I propose the following theorem (with corollary) regarding discussion of science fiction as a genre.
(ETA: Now revised for clarity! Plus a tentative title, and a suggested additional theorem.)
Theorem of Science Fiction Denial:
If an artist makes a point of asserting that a creative work is not science fiction, then (1) the odds that the work is science fiction increase to a near-certainty, while (2) the odds that the work's science fiction elements (e.g., world-building, science) are good decrease dramatically. Nb.: the work as a whole may still have artistic merit.
If an artist makes a point of asserting that a creative work is not science fiction because "science fiction is X", then the statement "science fiction is X" is almost certainly wrong.
Proposed Theorem of Genre Denial:
Take the above theorem (and corollary), replace "science fiction" with "genre," and remove the parenthetical.
* * *
prior comments, edits (kept because otherwise comments wouldn't make sense):
Comments, discussion, thoughts on snappy names? (While I believe it's traditional to name theorems after their originators, I rather doubt that I am the originator; moreover, "Nepveu Theorem" is awfully hard to properly spell or say.)
[Edit post-morning-dog-walk: I strongly suspect that this is generalizable to just "genre", but don't have the background to demonstrate it. For fantasy, the only example that's coming to mind is Phillip Pullman, and while the corollary definitely holds for him, I don't remember enough about the ways the His Dark Materials trilogy failed to say if the theorem does.]