Unpublished works

A few years ago I attended a workshop session on publishing for early-career scientists. One earnest delegate spoke up in favour of submitting work to local journals, especially if you work overseas. It helps build science in your host country, demonstrates willingness to engage with their institutions, and ensures that all your research gets published — even the bits that more prestigious journals might look down upon. For many natural history observations this is about the only way to get such findings into the literature.

I politely disagreed, specifically for early-career researchers, while accepting all the points they made. There is an important skill to learn, and it’s that of letting go. If you can write the big prestigious paper, then write the big prestigious paper. If you can’t, go back to the field/lab/computer and get the data you need to write it. Don’t waste time on the small stuff. It won’t help your CV, and all these noble intentions count for little if you don’t get a job. Recruitment panels won’t care about your lovely paper in the Guatemalan Nature Journal*.

Some people believe that all this unpublished work is a problem for science. Jarrod Hadfield recently wrote, in a provocative meeting report for the Methods in Ecology and Evolution blog, that preregistration of analyses would ensure that “the underworld of unpublished studies would be exposed and their detrimental effects could be adjusted for.” He notes, then dismisses, concerns about the extra workload involved or the frequent changes of plans that take place due to unforeseen circumstances.

Would you, as Orpheus, wish to venture into the underworld? Then look upon my file drawer and weep.


The cabinet of broken dreams. Beware: when you gaze into the file drawer, the file drawer also gazes into you.

This is filled with countless manuscripts at various stages of abandonment. Much sound data collected during my PhD with blood, sweat and tears (all quite literally) languishes here, almost certain to never see the light of day. Likewise there is still unpublished data from my second post-doc. Why have I allowed so many potential publications to rot? How can I live with myself while denying the wider scientific community access to this information?

There’s a simple answer — I had more important things to do. Every active decision you make in life to do something has a consequence elsewhere. Even writing this post. Sometimes I needed to work on another, better paper. The rest of the time I had to do all the things that keep me employed (teaching, administration, grant applications) or sane (sleeping, reading, holidays, drinking).

One thing I’ve learnt in recent years is that the hassle of publishing in a small journal isn’t that much lower than a large journal. There are several reasons for this:

  • Preparing the manuscript is no less time-consuming. Even though the expectations for data quality might be lower, the processes of analysing data, finding and reading the literature, preparing figures and putting everything together are much the same.
  • The quality of reviews is often lower for smaller journals (or at least the variance in quality is higher), increasing the amount of time it takes to respond to them. This shouldn’t be the case, but experience clearly indicates that it is.** Don’t vainly expect the journal to be simply grateful to receive your submission.
  • Lower-ranking journals employ smaller editing teams working with fewer resources. This might not seem like a big deal, but once your paper is accepted it makes all the difference. In a mainstream journal the proofs are turned around quickly and without fuss. It can be on the website in no time. In minor journals you might end up doing much of the legwork yourself. ***

There are sometimes good reasons to publish in a small journal. If you’ve put all the effort into writing a manuscript that was rejected higher up, then go for it, you’ve already invested the time ****. When moving into a new field I like to publish something small just to prove to myself that I can; it also helps with getting my head around a new literature. As a student there’s also great value in getting your first publication anywhere you can, just to experience the process.

What I advise against is writing a paper which you intend from the outset to submit to a small journal. Many studies in ecology don’t get published solely because there’s something better to do. Maybe the results were too complicated to tell a neat story, or couldn’t be easily explained. Maybe all the tests came out insignificant. Given a choice, any scientist should write up the paper with the greatest chance of getting published in a good journal. The small ones are unlikely to provide the same return on your time investment.

The file drawer problem doesn’t occur because we have something to hide, although this may well be true of medical trials or in some highly competitive fields. It’s mostly because we don’t have time. Learn to let go or else the ghosts of unpublished papers will haunt you for the rest of your career.



* Don’t get upset with me over whether they should, the point is that they don’t.

** The reason is pretty obvious. If I receive a review request from Big Name Journal then I know that (a) the authors thought it was important enough to submit there and (b) a specialist editor agreed with them. I’m therefore likely to be interested in it. On the other hand, if I receive a review request from Journal Named After Taxon, I might see which of the post-grads is checking Facebook and offer them a valuable learning experience.

*** In one case I’ve spent more time on editing post-acceptance than I did on writing the paper. I won’t reveal which, but let’s just say that their demands corresponded to neither the website’s Instructions to Authors nor the Chicago Manual of Style.

**** This is only true if your paper was rejected either for not being a good fit or for not quite being interesting or novel enough. If there were fundamental and irredeemable errors with the work then persisting would be a case of Concorde fallacy. Chalk it down to experience and concentrate on fixing the problems for the next manuscript.


10 thoughts on “Unpublished works

  1. aidankeith

    Hi Markus,
    an interesting post and probably sage advice for an early-career scientist who wants to land a job. I’m sure most of us suffer with data and writing leaking out our file drawers.
    It makes me wonder: why do not throw away (recycle) the contents of our file drawers? why do we not hand over the contents of our file drawers to someone with time?



