Puzzling authorship practices

A few weeks ago, in a half-serious entry, I expressed my amusement with the practice of stating explicitly, some place in a scientific manuscript, that a subset of its authors “contributed equally” to the work described therein. I received a flood of comments, as well as e-mails (as many as 5 altogether… no, I am not making that up… OK, maybe it was more like 3, but who’s counting), patiently explaining to me that the goal of the above operation is to grant “first author” status to more than one person.

Apparently, in some fields of science (e.g., molecular biology), it is customary for a researcher to receive significant credit for a co-authored, published piece of work, only if the name of the person is featured either at the beginning [0] or at the end [1] of the author list. Conversely, the contribution of an individual listed as second, third,…, (n-1)th author, will be normally deemed “essentially worthless“, as eloquently put by Mad Hatter. This assessment has been confirmed to me by a number of colleagues in bioscience. Given such a draconian state of affairs, in order to ensure that more than one person receive “first author credit” (i.e., some credit), the PI has ostensibly no choice but to make an explicit statement of equal contributions in the manuscript, and prominently so.
The obvious question is, why list all of those additional authors, then, if they will receive no credit anyway ?

In physics this does not seem to be happening much (yet). I have always thought it understood that the relative contributions to a research work vary with the level of seniority of the authors. A graduate student, a postdoctoral associate and an assistant professor can not be, and are usually not expected to have played comparable roles in carrying to fruition a research project described in a co-authored article. And, while greater recognition will surely go to the first (and maybe last) author, some recognition will also go to the other authors.
I have always been fairly dismissive of the importance of this author order thing; to put it bluntly, I think it is just a pile of malarkey, an exercise in futility. For one thing, very seldom will authors be in competition with one another for the same job, or grant, or promotion, or salary increment, or prize. If several people at the same stage of their careers (e.g., two graduate students) are part of the same group, and therefore could conceivably compete for the same job, different projects are usually assigned to each one, and first author status will go to the person “in charge” of that project, when the paper is written. There is usually no problem in doing that, even though the group functions as such, i.e. everyone contributes in part to all the projects, all the time [2].
And even in the case of work co-authored by collaborators at different institutions and/or with similar seniority, insisting with a particular order in the author list is probably a waste of time. My experience has consistently been that the community has a way of deciding which one or two authors in the collaboration are the main driving force behind the project; to them will go most of the credit, eventually, pretty much regardless of the order in the author list (my experience is also that the community gets it right most of the time).
In general, I think that the credit tends to go to the person who does the most convincing job at giving talks on the subject.

Be that as it may, my question is: if essentially no credit will be received unless one’s name appears at the beginning or at the end of the author list, what is then the point of including all other names at all ? Who benefits from a proliferation of names ? Would it not be easier and less confusing to list only first and last author ? Or, would it not make more sense to revert to a system in which each and every author in the list can be safely assumed to have played an important role in the project, i.e., the work would not have been brought to completion timely and accurately as it was, without the different contributions of each and every one of them ? After all, if stating that “authors contributed equally” does the trick, it seems a baby step to forget about order altogether (just use alphabetic order and, for [favorite swear word] sake, leave out those who did nothing).
I mean, seriously, is this just a case of “administrators gone wild” ?
Awaiting responses (no more than three, eh ? I don’t want to be overwhelmed…). Oh, but now it’s time for a nice home made meal [3].

Notes
[0] First author, supposedly main contributor. This is typically the person who did most of the actual research work, and possibly may have been in charge of writing the article as well.
[1] Senior investigator, intellectual leader of the group, holder of the grant off of which other investigators are paid, presumed originator of the idea for the project.
[2] Of course, in the case of two graduate students or postdoctoral associates, the assessment of their relative abilities given by the senior investigator will typically carry far more weight than their respective numbers of first-authored papers, in determining their success later on. This is another reason why making a big issue of author order is pointless.
[3] Contributors: Mrs. Okham, Okham, Okham’s neighbor, Okham’s buddy, spouse of Okham’s buddy and Okham’s mother-in-law (on leave from her regular institution… short leave… very short).

Tags: , ,

8 Responses to “Puzzling authorship practices”

  1. Anonymous Says:

    Starting the Comment flood….

    The opposite extreme are of course the mathematicians, who have typically fewer authors and order them alphabetically. If I have to choose one of the extremes, I’d take the bio approach, I think. (Of course, I do not have to choose and of course, the condensed matter physicists do it – as always – much better than anyone else and almost perfectly.)

