Sørina Higgins gave a talk recently at the Brazilian Mythopoeic Society about how time flows in fantasy. This is something that has interested me ever since I read Umberto Eco’s essay “The Woods of Loisy”¹. One of the techniques Eco used there to describe the temporal flow in a story was to make a graph of the “in-world time”, what the calendar on the wall says, versus the reader’s progression through the narrative. I’ll use “page number” to stand in for that. Eco uses it for The Odyssey, and Sylvie by Gerard de Nerval, and a limerick about a man from Peru.² I want to use it for stories about literal time-travel, instead of a narrative that shifts about in time while the characters all go forward.

Figure 1.

For experimental purposes, let’s construct a trivial time-travel story: A mad scientist in Texas invents a time machine. He uses it to go back to last February in Brazil. While he’s sight-seeing there, a butterfly lands on his shoulder and he brushes it off. Then he climbs back into his time machine and returns to the time he left. Well, we all know about the awesome power of butterflies in Brazil. When he returns, his lab has been blown apart by a tornado, the infrastructure for time-travel is wrecked, and so he sets about the job of rebuilding, one day at a time like the rest of us have to. The End.

Figure 1 is what that story looks like when it’s drawn as one of Eco’s diagrams. An upward-sloping line is what we all do all the time. A big jump straight up or down is when the time machine causes a change of the in-world time on a single page for the reader. These parts are the pure science fiction.

Among Sørina’s citations are Ted Chiang’s “The Story of Your Life”³ and Richard Feynman’s Nobel lecture. Though she didn’t mention it, these two texts share a deep structural element. Chiang’s protagonist learns an alien linguistic form from creatures who perceive language as a kind of variational principle, and ends up seeing her daughter’s life in that way, instead of via pedestrian linear time. Chiang, according to the end-notes in my copy of the book, is fascinated by Lagrangian dynamics, so he wrote them into a story.

Feynman earned his Nobel Prize for applying Lagrangian variational principles to Quantum Electrodynamics, and in the process inventing a way to compute preposterously-complex integrals⁴ without making your head explode. That method is now called a “Feynman diagram”. A Feynman diagram has solid lines with arrows for electrons, quarks, etc. There are wavy lines for photons and curlicue lines for gluons. Other bosons are represented by dashed lines. (E.g. the famous Higgs boson, but there are lots of smaller ones.) There are rules about how different lines connect at vertices, and if you follow all the rules, you can read the function you need to integrate off the diagram, and you’re sure to be doing a calculation that makes sense.

One of the key insights that made diagrams possible was that we can think of a particle of anti-matter as a particle of regular matter traveling backward in time. That’s because the critical parameter describing motion is the product of energy and time, so, mathematically, there’s no difference between something with positive energy going forwards and something with negative energy going backwards. -iEt = i(-E)t = iE(-t), right? But the corners of the red zigzag in Figure 1 all have one arrow coming in and one going out, which means they obey the most important rule of Feynman diagrams.

Figure 2.

In Figure 2, let’s fix the diagram in Figure 1 so that it obeys the rest of the rules, too. Those blue dashed lines are some kind of boson. They represent a force coming into the story from outside, which causes the time machine to turn on or off.

The upper left corner, when the time machine is first turned on if we read from left to right, has a forward-in-time arrow and an backward-in-time arrow if we read from bottom to top. That’s particle anti-particle annihilation. The bottom left corner is the reverse, called pair production. The third and fourth corners are good old scattering, as a particle gets kicked so it moves differently but doesn’t change into antimatter or anything. Another fun thing is that the internal lines don’t have to obey one law of physics (E=mc²); breaking one law of physics is very useful for someone in a science fiction story.

Now, if I were a French philosopher, I’d say something like we’ve drawn the role of the author into the story. And the next step is to add up the contributions from all the possible locations of the vertices and all possible trajectories of the internal lines, which means that all stories involving turning a time-machine on and off twice will be added together. Most of them will cancel each other out, but the ones that reinforce each other will be the enduring Ur-myth of the Time Machine.  Good thing I’m not a French philosopher!

