In academia, authorship can be a touchy subject that is all too easy to sidestep. However, it’s always simpler to have a contentious conversation prior to submission.
Cite as: Kamoun, S. (2022). Authorship — Let’s talk about it! Zenodo https://doi.org/10.5281/zenodo.7441483
Authorship can be a sensitive topic in academia, but let’s talk about it. My goal with this post is to throw out ideas and open paths for discussion, thus helping break the taboos around this topic and putting to rest some misconceptions. Discussion and prevention are always preferable to addressing issues after they arise. And authorship discussions don’t have to be as frosty as a meeting between Vladimir Putin and Emmanuel Macron.
Let me start with a disclaimer. Please don’t take anything I say as a hard guideline, but rather as a framework for getting a discussion going — an important discussion that all graduate students and postdocs should be having with their PIs and colleagues. More on when and how later. But first, some context.
Publishing: the principal currency of academia
As scientists, we do research, but we also must communicate our findings to our colleagues and our peers in order to advance our science and our careers. We publish primarily to communicate our findings, but also to have a formal record of what we actually contributed. Publishing is often viewed as the last step in a scientific research project and if a project is not published, or is not widely communicated, then it is a failure. Even if you have incredible data, if you keep it to yourself in your head and lab notebooks, it won’t be seen by other scientists and the impact on the scientific community will be nil. I would argue that by not communicating your results, you have failed your funder and your community.
Because publications provide the record of our contributions, publications are the principal currency of academic research— we don’t use money to reward each other, we use publications, and these publications advance our careers and get us new jobs, grants, and recognition. This is obviously why publication is often a source of tension — as with any currency, how the currency is divided can cause stress. The issue is more pertinent than ever because 21st century biology is all about teamwork — it is now nearly impossible to do any form of biological research without a team of scientists. So, we need to figure out how to divide the credit among all members of this team, which can lead to conflict.
If publishing and authorship are the principal currency of academia, this means that assigning authorship is one of the most delicate decisions a PI has to make and the one that’s most prone to result in tension among members of the team. My advice to PIs is to be open, transparent, and fair. Make sure to explain the basis for your decisions and the philosophy by which you approach authorship. That will take you a long way towards reducing the tension around this topic by managing expectations and providing a clear, consistent framework for decisions around authorship.
Who should be an author?
Some of you will recognise a younger version of the Beatles in this photo, but that guy in the middle, George Martin, has always been half-jokingly talked about as being the fifth Beatles. Maybe the Beatles were more than the four we all heard of? George Martin was the producer and was instrumental to the success of the Beatles, but I’m pretty sure that most of you have never heard of him although you’ve heard of the other four Beatles. So being an author is important — it makes sure you are recognized as being part of the band.
When I started my lab, I was concerned about how I was going to deal with authorship issues. I discovered a booklet called ‘On Being a Scientist: A Guide to Responsible Conduct in Research’, produced by the National Academy of Sciences. This little booklet is very clear in defining authorship — maybe as close to consensus as you can reach — and states that the authors included in a paper should have made a substantial contribution to conception, design, analysis, interpretation of data, or drafting and revising the manuscript in terms of intellectual content.
What this definition means is that there are a lot of different types of contributions that merit including someone as an author. I think it’s fair to say, especially when the list of authors increases, many authors generally contributed to producing or interpreting the data, or performing some of the experiments, or drafting and revising the manuscript. It’s generally unlikely that all authors contributed to all of the aspects listed above. For instance, if you have 20–30 authors, you can’t really expect (or want) everyone to write the paper. They may contribute a short section, or they may not contribute to the writing at all. This is the nature of modern research in the life sciences. The old definition that all authors have contributed significantly to the writing of the paper no longer applied once biology projects started involving large teams.
Nonetheless, as the ‘On Being a Scientist’ booklet states, a substantial contribution is expected of all authors. This is where different scientists have different views. What counts as a substantial contribution — that is something that must be decided by the PI.
Who should not be an author?
