Authorship decisions

Deciding who should and should not be on the author line of a science publication is not as simple as it seems. As we know, citations matter, publications matter and there are all sorts of implications for authorship of a science publication.

A question about this arose on the Twitts:

Of course, we start from a very basic concept. Authorship of a scientific paper is deserved when someone has made a significant contribution to that paper. I can't distill it down any more than that. Nice and clean.

The trouble comes in when we consider the words significant and contribution.

This is where people disagree.

I also rely on another basic concept which is that someone should try to match, to a large extent, the practices within the subfields from which similar work is published. This can mean the journal itself, the scientific sub-domain or the institution type from which the paper is being submitted.

On to the specifics of this case.

First, do note that I understand that not everyone is in the position to wield ultimate authority when it comes to these matters. @forensictoxguy appears to be able to decide so we'll take it from that perspective. I will mention, however, that even if you are not the deciderer for your papers, you can certainly have an opinion and advocate this opinion with the person in charge of the decision making.

My first observation is that there is nothing wrong with single-author papers. They might be rare these days but they do occur. So don't be afraid to offer up a single-author paper now and again.

With that said, we now move on to the fact that the author line is a communication. Whether you are trying to convey a message about yourself as a scientist or not, your CV tells a story about you. And everything on there has potential implications for some audiences.

ethical, schmethical. Again, you don't throw someone on a paper "just because", you do it because they made a contribution. A contribution that you, as the primary/communicating/deciderering author, get to determine and evaluate. It is not impossible that these other people referred to in the Tweet made, or will make, a contribution. It could be via setting the environment (physical resources, administrative requirements, funding, etc), training the author or it could be through direct assistance with crafting the manuscript after all the work has been done. All of these are valid as domains for significant contribution.

This scenario of a private industry research lab appears, from the tweets, to be one where the colleagues and higher-ups are not intimately involved in pushing paper submissions. It appears to be a case where the author in question is deciding whether or not to even bother publishing papers. Therefore, the politics of ignoring more-senior folks (if they exist) is unfamiliar. I can't do much but read through the Tweet lines and assume this person is not risking annoying someone who is their boss. Obviously if someone in a boss-like status would be miffed, it is in your interest to find some way that they can make a contribution that is significant in your own understanding or to have a bloody discussion about it at the very least.

Leaving off the local politics, we can turn to the implications for your CV and the story of you as a scientist that it is going to tell.

If all you ever have are first-author publications it will look, to the modern eye, like you are non-collaborative, meaning not a team player. This is probably an impression you would like to avoid, yes, even within an industrial setting. But this is easy to minimize. I can't set any hard and fast rules but if you have some solo-author and some multiple-author pubs sprinkled throughout your timeline, I can't see this being a big deal. Particularly if your employment particulars do not demand a lot of pubs and, see above, the other people around you are not publishing. Eventually it would become clear that you are the one pushing publication so it isn't weird to see solo-author works.

Consider, however, that you are possibly losing the opportunity to burnish your credentials. The current academic science arc has an expectation for first-author papers as a trainee (grad student, postdoc) which is then supposed to transition to last-author pubs as a scientific supervisory person (aka professor or PI). Industry, I surmise, can have a similar path whereby you start out as some sort of lowly Scientist and then transition to a Manager where you are supervising a team.

In both of these scenarios, academic and industry, looking like you are a team-organizing, synthetic force is good. Adding more authors can be helpful in creating this impression. Looking like you are the driving intellectual participant on a sub-area of science is good. This concern looks like it votes for thinning your authorship lines- after all, someone else in your group might start to leech credit away from you if they appear consistently or in a position (read: last author, co-contributing author) that implies they are more of the unifying intellectual driver.

This is where you need to actually think about your situation.

I tell trainees who are worried about being hosed out of that one deserved first-author position or being forced to accept a co-contributing second author this
; You are in for the long haul. If you are publishing multiple papers in this area of science (and you should be) then for the most part you will have first-authors and in the end analysis it will be clear that you are the consistent and most important participant. It will be a simple matter for your CV to communicate that you are the ONE. So it may not be worth sweating the small stuff on each contentious author issue.

In a related vein, it costs you little to be generous, particularly with middle authors that have next to no impact on your credit for this work.

If you only plan to publish one paper, obviously this changes the calculation.

Do you ever plan to make a push for management? Whether of the academic PI or industry variety, I think it is useful to lay down a record of being the leader of the team. That can mean being communicating author or being last author. At some point, even in industry, an ambitious scientist may wish to start being last author even under the above-mentioned scenario.

This is what brand new PIs have to do. Find someone, anyone to be the first author on pubs so that they can be the last author. This is absolutely necessary for the CV as a communication device. Undergrad volunteer? Rotation student? Summer intern? No problem, they can be the first author right? Their level of contribution is not really the issue. I can see an industry scientist that wants to start making a push for management doing something similar to this.

As always, I return to the concept that you have to do your own research within your own situation to figure out what the expectations are. Look at what most people like yourself, in your situation, tend to do. That's your starting point. Then think about how your CV is going to look to people over the medium and long term. And make your authorship decisions accordingly.

92 responses so far

  • DoctorD says:

    Good Advice. I agree on everything you wrote.
    Even when I worked outside of academia and my job did not require publication, I wrote because if I wanted to return to an academic environment I had to show publication productivity. Publication also documents the work that you did.

  • DJMH says:

    At the point that someone is writing 6+ tweets on one question, shouldn't they just get a fukken blogge already?

  • Established PI says:

    Polly Matzinger's solution to the single author question was to include her dog as second author.

    Matzinger P, Mirkwood G. J Exp Med. 1978 Jul 1;148(1):84-92.

  • KlassenLab says:

    For what it's worth, I've been told by more senior colleagues that the fact that I had 2 single author papers during my PhD (along with many other first- and co-authored pieces) was very impressive when applying for my current academic faculty position. In this case, it seems to have sent the message of intellectual maturity and ambition that my hiring committee was looking for. And I think that it was unique enough to memorably set me apart from the other candidates, never a bad thing. Millage elsewhere may vary.

