Surfing the Gnarl (6 page)

I'm a writer first and foremost, but for much of my life I had a day-job as a professor, first in mathematics and then in computer science. Although I'm back to being a freelance writer now, I spent twenty years in the dark Satanic mills of Silicon Valley. Originally I thought I was going there as a kind of literary lark like an overbold William Blake manning a loom in Manchester. But eventually I went native on the story. It changed the way I think. I drank the Kool-Aid.

I derived my notion of gnarl from the work of the computer scientist Stephen Wolfram. I first met him in 1984, interviewing him for a science article I was writing. He made a big impression on me, and introduced me to the dynamic graphical computations known as cellular automata, or CAs for short. The so-called Game of Life is the best-known CA. You start with a few lit-up pixels on a computer screen. Each pixel “looks” at the eight nearest pixels, counts how many are “on” and adjusts its state according to this total, using a fixed rule. All of the pixels do this at once, so the screen behaves like a parallel computation. The patterns of dots grow, reproduce, and/or die, sometimes generating persistent moving patterns known
as gliders. I became fascinated by CAs, and it's thanks in part to Wolfram that I switched from teaching math to teaching computer science.

Wolfram summarized his ideas in his thick 2002 tome,
A New Kind of Science.
To me, having known Wolfram for many years by then, the ideas in the book seemed obviously true. I went on to write my own nonfiction book,
The Lifebox, the Seashell, and the Soul,
partly to popularize Wolfram's ideas, and partly to expatiate upon my own notions of the meaning of computation. A work of early geek philosophy. Most scientists found the new ideas to be—as Wolfram sarcastically put it—either trivial or wrong. When a set of ideas provokes such resistance, it's a sign of an impending paradigm shift.

So what does Wolfram say? I'll break this into four points.

(1) Wolfram starts by arguing that we can think of any natural process as a computation, that is, you can see anything as a deterministic procedure that works out the consequences of some initial conditions. Instead of viewing the world as made of atoms or of curved space or of natural laws, we can try viewing it as made of computations. Keep in mind that a “computer” doesn't have to be made of wires and silicon chips in a box. It can be any real-world phenomenon you like.

(2) Having studied a very large number of visually interesting computations called cellular automata, Wolfram concluded that there are basically three kinds of computations and three corresponding kinds of natural processes.

Predictable.
Processes that are ultimately without surprise. This may be because they eventually die out and become constant, or because they're repetitive. Think of a checkerboard, or a clock, or a fire that burns down to dead ashes.

Gnarly.
Processes that are structured in interesting ways but are nonetheless unpredictable. Here we think of a vine, or a waterfall, or the startling yet computable digits of pi.

Random.
Processes that are completely messy and unstructured. Think of the molecules eternally bouncing off” each other in air, or the cosmic rays from outer space.

The gnarly middle zone is where it's at. Essentially all of the interesting patterns in physics and biology are gnarly. Gnarly processes hold out the lure of being partially understandable, but they resist falling into dull predictability.

(3) Wolfram's third tenet is that all gnarly computations are in fact universal computations. “Universal computation” is used in the technical computer-scientific sense of a computation that can in fact emulate any other computation. Universal computations aren't at all rare. Every desktop or smartphone computer is a universal computer in the sense that it can, given enough time and memory, model the behavior of any other such computer.

Given that physical processes are a type of computation, it's natural that the virtual worlds of our videogames support a kind of artificial physics. The objects in these little worlds bounce off” each other, the projectiles follow trajectories shaped by “gravity,” the race-cars skid and spin out when they make overly sharp turns.

Wolfram says we can turn things around. An interesting physical process is a gnarly computation, any gnarly computation is a universal computation, therefore any interesting real world process can, in principle, emulate any other naturally occurring process.

In some sense we're all the same: a cloud can emulate an oak tree, a flickering flame can model a human mind, a dripping faucet can behave like the stock market.

If this strikes you as a strange way to think, you're in good company. The universality of naturally occurring gnarly computations is something that the older generation of scientists finds baffling and outrageous.

(4) Nothing of any significance in the natural world is predictable. Science's dreams of ultimate mastery are self-aggrandizing horseshit.

How so? As argued in point (3), all the interesting naturally occurring computations are gnarly computations, and these gnarly computations are universal computations with the ability of emulating each other. Given these facts, it's possible, via some ironclad computer-science legerdemain, to prove that the interesting processes of nature are inherently unpredictable. The problem is that if you can predict the behavior of a particular universal computation, you run head-on into the Unsolvability of the Halting Problem, a paradoxical result proved by the early computer scientist Alan Turing in 1936.

What, by the way, do I mean by “predicting a process”? This means to have some procedure for determining the processes result very much faster than the time it takes to simply let the process run. The point of result (4) is that there are no quick short-cut methods for finding out what
a gnarly computation will do. The only way to really find out what the weather is going to be like tomorrow is to wait twenty-four hours and see. The only way for me to find out what I'm going to put into the final paragraph of a book is to finish writing the book.