There are some generally accepted guidelines about who should not be an author — but these are guidelines and not hard and fast rules. Authorship should not be based on factors such as seniority, position, or funding. You should not include people as co-authors just because you think that will improve the chances of the paper being accepted by a journal. Believe it or not, I’ve had people contact me saying “I’ll send you my paper and I’d really love to have you as a co-author”. Are you kidding me? In fact, that’s a form of corruption if you think about it. If you accept that authorship is a form of currency, it’s not much different from giving someone a pile of cash and saying, “can you do me this favor please?” You should not include your friends, or someone you want to be nice to just to help their career or get them another paper to add to their CV. That’s absolutely not ok. It’s unethical to say the least.
The guide to responsible conduct in research, ‘On Being a Scientist’, defines who should not be an author and includes as an example a category I had never thought about before: people who perform a paid service or a service for a fee. For example, when you send your gene, marker, or PCR amplicon to a company for sequencing; you don’t expect the people who did the work at the company to be co-authors on your paper. That’s a very simple way to think about it: the company is performing a service, and therefore the paid service people are not expected to be authors.
Gray areas
As I wrote before, I really get irritated when scientists, whether PIs, postdocs or students — take a cavalier approach to authorship. If you care about your colleagues and you care about developing their careers, then you should really take the time to think about whether they deserve to be an author on your publication. However, there are many gray areas in the topic of who should be authors and who should not be authors. The many gray areas mean that there should be an internal discussion about authorship set up by the senior authors of the paper or the PI of the lab. They should explain to their co-workers their philosophies around the issue of authorship and who should be and who should not be an author. It’s important to know that there is absolutely no community consensus on who should be authors, which again puts even more emphasis on the importance of PIs being consistent and open.
What can you do about that? Well, you can certainly discuss it with your PI and have internal discussions. You could also investigate and try to understand the guidelines in your community and those set by the journal of the scientific society you work with. Don’t forget that the key element of authorship is intellectual contribution. The booklet from the National Academy emphasizes intellectual contribution much more so than technical contribution. This is a source of misunderstanding because many people think that by working many hours on a project and spending a lot of time collecting data, they absolutely should be an author — and in many communities and many labs, this work is automatically recognised as being worthy of authorship. For example, big genome sequencing centres traditionally include most people who are involved in the project as authors, so these genome papers can end up with a very long list of authors.
However, technical contributions are not always considered sufficient for authorship, especially if there is no intellectual contribution. Of course, the question then becomes, ‘what is intellectual contribution?’ In my opinion, collecting data involves some intellectual contribution. We are not machines; we have to analyse and make decisions, and even some relatively minor technical contributions can be viewed as including some intellectual contribution. Say, for instance, the design of primers for a PCR experiment. Some may say that this is a minor technical contribution, but I would argue it’s an intellectual contribution, nonetheless.
In some cases, it may be appropriate for individuals who have made less significant contributions to be acknowledged in the article as contributors or acknowledged in the acknowledgements section. Again, there are no hard and fast rules for what a significant contribution to the research is. We just have to strive to be clear and consistent.
My point is to emphasize that there are many gray areas on where to draw the line on substantial and intellectual contributions. You should really understand the community you work with, the lab you work with, and clarify what the traditions are and what their standards are in terms of authorship. Don’t forget that somebody’s intellectual contribution may be to give you a very important idea, which should not be neglected in terms of authorship. That’s as important contribution, perhaps more important, than technical contributions, which can be viewed as simply someone doing their technical job and not automatically translated into authorship. There are a lot of different opinions here, but again there are no hard rules.
Author ranking
I wonder if the Beatles, when they took this famous photo for the Abbey Road album cover, had a lot of discussion about who should be first, second, etc. I am sure they did. That’s also true for academic papers, and again there are no hard rules here. The rationale for ranking authors varies depending on communities and topics. In fact, in the old days, the ranking was always from the most important to least important. Then molecular biology brought together larger teams and, all of a sudden, there was the idea that the PI should be listed last and that last position on the list became really important. These days it’s the first and last positions that are most coveted, but there are still communities and other areas of science where authors are ranked based on the most important contribution to the least important.
So, not only is it very important to discuss authorship with your project adviser and colleagues, but it’s also critical to discuss author order, to understand the traditions of your field and lab, and to be honest in your assessment of your contributions and the contribution of various authors. This is an ethical exercise that you’re undertaking, and you should be careful not to inflate or overstate your contributions and understand that you’re not the only one involved here.
It is also more and more frequent with larger teams to have shared first authorship. As an example, our recent paper “A pandemic clonal lineage of the wheat blast fungus” has four co-first authors. Four different scientists contributed with different experiments, and we decided that all four contributed enough that even though they are ranked from the first to fourth, there is a statement stating that they are co-first authors and contributed equally. It’s another way to recognize teamwork and that a multidisciplinary study involved important contributions from different authors.
Again, the key here is consistency. Have a consistent standard in your lab, in your group, so you are in control of the level of integrity you want to enforce. Consistency goes a long way towards giving you the authority to argue for whatever decision the PI or senior author has reached. This is imperative for large projects given that different fields, different communities have varying traditions.
Let’s illustrate this further to highlight the confusion that author ranking can bring. Although, the first and last authors these days are viewed, at least in modern biology, as most important, I still remember that in the Entomology Department at Ohio State, they didn’t understand that the last position was so important. They considered that authorship was in declining order of importance and they gave some of their younger PIs a hard time when they went for tenure because they didn’t have a first author paper. This again shows how confusing this business of ranking can be and how it varies between communities.
Here is another example. This is the first paper that describes the PCR method. Funnily enough, Kary Mullis is ranked number four there. This is THE Kary Mullis who later received the Nobel prize for inventing PCR. But on the first paper that describes the method, he was only the fourth author. The story goes that he actually didn’t do the first PCR himself, that was the first author, Randall Saiki. He was just slow in writing the methods paper that was supposed to come out at the same time as Saiki et al. This goes to show that even though he was ranked fourth in the first paper that describes the method he invented, Kary Mullis ended up being the only one to receive the prize because he was credited for coming up with the idea of PCR, not for performing the experiment.
The Kary Mullis example goes to show that authorship is not the only important thing to assign credit. There are other ways of handling credit besides the author order; for example, writing a methods paper in parallel with a research paper could make a clear statement that you are the one who invented this method even though it was used by colleagues and collaborators in the project.
Another progressive development that’s now common, unlike 15 or 20 years ago, is that most journals require a list of author contributions. That actually goes some way towards reducing tension around authorship because even if you’re listed third, fourth, fifth on the paper, in the middle of the pack if you like, you can now always say “look at the author contributions, this is what I actually did in this study”, and conversely, if you didn’t do much but happened to be listed in a big paper, there is also now evidence of this somewhat limited contribution. You can see in this 2005 paper that my colleague Kamal Gajendran is listed further down the list of authors. But if you look at the author contributions, he is actually the first-ranked author who did the bioinformatic analysis — that’s a very important contribution. Hopefully this sort of openness about who did what can reduce tension around the ranking of authors, at least for the positions in this particular paper.
Author responsibilities
There are duties that come with being listed as an author — it’s not just something that’s done for you, that you can add to your CV and forget about. Indeed, we take this topic very seriously in our lab and at The Sainsbury Laboratory where I work. We have a document available on our website detailing our policy on scientific integrity and including the following text. Anyone can read this text, allowing them to understand our position before joining the lab.
All authors should contribute as much as they can to improving the draft of the paper. Scientists are often anxious to be listed as co-authors, but then once they are listed, they don’t do much. That’s really not ok. Being listed as an author comes with responsibilities, including doing your best to improve the work, identify errors, read the proofs etc.
