Upon reflection I think I may be tilting a little bit towards proposing a system to get unjustifiably whining postdocs to just shut up and realize what time it is. To realize that “work” does not always equal “contribution” and that sometimes one gets screwed. Because after all the above proposal is simply what the “good” PI should be doing anyway in guiding the author list process. But extend me a little optimism. Even the “bad” PI who is willing to shuffle authors based on liking, who needs to defend, who is on the job market, who is currently crying in the office and other such factors maybe convinced (or at least shamed) by peer review making it clear that the intended author order is not scientifically justified. It needn’t always be a nuclear confrontation with the PI.

The benefit of the current system is that the corresponding author has effectively dictatorial power over authorship. The checks on this power are (1) the need to maintain some level of morale in the laboratory and among collaborators, (2) the reputation of the corresponding author, and (3) the threat of the “nuclear option”–an author pursuing institutional (or other external) oversight of the situation.

While this system definitely leads to disgruntled authors, it almost never leads to the nuclear option. Decentralizing the authority to determine authorship would, in my opinion, substantially increase the frequency of disgruntled authors going nuclear.

I don’t know how much you’ve been around painful authorship fights (?). It isn’t pretty. I thank my lucky stars that this is not a typical feature of my area of research. I am quite familiar with situations in which painful fights occur with just about every freakin’ paper.

While I agree with you that this would be better than the cockamamie idiocy that Nature came up with, it is still much, much worse than the current system, which is that the corresponding author has the final say.

Still, while yours is certainly a good idea, I’m still with writedit on this one: Adults should, in a perfect world, be able to figure this out themselves.

Perhaps they should hug it out.

For example, it may be the case that the first publication of a knockout or transgenic animal (even in this day and age!) is so important that it carries the day. Alternately, the animal may have been published, or is a me-too and therefore exploration of a novel phenotype is most important. You could also envision scenarios where the follow up (say in vitro expression, rescue, recapitulation, etc) experiments really make the paper important (“demonstrates mechanism”).

It wouldn’t solve everything, of course, but it is better than that absurd proposal in NatureJobs.