| HN Mirror

ELI5 attempt:

Suppose you're using Newtonian mechanics to model the behavior of a moving baseball that bounces off a brick wall. You write down some equations, do the experiment, the results match, yay science.

Now imagine your lazy coworker was supposed to do the same experiment, but he couldn't find any bricks, so instead of a wall he just built a tiny rocket engine inside the tennis ball that can exert exactly enough force to reverse its velocity. The ball's velocity changes in exactly the same way, except that it changes at a certain time (when your lazy coworker triggers it) rather than at a certain place (where the wall is). Would his results match yours?

For a single ball, yes. The behavior would be identical between these two scenarios. If you looked at a graph of the ball's location over time, you wouldn't be able to tell whether it had bounced off a wall or been "bounced" by a little rocket engine. Right?

Now consider a stream of baseballs. In your lab, your baseball cannon fires a stream of balls which all reverse velocity at a certain place in space, but your coworker is firing a stream of balls that all reverse velocity at a certain time. Would they also behave identically?

No, this time there are differences. For one example, the wall-balls would collide with each other (the ball that just hit the wall would run into the ball that's about to hit the wall) and the rocket-balls wouldn't (each ball reverses direction at the same time, so the spacing between the balls never changes). For another difference, they'd bounce back in a different order (the first wall-ball to be shot would be the first to return, whereas the last rocket-ball to leave the cannon would be the first to return).

That is more or less what this experiment is doing except they're using light rather than baseballs, and a substance with variable reflectiveness rather than tiny rockets. This might have repercussions for time symmetry, that's above my pay grade, but definitely doesn't involve time travel.