If it is a barbell, and if it formed by the joining of two bodies, I might expect the two to be more loosely connected than if it formed by the erosion of one body. In that case, around the time of greatest activity near perihelion I might expect differing jet reaction forces across the nucleus to lead to shifts in one body perpendicular to the line joining the two. That should lead to cracks (like we appear to have seen already) opening or closing near the neck. Depending on the details of the gravitational field and connection of the two joined pieces, we might even see vibrational modes - periodic opening and closing of cracks.

I can't wait for perihelion...

Just one more, though I can't remember if Emily didn't linked it from her blog. But if not :
Comet 26 09 14 NavCam Mosaic vc par 2di7 & titanio44, sur Flickr

I haven't been posting these to my blog because Elisabetta and Marco's colorization work is so artfully done that almost no one realizes that it is not actual color, yet we don't know what the actual color of CG looks like. I am holding out hope that at some point we'll see an actual color product from OSIRIS. Until we see that, I hesitate to post these artistic impressions because I want to make sure I and my readers understand what the true color of the comet looks like (inasmuch as we ever see true color from spacecraft cameras).

But I do think it's awesome work!

The contrast in brightness between the brighter neck region and the lobes is real, though, isn't it?

The rugged landscape of 67P is reminiscent of how the early space artists depicted the surface of the Moon.