But that means that I don’ t know what is represented by those blue dashed lines. I know they aren’t eternal; the number of them isn’t conserved. They can be created or destroyed by interaction with a plot.  What do you think they are?

Appendix

Here are Feynman diagrams for two simple scattering events.

time flows left to right

Have you ever wondered how exactly to define “science fiction”? I’ve seen lots of opinions, none of which sounded right.  In an article in PMLA, though, Eric Rabkin pointed me in an interesting direction.  He said “Carl Freedman wrote that his ‘thesis about critical theory and science fiction is that each is a version of the other’.”  Intriguing — they certainly have lots of things in common, since critical theory is constitutionally incapable of staying in one lane, and so wanders from literature to politics to economics to sociology to linguistics and back again.  SF started out that way on purpose.  Maybe there’s something there.

Via inter-library loan, Carl Freedman tells me that Darko Suvin defines science fiction as “the literature of cognitive estrangement.”  Whoa!  Assessing the analytical validity of that assertion will take some thought, but my immediate reaction was that it’s a perfect description of science-fiction fans. Suvin has got to be onto something.  SF fans are not only cognitively estranged from the mundane world, they positively embrace that status. (I say “they” as if I’m not one of them, but I doubt anyone will be fooled.) We’re proud of how differently we think, and judge our ways superior.  Somewhere in a box around here is a button that says “Reality is a crutch for people who can’t handle science fiction.”  We give non-fans scornful names like “mundanes” or “muggles”.

That last word brought me up short. Though SF fans use it, it’s from fantasy. I’m looking for a way to distinguish SF from fantasy.  Fortunately, Darko Suvin actually said something rather different.

SF is, then, a literary genre whose necessary and sufficient conditions are the presence and interaction of estrangement and cognition, and whose main formal device is an imaginative framework alternative to the author’s empirical environment.

“Estrangement” is easy, and obviously correct.  SF is set in a world that is not our own (though it may look just like ours, at the start).

The word “cognition”, once you work through the jungle of academic prose, isn’t quite right. As Suvin eventually explains, it means that the writer isn’t trying to give authoritative statements about characters, relationships, or phenomena. The writer is trying to pose them as problems to be solved, to expose them to inquiry.

The reader’s response is equally important.  The reader of SF is expected to think about the story in the same way as a scientist approaches an experiment: “How does this work? What processes are happening here? What can I learn from this?”  Fantasy is read completely differently.  “What a wonderful/terrifying world! Can I go there?”  This distinction in reader response is what distinguishes SF from fantasy, and from all other genres of estrangement.  In fact, it may be what gives SF the right to have the word “science” in its name.  Science fiction is not “fiction about science”. When I look at Robert Heinlein, for example, it’s hard to find any science in any book he wrote.  All the characters are engineers, solving engineering problems.  But they are solving problems, and the readers are right there working alongside them.

(As an aside, it’s interesting that when C.S. Lewis invented reader-response criticism, he talked for a whole page about SF [p.110], and said things that make it clear that he knew all this, but he never actually said it explicitly.)

So this is useful, and I shall use it.  But now I’m wondering why the terminology shifted.  The term that grabbed me was “cognitive estrangement”, not “estrangement and cognition”.  And I’m not the only one. Here’s how the two terms have fared so far:

So I conclude that the road to scholarly fame runs through steps:

1. Think an intriguing and useful thought.
2. Give it a clunky name that uses the same words as a good bumper-sticker.
3. Wait for somebody else to slip up and say the bumper-sticker version.
4. Watch the citations roll in.

Obligatory Quantitative Appendix

• My citation of Rabkin is at the top of a blog post.
• Rabkin’s citation to Freedman is on Page 2 of his 18-page paper.
• Freedman’s citation to Suvin is on page xv of his book.
• Suvin’s concise statement is on page 7 of his book.

We don’t have to read far into a work of literary criticism to get all the good stuff!

