Saturday, October 05, 2013
Fictional Worlds: True, but Possible?
In an essay titled “Truth in Fiction,” philosopher David Lewis argued that worlds created by fiction writers, worlds where “the fiction is told, but as known fact rather than fiction,” are not only possible but real.
“When I profess realism about possible worlds, I mean to be taken literally,” Lewis wrote elsewhere about his possible-worlds theory, which he called modal realism.
It’s an intriguing proposition for those of us who sometimes wake as if from a trance after spending time with imaginary characters who have sprung into being as we scribbled away or whacked at a keyboard. We thought we were just making it up.
Lewis uses Sherlock Holmes stories as a benchmark, set in a realistic world much like our own. Of other kinds of fiction, which may retail the exploits of superheroes from other planets or hobbits, he says huffily, “what a mistake it would be to class the Holmes stories with these!”
Nevertheless, a fiction is impossible if and only if “there is no world where it is told as known fact.” That leaves a lot of room. If in the first volume of a novel set in 1878 our hero has lunch in Glasgow and in the last volume shows up in London later the same day, he either used a then-unknown form of transit or, what may seem more likely, his author was careless.
In the latter case one could suppose a “minimally revised version” of the novel that makes the hero’s movements feasible, Lewis suggests, assuming “there were ways to set the times right without major changes in the plot.”
But here’s the kicker. “There might not be, and in that case perhaps truth in the original version – surprising though some of it may be – is the best we can do.”
Lewis joined the faculty of Princeton University in 1970, where thirteen years earlier Hugh Everett III had proposed the theory later known as the many worlds interpretation of quantum mechanics.
Quantum states are superpositions of possible alternatives – for example, a particle with both spin up and spin down, or, more notoriously, Schrödinger’s cat both alive and dead – which are only resolved when an observer makes a measurement, thus “collapsing” the wave function.
Everett’s solution to this non-common-sensical conundrum was to posit a universal wave function with an uncountable infinity of possible branches describing the whole of existence, in which all possible worlds are real. The collapse of local wave functions is illusory – all alternatives continue.
Scorned at first, not least by Neils Bohr, many worlds theory is taken seriously and remains the object of lively debate. For obvious reasons it’s a favorite of science fiction writers, first coming to wide public attention in a 1976 article in Analog.
Was David Lewis’s possible worlds theory influenced by Hugh Everett’s many worlds theory? It’s suggestive that Lewis followed Everett at Princeton, but nothing more than that. I’m pretty sure Everett’s presence at Sandia Base in 1956, when I was a teenager there (I don’t recall meeting him), had no influence on my 1981 novel Re-entry, which explicitly invokes many worlds. More likely it was Analog.
What the two theories don’t have in common is necessary physics. Everett’s many worlds exist in a universe with physical laws the same everywhere. Lewis’s possible worlds require only “known fact,” which may involve quite different physics.
To change one’s physics, one needs multiple universes. Brilliant Blunders, Mario Livio’s recent book about scientists’ mistakes, describes how multiple universes have been invoked to explain otherwise unanswered questions about dark energy.
Are even multiverses enough to encompass all possible worlds? More next time.