It's worth repeating this point. We will never find any magical tiny theory that allows us to make quick pencil-and-paper calculations about the future. Sometimes scientists—or science-fiction writers—have speculated that there's some compact master-formula capable of predicting the future with a few strokes of a pencil. And many still have an internal faith in some slightly more sophisticated restatement of this.

But, as Wolfram so convincingly argues, the world, being gnarly, is inherently unpredictable. We have no hope of control. On the plus side, gnarl is a bit better behaved than the random. We can hope to ride the waves.

Anything involving fluids can be a rich source of gnarl—even a cup of tea. The most orderly state of a liquid is, of course, for it to be standing still. If one lets water run rather slowly down a channel, the water moves smoothly, with a predictable pattern of ripples.

As more water is put into a channel, the ripples begin to crisscross and waver. Eddies and whirlpools appear—and with turbulent flow we have the birth of gnarl.

Once a massive amount of water is poured down the channel, we get a less interesting random state in which the water is seething. At this point I should caution that I'm using “random” the loose sense of “having no perceivable pattern.” It might be that a liquid or some other
complex process is in fact obeying a deterministic rule and is what we more properly call “pseudorandom.” But I'll just say “random” to keep the discussion simple.

Besides the flow of water, another good day-to-day example of a gnarly physical process is a tree whose leaves and branches are trembling in the breeze. Here's some journal notes I wrote about a tree I saw while backpacking near Big Sur with my daughter Isabel and her husband Gus in May 2003.

Green hills, wonderfully curved, the gnarly oaks, fractal white cloud puffs, the Pacific Ocean hanging anomalously high in the sky, fog-quilted.

I got up first, right before sunrise, and I was looking at a medium-sized pine tree just down the ridge from my tent. Gentle dawn breezes were playing over the tree, and every single one of its needles was quivering, oscillating through its own characteristic range of frequencies, and the needle clumps and branches were rocking as well, working their way around their own particular phase space, the whole motion harmonious in the extreme. Insects buzzed about the tree, and, having looked in the microscope so much of late, I could easily visualize the microorganisms upon the needles, in the beads of sap, beneath the bark, in the insects' guts—the tree a microcosmos. The sun came rolling up over the ridge, gilding my pine. With all its needles aflutter it was like an anemone, like a dancer, like a cartoon character with a halo of alertness rays.

“I love you,” I said to the tree, for just that moment not even needing to reach past the tree to imagine any divinity behind it, for just that moment seeing the tree itself as a god.

When we got home there were my usual daily problems to confront and I felt uptight. And now, writing these notes, I ask how can I get some serenity?

I have the laptop here on a cafe table under a spring-green tree in sunny blue-sky Los Gatos. I look up at the tree overhead, a linden with very small pale fresh green leaves. And yes the leaves are doing the hand jive. The branches rocking. The very image of my wandering thoughts, eternally revisiting the same topics. It's good.

The trees, the leaves, the clouds, my mind, it's all the same, all so beautifully gnarly.

As a reader, I've always sought the gnarl, that is, I like to find odd, interesting, unpredictable kinds of books, possibly with outré or transgressive themes. My favorites would include Jack Kerouac and William Burroughs, Robert Sheckley and Phil Dick, Jorge Luis Borges and Thomas Pynchon.

Once again, a gnarly process is complex and unpredictable without being random. If a story hews to some very familiar pattern, it feels stale. But if absolutely anything can happen, a story becomes as unengaging as someone else's dream. The gnarly zone lies at the interface between logic and fantasy.

William Burroughs was an ascended master of the gnarl. He believed in having his work take on an autonomous life to the point of becoming a world that the author inhabits. “The writer has been there or he can't write about it…. [Writers] are trying to create a universe in which they have lived or where they would like to live. To write it, they must go there and submit to conditions that they might not have bargained for.” (From “Remembering Jack
Kerouac” in
The Adding Machine: Selected Essays,
Seaver Books, 1986.)

In order to present some ideas about how gnarl applies to literature in general, and to science-fiction in particular, the table below summarizes how gnarliness makes its way into literature in four areas: subject matter, plot, scientific speculation, and social commentary.

In drawing up the table, I found it useful to distinguish between
low gnarl
and
high gnarl.
Low gnarl is close to being periodic and predictable, while high gnarl is closer to being fully random.

Keep in mind that I'm not saying any particular row of the table is absolutely better than the others. My purpose here is taxonomic rather than prescriptive. To this end, rather than using the word “predictable” and “random” to refer to the lowest and highest levels of complexity, I'll use the less judgmental words “classic” and “surreal.”

In order to spark discussion, I've positioned the names of some of my favorite fantasy or science-fiction authors in the first cells of each row. Note that some authors may write novels in various modes—Terry Bisson's
Pirates of the Universe,
for instance, is high gnarl and transreal, while his
The Pickup Artist
is a surreal shaggy-dog story. Also note that any given novel may have different complexity levels relative to the four columns.