Einstein asked himself what it would mean to have equivalence between accelerated observers. The question led him to a solitary route of about ten years’ hard work that took him from Special Relativity to the astonishing theory, prosaically called General Relativity, in which the notions of space, time and gravity got intimately linked.

The phenomenon of gravitation is a direct manifestation of the curvature of space-time. Energy and momentum do curve space-time, and the curvature of space-time affects the motion of matter and radiation. Einstein ‘liberated’ space-time from its role as the rigid arena in which physics took place. He turned space-time from a passive spectator into an active player, itself taking part in dynamical physical processes. General relativity explained numerous observed facts that Newton's theory could not, it predicted mindboggling phenomena like black holes and most importantly it opened the door to a completely new, dynamical perspective on our universe.

We all have the experience of being in an elevator that is accelerated upward or downward. You know when you are being accelerated upward, because you feel heavier. And if you accelerate downward, you feel lighter. Imagine now that somebody cuts the cables of the elevator so that it would be in free fall. In that elevator everything would be ‘weightless’: if you would drop something out of your hand, it would not fall down on the floor, but just move along with you. From these observations we may conclude that a theory of relativity for accelerated observers should include aspects of gravity. The freely falling observer would play a unique role because she is weightless; for her there is no gravity. Einstein discovered the key idea in 1907 with the principle of equivalence, in which gravitational acceleration was postulated to be indistinguishable from acceleration caused by mechanical forces. Gravitational mass, as in Newton's gravitational equation, was therefore identified with inertial mass (appearing in Newton's celebrated formula F-ma). Back then this was a phenomenological assertion, but not a logical necessity or a fundamental principle.

There is one other observation to be made in connection with accelerated observers: imagine a light ray.