    I would really not want to rely on the community’s guessing power, because all those bias and “pedigree” issues become more important in the absence of other information.

    If a group is organized as you describe, with each student responsible for one project and helping on others, first authorships are important, because they show that students were able to finish “their” project. (On the other hand, people who are always first author may not have been very good team members. Or may have been in a small group, so it doesn’t really tell.) And people do attach importance to first authorships. So I can understand that a PI may want to split that honor: The two authors may really have collaborated so strongly that one can no longer know who did more, or the paper may be the result of two merged projects.

    I am still waiting for papers listing “ispell” as an author, it did more that I am prepared to believe of some people on some papers. Particle physics comes to mind. But if ispell is co-first-author, I’d get suspicious.

    • Massimo (formerly known as Okham) Says:

      Re: Starting the Comment flood….

      All I am saying is that the following sample author list:

      “First author” Jane
      “Second First author” Joe
      “Oh-what-the-hell-first-author-too” Dick
      “Slacker” Frank
      “Who is he anyway” Oliver
      “Eh, we gotta put him too” Bill
      AND PI

      could be replaced by: Dick, Jane, Joe, and PI.
      Frank, Oliver and Bill will get no credit anyway, so no harm done by not listing them. As for Dick, Jane and Joe, the PI will eventually be the one to decide who should receive a greater share of the credit, IF the situation ever arises to where they have to be compared to one another.

      And people do attach importance to first authorship.

      Honestly, in physics I have never noticed a particular emphasis on this. I mean, an application from someone who has never been first author will probably raise eyebrows, I agree, but do you really sense that one applicant will be preferred over another simply, or mostly due to a greater number of first-authored papers ? I doubt it, frankly, and certainly my experience would not back that contention.
      I may be wrong, but it seems to me that there are so many other aspects to take into account, at that point… journals where papers are published, invited talks… some are subjective; for example, singly-authored papers certainly catch my attention, when evaluating tenure-track applicants.

      In my opinion the obsession with first authorship (which in some fields clearly exists) is a consequence of the widespread habit of including in the author list names of persons whose contribution to the work is flimsy, to put it mildly.

  2. Anonymous Says:

    A physical chemist friend who spent some time with me in lab was struck by how laborious biological research is. I think the sheer amount of physical labor that goes into a publication explains many of the differences in authorship assignation. It is very rare for a single scientist-PI pair to be able to work quickly enough to publish before getting scooped. Moreover, biological assays tend to be complicated, and as biologists become increasingly specialized, the need for other scientists’ help grows. What this boils down to is that biology is very much a collaborative science, and the way to encourage collaboration is to give collaborators some kind of credit. Not to mention the fact that it is the right thing to do.

    “why list all of those additional authors, then, if they will receive no credit anyway?”

    I was speaking in the context of faculty promotions. Middle-authorships are definitely worth less, but they do help junior scientists (grad students and some postdocs) who do not have many (or any) first-author papers. In the absence of being able to demonstrate that one can carry a project to fruition, demonstrating that one can at least contribute publishable data is better than nothing.

    “I have always thought it understood that the relative contributions to a research work vary with the level of seniority of the authors.”

    Absolutely not true in biology. As a grad student, I was first-author on several papers with middle-author postdocs because I did the bulk of the experimental design and work. As faculty, I am about to be middle-author on two papers by postdocs because, while I contributed intellectually and experimentally to the project, the postdocs did most of the work and are therefore the first-authors. Credit is given based on relative contribution, which I think is fairer than the “pecking order” model.

    “very seldom will authors be in competition with one another for the same job, or grant, or promotion, or salary increment, or prize”

    Also not true in my experience. Multiple PIs in my department have competed for the same grant. Multiple postdocs in my lab have competed for the same grant and/or job. Regardless, it’s not a matter of being “better” than one’s labmate so much as building one’s CV so as to be generally competitive among the pool of applicants in one’s field.

    “different projects are usually assigned to each one, and first author status will go to the person “in charge” of that project”

    Different projects are assigned, but projects can converge unexpectedly when student 1, who studies how X happens, and student 2, who studies how Y happens, discover that X and Y occur through the same biological pathway. It’s not uncommon at all. Another common occurrence is when the portion of a project which was initially thought to be the primary focus of the project doesn’t pan out, whereas a peripheral sub-project generates very interesting data. This changes the focus of the paper and, by extension, the person “in charge”. This is why authorship is almost never pre-determined, but rather decided once an outline of a manuscript is in place.

