The Puzzle of Possibility

10 January 2018

Happy New Year!  Now that we’ve launched into 2018, many of us are wondering what the year ahead has in store.  What might happen, to you, your loved ones, the nation or the world as a whole?  There seem to be a lot of possibilities, some to be hoped for and others to be feared.

Philosophers are as much concerned about the possibilities that lie ahead as anyone else is. But philosophers are also interested in possibilities for a different reason—or rather, in a different way. When we consider possibilities, most of us are curious about what is possible, but a lot of philosophers are also curious about what possibility is.  Put a little differently, a lot of philosophers are interested in the question of what it is that we’re talking about when we talk about possibility.

You might respond, “Well it’s obvious, isn’t it?  When I say that it’s possible that it’s going to snow in Savannah tomorrow, I’m just talking about the fact that it might snow in Savannah tomorrow.”  But this doesn’t really get us anywhere, because it’s just using different words to say the very same thing.  It’s like saying “Wealthy people have lots of money.”  Duh.

There’s been a whole industry in philosophy devoted to solving the puzzle of possibility. To describe this fully would require a book rather than a blog, so I’m going to confine myself to sketch just a tiny bit of it. There are plenty of great sources for those who want to learn more.

The most notorious, and most influential, way to address the problem of possibility was the brainchild of the philosopher David Lewis, and it’s really, really weird.  The idea is that true counterfactual statements aren’t really counter to the facts because if that were the case there’s no way that they could be true. So there simply must be facts that make these statements true. These are obviously not facts about our world (it didn’t snow in Savannah today), so they must be facts about other worlds—parallel universes or, in the philosophical jargon “possible worlds.”  When I truly say it could have snowed in Savannah, this is true because there’s some world—some parallel universe—where it snowed in Savannah.  For the theory to get off the ground, there’s got to be very many of these worlds (what Lewis called “an infinite plurality of worlds”) because there’s got to be at least one such world for every way that our world could have been (but isn’t).  Could there be porcupines that smell like crème brûlée?   Sure, why not?  Then there must be a world where these delightful creatures have that delightful smell.

I know that this all sounds pretty wacky.  But to say that something sounds wacky by commonsense standards isn’t a good refutation of it.  After all, quantum physics, which is possibly the most powerful scientific theory every developed, is super-crazy by commonsense assumptions, but that’s no grounds for rejecting it

However, there are other reasons to be unhappy with this approach to the problem.  Here’s a good one. Many people would say that Trump might have lost to Clinton in the last presidential election. This was a real possibility. According to the “possible worlds” idea, this means that there’s some world or worlds where Clinton really did beat Trump. But there’s a problem with this, because it can only work if Trump exists in two different places.  Imagine that The New York Times reports that Trump is in Trump Tower on election night and that Fox News reports that Trump—the very same guy—is in San Francisco at the very same time.  At least one of them has got to be wrong, because a single person can’t be in two different places at the same time.  

The same principle applies to possible worlds.  If a person is at one world, the very same person can’t be at a different one.  The same problem rears its head when it comes to winning or losing elections. At this world—the actual world where I’m writing these words—Trump is the one who won.  It can’t be that he won the election in one world and lost it in another, because that would mean that he both won and lost the very same election, which is impossible.

One way that philosophers try to fix this problem is to say that the braggadocious orange guy who lost the election in that other world isn’t really Trump. He’s just a guy who’s very, very similar to Trump—Trump’s counterpart. But this gets us into an even deeper mess, because it means that when I say that Trump could have lost, I’m actually talking about Trump’s counterpart rather than about Trump.  But when I say that it was possible for Clinton to have beaten Trump, I mean to be saying something about was possible for Trump and not what happened to some guy that resembles him.  So the fix doesn’t look a fix after all.

A different approach to the problem of possibility—one that I find much more attractive—is to deny that there really are any possibilities. Possibilities are a kind of conceptual illusion. Nothing could have been different than it was, and the future isn’t full of actualized possibilities floating out there in metaphysical space waiting to be actualized (or not, as the case may be).  So how come it seems like that possibilities are real and that some counterfactual statements are true?  Maybe the explanation is a pedestrian psychological one. Perhaps it’s because we are endowed with minds powerful enough to envisage a range of alternative pasts, presents, and futures that we fall under the spell of thinking that these alternatives are, in some sense, real.

As is true of virtually every philosophical problem, there’s no consensus about how to solve the puzzle of possibility. But whatever the solution turns out to be, I’m hoping that the best of what’s possible will be actual in the year ahead.


Comments (3)

Harold G. Neuman's picture

Harold G. Neuman

Wednesday, January 10, 2018 -- 11:28 AM

I liked the sense of "if so,

I liked the sense of "if so, then..." hypothesized by David Hume (see: Hume's Fork). These is some animal (maybe South American) that supposedly smells like popcorn---saw that on a nature show once. Seems to me that Possibility is a sibling in the family of Chance. It is slightly more likely because there is more that can be done to increase its likelihood of realization. There are innumerable possibilities for 2018, just as there were for preceding years. Some are desirable, others not so much. The agenda depends upon the beholder and one person's realization may be some else's nightmare. Chance, fate, possibility? Hume's Fork simplified things. Too much, though...

villageidiot's picture


Friday, January 12, 2018 -- 8:01 PM

Any serious philosopher - who

Any serious philosopher - who doesn't promote nonsense theories like "alternate universes" to impress fellow nerds - must read Wittgenstein. If the problem of possibility affects one to the point of torture, then they would give anything to dissolve the question, to free themselves from the stranglehold, and these are the souls for whom Wittgenstein is of great use. And the only way to dissolve the question is to look at the language that is used to pose the question in the first place.

ofinfinitejest's picture


Sunday, January 14, 2018 -- 10:57 AM

I think it makes the most

I think it makes the most sense to suggest that if the multiverse is true, then it means a very large number of alternate states of affairs and timelines exist, but absolutely not an infinite number (some physicists have ball-parked this number at ~10^500). Thus there is a world where Trump lost, its nearby variations are high in number, but there are also a small number of universes in which a transparent flying arctic giraffe named Rumpelstiltskin won the election. There are a much higher number of universes in which this giraffe is a more moral and intelligent leader than Trump, but I digress... That's a weird but not intrinsically impossible, and there are no universes where for example, modus ponens does not hold as a valid logical argument. We find ourselves in a more likely rather than less likely universe, but not adrift in an infinite space of the song "Anything Goes."