Advice Section

On Infinite Possibilities

by Lawrence Watt-Evans

Being an article explaining my attitude toward alternate histories, written for a book on the subject that never, to the best of my knowledge, saw print.

Last revised: March 15, 1999

The idea of other realities is not new. It's much older than science fiction. After all, what is Faerie, or Heaven, or Dreamland, but an alternate world? There have been stories of these other worlds since prehistoric times.

An alternate history, however, is a bit different. I haven't been able to find a clearcut beginning for the idea of a world that began like ours, and then diverged at some point. It seems like a natural enough development from the question, "What if?", but I haven't found any definite examples from before the twentieth century.

Interestingly, the first known collection of alternate histories was a book of essays by assorted writers and historians that appeared in the 1920s, and which had no connection at all to the world of science fiction.

It also had no imitators following on its heels in the mainstream, but not long after the notion crept into science fiction, with stories such as Murray Leinster's "Sidewise in Time" and L. Sprague de Camp's "The Wheels of If."

It makes sense for science fiction writers to tackle the idea; after all, they'd been writing future histories all along, extrapolating from the present. Is it really so different to backtrack and extrapolate from somewhere farther back?

That book of essays took a direct approach to the idea--what if some decision had gone the other way? Each author then described how history might have gone differently than it did, given that change.

One of them, Winston Churchill, got fancy--his essay was entitled, "What If Lee Had Lost At Appamattox?" and was a description of exactly that, written from the point of view of a person in a world where the Confederacy won the War Between the States--and of course, since he had grown up with different information and attitudes than a person would in our world, he got some of it wrong. That was a brilliant bit of innovation, one that's been imitated by Philip K. Dick in The Man in the High Castle, and by several other writers.

Essays, however--especially fictional essays--have a relatively limited audience. When Leinster and de Camp wrote their alternate histories, they presented them in the form of stories. And the easiest way to present a different history to a reader in the context of a story is to take a person from our world, and transport him to the alternate reality--that way, the reader will learn the differences along with the character.

This means, of course, that the alternate history becomes mere background, while the plot plays out in the foreground--just as it does with the invented worlds on other planets or in distant futures that made up most of the science fiction of the day.

That's fine for fictional purposes, and provides the model for a great deal of the alternate history that's out there now, but it also brings up a few points to ponder.

The idea that a person can travel from one version of history to another suggests that both versions actually exist simultaneously. It's not a matter of one having happened instead of the other.

In our normal, everyday lives, we assume that our reality is the only one that exists. In those first essays (as well as plenty of later stories), the alternate reality is also treated as the only one that exists.

But in stories where it's possible to travel from one world to another, then obviously, at least two histories exist simultaneously, and that's become one of the dominant models in alternate histories in science fiction.

And if two exist, differing because some decision could have gone either of two ways, then what of all those other decisions, throughout history?

It was obvious right from the beginning that there would be a huge number of possible worlds--in "The Wheels of If," in 1940, de Camp said, "The number of worlds is infinite, or almost."

It seems to me, though, that most writers haven't thought this through beyond that point. They've taken that to mean, "I can take any event in history, change it, and then work out the consequences from there!"

They're wrong.

Some writers have realized this. Ray Bradbury, in "A Sound of Thunder" (which is a time travel story, rather than a true alternate history), points out that there are so many factors to consider that you cannot predict the outcome of any change--that history is, in modern terms that didn't exist when he wrote the story, a chaotic system.

Larry Niven, in "All the Myriad Ways," pointed out that given all these parallel worlds, not only does everything become possible, but everything possible becomes necessary--if something could happen, then in some world, it did happen.

And if you start to think about this too much, it becomes unmanageable. You realize that everything happens.

You can take it clear down to the quantum level--in fact, you're required to, if you want to approach it logically. There's no reason that only differences resulting from human action should exist; the rationalistic universe of science fiction does not give humanity special privileges. Any time there is any event that could go more than one way--the breakdown of a specific atomic nucleus, for example--then it must go all the possible ways, each in a different universe. If something has a one-in-a-million chance of happening, then it does happen in an infinite number of universes--and it doesn't in 999,999 times as many.

Which means we're into transfinite numbers. Transfinite numbers are something that people can play with, but that nobody really comprehends. The whole structure of parallel worlds is getting out of hand here.

Some writers have recognized this problem and tried to explain it away--Fritz Leiber, for example, in his Change War stories, most notably "Try And Change the Past," posited a Law of Conservation of History. He argued that the universe resists change, that it wants to go along a particular path.

Unfortunately, there's absolutely no reason to believe in such a theory. It's rather obviously an artificial convention the author is using to keep things under control. Without it, just one little slip by Leiber's Snakes or Spiders would wipe out all of human history, including whoever made the change...

Paradoxes like that are inherent in time travel. Parallel worlds not involving time travel don't have that problem, at least.

Anyway, my point is that there's a lot more to parallel world theory than just figuring out, "What if Hoover had been re-elected in 1932?" and then telling a story.

In plain fact, you can't say what would happen if Hoover were re-elected--there are too many variables. There are, in fact, infinite possibilities.

Designing an accurate alternate history is no more possible than predicting the future.

You can design an alternate history, just as you can predict a future--and the odds of getting it right are just as bad. In each case you're extrapolating forward from a given point. We know what happened in our reality from July 1932, when Roosevelt got the Democratic nomination, up until 1994, but there are an infinite number of points where history could have diverged, and an author can't guess right at all of them.

I came up with one possibility when I wrote "Truth, Justice, and the American Way" for Mike Resnick's Alternate Presidents, but I knew, as I worked out my alternate history, that I was only working out one possibility from my chosen starting point, that there were any number of others. What if my guesses about what Henry Stimson would do as Secretary of State in 1934 were wrong? What if the conspirators against Hitler in 1939 would have backed down anyway? What would Stalin have done? What if, what if, what if?

And for each question I could choose an answer, but there was no reason to think it was the answer.

I found this unsatisfying.

Choosing one particular variant is all very well for fictional purposes, I suppose--but for me, it turns the story into pure fantasy. I have nothing against fantasy, but most alternate histories strike me as a rather drab sort of fantasy. That's why I haven't written very many, and why I declined invitations to submit stories to Mike Resnick's other "Alternate" anthologies.

Some authors have made the fantasies less drab--Randall Garrett's Lord Darcy stories take an alternate reality where not only did history diverge, but magic works. Poul Anderson's Operation Chaos and Esther Friesner's Druid's Blood, among others, offer us other magical alternate histories. I greatly enjoy reading those, but don't find myself inclined to write anything of the sort.

Instead of working out individual alternate histories, I find myself interested in the idea of all those transfinitely-many parallel worlds, and moving between them.

If there are an infinite number, how can you ever identify one specific universe?

And how could you be sure you weren't moving between them? Larry Niven touched on that in "The Fog."

So I've written parallel-world stories--"The Drifter," and "Why I Left Harry's All-Night Hamburgers," and "An Infinity of Karen," and several others--but I haven't gone in much for true alternate histories.

Not because I don't see enough possibilities in them, but because I see too many.


All contents and referenced pages are copyright by Lawrence Watt Evans except as noted.
All rights reserved
No reproduction permitted without permission of the author


That's it; here's your list of handy exits:

The Misenchanted Page
Front Page | Main Site | E-mail me!