OK, so you've heard of vacuum fluctuation, right? A particle and antiparticle are spontaneously generated, and most of the time, they come back together and annihilate. When that happens, they haven't interacted with anything else, and so they're called "virtual" particles. But because both the particle and its antiparticle appear, nothing "actually" happens; it's just an interesting nothing.
Now that I put it like that, I suppose it seems really weird if you're uninitiated. But hold on to your butts - it's about to get a whole lot weirder. First, though, we need to talk about symmetry. I promise, though: by the time we get to the end of this, minds will be blown.
So let's say there's a different Universe from ours, we'll call it the Buniverse (pronounce it the same, just put a B-sound in front). The Buniverse is just like ours, except for a few key differences: first, for every particle in the Universe, there's an antiparticle in the Buniverse (protons replaced with antiprotons, electrons with positrons, and so on); second, everything is swapped for its mirror image (we're all left-handed doppelgangers at the physical level); third, time flows backward in the Buniverse (the future is the past and vice-versa). The issue is, if you make all these changes, you actually can't tell the Universe apart from the Buniverse. They're exactly the same. (Unless, of course, you yourself actually traveled there - you'd be annihilated just like your antimatter doppelganger would here in the Universe. Silly.)
My own private speculation is that these are actually three different ways of looking at one and the same thing. Kant said all three formulations of the categorical imperative were really saying the same thing (and if what he thought they were saying was, "You shouldn't make an exception just for yourself," then he was right but he went about it in a really confusing way), and it's kind of weird, but stick with me here. Particles are oscillations in fields, right? And time is also a dimension, right? So we're really just taking the mirror image of everything, quantum fields and time included, and saying, "Look! It comes out the same!" Surprise? Stuff also comes out the same when you flip it upside-down, then backwards, then rotate it halfway around.
But that's just me. I don't know. I'm not a doctor.
But you know who is a doctor? Stephen Freakin' Hawkin'. Hawking. Sorry. I got carried away there. Also, Freakin' isn't his middle name, I'm pretty sure. Anyway. I recently read A Brief History of Time during my breaks at work - my best friend got it for me years ago, but I got distracted in chapter four and never came back to it, and so I finally came back and finished it - point being, he points out an interesting consequence when you take vacuum fluctuation and symmetry together. Check this out.
So you've got your particle and antiparticle, created in a vacuum fluctuation, then coming back together and annihilating each other. Poof, nothing happened. However! The thing about antiparticles is that you can't distinguish an antiparticle moving forward in time from a particle moving backward in time - when you take symmetry into account, they look so goddamn the same, there's just no telling. So another way of looking at it is: a particle appears out of freakin' nowhere, gallivants about the Universe for a bit, then travels back in time to its original point in space, forming a closed time-loop. We see a mirror-imaged vector for an antiparticle going forward in time (so we think), but that's indistinguishable from its own mirror-imaged anti(anti)particle going backward in time to the moment of its own creation.
All right: so it's a particle/antiparticle pair spontaneously arising and then mutually annihilating in an interesting nothing, or it's a particle spontaneously arising and then being yanked back through time to the moment of its own creation on a closed time loop in what is still an interesting nothing. Since they look the same, how can we tell the difference? How can we tell whether nothing happens "first" and then resolves into nothing, or if something happens first and then goes and retcons itself back into nothing having happened all along?
Here's the killer: trying to "tell the difference" is a fool's errand. Symmetry, remember? Oh, also, remember that time is relative and so there's no privileged way of looking at it. The "different" scenarios as summarized in the last paragraph aren't actually different at all: they're just two different ways of saying the same thing. Asking which way it is, is like asking whether you're at 1600 Pennsylvania Avenue or the White House. Samey-samey. If you say it's a particle/antiparticle pair traveling forward in time, you're right; if you say it's a particle traveling along a closed time loop, you're right; if you say it's one but not the other, that's when you're wrong. It's a freakin' Necker Cube, man: it just is what it is, there for all to see. Seeing it one way or another is an artifact of your own perception - it's totally unproblematic being uninterpreted all by itself. But no one way of looking at it (or talking about it, for that matter) is privileged over another, so long as you take all them liney-lines into account.
So think about that for a minute: just chew on it for a good, solid sixty seconds. What we call "vacuum fluctuation" may as well be tiny closed time-loops - and we know near as dammit that they happen because we've observed the Casimir effect.
By the stars, perspective is a Hell of a thing!