I think if I did your idea of having the camera inverted, the same spots likely wouldn't occur in the sky as it would now be in the lower half of the frame. If there is something with sufficient tonal range (which is nearly anything that isn't a single, deep shade like a blue sky), the top half will show no evidence. To better give you guys an idea of what I'm talking about, take this image for example. I didn't digitally remove anything other than the usual dust marks and shows no evidence of the smudges because it appears to be blocked by the different shades. Here's another one, which actually has somewhat even-toned skies and again no smudges because the sky is too bright (I'm assuming). Back to the camera inversion technique, I may end up finding completely different smudge marks if the sky was somehow at the "bottom" of a frame because that is usually covered up by something else. But that just goes to show that regardless of positioning, the smudges will occur on the same parts of the negative.
Oh and as for the negs,
It is definitely on the negatives.So I decided to closely inspect the negatives by hand and of course, I saw the smudges very clearly in the same spots as I see them as a positive when scanned. So that rules out the scanner.