Pros and Cons of Platter Mass

The point made about Sacrificial Surfaces are important and within the Bearing Housing there are designs in place to address this.
The Spindle is usually designed to be the part that has the Hardest Surface and in a good design will usually be produced with a Case Hardened Surface.  
Bush Materials are selected that are produced to have lessened surface hardness than the Spindle.
A vintage TT that has not had a life of function with a Spindle / Bush Interface where a Hydrodynamic condition is present is at risk of having produced a wear to the Bush Material. When the Bush is produced from a Metal, the risk also becomes that abrasives may be in the Bearing Housing and having an unwanted impact on the important parts and interface surfaces. The Wear can impact on other important areas of function, especially increasing the likelihood that an eccentric rotation will develop.

A Spindle does not usually interface directly with a Thrust Pad, there is a intermediary part used that is Sacrificial, in many cases this part is a Metal Ball, that can be found to be a interference fit into the base of the Spindle or in other cases a fully exposed ball that is a standalone part.
These Balls have been witnessed scored on the surface as a result of being in an environment that is abrasive. 

The modern designs for the sacrificial parts used at the interfaces are more commonly moving away from the use of Metal and are leaning towards the use of Thermoplastics as a Bush and Thrust Pad, and the Metal Ball is more commonly seen removed from a design and exchanged for a different material that has formed the ball.

In relation to vintage TT's with long periods of use behind them where a bearing housing is using Metal Sacrificial Parts, it is difficult to condone that a application of a new lubrication only is the required treatment to produce a environment for the Spindle to function without impediment.

There are many grades of Thermoplastics with different properties for the resilience to wear.
POM and PEEK are two examples and the PEEK is common choice to be used when a Platter is going to be above a certain weight.
It does not have a property that would wear a correctly selected interface ball or will it cause damage to the Spindle.
There are more modern Thermoplastics that can be selected with properties that are further suited to the use within a bearing housing.     
 

@lewm , you need to add motor torque to that list.

I think as everything in life there has  be a balance and it depends on what type of drive is being used along with it's performance. The best direct drive motors do not require as much mass as the best belt drives.
Mass does not protect you from rumble. It might lower the frequency a little but you are also adding more thrust to the bearing, increasing noise and wear. 

How much mass is required? Whatever it takes to get the best performance and no more.

What confuses me the most is how a topic sat dormant and was revived 19 years later. I had to check to see if the OP was still with us and certainly glad he is. 
My TTW table’s plater separates into two parts for convenience, but the total weight is 70lbs. The copper weight and periphery ring add more and I’d love to know how much the the flywheel effect adds since the TTW periphery ring has brass weights.
I can say going from a less massive table to the TTW added dynamics and bandwidth to the sound. Also I feel the mass allows vibration from the stylus in the groove to sync, and the center weight and periphery ring holding the LP down essentially make the LP weigh 70lbs lowers distortion tremendously.

It is becoming more common for manufacturers to use a magnetic thrust mechanism. No ball and no thrust pad. Spindle wear is much less of a problem because of the greater surface area involved. Here, lubrication is quite effective if the bearing is designed properly. My subwoofers tell me there is most definitely less rumble with magnetic thrust bearings, at least the one I am using. 

Magnetic thrust bearings also allow more platter mass without increased wear. The same is true of real air bearing turntables. Frank Kuzma redesigned his big table with an air bearing and increased the mass of an already huge platter by about 1/3rd. I wonder if all that air rushing around increases noise. What if it blows on the tonearm?

The best direct drive motors do not require as much mass as the best belt drives.

No.
If you look at the Technics SP10 series - the most powerful SP10mk3 has a mucher higher platter mass (25lb ) relative to the weeker motored mk2.
Similarly with the Kenwood L07D - highish platter mass was oart of the design - the argument was high inertia, resistance to stylus drag.

Finally the most powerful direct drive motor built by Technics for lathes - the SP02 - dwarfs the motor in the SP10mk3 - but it was specified to be used in the Neumann lathe with an added 70lb flywheel.

If you build a platter out of weetbix, it will still sound like weetbix irrespective of drive topology.