It’s important that even if you are not the corresponding author, you respond to communication about the authorship. For example, if you are asked to approve a paper for submission, you shouldn’t ignore messages and then start complaining months later. Believe me, these things have happened. If you don’t reply to emails about the paper, don’t provide feedback on the paper, then the question can be raised of whether you deserve to be listed as a co-author even if you did a lot of work. Doing a lot of lab work does not protect you from contributing to the other activities that are required for production of the paper. There is a joke (that might be too true to be funny) that it takes 95% of the effort to finish 5% of the project. I think all of us involved in the production of a paper realise how difficult it is to take a research project, even if it is 95% completed, to publication; that last 5% can take a lot of effort. So, if you contributed to the first 95% and think your job is done: no, it’s not done. You are still a part of that study, part of the team; you should really put in the effort to help complete the paper. Sadly, too many co-authors feel as though they did their bit and now they can just ride the wave and wait for the paper to be accepted by a journal. There are responsibilities that come with being an author, whether you’re the senior author or whether you are listed in the middle of the pack.
Issues around authorship may also relate to issues of scientific integrity and misconduct and therefore, authors should be ready to take public responsibility for the content of the paper. That’s very important because these days there is a track record for that paper. The paper will be online, people will be looking at the data, people will be scrutinizing the figures, people will be trying to reproduce the work, and so on. If you are going to be an author on the paper, you have to agree with the content of the paper. You don’t want to have your reputation and career affected by you being co-author on a very poor paper, or a paper that includes some fake information or wrong information that may reflect badly on you. The life of a paper doesn’t end the day it’s published. Post-publication peer-review, on platforms such as PubPeer, is becoming standard procedure in the life sciences.
I know authors who refused to be on a paper because they didn’t agree with the content. That’s perfectly fine; if you’re not happy with the content of the paper, you can choose not to be an author. If you really feel strongly about the topic you work on, of course, you can write a commentary, you can write a review on the paper stating how you disagree, but if you agree to be an author, you cannot come back a year or two later and say you didn’t really agree with that experiment but still want to be an author.
Some authors have very specific duties. For example, in addition to first and last authors, one (or more) authors are designated as corresponding author — that is, the author to whom correspondence should be addressed and are primarily responsible for the content of the paper. Other journals require an author be designated in charge of distributing materials. Corresponding authors ensure that authorship and the order ranking of authors are distributed in a fair and transparent manner.
“You do not submit anything with my name on it without first running it by me”
Being included as an author requires consent of each included author. To illustrate this, I’d like to share my first experience of tension around authorship, which I detailed previously on my blog. I was a naive PhD student working with my supervisor Cal Kado at UC Davis. Cal was my PI, and I applied to attend a graduate student–organized conference. I wrote an abstract with his name on it and I sent it to the organisers without running it by him. When he found out, he gave me a very serious dressing down and told me that nothing with his name on it should ever be submitted anywhere or published without his prior approval. That was a very useful reprimand because it allowed me to appreciate the importance of the topic. My naivety might have reflected how many of us approach authorship initially, not thinking it’s that important. However, it something we should keep in the forefront of our minds whenever we submit an abstract, a poster, a dataset, a preprint, and of course a journal article. We should never forget that authorship is important to our colleagues and co-authors.
Here are some excuses used over the years by people who did not run by me what they were submitting. “You should be happy to be listed anyway so why should I even run the abstract or the paper by you?” “I assumed you would be too busy to reply so I didn’t send you the manuscript.” “It’s just a conference abstract, who cares?” These aren’t valid excuses for neglecting to run the abstract or paper by all co-authors. To start with, if an abstract or poster has my name on it, it reflects on me, so I need to approve it. That means I need to look at it and make sure I agree with the content and fix any typos — or worse factual errors.
There is another reason why you should always run your work by your co-authors before submitting it. You want to show your co-authors that you have a professional attitude, that you actually care for them. Please, always send your co-authors an email to say, “you know, we’re thinking about submitting this abstract, what do you think? Do you have any comments?” Give them plenty of time. What you’re doing is showing them that you care about the issue of authorship and that you take it seriously. They will gain more respect for how you approach this topic, which will ultimately reduce the risk of having issues later on. I think that’s really important because of the drama around authorship. We’ve already got too much drama around authorship in academia!
