It all depends on the geometry. If you're talking about a differentiable manifold or something, then the geometry is intrinsic and locally defined by a metric. Curvature and topology are some of the most weird and difficult areas in mathematics.
If you're assuming a well-behaved system with infinitely long planes in euclidean 2-space, then I believe it's true. (I think it might always be true if you reduce the system enough.)
The biggest thing in pure mathematics is defining the system. I would think it would be relatively straightforward to formulate a layman's "proof" in E2 using a few postulates from high school geometry. You could probably extend the proof to higher dimensions, too, since they're all orthogonal.