    “the community has a way of deciding which one or two authors in the collaboration are the main driving force behind the project”

    I don’t see how this could be possible unless one were in a very small subfield in which everyone knows everyone else. In my department alone, there are >40 tenured faculty (not including Asst. Prof. or non-tt faculty). And my department is one of 10 similarly-sized departments at my institution that all fall under biosciences (excluding biostatistics, bioinformatics, and other more divergent departments). I haven’t met all the faculty in my department, let alone be able to tell who has contributed what to a paper from someone else’s lab.

    “insisting with a particular order in the author list is probably a waste of time”

    In the 12 years I’ve been in academic research, I have only witnessed one instance of a fight over authorship which delayed publication. Authors typically agree amongst themselves in very little time.

    • Massimo (formerly known as Okham) Says:

      I am being flooded again… oh, well, that’s life…

      [Patient and detailed explanation omitted]

      I see. Your field does seem to work quite differently than mine then, and of course, it is true that the sheer size of the bio sector, much greater than condensed matter physics (or all physics, for that matter) may well account for most of the difference, and makes a direct comparison improbable.

      A few comments:

      What this boils down to is that biology is very much a collaborative science, and the way to encourage collaboration is to give collaborators some kind of credit. Not to mention the fact that it is the right thing to do.

      I have to disagree with this. I see this phenomenon in experimental high energy physics, very much a “collaborative effort” too, with papers co-authored by hundreds of people. Result ? Publications have become essentially irrelevant to one’s career.
      That a contributor deserves some credit is clear. That authorship is necessarily the only or the best way to give that credit, is far from clear to me. High energy physics is perhaps an extreme case, but if a research enterprise grows to a size and complexity that make it similar to industry, then maybe an industrial type model is more appropriate to determine career advancements. And industry certainly does not base its assessment on publications.

      Credit is given based on relative contribution, which I think is fairer than the “pecking order” model.

      I must not have made myself clear, that is exactly what I meant. And that is unrelated to author order. In other words, upon seeing a paper co-authored by a student, a postdoc and a senior person, I think most people in my field will sort of assume a particular division of labor, almost regardless of the order in which authors are listed.

      • Anonymous Says:

        Heh…that’s what you get for writing a blatantly provocative, controversial, and inflammatory post! 🙂

        I vote for cash bonuses instead of middle-authorship for contributors. Think academia will go for it?

      • Massimo (formerly known as Okham) Says:

        Cash bonuses

        Heh…that’s what you get for writing a blatantly provocative, controversial, and inflammatory post! 🙂

        Yeah, plus half of the comments are my own. It’s a bit like artificially raising my blogger h-index through self-citation… I am playing the political blog game, so to speak.

        I vote for cash bonuses instead of middle-authorship for contributors. Think academia will go for it?

        I would not be surprised in the least if that became actual practice, and in many respects I think it’s already happening, except that it’s called something else. Obviously, we are not talking new or extra money, but merely a different way of distributing the same amount.

        To go back to experimental HEP, which may be breaking new ground in this regard, given the size of the collaborations, I see quite a few papers these days with author lines that say, for example “So-and-so on behalf of the Some-cool-acronym collaboration” (e.g., this paper), without mentioning all the (several hundred) authors.
        That means that one’s role in the collaboration is assessed differently than through authorship, mostly through a (largely internal) mechanism of promotions and increased responsibility. I suspect university departments will start rewarding faculty based on whether they move up the managerial ladder of the collaboration.
        Think it’s crazy ? I honestly don’t think it is nearly as far-fetched as it may seem.

      • Anonymous Says:

        Re: Cash bonuses

        Is there no way for livejournal to remember who I am so I don’t have to enter my identity URL every single time?

        Anyway, I don’t think it’s a crazy idea, but a majority of institutions would have to adopt the system and agree on a common non-authorship-based reward in order for it to work. Otherwise, the reward would be unhelpful for grad students and postdocs who tend to move from one institution to another for their next positions…like being paid in a currency that is only legal tender at one’s institution.

      • Massimo (formerly known as Okham) Says:

        Re: Cash bonuses

        Is there no way for livejournal to remember who I am so I don’t have to enter my identity URL every single time?

        Of course not. Are you kidding me ? They are already letting me pay for keeping a journal with them, something which anywhere else I would have to do for free, and now I should even ask them to provide some functionality ? Dream on….
        This is why, I would contend, the number of comments that I receive should be corrected by some multiplicative factor greater than one, given the difficulty of leaving a comment. Unreasonable ? I don’t think so…

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: