Curious. Why not? Its very odd that MF and 35mm used the same film yet exhibited differences and yet digital doesn't!!! Odd too that cutting out a 35mm transparency from a medium format slide never had quite the same feel to it as an original 35mm slide. Odd too that the images from my Contax 645 produced substantially better scans tonally than those from my 35mm Nikons and yet apparently FF Canon produces an image identical to that from cropped formats. Its somewhat intriguing don't you think???
I'm open to any technical explanation as to how a imaging technology, be it film or digital, can possibly know to perform differently depending on its surface area. I am not doubting your, or Paul's, observations but rather the explanation for them. There is no way that an identical film can produce different tonal variations based solely on how much of it is used. Other factors are in play, almost certainly the predominant one being the manner in which the results are viewed.
I think that if you were to mount a lens on a FF body then a cropped body and examine the resulting file at 100% then they would (assuming the same sensor pixel size) appear prettywell identical.
That is true and has been demonstrated. Knowing that fact, how can you argue that there are differences due to how many pixels exist? It cannot be true!
In the real world different formats are used to capture similar image content by using different lenses and as a consequence the image 'quality' is different and this certainly includes tonality! This is true of FF and cropped sensors and the results ARE accordingly somewhat different.
Yes to the first and no to the second. Different systems will produce different images for a number of reasons but one of them is not that the imager itself is performing differently because of its surface area. Regardless, film and digital are completely different in this respect and I've already explained why that is.
...and by so doing the tonality within them will show a 'different gradation' than those from a smaller format using a shorter lens to produce the 'same' image (or should I say an image with the same compositional content?). We could continue to engage in argument for ever but for me this topic has now ENDED!
I can see that you believe this to be true but you cannot offer an explanation for why this should be true. I certainly wouldn't want to defend that position either!
It is important to know what aspects of larger formats contribute to better IQ and what aspects do not. A larger sensor does not automatically mean that the performance of individual "pixels" is improved whether film or digital. The fact is that the D80/D200 is capable of maximizing the performance of a 12 bit ADC (see Fig. 5). What this means is that no other dSLR can outperform it in tonality unless an ADC offering more than 12 bits is used. Cameras such as the 5D can match it and sustain that level of performance through higher ISOs but they cannot beat it. This is why the 1D3 is getting a 14 bit ADC and why they specifically say it is done to improve tonality.
Ergonomic do play a role in the picture taking process, if you have to juggle with your housing every time you wish to change aperture or speed, you are not concentrating on taking the picture. So yeah I feel that having independent easely accessible controls do make a hell of a difference.
I agree with this but it's more a function of the housing than the body. I have, in the past, argued that the best system and body to use is the one that is supported by your preferred housing and ports. This is less an issue now with so many bodies being supported by so many quality manufacturers. Nevertheless, the D80 has two control dials just like the 20D/30D so I'm not sure what bearing this has.
so care and wisdom in your lens selection will make a long term difference. 
