Well, I'm flummoxed...
If the film holder is inserted correctly and it is properly locked into the gate, no light should be able to get around the image aperture. (of the film holder.)
Vignetting from a misaligned lamp is eliminated. The lamp holder is fixed.
If you have blocked the light from coming out of the enlarger except for what comes out of the lens (using cardboard) then there should be no stray light coming out of the enlarger, itself. All the light hitting the paper must be coming through the lens and, by extension, through the film.
If you had internal reflections from the lens, hitting the paper where it shouldn't, it isn't very likely that they would form rectangular bars. Most likely, they would be oblong blobs. Neither would they be likely to be evenly spaced on all four edges of the image proper.
If the bars are blurry along the edges, that tends to rule out gaps between the easel blades but it doesn't completely eliminate the possibility.
If the blades were down, tight against the paper, they would form sharp-edged shapes. But not all easels hold their outboard edges perfectly flat to the paper, either. Even if they are a millimeter above the paper surface, they'd form "mostly sharp" lines but they might be a hair bit blurry. Regardless, blurry-edged lines score a point against this theory.
So, what's left? Stray light from the enlarger is eliminated. A misaligned film holder is eliminated. Enlarger blades are partially eliminated. There's not a whole lot left that I can think of.
Does the size and shape of the rectangular aperture in the negative carrier exactly match the size and shape of the negative? It can be a millimeter or so larger but that's it. Otherwise rectangular bars of light might escape around the edges of the image and they might, very well, have blurry edges.
I assume that this is what Terry is getting at. No?
A scan or a digi-photo of an actual print that exhibits this problem would be very helpful at this stage of the game.