The voltage is a difference between the electric potentials of two conductors. Hence, to change voltage, only one of the potentials has to change (although both can).
In AC power only one of the wires (live/phase) changes it's potential, while the potential of the other one (neutral) remains constant.
In the picture above, the orange horizontal line represents the potential of the neutral line (marked as zero for convenience), while the blue curve shows the constantly changing (in relation to zero) potential of the phase line.
Since in properly constructed power network the neutral wire is maintained at a potential level close to ground potential, there is nearly no voltage between the neutral and the ground. Hence, touching neutral will not cause current to flow through human body into ground.
Live line, however has a potential that rapidly changes from highly positive relative to ground potential, to highly negative. This difference in potentials (voltage) of the conductor you're touching with your hand and the one you're standing on causes a current flowing through you and at typical outlet voltages can be very deadly.
Analog Ground is a reference to a constant potential wire, that all other signals (voltages) relate to. Is is what you name your "0" when measuring other signals. For example, in most battery-powered devices it is the wire connected directly to the negative terminal of the battery. Naming "ground" or "0" is a matter of convenience, however in outlet-powered applications the designations are often separate, since "ground" is electically connected to actual ground.
See also http://en.wikipedia.org/wiki/Ground_and_neutral
As the comments already said, this is most likely due to a break in the neutral:
A floating cable next to the phase wire collects some voltage due to capacitive/inductive coupling. As your measurement device also plays a role, you usually will not measure 77V, 151V and 228V (=151V+77V) between the three wires. However, 77V is a typical voltage measured in this case.
Finally, neutral is broken, because you measure the expected voltage between earth and phase, and neutral gives strange results. If you connect any load to the circuit, you will see that neutral will be at 228V to earth an 0V to phase.
Once I had the same problem. Finally, there was a single luster terminal where all neutrals of a room (five) were connected to. One of them was pushed in too far, the part inside the terminal was covered by insulation. The screw bit into the insulation, and it was working for 20 years. From one day to the other, the connection became loose. This could have ended up in a disaster, but there was not even a sign of heat at the terminal.
Best Answer
simulate this circuit – Schematic created using CircuitLab
Figure 1. The likely cause of the low voltage reason on the blue phase.
It sounds as though your supply is from a star (European term) or wye (American term) connected transformer. Normally the star point, marked 'N' for neutral, is connected to earth to "neutralise" it.
In your case it seems as though either
The second arrangement can be used in the right circumstances to give single-fault tolerance of a ground fault. The circuit will continue to operate (as you have noticed) until a second ground fault occurs. When this is done it would be normal to add ground fault detection to warn of the event. Three bulbs in place of the voltmeters would suffice: when a ground fault occurs that lamp would go out.
The problem with this arrangement is that devices rated for 230 V to ground now have 400 V to ground. This may stress the insulation, causing early failure, and also provides a higher shock hazard to maintenance personnel.
simulate this circuit
Figure 2. Earthing of star / wye point on transformer. This is standard European practice.
I'd recommend understanding the design intent, examining the system and repair, if required.