I rarely find that reading fiction makes me upset. Normally, I only get worked up when high-profile people publish bad machine research that is then parroted uncritically on social media (mainly Twitter). Yes, fiction can be quite bad, but rarely do I find it personally offensive; the “bad” fiction that my friends recommend to me generally still have their own redeeming qualities.
But Greg Egan’s short story “Didicosm” managed it anyway.
Spoilers ahead.
A standard take on Greg Egan’s writing is that the science part of his science fiction is quite good, but the fiction part is comparatively much worse. His skill lies in coming up with interesting alternative physics or integrating interesting math to create an alternative world, but he often struggles to populate the world with characters with satisfying character arcs. “Didicosm” is no exception to this.
The core scientific conceit of the piece is the following: (in reality,) we seem to observe that the universe is flat and spatially unbounded. A natural conclusion (often made in modern cosmology) is that we exist in an infinite, flat universe.
However, this does not necessarily follow. A 3-torus, for example, is locally flat and has no boundary, but of finite volume. In fact, there exist 10-such closed flat Riemannian 3-manifolds, which John Conway dubbed the platycosms, from the Greek platys-, meaning flat, and kosmos/cosmos. (See Conway and Rossetti’s Describing the platycosms for a full discussion of the 10 platycosms.)
The way you’d distinguish between a spatially infinite flat universe and any of the platycosms is by looking for places where the universe seems to repeat. We don’t seem to observe any such patterns in the night sky. But strictly speaking, our observations of the observable universe only strictly rule out platycosms that are small; if our universe is a platycosm with spatial extent much larger than the observable one, then this would be consistent with our observations (even though this might matter for predicting the shape of the far future).
As the title suggests, the universe in “Didicosm” takes on the form of a didicosm, perhaps the most interesting of the platycosms.
So what, then, is the plot of “Didicosm”? How does one turn this interesting mathematical observation into an interesting story?
The plot of didicosm is less about the shape of the universe, and more about the effectiveness of scientific critique in a world where the public’s understanding of science is mediated through charismatic but unaccountable science communicators.
The story starts with the protagonist Charlotte’s father giving her a lecture on how he believes the universe to be spatially infinite, before committing suicide to “live in a better world”.
Charlotte comes to believe that her fathers suicide resulted from claims in a popular science book, Everything Happens! (a parody of Max Tegmark’s Our Mathematical Universe):
At this very moment, countless light-years away, on a planet that looks exactly like the Earth, a person who looks, thinks and acts just like you is reading exactly the same sentence as the one you are reading right now.
Following a confrontation with the book’s author Derek Linderman (a mixture of Max Tegmark and Michio Kaku), she then dedicates her life to proving this claim wrong.
Based on all we can observe, the universe does not contain the repeated patterns that would serve as a smoking gun for a platycosm over the spatially infinite physics, even if we use the cosmic microwave background (CMB), which allows us to map the universe as it existed ~400k years after the big bang.
Charlotte’s idea is to measure the cosmic neutrino background, which would allow us to map the universe as it existed around a second after the big bang. (This allows us to measure the shape of the universe in a volume 3% larger than the CMB data dose). After some effort, she eventually contributes to a new scientific project called NuWave that successfully does so, and her collaborators find that the universe turns is a didicosm. (How exactly NuWave functions is neither described nor important for the plot).
After they announce this discovery to the world, Charlotte becomes disheartened by seeing Linderman refuse to concede defeat and instead pivots to arguing for an infinite greater reality, composed of finite volume didicosms.
But eventually, an undergrad at her university comes to her after class with a quantum gravity based explanation for why the universe takes on the form of a didicosm as opposed to any other platycosm. Charlotte takes comfort in the fact that, even if she cannot change the behavior of science communicators, she can at least inspire the next generation of scientists:
She had to stop thinking of the NuWave results as a failure. Even if nothing was settled, even if people kept disputing them for another thirty years, she had helped to open the door for the next generation to continue searching for the truth.
Sprinkled alongside this main plot are conversations between Charlotte and her partner Vince. Their relationship itself matters little for the plot, and Vince’s main role is to serve as the uninformed outsider that Charlotte and her fellow cosmologists can dump exposition at.
—
If I had to pick one sin in science fiction writing, it’s in writing a story in which the plot does little to add to a description of the central conceit. Despite my complaints about the short story, I found both Conway’s platycosm paper and Egan’s notes on didicosms fascinating. But I think “Didicosm” avoids this sin to some degree – while yes, his characters are relatively flat, and yes, the plot is barebone and not dependent on the specific, there’s a fair amount of exploration
The reason that “Didocosm” made me upset was because it felt like a story of Greg Egan taking potshots at scientific communicators as morally bankrupt while strawmanning their arguments, and also casually inserts some fun facts about flat Riemannian 3-manifolds that matter little for the plot.
First, Egan doesn’t actually present Tegmark’s arguments from his work – instead, his Tegmark stand-in Lindermann first only argues that the spatially infinite universe is the null hypothesis, that Charlotte has failed to reject:
“The science is what it is,” [Lindermann] insisted. “The universe is spatially flat, within the error bars of every measurement we’ve made. So the null hypothesis must be that it goes on forever.”
“Must be?” Charlotte spat back. “There are no less than six kinds of finite flat space that would work just as well.”
“None of which we have evidence of inhabiting.”
“None of which we’d expect, if they were large enough.”
Linderman shook his head stubbornly. “You can dream up as many hypothetical properties for the universe as you like, but if they’re undetectable, no one has any reason to believe in them.”
After the universe is shown to be spatially finite, his arguments turn even more cartoonish:
“We couldn’t ask for starker evidence, really,” Linderman continued, while the interviewer on the split screen nodded encouragingly. “What NuWave revealed might as well have been instructions from some alien Ikea assembly sheet. To build your pocket universe, step one, join tab A to tab B. And now we’re sitting in someone’s bedroom, like a fishtank! One of millions of fishtanks – and that’s on just one planet, in the infinite parent universe.”
Second, Egan takes aim at not just physics/cosmologists, but also other speculative ideas that are obviously ridiculously:
“But even if that book didn’t kill him, it’s part of a whole corrosive trend, where bad pop science click-baits its way into the wider culture. Remember when random celebrities would proclaim that there was a 90 percent chance the universe was a simulation? Or when people with actual political power believed that AI was on the verge of bootstrapping itself to superintelligence?”
Oh hey, that’s me.
For all that Charlotte demands epistemic humility of pop science cosmologists in his story, she sure lacks the same epistemic humility when it comes to other areas of scientific communication.
I think I would be interested in an essay from Greg Egan responding to Tegmark I arguments, and also against the simulation hypothesis and the possibility of ASI. And as previously mentioned, I found his mathematical notes on Didicosms fascinating.
But “Didocosm” is neither. And its lack of charity toward those espousing ideas that Greg Egan finds ridiculous (such as myself) made me upset enough to write this piece.
Also, if you want to come meet me or other InkHaven residents, InkHaven is hosting a fair this Saturday that’s open to the public! See the Partiful for more information.
Discuss