Adam Roberts’ Gradisil (2006) opens with Miklós explaining to his daughter Klara that 20th-century rocket scientists were stupid:

“If it hadn’t been for NASA, and Von Braun, and Korolev in Russia, and their mania for rockets, mankind would have colonised the planets by now. […] Speed is where all this went wrong. Humanity became hypnotised by eleven kilometres a second, and all its space research was oriented towards achieving that ridiculous speed.1 Only rockets could do this, so rockets were what the space programme became. […] The Earth has already provided for us. The Earth possesses something called a magnetosphere,2 created by the differential in rotation between the Earth’s molten core and its solid mantle. Really—think of it like a bar magnet, though on a huge scale. The line of magnetic force run out from the North Pole in a great sweep through space, and in again at the South Pole. Ions from the sun stream down the branches of this tree, my princess, at the north and the south poles, to create the auroras. Better, we can climb up the same branches to space.”

At this point the reader says, “Hold your horses, mate! Perhaps the footnoted errors can be attributed to the speaker, but are you seriously going to be expecting me to swallow a book in which people get to space by magnetic levitation?” The magnetic lift equation is $$ P_{\rm mag} = {B^2 \over 2\mu_0}. $$ Taking \(B = 6×10^{-5}\ {\rm T}\) (the strength of the Earth’s magnetic field at 50° latitude), and \(\mu_0 = 4π×10^{-7}\ {\rm N}·{\rm A}^{-2}\) (the permeability of vacuum), we get \(P_{\rm mag} = 1.4×10^{-3}\ {\rm Pa}\).

To lift a small person with a minimum of equipment—let’s say 150 kg—against gravity takes 1470 N, which (under the unrealistic assumption of 100% efficiency) would require the magnetic lift from a slice through the Earth’s magnetic field of a bit over 106 m2. That’s more than a square kilometre. To look at it another way, 100% efficient magnetic levitation against the geomagnetic field could in theory lift objects with a cross-sectional density of about a tenth of a gram per square metre. And of course we have nothing like 100% efficient magnetic levitation. So who’s looking stupid now, Miklós or Korolev?

Of course it’s conventional in science fiction that authors can introduce new scientific laws and new technologies—even implausible or impossible ones—as needed for their story to work. But it matters how this is done. If done well, the reader politely agrees to suspend their disbelief and imagine some way it which the impossible can be made possible. But shove the impossibility in the reader’s face and they might not be so polite. They might even refuse to put in the necessary imaginative work.

The problem here is not the implausible magnetic levitation—on its own this would be an acceptable point of departure from reality—it’s the combination of the implausible magnetic levitation with the sneering at the rocket scientists. The sneering calls attention to the problem with the magnetic levitation idea, and closes off all sorts of ways in which an imaginative reader could have justified it. You might, for example, have supposed that this is an alternative universe in which the Earth’s magnetic field is thousands of times stronger than ours. But then you’d have to also imagine NASA being so stupid as to fail to notice that. (Children would notice it when playing with magnets. Ancient civilisations might have invented magnetic levitation before the wheel.) You might instead imagine that some amazing new technology allows the gathering of magnetic force from a wide area without having to physically build an object that big (some kind of superconducting magnetic sail, perhaps, though nothing like this is described in the book). But then if this technology is amazing and new it would hardly have been stupid for Korolev to fail to use it.

There’s worse to come. A character gets tired of ferrying fresh air up to his house in orbit, and buys fifty kilometres of hose.

[Teruo] rigged up a little motor and some runners and automated the process, spooling the capillary pipe out and out.3 It was slow, and the motor was noisy which made life in the house less pleasant, but eventually it was all done. We anchored the pipe firmly inside the room, and fitted a small pump to the end of it. Then we set it running. […] And there it was: a tiny stream of new, cold air, coming out of the pump.

This is so blatantly absurd that it brought the story to a juddering halt while I tried to work out what’s going on. Is this a literary device designed to highlight the contradictions in the ‘hard’ sf genre? Or is it just that the author has no understanding of the physics of air pressure?

Well, luckily I don’t have to guess, because Roberts reveals that it’s the latter in these two threads on James Nicoll’s blog. Roberts:

You [Nicoll] don’t like my novel; you’re not alone in that, by any means. It’s not something that admits of argument by (for instance) me, and it almost certainly means that I haven’t done my job as a writer properly. But: saying I haven’t done my sums … saying, indeed, that I’m incapable of doing my sums. That’s something else.

Roberts goes on to reveal, in an admirable display of honesty, that he is indeed incapable of doing his sums:

Imagine a hose with a diameter of, say, a centimetre and a length of 100 km. Such a hose would contain, give or take, about 80 cubic meters of air; about 100 kg of air. Too much for a lung to suck up, sure. But too much for an engine to pump?

[…] I honestly don’t mean to make you wave your hands; and I daresay I’m completely wrong. You can’t apply more suction than vacuum? Isn’t suction a function of the differential in pressure? A differential from 1 ppsi to vacuum isn’t going to apply as much suction as a differential from 10ppsi to 1 ppsi, surely? Or am I just being an enormous dufus, here?