  • drugmonkey says:

    That tweep has one DJMH. At what point should a commenter who has lots to say on multiple blogs restart her own much-missed blog?

  • E rook says:

    I think having a single author paper during phd means you had a very generous and/or confident advisor.

  • Susan says:

    I just went through this decision as a new PI, and ... my grad student is getting a big gift in terms of a first authorship, but it is most important, for the long-term and cv reasons, that I am last author. Sole authorship would have been less preferable, as a new PI.

  • drugmonkey says:

    Absolutely, Susan.

    E-rook- just depends on the demands that are upon the PI. And the subfield expectations.

  • FTG says:

    As always, great suggestion and thoughts, Drugmonkey.

  • ecologist says:

    OK, I'm a bit surprised that no one has yet mentioned that there actually are standard criteria for authorship, that have nothing to do with "long-term and cv reasons" or "generous and/or confident advisor[s]." The most commonly cited one is that from the International Committee of Medical Journal Editors, which are also adopted by NIH and COPE, and lots of journals:

    The ICMJE criteria (emphasis in the original):

    Authorship should be based on the following four criteria:

    "Substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; AND

    Drafting the work or revising it critically for important intellectual content; AND

    Final approval of the version to be published; AND

    Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. "

    "All those designated as authors should meet all four criteria for authorship, and all who meet the four criteria should be identified as authors."

    I know from experience that authorship decisions can still be controversial, but at least there is a context (in some fields at least) in which they can be discussed, and sometimes that can be helpful.

  • drugmonkey says:

    That's because that is ridiculous bullshit that humanities majors who do not understand science publication are trying to foist on us.

  • drugmonkey says:

    (And yes, we've discussed this nonsense on the blog before)

  • The Other Dave says:

    Where I'm from, and on the hiring committees I know, single author publications are very impressive. I think it's a serious mistake to encourage adding co-authors just to add co-authors. That doesn't make you look like a team player. It just makes you look like a suck-up with flexible morality.

    When it comes to authorship, my rules are:

    1. "Authorship is not decided until the paper is done". Which means that I don't care if you worked really hard for 12 years on something. If it isn't mentioned in the paper, then you didn't contribute to the work described in that paper.

    Related to that...

    2. "Authorship is not just credit, it's responsibility". You get authorship if there is something in the paper that you can point to and say 'See that? I take primary responsibility for that.' Responsibility means blame, if it doesn't stand the tests of time.

  • drugmonkey says:

    Your 1. is bullshit and wrong, particularly in an era where the hard work leading up to a pizzazz figure isn't respected. If someone labored mightily doing work that allowed the fancy stuff in the paper to happen, it can very easily be a significant contribution to the paper. Even if none of that hard won preliminary stuff ends up published.

    Having a lab standard like this is not a recipe for solid progress and teamwork either.

  • drugmonkey says:

    Also, please explain your "flexible morality" charge. This always fascinates me when someone starts in on ethics and morality based on their training in publication traditions from one tiny perspective.

  • The Other Dave says:

    My rule number 1 is the way it is in my lab and I think the most common in my field. I'm surprised you don't agree, DM. Doesn't giving authorship for effort lead to all sorts of headaches? How do you decide what level of non-contribution deserves authorship? Do I get authorship for thinking really hard about the stuff and bringing cookies to all the lab meetings?

    Note that my system allows for authorship for people whose work maybe didn't lead to a figure but nevertheless makes it in the paper, like "We examined 432 candidates before observing significant effects with blah-blah, which we pursued further by..." The people who examined the candidates can get authorship, as long as those candidates are noted (usually in supplemental data). Saving other labs work is part of what publishing is for, and those authors can easily point to that sentence (and S.I.) and say "I helped rule out all those other candidates". That's a real contribution (and responsibility, because if any of those 'ruled out' candidates turns out to be paydirt later on, someone deserves blame for missing it).

    But someone who worked sort of on the project but never contributed to any knowledge or exclusion of possibilities? Is that what you're talking about, DM? if so, I think that's just honorary authorship based on what must ultimately be very subjective criteria. You want to know why science is still an old white boy network instead of a meritocracy? It's because of crap like subjective honorary authorship. Worst. thing. ever.

    The rubric for authorship has to be clear and convincing. No political games. No social niceties. Otherwise, it's unfair, misleading, and facilitiates the destructive power of unconscious bias.

  • The Other Dave says:

    Re 'flexible morality': If someone is willing to give authorship (the 'currency' of academia) to anyone for any reason other than a legitimate contribution, it's arguably the equivalent of bribery. Plus, as others have mentioned: Journals and societies typically do have pretty straightforward rules about authorship, and none of those lists includes 'political/careerism concerns'. What is bending the guidelines to advance one's political/career goals if not an example of flexible morality?

    You may be fine with honorary authorship. I see it as a sign that you may also be fine with 'you scratch my back I'll scratch yours' practices of all sorts, like I'll review your papers easy if you review mine easy. Or I'll put you on this one as author if you put me on that one as author. Or I'll send you those reagents if you can maybe give me some feedback on what people said at study section. Or maybe 'Well we were working on it anyway and would have come up with the idea eventually so.. so what if we did the experiments described in their proposal and scooped them?'

    There is no 'OK' amount of shenanigans. Authorship games is not innocent or generous or clever.

  • Which means that I don't care if you worked really hard for 12 years on something. If it isn't mentioned in the paper, then you didn't contribute to the work described in that paper.

