Skip to main content


"Why do mirrors flip text to be right-to-left, and not top-to-bottom?"


in reply to Ben Weinstein-Raun

Yeah, I sometimes wonder if we had evolved in zero G, and without some other thing to break symmetry, we would not have developed the mirror left-right switching intuition.
in reply to Rick Korzekwa

I think part of the deal is that people normally stand in front of us the way we would stand if we were rotated around the vertical axis, which makes the left-right switching thing feel natural when you're looking at yourself.
in reply to Daniel Filan

and in zero G i feel like this is probably how we'd do it anyway, since it's nice to match face bits with face bits?
in reply to Daniel Filan

I think this falls under "some other thing to break symmetry" though
in reply to Daniel Filan

it's nice to match face bits with face bits


is that what they're calling it these days

in reply to Ben Weinstein-Raun

Honestly, the part where I see myself as being left-right switched rather than up-down switched still feels confusing to me in this frame.
in reply to Ben Weinstein-Raun

Like, if I think it through all the way, I can see why I am actually mentally privileging the vertical axis, but it's hard to get my system 1 to understand this.

Ultimately I think the biggest thing is that humans are roughly bilaterally symmetrical but not top-bottom symmetrical, and my guess is that this is sufficient to do the symmetry breaking. Like, if you parity-flip me, I still visually "make sense" as a left-right-swapped person, but I don't "make sense" as a top-bottom-swapped person; people don't have their heads on the bottom.

in reply to Ben Weinstein-Raun

So there are at least two interesting thought experiments about putative zero-g humans:

  1. What if humans were radially symmetrical? and
  2. What if humans had no visual symmetries at all?
in reply to Ben Weinstein-Raun

I think non-symmetric people in an environment that has gravity might still perceive mirrors as flipping things left-to-right. As we've established, it isn't that the mirror flips things, it's that you flip things when you turn them to face the mirror (or you imagine them having flipped when they face you in the mirror), and so the question is "what is the default way to flip things?". And I think in the presence of gravity, it's most natural to use the direction of gravity as your rotation axis, so that e.g. your mirror self's feet are still on the ground.
in reply to Ben Weinstein-Raun

Yeah, I was about to say the same thing about bilateral symmetry.

Another thing I just realized is that if I look at a mirror on a ceiling it intuitively feels like it's reflecting across the mirror and not switching left/right.