It's not "idiosyncratic" to roll dice to make these kinds of decisions, either, but you wouldn't call dice rolling "systematically rational," would you?
If the roll of the dice selects for a coherent system which is universally implemented, then that is about as impartial as a decision can be. Moreover, the system selected is systemically rational in that it will universally generate, more or less, the same result (contingencies of viewing not withstanding). It is rational in the sense that it subjected to the same rules which generates consistent results. You know exactly what you're going to get.
It seems to me, logically, that meddling with existing art via a system can only work along two axes: either it adds random noise, or it moves the idiosyncrasies earlier in the process, IE: I subjectively choose these algorithmic inputs upfront and then a bunch of stuff happens, but ultimately the output is still the inevitable result of those inputs. Or some combination of the two.
If the idiosyncratic choice is standard and predictable, then it can be filtered like noise. It doesn't matter if the noise is random or a procedural "enhancement" (like stereo-widening, upping the bass on an EQ, changing the color settings on a TV). These we can filter out or adjust to, like hiss on an old cassette tape. On the other hand, cutting out half the film and replacing it with a dog jumping on a trampoline (or whatever your example was), isn't just filtering, because cannot tune out the dog to see what was once in the image.
How many logs can we change before we no longer have the ship of Theseus? Well, interpolation doesn't remove a single log from the ship. It doesn't paint over the ship in the drab zombie colors of the colorizers. On the contrary, it just lets us see what was intended to be seen, sailing more smoothly. That is hardly the vandalism or violation of a re-edit.
Preemptively, if by "systematically rational" you just mean "done with an internally consistent system," then I agree that the process is systematically rational, but by that definition it really doesn't do anything to defend the practice.
My argument is that it does not involve the obscuring/mangling which is feared, because it functions as a filter (like a noise we can tune out or an enhancement that allows us to tune in better).
It is impartial, because no individual is making odd choices which change a single frame of the original movie (it is just statistically averaging between frames). It is objective because it is an impartial standard (you know what you're going to get every time). The image moves smoothly rather than jaggedly.
My argument as to why it is better is not to be found here. Rather my appeal is to the old standard of J.S. Mill. Let people who are unbiased judges with experience of both decide for themselves whether to select a black and white television or a color television, a VHS or a DVD, at CRT or LCD, 24 FPS or 48 FPS. History has shown that people, over time, express a preference for more information, clearer information, and smoother information.
...and yet we bear collective responsibility for its outputs, as the creators of its inputs, which is the thing germane to what we're discussing.
Right, it may be objective (in the mere sense that this is how it is everywhere), but you still can't derive an ought from an is. That this is objectively so does not mean it should be so. Even so, however, objectivity is a desirable thing to have for a standard, if it can be secured. And we can indeed, objectively/impartially bump up frame rates. And as for whether we should see J.S. Mill (above). If people with experience of both (without a lifetime of conditioning in either) prefer the new to the old, then the new is "better." And again, history has born out that people prefer to see more information, to see it more clearly, and to see it more smoothly. I expect that history offers us a plausible means of inference here, which means I have a warrant.
Yes, but you don't prosecute people under the old law retroactively, which is more or less what you're doing when you use new technology to alter works made without it.
Every new translation of Beowulf prosecutes Old English under the new. Every "new edition" of the Original Trilogy of Star Wars was a retroactive re-writing under a new standard of SFX and sound and video quality. Moreover, judging the past by modern standards is quite the pastime these days.
Are there limits? Yes. What Lucas was doing with Star Wars was getting close to that dog on a trampoline. Upping the frame rate, however, doesn't change the aspect ratio, shot length, composition, sound, lighting, dialogue, does not change a single original frame. But again, it is not for me to judge, but for the great collective, with experience of both, and obligation to neither, to make their own decision as to whether we should move on from 24 flickering frames. And I think we both which side will win the contest in the long run.
you seem to be arguing, either explicitly or implicitly (hard to say), that the interpolation of these frames leads to a fuller or more complete realization of the creator's vision, which is a totally different argument than just saying you prefer it or you think it makes some things better.
I think it makes somethings better because it realizes the creator's vision at the speed of life, the very thing the filmmaker (in most cases) is trying to capture in the verisimilitude of film. The maker wanted to fool us. And for those of us who don't have 24 FPS "on the brain" more frames per second is more lifelike. Again, I agree that there are use cases (e.g., horror, war) where a more stuttered frame rate produces a desirable aesthetic effect which amplifies the theme of the film. In the main, however, the default expectation is for life on the screen to be lifelike, to achieve the magic trick convincing us that those images really are moving.
I'm a little curious as to why you're so rah-rah on frame rate increases but so down on colorization. An awful lot of the arguments you make (wouldn't most of them have wanted to shoot in color, if they'd had the option? Etc), though certainly not all, would seem to apply to both.
I see the two cases as being quite distinct.
Upping the frame rate does NOT alter a single frame of the original. Colorization does. Interpolation just adds in averages to create the illusion of life's continuity. Everything that was in the original is still there. And you can still watch it at 16 or 24 FPS if you wish. Colorization, on the other hand, has destroyed some films - the preserved copy now a copy that does NOT have the original frames.
Moreover, colorization does NOT show us what the original virgin film stock would have "seen" as light entered it from the lens of the camera. The colors are not the colors that would have entered it from the camera lens, thus the image is not the image that the camera would have captured if the camera had been recording with color film stock. On the other hand, upping the frame rate with an algorithm simply show us what the film would have been exposed to if the film were being cranked at a faster rate. That's a big difference.
Finally, it damages information on the original print, muddying the contrast values.
If you find a flaw here, then by all means let me have it. I will either have to commit to endorsing colorization (which is about as likely as Crumb coming around to 60 FPS) or I will have to admit that upping the frame rate is a violation and should not be thought of as an innocent intervention with an old film (but what could be more innocent than watching an old film which can interpolate into a higher frame rate, if you desire--the original is not altered and you can switch back and forth).