Consider the rotating magnetic field as equivalent to physical rotation of two stator poles N1 and S1.
Consider an instant when two poles are at such a position where stator
magnetic axis is vertical, along A-B as shown in the Fig. 1(a).
At this instant, rotor is stationary and unlike poles will try to
attract each other. Due to this rotor will be subjected to an
instantaneous torque in anticlockwise direction as shown in the Fig.
Due to inertia, before rotor hardly rotates in the direction of
anticlockwise torque, to which it is subjected, the stator poles change
their positions. Consider an instant half a period latter where stator
poles are exactly reversed but due to inertia rotor is unable to rotate
from its initial position. This is shown in the Fig. 1(b).
At this instant, due to the unlike poles trying to attract each other,
the rotor will be subjected to a torque in clockwise direction. This
will tend to rotate rotor in the direction of rotating magnetic field.
But before this happen, stator poles again change their position
reversing the direction of the torque exerted on the rotor.
The question is obvious that will happen if by chance the rotor
position is in such a way that the unlike rotor and stator poles are
facing each other ? But owing to the large inertia of the rotor, the
rotor fails to rotate along with the stator poles. Hence again the
difference of position of magnetic axes gets created and rotor gets
subjected to quickly reversing torque. This is because the speed with
which rotating magnetic field is rotating is so high that it is unable
to rotate the rotor from its initial position, due to the inertia of the
rotor. So under any case, whatever may be the starting position of the
rotor, synchronous motor is not self starting.