My maths teacher explained this to be me by way of analogy: a car driving around a perfectly circular track would be constantly changing its velocity (while the magnitude of the velocity is not changing, the direction is). Because acceleration is the rate of change of velocity, and the object is changing direction, it is said to be accelerating. This strikes me as an odd definition of acceleration, as surely it still equals $\mathrm{0 ms^{-2}}$, even if the object is changing direction. Nevermind, I thought, it's just a definition.
What is strange is that this seeming technicality actually tells us information. Because the object is accelerating, there must have been a resultant force acting on it (since $F=ma$). This is completely baffling to me — yes, the object must have been 'accelerating', but if the magnitude of that acceleration = $0\mathrm{ms^{-2}}$, it seems certain that $F$ = 0 as well. What am I missing?