Works Cited

Freedman, Carl. Critical theory and science fiction. Wesleyan University Press, 2000.

Lewis, Clive Staples. An experiment in criticism. Cambridge University Press, 1992.

Rabkin, Eric S. “Science fiction and the future of criticism.” PMLA 119.3 (2004): 457-473.

Suvin, Darko. “Estrangement and Cognition“, in Metamorphoses of science fiction: On the poetics and history of a literary genre. Yale University Press, 1979.

Mythmoot V is coming up, and the theme is “Fantastic Frontiers”. That’s got me looking at all kinds of boundaries, frontiers, and demarcations in speculative fiction.

Isaac Asimov once wrote this about how to write a science-fiction detective story (which his editor said couldn’t be done):

The writer must carefully explain to the reader all their boundary conditions of the imaginary society. It must be perfectly clear what can be done and what can’t be done and with those boundaries fixed, the reader must then see and hear everything the investigator sees and hears, and he must be aware of every clue the investigator comes across.

From the hundred-or-so Asimov stories I’ve read, I infer that this is a good rule for all SF stories if you replace “investigator” with “protagonist”. Readers are willing to make great leaps of faith to enjoy a story, as long as the boundaries are fixed [1] and we know how far the author wants us to jump.

There is one side, though, on which the boundary is fuzzy and permeable. That’s the side that faces the reader. Anything that’s familiar to us in our mundane world can be in a science fiction story. In fact, it’s almost essential to include a bunch of things from our world to keep from weirding the readers out.

It amused me to notice that the theater works the exact opposite way. The wall that faces the audience is the only fixed one, there. The other three walls are almost infinitely mutable. From one minute to the next they can be DUKE ORSINO’s palace, The sea-coast, OLIVIA’S house, A street, and Before OLIVIA’s house. The fourth wall can be broken, sure. When it is, the playwright is doing it consciously, in exactly the same way that a science-fiction writer might deliberately minimize the connections between the fictional world and the mundane to build a sense of alienation.

The theatrical version of boundaries seems to apply to television and movies, too. So when the subject of a movie is science-fictional, which one wins? Interestingly, it seems that all the rigid boundaries apply. A science-fiction movie is truly in a box.

[1] This is not to say the boundary is finite. Borrowing from complex analysis [2][3], I consider a boundary at infinity to be fixed.

[2] This concept is also useful in real analysis, but this is a humanities blog where the phrase “real analysis” can easily be misinterpreted, so I bumped it up a level.

[3] I’ve read so many Terry Pratchett stories that nested footnotes don’t look odd to me any more.

Since the Mythgard Academy has just wound up its class on The Hitchhiker’s Guide to the Galaxy, I think it’s a good time to tell you all what the Question really is. Several characters in the book got themselves all worked up because they know the Answer is 42, and there’s no way that such a big question can have such a simple answer. Idiosophy is here to help.

You know that the universe is really just a simulation, right? Well, most computer simulations have a pseudo-random number generator in them. The randomness does two things for us: it lets us account for unpredictable things like wind, and it lets us explore the whole range of things that can happen. For example, I might be trying to figure out how many checkout lines they need at the supermarket, and I might want to know how long the wait would be in the worst-case scenario, even though nobody can describe for me exactly what that is. I can randomize arrival times, number of items purchased, etc., and see which combination causes the worst (or best) result.

We can’t use real random numbers in simulations (assuming that such things aren’t a mere mathematical fiction) because we’d never be able to make sure the software was working correctly. We have to be able to run the program, fix problems, and then run again with all the same inputs to test that our repair job didn’t cause new problems in some other part of the system we’re studying. We use a pseudo-random number generator because it has a “seed” that we can set, which makes the numbers come out the same every time. When the time comes to really learn something, you change the seed to lots of different values.

So here’s the Ultimate Question to Life, the Universe, and Everything: “What seed should we use to start the simulation we live in, to make everything come out all right?”

For what it’s worth, I ran this idea past the smartest engineer I know. He told me he never delivers a simulation without testing it with “42” first.