[BasilG]

This function will never produce a negative intensity.  Such a thing is impossible, you cannot have negative light. 

 

[...]

 

It appears black.  It can never appear less than black.

 

You seem to imply that I suggested that "negative light" and "negative intensities" exist - which of course I did not. What happens - according to my understanding - once the (real part of the) optical transfer function crosses zero is that the contrast inverts - black becomes white, and white becomes black (and nothing becomes "blacker than black"). As explained in the Wiki entry I linked earlier, the MTF is the absolute of the real part of the OTF, so the former never drops below zero even if the latter does.

 

Thinking about the issue at hand a bit more, I do realize that there may be some aspects I have not considered enough, in particular that transmission illumination may in fact have some effects I have not considered. Nevertheless, given that the Wikipedia entry agrees 100% with my intuition, I'm still not convinced that my interpretation was wrong, and in particular, I'm absolutely not convinced by your explanation that involves simple interference. But it might be better to give this a rest, as there does not seem to be any progress in this discussion...