlying point here. What makes the generalized confusion possible are the unexpected consequences of a set of technologies and social practices. Communications built on open, trusting protocols such as TCP/IP. The confluence of real life and online activities. People’s willingness to do pretty well anything as long as they think it’s a ‘game.’ Distributed computing and public key cryptography. Services that provide electronic updates from imaginary family members, for people who would otherwise be hopelessly lonely. Cars run by expert systems embedded in the communications net. All of these things enable what Larry Lessig might describe as ‘architectures of control.’ But the architectures don’t really control people any more in the ways that they used to. Instead, they create a toxic mixture of ubiquitous surveillance and official cluelessness about how to use the information and opportunities it creates to deal with new security threats.
More generally, (and herein, I think, lies the pun in the book’s title), it is practically impossible for traditional authority-based politics to cope with a world of this kind. The traditional state, if it is to work at all, needs an underlying system of responsibility and accountability, with clear lines of command. There may be areas in which these rules are deliberately relaxed of course, most obviously intelligence and counter-intelligence. Here, as multitudes of spy novelists from John Le Carre to Alan Furst tell us, ambiguous motivations, uncertain information and deliberate deception create a miniature world which is rife with confusion. But even this shadowy world has its own informal rules and mutual understandings about what can and can not be done. The world that Stross portrays is one where (a) the confusion of intelligence operations has become ubiquitous and (b) the spymasters have completely lost their grip without realizing it. States simply aren’t in control any more, to the extent that they ever were.
This comes out most clearly in state authorities’ interactions with the most characteristic phenomenon of Stross’s world - massive, decentralized networks of game players. Both Chinese and British intelligence use gamers as agents - the Chinese use a clan of rabid online gamers while the Brits use a crowd of people who think that they are playing an elaborate real life role playing game but in fact are auditioning for a starring role in the new world of international espionage., they don’t realize that the networks are beyond their control. The traditional state has reached an endpoint in which the world simply Does Not Compute. As the viewpoint character whose worldview is probably closest to Stross’s own describes it:
The spooks in Guoanbu probably are professional, they wouldn’t mess with the European SCADA infrastructure short of an outright shooting war … but are they likely to realize that they’ve almost certainly been Pwn3d by their own pet griefer clan, and all their electronic armoured divisions are in the hands of a dozen Asperger’s cases with attention-deficit disorder and a quantum magic wand? It’s not a risk you can take. And it’s not a risk you can explain to Barry Michaels, because you know his type and after seventy years of data processing they still think that coders can be hired and fired; that the engineers who ripped out the muscles and nerves of the modern world and replaced it with something entirely alien under the skin are still little artisans who will put their tools down and go home if you tell them to leave the job half-done.
Massively distributed networks of information exchange (with anonymity, pseudonymity etc built in) empower networks to do all sorts of things that previously took top-down hierarchical organization. This can sometimes be convenient for states - it is hard to hold a state accountable for the behavior of agents in networks who are sort-of affiliated with them (this is a real problem - security types here in Washington DC are perpetually muttering about Chinese hackers trying to infiltrate USG systems - but no-one knows for certain whether these hackers are state-sponsored agents or just kids out for an electronic joyride at the expense of Uncle Sam). But it has deep, underlying, fundamental problems for state agency, because (contra Michaels and his Chinese counterparts, these networks can’t just be turned on and off as the state likes. They’re autonomous, have their own internal logic, and are inherently unpredictable. A world dominated by diffuse shifting social networks is a world that states aren’t going to be able to control any more.Thus, in Halting State, open technologies and the networks they permit to form are creating a kind of low-level Singularity. States think that they are able to control these networks - but they are instead helping to build forms of social organization that challenge their underlying logic of organization.
Stross’s vision of the near future isn’t as stylized as, say, William Gibson’s or Neal Stephenson’s. Nor does it involve the mixture of worries and individualist wish-fulfilment that, say, Vernor Vinge’s Rainbow’s End does. Instead, he presents a world which is (if it’s possible) even more muddy and complicated than the one we inhabit today. I have no doubt whatsoever that this book should have won the Hugo a couple of months ago (not that I begrudge Michael Chabon, but TYPU wasn’t his best book by a fair stretch). It sketches out a new way of thinking about SF that I suspect will be far more influential in the future than self-conscious movements like Ryman’s Mundanes. Rather than engaging with the futures of the past (as lots of SF today does, it tries to set out the futures of the present, engaging with a bristlingly complex set of social developments and reaching out to a new set of readers who are embedded in SFnal media products but rarely read SF (an entirely separate essay could be written on the new ways that HS tries to engage with readers). I think that this is the first genuinely successful SFnal take on the social changes that we’re facing into - not, of course, because it is going to be right - but because it takes some of the core dilemmas of an IT based society, plays with them and extrapolates them in ways that challenge our basic understanding of politics in a networked society. About two thirds of the way through reading this book, my mind was completely blown, in a good way. I’ve been thinking about it ever since.
1I suspect that the name is a sly dig at Hayek rather than a tribute to him- the decentralized actions and knowledge of multiple agents produces chaos rather than catallaxy.