In a genre dominated by fantasy dressed in a spacesuit instead of a wizard’s robe, Greg Egan stubbornly sticks to extrapolations from the physics we know. He rules out from his fiction faster-than-light travel or communication, or spaceships that couldn’t possibly be fueled by any form of energy we know. This self-denying ordinance deprives him of the props and conventions of the genre: no galaxy-spanning empires or interstellar wars for Egan. So he is forced to invent new kinds of plots and conventions.

You can see his commitment to consistency of fiction with fact in his essay on the flaws in his 1992 novel Quarantine, and here:

I certainly knew when I was writing [Quarantine] that it was utterly implausible. But I wasn’t aware of quite how many things I was getting wrong—and in particular, one theorem which was proved a few years after the book was published1 really drives a stake through its heart. [rec.arts.sf.written, ]

Which is not to say that his extrapolations are always plausible extrapolations. On the contrary, they can be highly baroque. He often starts with just one departure from what we know, one speculative extension of current knowledge, the consequences of which are worked out in detail to extraordinary and absurd conclusions.

His characters are usually mathematicians, physicists or other scientists. Egan has little interest in the idea that science fiction is about exploring human responses to changes in technology and society: his characters are often post-human or non-human, and rarely particularly engaging or sympathetic:

Some writers are so obsessed with creating characters that readers can “relate to”—even when they’re living in virtual reality, or a thousand years in the future—that they pretend that nothing important will change. I didn’t want to do that. With the power to reshape themselves as much as they like, no one can seriously expect the inhabitants of VR to spend century after century just imitating us. And once they stop doing that, a lot of things that are central to our lives, right now, will either vanish, or come to be seen in a very different light. [Interview with Carlos Pavón, Gigamesh, July 1998]

After all, there are plenty of other authors writing about 20th-century people in space.

His left-wing politics inform his writing. He is amazingly optimistic about the ability of intelligent beings to solve the problems of exponential growth. Malthusian fights over resources don’t feature in his work: his message, loud and clear, is that so long as we eschew exponential growth then there is plenty of time and space for intelligent beings to build utopias (though Egan’s utopias involve their citizens studying more mathematics than your average utopia):

Exponential growth always hits a brick wall, eventually. There might be situations where it’s a good idea for a short period, but as a general strategy, finding ways not to need it is the best investment a civilisation can make. [rec.arts.sf.written, ]

This Usenet debate appears, slightly disguised, in several of Egan’s books. The position of Mark Atwood (that sustainable civilizations will inevitably be overrun in the Malthusian struggle by—“end up as food for” is Atwood’s pithy phrase—their exponentially-growing neighbours and therefore we must commit ourselves to the fastest possible exponential growth lest our descendants be destroyed by beings that achieved a higher exponent) appears as a debating point several times in Egan’s oeuvre, usually scorned as a savage and primitive view. (Way to win your Usenet arguments!)

However, the political and philosophical aspects of his books are unsatisfactory because Egan has trouble in making sympathetic characters with opposing points of view. He presents these characters, he tries to present them fairly, but to me they seem like puppets, mouthpieces for the opposing viewpoint. (Prospero in “The Planck Dive” is a typical example.)

So he’s a bit of a minority taste, but I’m in that minority and I enjoy his works more than I do the better-characterized, more colourful science fiction of his peers.

Which brings me to Incandescence, Egan’s latest novel.

Incandescence is a book built around a classic science-fictional pattern: the characters are placed in a situation that forces them to learn the laws or rules that govern their world, and then exploit these rules to survive and escape their predicament.

(I don’t normally worry too much about spoiler warnings, because most books and films have very little that’s original or surprising in them. But Incandescence is written in the form of a puzzle, so if you would prefer the pleasure of working it out for yourself, then here be spoilers.)

There are really three interesting puzzles in the book, all standard sf tropes, but executed with style and in a manner that respects the reader’s intelligence and gives you the clues to work out some of the answers for yourself. First, what manner of beings are the protagonists and where do they find themselves? Second, what are the rules or laws that govern the protagonists’ world? And third, how did the protagonists’ rather unlikely predicament come to pass in the first place?

The protagonists are alien beings living in the Splinter, a rocky planetoid orbiting in the Incandescence, the accretion disk of plasma surrounding a massive black hole. The Splinter’s orbit is decaying through friction with the Incandescence, and unless the protagonists can discover their predicament and work out how to steer the Splinter further away from the black hole their world will be torn apart by tidal forces.

The rules governing their world are the laws of gravitation as described by the theory of general relativity. Starting with simple observations of the behaviour of pebbles floating in free fall at the centre of the Splinter, the protagonists work out that their world is in orbit, and the dynamics governing their motion.

It’s entertaining and challenging to follow along, like auditing a course in orbital dynamics being lectured in an alien language. My only complaint is that there aren’t enough diagrams: some of the dynamical behaviours described are really difficult to visualise from the prose descriptions, and more diagrams would have helped a lot. Greg Egan’s web site has diagrams and animations aplenty, but it assumes that you’ve read that book or don’t mind the plot being spoiled. So let me appeal for a second edition of the novel with twenty or so well-chosen figures.

This is science fiction that really makes you work: many things about the background are classic sf tropes that are simply mentioned in passing, with the assumption that the reader is paying attention and recognizes the idea from its many previous expositions: software personalities, personality backup and duplication; panspermia, the prime directive, immortals dealing with boredom, the difficulties of arranging meetings on the other side of the galaxy when information travels at light speed. (For a taste of this, try the short story “Riding the Crocodile”, set in the same fictional universe.)

An interesting aspect of this book is that you can see the construction lines. It’s clear that Egan started with the thought, “How can I make a book in which the survival of the characters depends on their accurate understanding of general relativity?” and that thought leads fairly directly to the scenario in the book, since only somewhere like the close environs of a black hole can the difference between Newton and Einstein make the difference between life and death. But the resulting situation has a number of features that need explaining.

First, how did the alien civilization end up closely orbiting a black hole in the first place (in the crowded neighbourhood of a large black hole in the galactic centre, stars live for too little time for civilizations to evolve on their planets). And second, how come the alien civilization has the intellectual capacity to discover general relativity and work out an escape mechanism just at the time when it is needed? Both of these are rather implausible.

A lesser writer might have hand-waved over these gaps in the plot, but Egan confronts them head-on, resulting in a second strand to the novel that explains the contrived situation from the first strand, though the two never intersect. And the second strand gives a fine finale to the novel, with a compelling moral dilemma.

  1.  The theorem Egan’s referring to is from Bennett et al. (1996), “Strengths and Weaknesses of Quantum Computing”: relative to a random oracle, NP ⊈ BQP. Roughly speaking, this means that quantum computers are probably unable to solve NP-hard problems in polynomial time (as they can in Quarantine).

    I think Egan is being slightly hard on himself here. Random oracle results are by no means conclusive, for example relative to a random oracle, IP ≠ PSPACE, but back in the real world IP = PSPACE, as shown by Chang et al. (1990), “The random oracle hypothesis is false”. So there’s still a chink of doubt through which the computational speculation of Quarantine can slip.

Update 2009-03-20. I wrote a detailed response to a review of Incandescence by Adam Roberts.