Last words: Transparency, openness, consistency
Publications, as the key currency in academic careers, tend to stir up tension and managing the process of determining authorship and author order is a big challenge that scientists have to navigate throughout their careers. This tension can be reduced by more openness and transparency as well as care among senior authors to share papers with their co-authors well in advance. Share work in a timely fashion so that co-authors can offer feedback — don’t send it on a Friday telling them we’ll submit the paper on Monday. Explain the rationale for listing authors and ordering them in the way you did, especially when it’s a complicated paper with many authors. Highlight any difficult decisions that may invite debate. I often explain that we really struggled with ranking authors X, Y and Z, but this is the rationale, does anyone have an issue with that? I am happy to discuss with anyone. For senior authors out there: be consistent. Create an accurate record of consistently using the same philosophy in your approach to authorship. For example, some PIs include technicians as authors, while some don’t. This is open for debate, but as long as you are consistent, you are less likely to face issues.
Also, it is always easier to have a difficult discussion before everything is public than afterwards. The worst thing that can happen is that you publish the paper, you post the preprint, you display the poster and then somebody at the conference is very angry because they only just realised they are not happy with the authorship. Keep in mind that this would be much more of a disaster than having the difficult discussion before submission. So, preventing and pre-empting difficult discussions is always good. It’s always easier to have a difficult discussion before submission than after.
I hope that the issues I’ve raised here help you have open discussions about authorship on your papers and make equitable decisions, no matter what stage you are at in your scientific career.
Appendix — when to discuss authorship
This email to my lab spells out the authorship philosophy and approach I have taken throughout my career as a PI. A key point is when to discuss authorship. Although it’s important to have an ongoing discussion about authorship philosophy between a PI and their team at all stages of a project, I categorically refuse to make any pre-arranged deals prior to the completion of the work. This is also the policy of The Sainsbury Laboratory as discussed in this post. Anyway, here is my message:
Dear Labbies,
Our goal is to publish the highest impact papers possible. This would not be possible without some sort of teamwork and collaboration between lab members. My philosophy is to encourage team work as much as possible and to eliminate any barriers that would prevent lab members from working synergistically. You can probably judge that this is working well based on the number of synergistic interactions between different lab members and the number of multiauthored papers we have published over the years. We would have never achieved all this if each one worked on their own in a corner.
Following on the above logic, I categorically refuse to make any deals or agreements on authorship until an advanced draft of the paper is available. Once an advanced draft of the paper is ready, then I will make a decision based on my evaluation of the overall contributions of all participants taking into account not just the execution of the work but also intellectual and conceptual contributions to the study. I typically will consult with the key people to gain a better insight into who contributed to the study. I will justify my decision on the authorship order to the authors. I am always happy to discuss the order further with all authors especially if it appears that two or more authors made similar contributions. In terms of who should be an author, my philosophy is to be inclusive but there has to be a documented contribution of some sort. Note that the importance of intellectual contributions cannot be ignored.
In essence, you will have to trust that I will make a fair decision at that advanced stage. I’ve always strived to make a fair and honest decision. We have had minimal issues with unhappy authors despite the number of papers we published and I stand proud of pretty much all of the decisions.
Thank you for your understanding,
Sophien
Disclaimers
No co-authors were harmed in the making of this article.
Acknowledgements
This post is based on a talk I gave to the CIMMYT Global Wheat Program on November 28, 2022. I thank Alison Bentley, Director the GWP for inviting me in the first place and for her dedication to improving academic culture. Here is the link to the slide deck — Authorship — Let’s talk about it!
I’m grateful to Joe Win for producing the transcript, and to Victoria Hurr and Jennifer Mach at PlantEditors for turning the transcript of the talk into the readable version posted above. I used QillBot and to ChatGPT to improve the text.