    This is absurd. Shittio that didn't work informs figuring out what did work. If someone participated in conceiving, performing, analyzing, etc, the shittio that didn't work on the way to what did work, of course they are legitimately an author. And BTW, to amplify on DoucheMonkey's critique of the ICJME "standards" for authorship, at least two of them are absurd on their face:

    (1) "Drafting the work or revising it critically for important intellectual content"

    These are scientific manuscripts reporting data and data analysis. What matters is the experimental design, experimental results, and data analysis. The particular words and graphic design don't mean jacke fucken dicke, and the idea that if you didn't write or revise words of the manuscript or diddle around with photoshop to design a figure you aren't a legitimate author is ludicrous. This kind of rule could only be conceived by someone who has no fucken clue what the fundamental purpose of scientific publication is.

    (2) "Agreement to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved."

    Again, this is fantastical gibberish. The grad student who designed, ran, and analyzed a western blotte experiment is supposed to be accountable for ensuring that questions relating to the accuracy or integrity of the high-throughput sequencing experiment are appropriately investigated and resolved? Obviously, this standard could only be considered other than laughable by someone who has no idea how collaborative multidisciplinary research projects are designed and implemented. The only authors who could possibly be held to this standard are corresponding authors and, maybe, lab heads of collaborating labs.

  • There is no 'OK' amount of shenanigans. Authorship games is not innocent or generous or clever.

    Dumshitte, learn to fucken read. No one is suggesting playing "games" or "shenanigans". The particular standards you choose to apply for deciding on authorship are not given on from high, like the ten commandments, but rather are your personal judgments. As such, they are subject to substantial good faith disagreement by others.

  • The Other Dave says:

    Comrade, regarding your first comment: Please read MY whole answer. I totally advocate authorship for people whose work led to the accomplishments in the paper, as long as that 'supporting' work is adequately described. But if you can't point to something in the paper and take responsibility for it, you don't get authorship. Period.

    "Shittio that didn't work informs figuring out what did work. If someone participated in conceiving, performing, analyzing, etc, the shittio that didn't work on the way to what did work, of course they are legitimately an author."

    Yea, but only if those negative outcomes are reported in the paper, in which case they are legit contributions.

    "Doing shit for nothing but having people feel sorry for you" is not, in my opinion, a valid criterion for authorship. And I think it's dangerous, because, like I asked in my earlier comment: What level of NON-contribution justified authorship?

    Does failing for one day justify authorship? One week? Or does one need to slave away in your lab on a shit project accomplishing nothing for a whole year before they earn some sucker honorary authorship? What if they accomplish nothing and you don't like them, making it easy for you to justify that their non-contribution didn't lead to the stuff that's in the paper?

    With regard to 'shenanigans', I think Honorary Authorship *is* shenanigans.

    You want to know why middle authors don't get no respect? It's because guys like you apparently hand out authorship to everyone with a heartbeat. No one else can take that authorship seriously any more. I read someone with lots of middle authorships, and I don't know whether they're really a brilliant team-player scientist, or someone who the PI just feels sorry for.

    I knew a PI who routinely gave authorship to the cute grad student he was banging on the side. Everyone knew she didn't have anything in the papers to take credit for. But certainly she was a valued member of the 'team', whose 'work' definitely contributed in *some* way to the success of the projects.

    I think it's way better for every author to be able to point to something definite that they contributed to the paper, rather than rely on what people might otherwise think.

  • The Other Dave says:

    Anyway, back to the original post: Telling the tweeter to add authors even they didn't contribute significantly to the paper is bad advice. Bad. Bad bad bad. bad for the tweeter, bad for science. bad bad bad.

    What DM should have said was 'You have a single author paper? Congratulations!' Single author papers are damn impressive. I have never ever ever heard anyone say otherwise.

    Then it would have been fine for DM to suggest that the tweeter consider working more as part of a team for other projects. But adding authors to a paper for no good reason? Bad advice. Really bad. Bad in multiple ways. That's where DM screwed up on this one.

  • Dumshitte, your "no good reason" is someone else's "significant contribution". You simply disagree with what constitutes a significant contribution, and that's fine. But enough with the histrionic assertion that DoucheMonkey is suggesting adding authors whom he himself agrees didn't contribute significantly.

  • The Other Dave says:

    Thanks for the 'Dumshitte' citation, Comrade. But please henceforth acknowledge my contributions to your work with co-authorship.

  • Busy says:

    Here's a real life scenario: a group of researchers were stuck in some resolution threshold past which they couldn't get. A team of really smart people tried any way around it, and couldn't. One day a junior member had a Eureka moment and pushed beyond this limit by a modest amount. Once this barrier had been removed a super bright young graduate student ran away with it and she completely obliterated the threshold.

    The techniques she developed, while inspired by the first breakthrough had not tell tale signs of the first effort. So when you read that paper there is no place where you can point to the Eureka moment, since what's the point of reporting a method that allows you a 20% improvement when we have another that gives 40%.

    According to TOD "if you can't point to something in the paper and take responsibility for it, you don't get authorship. Period."

    It goes without saying that I disagree, and that I believe we did the right thing by giving co-authorship to the junior researcher.

    p.s. adding "Period" at the end just makes it sound that you have no arguments and are just trying to push this on an authority basis alone.

  • The Other Dave says:

    Busy: I don't understand. It sounds like the 'Eureka moment' is what led to the procedures and results in the paper. That's co-authorship material. No problem.

    And it's not true that you 'weren't able to point to something in the paper'. You did. Quite easily.

    I'm talking more about hard work that did not lead to anything in the paper. In your example, the 'Eureka moment' clearly did. I have no problem with that.

    Go back and read the tweets as posted by DM. Some of the tweeters colleagues played no role in the work, but merely suggested tweaks to the manuscript afterwards. That's acknowledgement material, not co-authorship material. I think DM was wrong to suggest they might deserve co-authorship.

  • drugmonkey says:

    I merely said that what you have reduced to "tweaks to the manuscript afterward" may in fact be significant contributions. Your refusal to consider that such contributions might be more than your dismissive "tweaks" is simply your lack of imagination talking.

