The vertical element is connected to the centre conductor (RF live). The
radials are connected to 0V on the transceiver.

No. They're connected to the shield/braid of the feedline. There's
no assurance that this point will be at "0V" with respect to anything
in particular except itself, and in particular it usually won't be at
0 volts with respect to the transceiver's chassis / output jack
(except perhaps momentarily, twice per RF cycle).

If only the vertical is connected, the antenna still radiates although not
as well.

.... because the outside of the feedline will tend to act as a
poorly-tuned radial/counterpoise.

If only the
radials are connected, the antenna does not radiate because the radials are
connected to 0V and not a varying voltage.

With both vertical and radials connected, the vertical element radiates the
wave. The wave is reflected by the radials as boundary condition of
Maxwell's equations. The reflection induces a current in the radials. This
current has a standing wave on it.

Do you think the above is correct?

Not really, no. It's a mistake to think that the radials "are
connected to 0V and not a varying voltage". You're falling into the
trap of thinking that "ground" is some sort of magical "zero volt"
reference which is the same everywhere. That isn't true even at DC,
and it's certainly not true at RF!

All parts of the antenna form the
impedance. Without radials, the impedance is poor and the vertical element
does not radiate well.

The vertical element radiates very well indeed... it'll radiate all of
the power which is fed into it, except for a small amount of loss.

The problem isn't that it doesn't radiate. The problem is that it's
difficult to feed power into it, much of the time.

--
Dave Platt AE6EO
Hosting the Jade Warrior home page: http://www.radagast.org/jade-warrior
I do _not_ wish to receive unsolicited commercial email, and I will
boycott any company which has the gall to send me such ads!