This sounds like the Novikov self-consistency principle.
Say you have a wormhole on a billiard table. It curves around and goes three seconds backward in time. You roll a ball into the wormhole, aimed such that after it exits the wormhole, it will knock its earlier self off the path so that it never enters the wormhole. Paradox.
Except when you try it, instead of emerging along the pathway you aimed, it emerges along a slightly different path, and strikes its earlier self only a glancing blow. And why did it emerge along a different path? Because it was struck a glancing blow.
David Deutsch also proposed a similar resolution for closed timelike curves, which say that if you have a probability distribution of the present, all that's necessary is that when you travel back in time, you get the same probability distribution. In short, if you flip a (true!) random coin to before shooting your grandfather, no paradox exists:
The problem with this is the universe is no longer causal. Events can happen which cause themselves. It requires you to try every possible universe and then "throw away" the ones that aren't consistent. And if you simulate a universe, doesn't it cause that universe to exist, even if you later decide you don't like it? Who is to say we don't exist in one of the many "throw away" universes vs the one where the particle causes itself to be consistent.
Say you have a wormhole on a billiard table. It curves around and goes three seconds backward in time. You roll a ball into the wormhole, aimed such that after it exits the wormhole, it will knock its earlier self off the path so that it never enters the wormhole. Paradox.
Except when you try it, instead of emerging along the pathway you aimed, it emerges along a slightly different path, and strikes its earlier self only a glancing blow. And why did it emerge along a different path? Because it was struck a glancing blow.
http://en.wikipedia.org/wiki/Novikov_self-consistency_princi...