  • drugmonkey says:

    And I return to my usual question- where is your evidence that true "honorary" authorships actually exist in numbers large enough to worry about. Even if they do exist, where is the harm and to who?

  • Grumble says:

    I once had a single-author paper, but unfortunately it was a courtesy authorship.

    Seriously, though, I don't think it hurts to have one or two single-author papers, especially early in your career. First of all, it catches the eye because it's so rare nowadays - colleagues in my department made a point of mentioning the paper to me, more than with my multi-author papers. Second, single-authorship says, "I did all this shit myself." It establishes you as someone who can handle every single aspect - conceiving, doing, analyzing, writing. Finally, being a team player is not ALWAYS the most important thing. Many great ideas are the result of a single person's singular inspiration. If a paper then gets a reputation for having proposed a startling new idea, and if that filters through to hiring or promotion committees, the message a single-author paper sends is: "That guy/gal thinks". Always good to be known for that in this business - and all the better if there's no question about who did the thinking.

  • jojo says:

    I'm with The Other Dave.

    I think it's better to err on the side of "actually contributed significantly" to the work. Yes, this has led to me not being 4th author on a couple of papers where I helped with the grunt work (e.g. took care of animals in someone else's experiment, helped a close colleague figure out how to use some software I developed). I'd be embarrassed to have those on my CV anyway, since I was barely involved. My CV has only a single paper which is not primary author. It doesn't mean I'm not a team player, it means that I don't take credit for work I didn't do. If someone asks about those papers where I am put on for no reason, what am I going to say? My integrity would be forfeit.

    I was "offered" Honorary Authorship once and declined vociferously. I literally had contributed none of the data and did none of the analysis. They didn't even tell me I was going to be on it until after they submitted the paper. This kind of thing needs to stop.

  • The Other Dave says:

    jojo recognizes that paper count isn't everything. In the long run, reputation still counts for a lot. If there were more jojos in science, a lot of problems would be solved.

    DM: I am merely going by what the tweeter said. Re-read those tweets you posted. It's pretty clear that the others didn't contribute to the paper in any way that deserves co-authorship.

    As for harm, I said it already: Honorary authorship minimizes the accomplishments of middle authors, because it weakens the meaning of middle authorship (how do we know so-and-so really did anything meaningful?)

    Is honorary authorship widespread? Well, I think there are certain otherwise sensible bloggers advocating it...

  • drugmonkey says:

    I don't know that, given attitudes like yours, TOD. Also given a pronounced inability of many trainees to understand that the environment they have been given in which to do science matters.

  • AcademicLurker says:

    I'm with DM in that, at least in my personal experience, true "gift authorship" is vanishingly rare. By "true" here I mean "genuinely did nothing". I can't think of a paper I've been on where that was the case for any of the authors.

    The one glaring case I'm aware of is a big cardiology center at my previous institution. They brought in a Harvard big wig to run the place, and as part of the deal he is corresponding author on every paper that comes out of it. This came up because we had a "2 first author paper" requirement for our graduate program, and this made it hard for our graduate students to work for any of the PIs in the center, since the PIs were always first authors (since the corresponding author spot was reserved for Mr. big wig).

  • drugmonkey says:

    Honorary authorship minimizes the accomplishments of middle authors, because it weakens the meaning of middle authorship (how do we know so-and-so really did anything meaningful?)

    For this to be true you have to assert that all middle authorships are less than meaningful contributions. Or at least that that vast majority are. They are not, so it is hard to see where the subjective views on the relative importance of first and last versus middle authors has been built upon true courtesy or honorary, non-contributing authors. Subjective evaluation of middle-authors has, for better or worse, been built upon middle authorships that everyone thinks of as an author-worthy real contribution.

    I see no conceivable future where cutting average author lists, e.g. from 7 to 5, is going to magically convince everyone that five papers in slot 2-3 are as good as two first-author papers.

  • drugmonkey says:

    Is honorary authorship widespread? Well, I think there are certain otherwise sensible bloggers advocating it...

    I have yet to see one single person advocating including authors who have not made a significant contribution to the manuscript.

    You are, again, allowing your personal opinion on what constitutes a significant contribution and your rectitude in defining that for everyone else to contaminate your reading of what other people are saying.

  • drugmonkey says:

    this has led to me not being 4th author on a couple of papers where I helped with the grunt work (e.g. took care of animals in someone else's experiment, helped a close colleague figure out how to use some software I developed). I'd be embarrassed to have those on my CV anyway, since I was barely involved.

    You are more then welcome to have your own standards when it comes to your participation and, indeed, when you are the head a lab with the decision making role. So long as you behave consistently as the lab head, this is your right. The trouble comes in when you are 1) significantly out of step with the real-world practices of your subfield AND 2) insist that everyone should share your view. Especially if you start spouting about practices that vary from your preference being unethical, immoral and the like.

    My CV has only a single paper which is not primary author. It doesn't mean I'm not a team player, it means that I don't take credit for work I didn't do. If someone asks about those papers where I am put on for no reason, what am I going to say? My integrity would be forfeit.

    That's fine that you think that. But the fact of the matter is that we all know quite a bit more about our contributions and unique talents than is readily apparent to the reader of our CVs and Biosketches. We will not always be in a position to argue chapter and verse about how awesome we are and how our unusual constellation of single-author and two-author papers really means that we have the utmost integrity. Sometimes, our personal standards for what represents authorship will be in line with the subfield-typical expectations of the reader and sometimes they will not. And I'm here to tell you that if you have only first author pubs then you look like not-a-team-player.

    So it behooves us to think a little tiny bit about the perception we will be generating. Sometimes this is only so that we are aware of the need to emphasize things in the Cover Letter or the Personal Statement on the Biosketch because we really aren't in a position to change anything. Sometimes this means we indeed change how we behave by seeking the opportunity for someone else to make a material contribution to the paper.

  • Dennis says:

    Hi,

    this is a really interesting discussion!