There should no particular shame in lacking good intuitions about air pressure. What’s more important is having some sense of what your limitations are. If you know that your physical intuition can’t be trusted then you can arrange for experts to check your work.5 But lack of skill and lack of knowledge about lack of skill often go together.

Does this kind of thing matter? Well, it depends on what you’re trying to achieve. From a 2007 interview with Sci Fi Wire:4

Roberts said that Gradisil came about as a result of his working his way through the different iterations of SF. “You know the kind of thing: time-travel story, planetary romance, alternate history, cyberpunk, first-contact story, etc., etc.,” he said. “So my last novel [The Snow] was an end-of-the-world story, inflected via my rather peculiar perspective on J. G. Ballard. Previous novels have done Jack Vance-[like] interstellar excess, Ursula Le Guin planetary grimness and so on. When I set down to write Gradisil, I wanted to write something hard SF, something near-ish future, something Robert Heinlein or Stephen Baxter-like. As with all my novels, I think it’s fair to say that something weird and dislocating happens to these great authors when I force them into the woodchipper of my own imagination, but there’s something tech-SF-y and war-story about this particular novel.”

Roberts is very skilful at pastiching style and subject matter, and Gradisil ticks many ‘hard’ sf boxes. The anti-NASA rants capture the bitterness of similar passages in Stephen Baxter; the treatment of politics and government is reminiscent of the shallow libertarianism of Jerry Pournelle, Charles Sheffield and many others; and the narrative style of the first part of the novel has the fluent matter-of-factness and skating-blithely-over-unpleasantness of Heinlein. But there’s something missing from the mix that indicates that Roberts has a blind spot when it comes to the understanding of how this particular subgenre of science fiction is actually read by its fans.

First, there’s the simple aesthetic pleasure of reading fiction that convincingly portrays the physical world. (Analogous to the pleasure of reading novels which have convincing and insightful psychological portrayals of characters.)

Second, the use of this kind of novel for political propaganda relies on the author at least appearing to get the essential details right. When one of the points of the novel is to say, the future will be like this (or, increasingly, the future would have been like this if NASA hadn’t messed it up), and that shows why we must support my politics, it’s important that the author can sustain at least the illusion of accuracy, because otherwise the reader will say, “no, the future won’t be like that, because in the real world you can’t just pull free energy out of your arse.”

Merely pastiching the surface features of this genre won’t by itself produce something that’s recognizable as ‘hard’ sf to the aficionados. Carlos Yu puts it clearly:

It’s a peculiar and demanding stricture: the accurate mimesis of the physical universe in a story which requires it. The other trappings of ‘hard’ SF—libertarianism, manifest destiny, the military—those are contingent details in the form’s evolution.

There’s a tendency in ‘hard’ sf that is sorely in need of criticism-by-pastiche: namely, the gap between true mimesis and mere conventional gestures in that direction, and especially the tendency to conveniently ignore that gap when using the novel for political propaganda. But to criticize that tendency, you have to understand that it’s there, and that means at least being able to understand how the fans read the genre, even if you don’t read it that way yourself.

  1.  This is complete nonsense, of course. To enter orbit, any object (rocket-propelled or otherwise) has to reach orbital velocity (which is 6.5 to 8 km/s in low Earth orbit). Rockets also need to get to orbit as fast as possible, to minimize the loss of kinetic energy: the Earth’s gravity robs a rocket of 9.8 m/s each second, so every second counts.

  2.  I think the speaker means ‘geomagnetic field’ and not ‘magnetosphere’ (which is the region of space containing ions controlled by the magnetic field).

  3.  No explanation is given for how the other end of the pipe is made to enter the atmosphere, as opposed to just floating around next to the orbiting house. I get the impression that Roberts doesn’t realise that the pipe wouldn’t just dangle downwards as it would from a balloon.

  4.  Sci Fi Wire is defunct, but I found a copy of this interview here.

  5.  James Nicoll wrote to point out that “Roberts did get someone to check at least parts of his book over and that someone was Stephen Baxter.”

    Roberts: The problem is that rockets are expensive. No way round that. So I needed a plausible means of getting into orbit that didn’t involve rockets. I imagined a technology you could fit to any regular jet-plane or private flier that used resistance to the Earth’s magnetic field to ‘fly’ into orbit. When I ran the idea past Stephen Baxter6 he did not say that it was not without the possibility of never working in real life. At all. I took this to be a ringing endorsement.

    I guess this shows the problem with asking for advice on a narrow technical question without considering how the answer will fit into the whole work. In this case, Baxter is right that magnetic-levitation-to-orbit would be fine on its own: sure, it’s applied phlebotinum, but it would have been fine in some other type of story: in particular, one that didn’t also have the anti-NASA rhetoric.

  6.  Baxter was perhaps not the best of choices, since he’s not exactly immune to making the same kind of mistake himself. His use of the Doomsday argument in Time so troubled reviewer Dave Langford that he contacted Baxter to check whether he believed in the argument himself.