    1. Markus Eichhorn Post author

      Thanks, it’s a reasonable comment. In my case, I wouldn’t compel a post-grad to write up my leftovers when they could be out getting bright shiny new data of their own. If I had cash to employ someone, I’d rather pay them to go out and get more data instead, in the hope of better papers to come. The same opportunity costs would apply to the use of someone else’s time.

      Perhaps one bright light is the growth of macroecology and the collation of large-scale databases. Each individual small study may not get published, but when compiled together they can form part of something worthwhile. In that case not all contributors end up as authors, but I’m still happier to see my data being used and valued by someone.


  2. Bob O'H

    I agree that the problem is a problem, but I don’t agree with the solution. Perhaps we should try to lower the barrier to publishing these results. In particular, the introduction and discussion should be cut down (“We tried to do this: Darwin (1859) explains why it is interesting”), so it’s easier to write and for the reader to digest.


    1. Markus Eichhorn Post author

      Thanks Bob. I spend a lot of time working with physicists, mathematicians and engineers, whose papers make ours look like bloated waffle. I used to feel self-concious about this, but I’ve increasingly come round to the view that the contingent nature of ecological systems requires us to spend more time on making careful comparisons. Changing the culture of publication in ecology would help us to finish more papers (and journals to fit more in) but what would be lost as a result?


      1. Bob O'H

        In a lot of cases, I think little would be lost. I wouldn’t advocate that all papers should be really short, but for some work the gain from publishing the work (and the data) must outweigh the cost of not making all comparisons. If it’s important, someone else can make than comparison and it can be made clear that any conclusions are contingent on the system.


  3. Jan Klecka

    Hi. This is a problem I have been dealing with for a long time now and I understand your position, but I do not agree. I think that not publishing the results of fieldwork or experiments I have already done is always a pity.

    I too believe that publishing simpler papers would be a good solution. I started to write shorter manuscripts after I read a few amazingly brief papers written by physicists. Writing different classes of papers would be efficient: some big comprehensive papers, but also short notes with just a few paragraphs of text. The costs of writing the papers would be reduced also if we submit such short papers to “non-selective” journals. I agree that submitting to small specialized journals can be quite frustrating; I started to avoid these…

    From a career perspective, the advice of focusing only on papers which have the potential to end up in respected journals makes sense. However, I think we would all be better of if we focused more on doing science (big and small) and less on worrying about careers, strategically building our CVs, landing jobs at the most prestigious institutions we can, and all that. I know, I may be too idealistic. So, I have some old data which I believe are interesting, although they are not cool enough to be published in top journals, and I’m trying to publish them anyway. It may not be the best strategy for my career, but I feel that this is the right thing to do.

    I think it’s really important to learn to choose new projects wisely and to estimate how much time we need to complete them, so that we are not collecting more data then we can analyse and publish. I think this is one of the most important skills a supervisor can teach his/her students, although in many cases the reality is very different… Collecting data, but not publishing them means that we wasted a lot of time which could have been better spent on something else. Having a file drawer full of unfinished manuscripts can be quite depressing, so the right strategy is prevention! I have been recently able to carve out some time to work on a few old manuscripts. Publishing them will not make me famous, but it will make me feel good:-)

    Liked by 1 person

    1. Markus Eichhorn Post author

      Thanks for your comment. I think we’re more in agreement than you realise. I’m also spending time writing up several ‘little’ papers this year, but I’m fortunate enough to have a permanent academic position and can choose how to spend (some of) my time without constantly fretting about my career. Unfortunately where we disagree is on worrying less about building CVs and landing jobs. A post-doc looking for their first faculty position doesn’t have the luxury of being idealistic. Search committees are often driven by metrics, at least in the first filter of applicants, and thinking tactically at that stage is essential.

      Your comment about choosing new projects more tactically is an interesting one, and not something I’ve given much thought to before. Usually I dive in to see whether something works before thinking about whether it might eventually lead to a good paper. This is more fun but also leaves a lot of loose threads behind. I’m also still absurdly over-optimistic in my assessment of how much progress I can make in a given amount of time 🙂

      The one issue I have no solution for yet is unpublished work from post-graduates who choose to leave science. If they have no incentive to do it, and I have no time, then it’s a loss to everyone. And so the file drawer fills up.


  4. Pingback: Friday links: #0papers, and more | Dynamic Ecology

  5. Gordon

    Why not dump those papers on the internet with a big disclaimer? It would help, even if it is just to inform the early stages of formulating a hypothesis.


    1. Markus Eichhorn Post author

      Would anyone read or care about them? Their validity would be to some extent judged by the authority of the author, which in turn depends on what they have already published. A senior colleague (now sadly passed away) deposited several of his unpublished studies on his blog site post-retirement. I don’t know whether they proved useful to anyone but at that point he had nothing left to prove as a recognised authority in his field. For a relatively unknown ECR they might as well drop their data in a hedge. It’s still work, and will do nothing to build their CV which is (sadly) the main objective, even if the data turn out to have long-term value.



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 )

Facebook photo

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

Connecting to %s