The Quantum Thief


Jean le Flambeur1, gentleman thief, is sprung from prison by the beautiful Mieli, who persuades him to undertake one last job. Author Hannu Rajaniemi works hard to obscure the familiar outline of this hoariest of heist plots, by layering on the worldbling2 with a trowel:

The hidden Sobornost tech beneath the Oortian sapphire coral wakes up. The spidership reconfigures itself. The scattered modules pull themselves together along their tethers and fuse together into a tight, hard cone. The q-dot winglets transform from a perfectly reflective material into a diamond-hard firewall. Just in time, before the Archon’s nanomissiles hit.

I do like the experience of being confronted by a complex fictional world and having to make sense of it. But I also like to feel that there is in fact some sense to be made, and in this novel I frequently failed to make it.

What kept bothering me is Rajaniemi’s lack of clarity about the ontological status of the events described, a failure to distinguish the virtual from the real. If I’ve understood the book correctly (and I am not at all confident that I have), some of the scenes take place in the real world, and some in virtual or simulated environments. The Dilemma Prison of the Archons in the first chapter is a simulation, and so is the Realmspace of the zoku, whereas the Oubliette is a physical city on Mars. When Jean le Flambeur’s mind is extracted from the Dilemma Prison, he gains a “new body” (some kind of cyborg or android grown or constructed for him) and his mind is transferred to it. But later on, characters are described as moving between real and virtual environments without any clear explanation of how they do so. When the narrator enters Realmspace via a ‘Realmgate’, it’s described like this, which I think is meant to convey the experience of a software personality being assembled:

The darkness rebuilds us. For a moment, I feel like I’m being sketched by a pen, feeling returning to my flesh and skin and limbs, one by one. And then I can see again.

Presumably the idea is that all this is so natural to the characters that they don’t pay much attention to it, thus explaining how, at a couple of points in the plot, the virtual/real distinction turns out to have been muddled and the characters failed to notice the switcheroo.

I also understand that this is a natural evolution of the genre: early novels featuring some new science-fictional idea (here, software personalities) have to spend some effort explaining how it all works, but later works can take it as a given part of the background. But I had a few problems with the way Rajaniemi goes about it:

  1. There’s some careless deployment of annoylogisms. For example, in the Oubliette, interactions between characters are modulated by ‘gevulot’3, whose meaning slips back and forth between two things: first, a cryptographic protocol that allows people to exchange limited personal information while preserving privacy, and second, some kind of ‘utility fog’ that implements the protocol but also obscures the physical appearance of things. It’s not on the face of it implausible: I suppose you might end up calling your phone a ‘feed reader’ if reading feeds was the main thing you used it for. But as an authorial choice, it’s one more mystification in a book that already has quite enough.

  2. Different things matter in different environments. In reality, it matters if someone is pointing a gun at you. But in a simulation, it may matter or it may not. Readers need an explanation of the simulated consequences of being shot before we can understand why it matters: do you lose the game, or do you just lose some points and play on? In the opening chapter in the Dilemma Prison, Rajaniemi is careful to show in what way the simulated events matter. But later on, the explanations are omitted but characters carry on doing things like waving virtual swords at each other when it’s not clear what the swords can actually cut. Maybe we are supposed to invent our own explanation for why these simulations matter to the characters? In which case, my best guess is some kind of combination of role-playing game and amateur dramatics.

  3. Software can be copied and therefore software personalities can be copied too. This is made explicit in the first scene, where Jean le Flambeur is made to play the prisoner’s dilemma against a copy of himself. So how can he possibly be ‘rescued’ from the Dilemma Prison as described? Won’t the Archons just restore him from backup and continue as they were?

  4. There’s no excuse for the characters not paying attention to the virtual/real distinction. The base levels of Maslow’s hierarchy of needs are physiological (which, for a software personality, means access to computing power) and safety (which, for a software personality, means software integrity: running your own mind and being safe from viruses, trojans and hackers). But it’s clear from the events in the book that these needs are unmet, and yet the characters just continue as they were, only noticing the gaping holes beneath them when the events of the plot require them to do so. It’s distracting for this reader, like watching people play Hamlet on a tightrope.

    It’s instructive to compare this with Diaspora by Greg Egan, a 1997 novel also featuring software personalities. In Diaspora, events follow the hierarchy of needs in a plausible way: it’s clear that the characters’ physiological and safety needs are met before they are motivated by the higher levels of the hierarchy.

The heist plot relies for its dramatic effect on our appreciation of the cleverness and daring of the criminals, but this depends on understanding the difficulty and risk of their actions. We can be impressed by the quick-talking con artist’s fluent deception because we know how hard this is to pull off in real life. But here’s how Jean le Flambeur gets away with the crime:

… everyone at the party remains standing, gazes fixed on something far, far away, but unseeing, as the fireworks fizzle and die above us. Another trick from the gogol pirate handbook: an opto-genetic virus that makes brain cells hypersensitive to certain wavelengths of light. It was not hard to customise it not for the purposes of uploads, but for creating a period of inactivity. … And there are only so many fireworks manufacturers in the Moving City: bribing them with the pretence of a little innocent surprise for M. Unruh was the easy part.

Here “gogol pirates” are criminals who upload people without their consent, and “opto-genetic virus” has previously been established as a tool for brain analysis:

It’s an old trick, pre-Collapse. They used to do it with rats.4 You infect the target with a virus that makes their neurons sensitive to yellow light. Then you stimulate the brain with lasers for hours, capture the firing patterns and train a black box function to emulate them.

So how impressive is Jean le Flambeur’s trick? It’s not at all clear. As far as I can tell, he bought a vial of opto-genetic virus from some gogol pirates and squirted it around the place with an atomizer. Is that a big deal? The episode makes me feel as if Rajaniemi is merely gesturing at dramatic effect, hoping that we’ll fill in the details based on identifying the cliché and remembering the effect it produced in other books and media.

Also, it makes no sense. If you can reliably infect the partygoers with a neural virus then why the business with the fireworks? Why not just infect them with a virus that sends them to sleep at a given time? As described, the plot would only make sense if neural viruses were limited to causing sensitivity to yellow light, which seems a curiously specific limitation. And how is the light from the fireworks getting to the neurons anyway? Those pre-Collapse rats had to have holes drilled in their skulls for the lasers to shine through, but that doesn’t seem to be happening here, so presumably the “opto-genetic” business is just a metaphor: the light is reaching the infected subjects via their eyes and then something consequent to this experience is triggering the paralysis via the infected neurons.

So the whole explanation is a bait-and-switch on the reader here: “opto-genetic virus” looks at first glance as if it is supposed to explain what happens in the novel, but when you investigate more closely it doesn’t remotely meet the needs of the case. It’s a mere gesture at explanation, the modern equivalent of E. E. Smith’s “sub-ethereal rays”.

  1.  I couldn’t help being reminded of the master criminal Flambeau in the ‘Father Brown’ stories of G. K. Chesterton.

  2.  See Adam Robert’s review of Neal Stephenson’s Anathem for this and other annoylogisms deployed here.

  3.  Hebrew for ‘borders’ or ‘limits’.

  4.  Optogenetics was Nature Methods’ ‘Method of the Year’ in 2010, the year The Quantum Thief was published.