    I have a (at this point) more hypothetical question. I am a postdoc hoping to become a PI within the next few years (no more than 4, I hope). In my PhD lab my adviser was running experiments 'on the side' that our technician conducted under his supervision.

    I'd like to adopt this. So, if I find a technician capable of conducting experiments for me and I find the time to analyze the data myself, etc., should I consider putting the tech as first author or rather solo author possible papers coming out of this work? Or does it look weird when the first author isn't a student or postdoc?

    cheers,
    D.

  • As a computational biologist I end up with a middle authorship more often than not. First author is the grad student/postdoc who did the bench work, last author the advisor of the first author. People who analyze the data, do statistics and so forth? Middle authorships. It's just how it works, unfortunately.

  • drugmonkey says:

    So, if I find a technician capable of conducting experiments for me and I find the time to analyze the data myself, etc., should I consider putting the tech as first author or rather solo author possible papers coming out of this work? Or does it look weird when the first author isn't a student or postdoc?

    I don't see where author lines list the person's job title in pub med or on a CV. occasionally journals will print degrees but a grad student and a tech often have the same bachelor's degree.

    When it comes to technicians as authors I know of some people that think that a tech is never an author, no matter what. Some other people think it is perfectly fine to put techs on papers, even as first-author in some situations. I think a lab that sets a "no-tech-authors" rule is kind of jerky but that is the PI's right to decide (as long as it is applied consistently).

    Would it look weird? Yeah, I suppose if there was a lab consisting of a PI and a tech and everybody came to know the tech was nothing more than dumb hands and never knew anything about the science but was always the first author....that would be unusual. ...in most of the situations like this that I can think of the PI is in the first author slot.

  • drugmonkey says:

    jojo recognizes that paper count isn't everything.

    It is important though. I can't recall hearing too many papers being a knock on anyone although I sure as heck have heard people be criticized for having too few. And I'm here to tell you, even if the obvious conclusion is that the person with the low number of pubs clearly did most of the heavy lifting on each and every one of them this doesn't help much.

    There are some parallels here with multiple grant attribution to every paper. If, as an author, you are able to "just get on" a bunch of collaborative works by providing a reagent, mouse, software package, key analysis or whatnot to many different people, you are going to look good. If your every paper is your own solo work of broad sweeping excellence, you necessarily are going to have lower output. Eventually, that will bite you.

  • The Other Dave says:

    And I'm here to tell you that if you have only first author pubs then you look like not-a-team-player.

    And I'm here to tell you that if you add authors to publications where you *should* be sole author, then you are selling yourself short.

    I have a (at this point) more hypothetical question. I am a postdoc hoping to become a PI within the next few years (no more than 4, I hope). In my PhD lab my adviser was running experiments 'on the side' that our technician conducted under his supervision.

    I think this is a good idea. Someone needs to do the exploratory stuff that may or may not turn into projects worth giving to a student of postdoc. And in a mature lab, there's an awful lot of unfinished results that will go wasted unless someone spends full-time fleshing them out for publication.

    I'd like to adopt this. So, if I find a technician capable of conducting experiments for me and I find the time to analyze the data myself, etc., should I consider putting the tech as first author or rather solo author possible papers coming out of this work?

    Job titles don't matter. If a tech or your grandmother or dog or left shoe bears responsibility for the paper or any reasonably significant part of it, they're an author. Otherwise, they're not.

  • The Other Dave says:

    Wait, DM, did you really mean to type 'only first author pubs' above?

    Are you really telling people to seek out middle authorships above first authorships?

    Please don't. That's cruel.

  • AcademicLurker says:

    Are you really telling people to seek out middle authorships above first authorships?

    Forget middle authorships. What you should really be aiming for are mentions in the acknowledgements.

  • drugmonkey says:

    That is the only moral target, AL.

  • jojo says:

    Thanks for replying DM

    We will not always be in a position to argue chapter and verse about how awesome we are and how our unusual constellation of single-author and two-author papers really means that we have the utmost integrity.

    I'm not sure my CV is that unusual for my field based on what I've seen. The thing I want to avoid is someone asking me "Ah, yes I saw this paper... what did you do on this project?" and either having to lie or say the truth, which would make me look like a jerk.

    Is it likely that would never come up? Probably. But I don't want to be in a position for that to happen.

    Sometimes, our personal standards for what represents authorship will be in line with the subfield-typical expectations of the reader and sometimes they will not. And I'm here to tell you that if you have only first author pubs then you look like not-a-team-player.

    I don't see the point of pushing myself onto papers that I don't deserve (and I think I'm following subfield-typical expectations here on the definition of deserve). I think reputation does matter in terms of future productivity. If I have a reputation for being a grabby jerk, no one will want to collaborate with me. There are a handful of people I have decided not to collaborate with for this and other "integrity-based" reasons. At the same time I see no reason not to continue to help my colleagues - they help me too after all.

    I can't recall hearing too many papers being a knock on anyone

    I have heard this in the case where there are only a small handful of primary author pubs. AKA if you have 15 papers and only 1-2 primary that's a red flag. Not saying that the person who said that is right universally - it's probably subfield specific, I imagine?

  • drugmonkey says:

    it is true that first, middle and last author papers are not substitutable quantities. in your scenario, "too many middles" is really just saying "too few first-author" is it not?

  • The Other Dave says:

    I'm not sure my CV is that unusual for my field based on what I've seen. The thing I want to avoid is someone asking me "Ah, yes I saw this paper... what did you do on this project?" and either having to lie or say the truth, which would make me look like a jerk.

    Explaining stuff like that, if needed, is what letters of recommendation from your advisor are for. After a while, people should know you and what you do, and be able to just figure it out.

    AKA if you have 15 papers and only 1-2 primary that's a red flag.

    Depends how long it took you to get those. If that was all in one 4 year postdoc, then you're going to be a damn fine PI. If that was over 15 years, you look like tech material.

