Something is bugging me... Is there something wrong with thinking that if an object A exerts a constant force F on object B, through a distance d, then A transfers F*d joules of energy to B? Consider a situation in space... an astronaut mass m, and a rock m (no other objects anywhere). The astronaut exerts a force F on the rock (let's say along the x axis) for some time t. Let's say the mass acquires a velocity v, so the astronaut acquires velocity -v. If I were asked, how much chemical energy does the astronaut lose... The answer seems to me to be: mv^2 (sum of the final kinetic energies of both the astronaut and the mass). There's no other source for the kinetic energy to come from. But if I were to use this idea of energy transfer (force through distance = energy transferred) then, I'd get the result that the astronaut transferred 1/2(mv^2) to the rock (no problem), and the rock transferred 1/2(mv^2) to the astronaut (big problem here!)... Thanks for your help.