Jeholornis Tail Feathers, How Do They Work?

There is a new paper out that suggests there's something interesting going on with the tail feathers of Jeholornis. In multiple specimens of this long-tailed avialan there appears to be a fan-shaped feather structure at the base of the tail.

Based on my reading of the paper, the authors interpret this fan as lying along the dorsal surface of the tail with the feathers pointing back (represented by the individual on the left in my drawing). Some have suggested that they probably represent specialized tail coverts in this interpretation, in which case the fan would be similar to a peacock train (although much smaller).

However, it appears that many artists ( to name a few; in fact I haven't even seen a serious life restoration depicting the authors' interpretation!) have independently come to the conclusion that it's more likely the fan extended laterally and were similar to the long feathers at the base of the tail in Archaeopteryx (represented by the individual on the right in my drawing).

That's an entirely reasonable interpretation, but I'm starting to think there may be something wrong with me because I don't see what necessarily makes it a more likely interpretation, other than Archaeopteryx setting a precedent for such a structure being present in basal paravians. I find it interesting that in all the specimens the fan is preserved closer to the dorsal surface of the tail and, assuming they were lateral structures, only the feathers on one side of the tail is preserved, which as far as I'm aware is rather atypical of retrix preservation in other aviremigians. (The Berlin Archaeopteryx, which is the one that preserves the longer tail feathers, has them only on one side as well, yes, but the leg feathers that were originally present would have blocked those on the other side from view had they preserved.) Maybe we're seeing the feathers on both sides bunched up? But then each side should only be depicted with 2-3 feathers rather than the full set of 4-6. In addition, the authors imply that the fan is bilaterally symmetrical, which would be consistent with it being oriented across the body... but I suppose there's nothing preventing a lateral tail feather configuration from achieving that kind of symmetry as well (which I've tried to convey in my drawing). Maybe there's some sort of taphonomic quirk at play with Jeholornis, which is certainly not out of the question. As usual, a decisive answer will probably have to come with more specimens, but in the meantime it's certainly good for generating discussion.

has speculated that all the Jeholornis caudal integument we have are really just parts of a full frond along the entire length of the tail, which is interesting to say the least. If he's right that the proximal fan in different specimens is really attached to different parts of the tail, I may be more inclined to say that the lateral structure interpretation is indeed more likely.

At the very least, I don't think anyone agrees with the vertical tail fin that is depicted in the illustration accompanying the press on this discovery.