  • E rook says:

    I'm 4 years into my adjunct prof position and the "PI" who lets me & my tech occupy the lab space gets last author on all my papers (regardless of contribution), the one time I tried not to, shit went nutso. Has this hurt me? Yes ... Reviewers on R01 apps have more than once commented on the fact that there's always another senior author on my papers. Sucks but I understand this "PI" need wind in the sails to keep the ship afloat.

  • drugmonkey says:

    How about communicating? Will the big cheese give that up?

  • The Other Dave says:

    Ha. 'Corresponding author'.... Now we are onto another topic where no doubt DM and I will likely disagree. (And I think we did already on this, years ago) But, well, just to get it going...

    I think 'corresponding author' is for the readers, not the authors. Corresponding author is not a statement of author contribution. It is just a job to do after the paper is published. Corresponding author takes primary responsibility for dishing out info and reagents. Corresponding author should have a reasonably stable position and be able to be contacted fairly easily. Corresponding authors agree to... well... correspond.

    If someone is corresponding author on something and they are not the PI, it doesn't make me think more highly of the corresponding author (if indeed I even notice). It makes me think the PI is just somewhat lazy/irresponsible. If corresponding authors point that status out to me as if it's something special, I think it's just lame. It makes me think they're petty claimants to meaningless 'status'.

  • The Other Dave says:

    Didn't mean that whole last paragraph to be italicized. Just forgot to close the italics after 'correspond'. Wish there was a preview or edit comment function here.

  • K99er says:

    Nobody has ever asked me in my process of job interviews, meeting guest speakers, etc. what my individual contribution was on one of my middle author papers. I think this worry is based mostly on a myth that anybody reading your CV is going to care about your middle author papers beyond them adding to your total paper count. Yet, I am always prepared for this question regarding all of my papers even if it is something really simple like, "I was an undergrad at the time and I designed qPCR primers for figure 3 and tended the cell lines used in figure 5." As long as you did SOMETHING, what's the big deal?

  • I think 'corresponding author' is for the readers, not the authors. Corresponding author is not a statement of author contribution. It is just a job to do after the paper is published. Corresponding author takes primary responsibility for dishing out info and reagents. Corresponding author should have a reasonably stable position and be able to be contacted fairly easily. Corresponding authors agree to... well... correspond.

    If someone is corresponding author on something and they are not the PI, it doesn't make me think more highly of the corresponding author (if indeed I even notice). It makes me think the PI is just somewhat lazy/irresponsible. If corresponding authors point that status out to me as if it's something special, I think it's just lame. It makes me think they're petty claimants to meaningless 'status'.

    This may be what *you* think, but it is completely out of the mainstream in the biomedical sciences, where the corresponding author(s) are (correctly) understood to be the senior leader(s) of the project. And more importantly than what happens after the paper is published, the corresponding author is solely responsible for communicating with the journal editor(s) regarding the paper.

    Honestly, your comments in general in this thread give the strong impression that you have been an author on few--if any--scientific manuscripts in the real world. You sound like one of these humanities goofaloons who seeks to impose standards on scientific publishing without even the slightest understanding of how publishing works in the context of the scientific enterprise.

  • jojo says:

    in evo-eco land corresponding author is often the first author (grad student), not the PI, though many times both are. Mostly depends on who is really steering the project and is therefore in a position to correspond most effectively (aka is it the PI's project the student is working on, or is it the grad student's project and the PI is advising).

  • drugmonkey says:

    Splitting up senior and corresponding is a way a generous BSD heading a huge lab can help the trainee establish their domain of focus in the group. It is worth asking for.

  • AcademicLurker says:

    @jojo: For areas of science that involve field work, shouldn't the corresponding author be a secretary back at the university, so that someone can keep up correspondence after the researchers are eaten by wild animals?

  • jojo says:

    @jojo: For areas of science that involve field work, shouldn't the corresponding author be a secretary back at the university, so that someone can keep up correspondence after the researchers are eaten by wild animals?

    Oh for sure. An alternative is to make sure only one of the corresponding authors is in the field at the same time.

  • E-Rook says:

    Yep I am corresponding author on these pubs which Dr Cheese gets last author on. Though I suspect RoD was a reviewer of my last grant app. Or at least shared his sentiment. I also think this reviewer was as out of touch as CPPs description of ToDs comments. I think some people just think Thier career trajectory and experience with appointments and promotion represent The Way Things Are without consideration for the variety of circumstances that PIs (of the lesser gray haired breed) find themselves in. My local P&T committee gets it, but the wider world's perception is what concerns me ... since it matters more.

  • The Other Dave says:

    CPP, dude, it's like you are going out of your way to disagree with me. Relax.

    This may be what *you* think, but it is completely out of the mainstream in the biomedical sciences, where the corresponding author(s) are (correctly) understood to be the senior leader(s) of the project.

    I agree. In practice, the corresponding author is almost always the senior leader of the project. That is because, as I said above, the senior leader of the project typically is best equipped to answer questions about the project and distribute reagents. 'Corresponding author' should NOT, in my opinion, be used as an authorship consolation prize, which is what DM seemed to be advocating.

    And more importantly than what happens after the paper is published, the corresponding author is solely responsible for communicating with the journal editor(s) regarding the paper.

    Agreed. Which, except for crafting rebuttal letters, is mostly waiting for shit to upload and the PDF to be completed for approval. I like to do this with grad students at least once so they can see how the system works (until they change it again in another few years), and because it is exciting for them.

    Honestly, your comments in general in this thread give the strong impression that you have been an author on few--if any--scientific manuscripts in the real world. You sound like one of these humanities goofaloons who seeks to impose standards on scientific publishing without even the slightest understanding of how publishing works in the context of the scientific enterprise.

