I try to keep relatively current on science / speculative fiction because of the reading circles I lead in that genre for our accreditation-driven “Quality Enhancement Plan.” So I just finished reading a short novel by Connie Willis (better known for The Doomsday Book, up next), Bellwether (1996). Why did I read this one? Well, it’s what Overdrive offered as immediately available while I waited for any of the other books I had cued up to check back in. An accident. But as it turns out, it was very interesting, because Willis is trying with some success to work with chaos and complexity and the dynamics of scientific discovery.
As a story, there’s not much here. The main character, Sandra Foster, studies the dynamics of fads and fashions for a research corporation called HiTek. One of the running gags in the book is all the faddish corporo-bureaucratic acronymized gyrations “Management” forces the researchers through, complete with “simplified” paperwork in dozens of pages and cryptic, jargony instructions. It’s good for about one giggle for a guy reading the book because of a Quality Enhancement Plan, but is quickly overdone and tiresome, like these gyrations themselves. There’s also a surly, self-absorbed, incompetent millennial gofer named Flip who keeps showing up and driving the plot. Because she’s an actual presence, she’s among the most irritating devices imaginable, even worse than white whales and old bears deep in the Yoknapatawpha woods. Sandy cluelessly circles and then unsurprisingly hooks up with another geeky scientist, Bennet, a biologist who works on group behavior and learning, and yet is personally immune to fads and fashions. “It isn’t that he doesn’t care what other people think. It’s just that he doesn’t see what it has to do with him.” Well that sounds familiar.
And it goes on and on, with Sandy meeting obstacle after infuriatingly gratuitous obstacle in her quest to figure out why hair bobbing became a thing during the 1920s. It’s actually a great cultural studies topic and anecdotally well-researched by Willis, but Sandy’s field awareness and analytical level are somewhere around advanced undergrad, which gets annoying real quick too. She semi-obsessively collects data on fads both historical and contemporary, for some reason trusting Flip to make copies of her clippings over and over again despite impressively comprehensive evidence that it ain’t gonna happen and they’ll probably get lost, trashed, and/or drowned in Coke. Sandy is, in fact, pretty much just as incompetent as Flip, consistently failing to perceive and act on the obvious. “Just because people are experiencing things doesn’t mean they have any insight into them,” she observes. Yes indeedy. At least she’s relatively good-natured about this, unlike many academics I know who this description would fit. In the climax, via corporate trials and tribulations Sandy and her boy combine forces to study the learning behavior of sheep, which act just like the cartoon sheep cynics compare people to. Because Willis is comparing people to sheep, get it?
The slapdash device-mongering of character and story kept trying to throw me out of the book. But alongside all of that crud the book is explicitly and substantively complexity-themed, and that definitely changes the game. I wanted to see what Willis was going to do with that. She’s interested in how science works, and she’s post-Kuhnian in thinking that it works chaotically. How do fads happen? How does science happen? By luck; by accident.
Scientific discoveries are the same way. People like to think of science as rational and reasonable, following step by step from hypothesis to experiment to conclusion. Dr. Chin, last year’s winner of the Niebnitz Grant, wrote, “The process of scientific discovery is the logical extension of observation by experimentation.” Nothing could be further from the truth. The process is exactly like any other human endeavor—messy, haphazard, misdirected, and heavily influenced by chance.
So, the kind of luck and accident that need all sorts of conditions and processes to self-organize into emergent transitions of practice and understanding. As it turns out, every damn irritating thing in the book, all the distractions and confusions and derailments and frustrations, ends up being conditioned through a discovery agenda into a breakthrough:
Scientific breakthroughs involve combining ideas no one thought to connect before, seeing connections nobody saw before. Chaotic systems create feedback loops that tend to randomize the elements of the system, displace them, shake them around so they’re next to elements they’ve never come in contact with before. Chaotic systems tend to increase in chaos, but not always. Sometimes they restabilize into a new level of order.
I especially like this theme of disruptive cross-connecting and field-hopping as the means to creative emergence. She cites all sorts of examples of this, including Newton’s apple, Poincare’s bus, and Feynman’s bongos. In fact, to hear her tell it, ‘normal’ science never discovered anything. It takes a trip outside of the box.
So far so good, if we accept for a second the implicit dissing of field competence and condition-setting in the run-ups to transformations. But it turns out that Sandy and Bennet, despite being able to reel off impressive accounts of complex systems, interaction and scale dynamics, non-linearity, iteration and feedback, and irreducibility, are both committed to the view that emergence requires a single essential catalytic factor, the ‘bellwether’ of the title. This is dramatized first by their discovery that to get the sheep to do anything they need an actual bellwether, a sheep that’s just a little out in front of the herd, that wanders just a little more purposefully than the rest. It is then cemented for the whole discussion by the ‘discovery’ that [SPOILER ALERT] the worse-than-useless FLIP was actually the catalytic agent who, by her ubiquitous bumbling, maneuvered and forced and prompted and enabled the systems breakdowns that so productively brought Sandy, Ben, and sheep together in just the right alignment to figure out how new things happen.
Well. At least this deus ex machina of progress is not an intelligent designer or the Spirit of Reason or the Great Man / Woman or even self interest and the invisible hand, although the latter probably does a better job of defocalizing the dynamics of systems. But it’s still smuggling back in that dream of the single effective variable, right when we finally have the tools to get over that once and for all. It’s so much easier that way though; so much harder to imagine how the dynamics themselves could assemble into outcomes without anything in particular ‘driving’ or ‘causing’ or even in an older sense ‘ordering’ them.
But, I also wonder if this is the genius of the book. What if Willis meant the book itself to be a bellwether? She wouldn’t be able to go all the way to complex systems in one step, although she could gesture at that. The deus ex machina of the bellwether may be about as far as the herd of regular sci-fi readers are ready to go at the moment. As I think about how to teach complexity, I can see how packaging it in familiar forms and pulling its teeth for a second might just be good marketing. A dopey romance? Frustrating workplace bureaucracy? Overworked pseudo-critical geek cliche’s? I can see how this would be just the strategy to move some readers closer to a tipping point of enlightenment. Now, is that how learning and persuasion actually work? The book says so, and maybe also practices so. I’m not so sure, but from the standpoint of complex systems without the deus ex machina, I’m not sure how possible it is to make any kind of change on purpose.