These issues with quantifying authorship have likely been around for a long time, but they are particularly problematic in academia. When promotion and tenure depend on the number of publications on one’s CV, the pressure can cause individuals to act unethically. Largely, I agree with the overarching guidelines described in Osborne and Holland’s article.
To put it as concisely as possible, an author should be someone who directly contributes to the composition of content for a text. Beyond those individuals, people who contribute by providing funding, space, or other support should receive credit in the acknowledgements, and anyone who was not involved in the composition of the text but contributed ideas or previous research should be credited in the reference list and possibly the acknowledgements. The role of individuals who contribute in a limited way by analyzing the samples, statistically evaluating the data, typing the paper should be assessed on a case-by-case basis. If the contribution makes it into the final product, then the author should be listed as such, but if the contribution is the basis for the work of others and is not used directly, then I would argue that contribution should be cited in the references, whether the text has been published or not. All authors should take part in the writing/revising/submitting process. This means writing or co-writing portions of the text as well as reading the full text at multiple stages of the process. This does complicate the process, but it may be the only way to ensure that all authors maintain an appropriate amount of agency in their publications.
As I said above, I think the constant pressure to publish is a major contributor to these problems with authorship. It leads to issues with fairness that are not easily addressed. As Osborne and Holland discuss, other academics in various disciplines are using these imprecise guidelines for authorship already, which gives them unjustly obtained publications to fill up their CVs. In the competitive job market, it is difficult to fault a graduate student or junior faculty for either trying to get their name on a paper they didn’t directly “write” or putting an established professor’s name on their paper to get it better respect and recognition. To me, this seems like a systemic problem that will have to be addressed at a higher level, perhaps through the journals or through the universities themselves.
Within my own discipline, I am aware of students writing papers on their own and then collaborating with professors when they try to get the work published. In some cases, this is a legitimate collaboration where the professor builds upon the work of the students and they collaborate to craft the original paper into one that is worthy of publication. However, I have also heard of students writing papers and attempting to collaborate with professors only to find that the professor sees his or her role as purely providing feedback that the graduate student will then implement before publication. This is not truly co-authorship, but I would that it happens in departments all over.
I think the best way to address this in the future is to establish expectaitions early on in a project. As a graduate student, one may not have the ability to “lay down the law” and get a professor to agree to say explicitely what he or she will contribute and how the authorship will be communicated; however, that would be my goal as a professor in the future. I think all collaborative work should start with a discussion of expectations, goals, and projected outcomes. This would be a perfect time to establish how the co-authors want to complete the research and write the text.