    That is an interesting conclusion. Since it is in fact completely wrong (I am a nascent full professor at a RU/VH university with large medical school, plenty of publications, plenty of reviewing, and even some editing for the leading journal in the subfield), I must conclude that your blog comment analysis methodology is flawed. You may want to revise your conclusions and reconsider my opinions. Perhaps it's possible that both of us represent 'mainstream' thinking?

  • The Other Dave says:

    Dear junior scientists out there:

    You can believe DM and CPP and whomever you want to about the value of corresponding authorship. Go ahead and argue for that designation on your next pub if it pleases you.

    But please also recognize that people like me out out here, reading your proposals, looking at your job application files, reviewing your promotion & tenure packages. I don't think corresponding authorship is a meaningful scientific accomplishment. My advice is to concentrate on more meaningful things. Work to make sure that you are in undisputed first author position, or senior author position, whichever is most appropriate for your career level. Then, when you are PI and have some whiney student or postdoc who won't shut up about their authorship position, consider using the 'corresponding author' scam to quiet them. I don't like it, but it seems to work for others.

  • E-Rook says:

    I am well aware that ToDs sentiment is out there. The problem for me is that I occupy the first author position on lots of pubs since my adjunct faculty appointment and I am also the corresponding author. This is regardless of who was the driving force or leader behind the study. Or who really pushed it over the finish line and got the damned thing out. Generally, the last author goes to the big cheese who allows me and my tech to occupy the lab space. The arrangement allows Dr Cheese to continue to demonstrate productivity and justify keeping space but hinders me because of prejudices like those expressed by ToD. The other option is to not have a lab. Or move. Without burning bridges. Without startup funds and my own space or an R01 as PI, I don't see this happening and just try to chug along. I don't think dr cheese is behaving ethically but it's what I'm working with.

  • jmz4 says:

    I think that there's little harm in adding middle authors for minor contributions, negative results, etc. Politically, it makes people more likely to collaborate with you, since they know they'll get something out of working with you. Also, only real alpha lab-tough guys care that much about how many authors are on a paper, since that can vary hugely by field and scope of a paper.

    A possible point for discussion is whether the PI's of people from other labs who did some experiments for you should get to be on the paper. What if they didn't even cover reagents?

    Secondly, and less related to the original post, what about co-first authors?

    We actually just went through the authorship class at our responsible conduct of research course, and there was a HUGE disconnect between PI's and postdocs on the merits of various forms of authorship, but that was the biggest dispute.

    Most PI's saw co-first authors as completely equivalent, while virtually all postdocs insisted that the first co-first author was the "real" first author, and the second one was likely tacked on for political reasons (or because the value of middle authors has been so degraded that promotion to second co-first is the only way to signal real, substantial involvement).

  • drugmonkey says:

    "Most PIs *claim* to view co-first as completely equivalent..."

  • Comradde PhysioProffe says:

    "Most PIs *claim* to view co-first as completely equivalent..."

    But all of them know they aren't.

  • drugmonkey says:

    And the scam rolls on.

  • drugmonkey says:

    Jmz4- as far as the other lab's PI, why is the collaborating postdoc working with you? How is she funded, what resources is she using?

  • The Other Dave says:

    If I see two people listed as co-first authors, I assume they contributed equally. Otherwise, why would the authors go to the trouble of agreeing to joint first-authorship?

    I have even advocated in the comments of this blog, years before, that the co-first author name order is interchangeable. After all, if A=B, then B=A. The harshness of the replies convinced me that, although I thought it was perfectly fine, others did not, and therefore I stopped telling trainees that it might be OK. It was one of those 'yea, I know, but this is what people think...' moments. The real world sucks.

    One of the reasons I've visited this blog now and then for about a decade is because I like DM's approach (then also BM and CPP) to explaining the real world of science to readers.

    Unfortunately, it's attitudes like those by DM and CPP above that ruin it for second-listed co-first-authors everywhere.

  • Comradde PhysioProffe says:

    "If I see two people listed as co-first authors, I assume they contributed equally. Otherwise, why would the authors go to the trouble of agreeing to joint first-authorship?"

    The question you *should* be asking is, why would the authors have agreed to list the first-listed author first and the second-listed author second?

  • jojo says:

    Unfortunately, it's attitudes like those by DM and CPP above that ruin it for second-listed co-first-authors everywhere.

    I mean, when your paper is cited it will be as First A et at, not First A and Second A (unless it's a 2 author paper...). Of course this matters - as a trainee it's vital for people to learn your dang name. That was the whole objective of putting trainees on as First Author - to get their name associated with their work as it is cited.

  • AcademicLurker says:

    In the digital age, the co-first author problem could be solved by have the order of the 2 names be randomly determined every time the pdf is opened...

  • Juan Lopez says:

    ToD,. Isn't it always the case that someone's attitude ruins it for all others?

    See for example your own comment

    "If someone is corresponding author on something and they are not the PI, it doesn't make me think more highly of the corresponding author (if indeed I even notice). It makes me think the PI is just somewhat lazy/irresponsible. If corresponding authors point that status out to me as if it's something special, I think it's just lame. It makes me think they're petty claimants to meaningless 'status'"

    I see corresponding authorship as important. In my area I am sure other people do too. But you don't, and you even take it against us.

    In my opinion it's all those who give nil credit to middle authors, but that also want collaborative science, who are the problem for everyone.

    I also come to this blog to hear what others think, whether I like it or not. Sometimes their arguments have changed my mind. It's awesome when that happens.

  • Juan Lopez says:

    AL, ha ha. Yea, what we need is software that randomly changed documents. That's a really really bad idea.

  • AcademicLurker says:

    Come to think of it, why stop at the order of author names? We could randomize the p-values as well. Disruptive innovation!

  • jmz4 says:

    "Jmz4- as far as the other lab's PI, why is the collaborating postdoc working with you? How is she funded, what resources is she using?"
    -This was a hypothetical, but I do work with someone who I know is using their K99 award to fund a lot of the research we're collaborating on (it's a c. elegans project, so this is totally doable). As such, his time and most of the materials are through himself, but things like microscopes and TC hoods are obviously not. I've really got no problem with the practice of putting both authors on (though I think it would be better if there was a way of obviously pairing them). I was just wondering if there was a situation anyone has run across where that didn't happen.

