In big science fields, such as particle physics, these problems have been around for several decades. What is needed is a careful statement of the contribution of the candidate, and testimony from co-workers about that contribution (this is not so much a reference evaluation as a testimonial as to contribution to joint work). To lead a sub-team is one such role, another is to take charge of a major part of the endeavor, another is to be the theorist consultant to the whole project, and I am sure there are other such roles. Some of this comes up in statistical work. In these roles, independence and contribution are readily discerned by those inside the work.

Analogies with army ranks abound.

Usually, the problem in promotion dossiers is when these issues are finessed or little evidence is offered, or the number of publications is low. In the latter case, in fields where people count publications I wonder why we don't just divide by N-1 or some such. There are good reasons for doing so, or not doing so. I note that the multiple authored papers in many fields do not seem to involve more work than singly-authored work, or surely not much more work. As for little evidence, I have seen dossiers turned around by the testimonials I referred to above. Typically they say, we could never have done our project without Joe Schmoe, he took a major role in the analysis (or whatever), and he has contributed more than his share to the project's formulation. When you have 30 members of a team, you need more information. When you have 3 or 4, often the testimonials of the collaborators work fine.

There may be issues about who raised the grants, but even there grants might well be written by teams.

In general, teams can be 2-4, 5-12, and 13-100. Different issues arise for each.

As for interdisciplinary work, I observe that we need to see some hard cases where evaluation is proving difficult. Usually the problem is that the preparers of the dossier make excuses, rather than find the people who recognize the value of the work. They make excuses for why the work is not published in prominent journals, etc. If the work is valuable and a contribution, there are people of significance who will recognize it, tell us why it is important, etc. Sure, some will say it is not chemistry, so they can't judge it. But there are others whose enthusiasm and detailed analysis tell us all we need to know. It will be useful, again, to see some hard cases.