  • drugmonkey says:

    Let us be clear ToD- I advocate a solution outlined by AL. No reason authors cannot be randomly shuffled on pdf download. I say that authors *should* be allowed to list themselves first on the CV and anyone who has a problem with that and yet engages in co-equal shenanigans on their own papers is a hypocritical asshole. And it is a crying shame that equal contribution is not indicated on the Medline record. In short, I do not endorse the notion that there can be only one.

    However.

    This blog is as much about succeeding in the world as we find it as it is about complaining about the bad things. And the simple matter is that on average, the academic crediting systems do not view listed-second co-equal authors the same as the listed-first one. Misunderstanding this will lead a trainee to make suboptimal career decisions.

  • AcademicLurker says:

    I advocate a solution outlined by AL.

    That includes the p-values part, right?

  • drugmonkey says:

    No.

  • AcademicLurker says:

    More seriously, my opinion of co-first authorships has been influenced by being one. It really was a a very close 50/50 split where the work could never have been done without both contributions. I was happy to take the second slot because I was at a somewhat more advanced stage of my career at the time than the other author. But would have had no qualms at all about being listed first.

  • Drugmonkey says:

    The only important question is whether you understood the way it would be perceived by others. As long as this is clear, nobody is being abused.

  • The Other Dave says:

    jojo: You're right. We should all start citing co-first author papers using BOTH first author names, as in: "J. Blow/J. Doe et al".

    In the digital age, the co-first author problem could be solved by have the order of the 2 names be randomly determined every time the pdf is opened...

    I like that idea.

    Every joint first author paper submitted from my lab gets submitted with a slash between the names, as in:

    "J. Blow/J. Doe (contributed equally), F. Snow, B. Low, and T. O. Dave"

    But in the end every publisher always changes it to their standard format, after which I ask why they did that, and they cite some weird standards or something. But if enough of us do subtle pushes like that, things can changed.

    Juan Lopez: My position on corresponding authorship isn't because I want to discredit the person who is corresponding author. It's because the position of corresponding author already has a specific meaning, and it is NOT 'person who should be PI on this but can't for some reason'. it is 'corresponding author' -- the person in charge of handling the manuscript and subsequent questions and reagents and stuff. You know you can be joint LAST author same as joint first author. If you think you are getting shafted by someone who wants last position on a project you really led, then get asterisks by both your names stating that you contributed equally. Or state the roles clearly in the acknowledgements. There are ways to deal with it that make it clear to readers. Secret meanings attached to 'corresponding author' status are not, in my opinion, the way to do it.

    DM: I (reluctantly) sort of agree with you. But I think part of the problem is our own doing, and related to the original post and discussion here. As long as authorship, or authorship position, gets handed out for anything other than significant contributions to the paper, then people are going to wonder whether any particular author really deserves their position. When some game the system, we are all left with a system that everyone assumes is being gamed. Sucks. Not much we can do about it except recognize reality and try to avoid playing the game ourselves, or encouraging it. That's why maybe some of my opinions seem to take a hard line. This is the internet. We have some responsibility. If we say it's OK to fiddle with authorship in fishy ways and throw away outliers because they don't fit the prediction, then people are going to start to assume that everybody does that stuff.

    So, yea, DM, when you asked me earlier whether honorary authorship was a big problem... I didn't really give an answer. That's because I think it doesn't matter. All that matters is whether it is a perceived problem.

    The only important question is whether you understood the way it would be perceived by others. As long as this is clear, nobody is being abused.

    Yea, I think this nails it. Publications, and everything on them -- including authors and author order -- is a communication. You've got to know exactly what you're communicating.

  • clueless noob says:

    Does the importance of the last author vary at all by subfield? I see some fairly senior people in my area who still assign the authors in descending order by contribution (e.g., PI=first, postdoc=2nd, hardworking RA=3rd, ..., summer intern = 6th). It would be mildly entertaining if the also-rans on these papers strings were getting senior author credit.

  • qaz says:

    @CNoob: Absolutely!

    Each field has very different authorship assumptions. Biology in general is based on the first-author(s) is (are) the people who led the actual study (the boots on the ground so to speak) and the last-author(s) is (are) the supervisor who oversaw it (the general).

    But computer science used to be listed in terms of contribution. (Which made for very interesting authorship discussions when the computational neuroscience crowd started to come in. CS who has only contributed one modeling figure: "Sure, just put me last, that's fine." Neuroscience colleague: "WHAT?")

    My favorite is the physics report of (I think it was) the top quark that had 900+ authors, listed alphabetically, all of whom where X. Y. Zee, except for the lead author who listed his full name spelled out (somewhere in the middle of the 900 authors).

    Authorship order is not a reliable communication of contribution across fields.

  • drugmonkey says:

    But no co-equal authors? Very strange.

  • Juan Lopez says:

    Drugmonkey December 17, 2014 at 12:08 pm
    "The only important question is whether you understood the way it would be perceived by others. As long as this is clear, nobody is being abused."

    DM, perhaps you mean that nobody is being tricked or misled. That's very different from not being abused. Your own blog covers plenty of cases where people abuse others, often with everyone fully aware of the abuse, just unable or unwilling to do anything about it.

  • drugmonkey says:

    Fair enough.

  • @qaz
    Actually, in my experience as a Computer Science postdoc is that the CS system is more often simple alphabetical order as they think all this fighting over authorship position is juvenile. That being said, having a surname starting with "B" wasn't altogether unhelpful in regard to biologist interpretation...

  • drugmonkey says:

    What are the measures of academic credit for a CS postdoc?

  • […] who should and should not be on the author line of a science publication is not as simple as it seems,” writes […]

Leave a Reply