What’s Wrong With “Publish or Perish”?

By Neuroskeptic | March 28, 2013 10:41 am

‘Publish or perish’. Everyone seems to agree that this phrase describes how science works today, and no-one much likes it. But what exactly is wrong with it?

As I see it, there’s two issues here – the publishing, and the perishing. They’re separate problems, and only loosely linked.

The perishing is really pretty straightforward. In science today, there are more people at the bottom (PhD students) than there is room at the top (Professorships). At every step up on the career ladder, there are less posts available. So not all of the people entering science will be able to succeed in this career, and some of them will ‘perish’ and end up without a career in science. So there’s competition for jobs and advancement.

Perishing is an inevitable consequence of the demographics. It’s linked to publishing only by accident, as it were; today, scientists happen to be assessed mostly by their publications, so it’s publications that save you from perishing. But you can’t blame publishing – there’s just not enough room for everyone. Some people will drop off the science ladder, until we either stop awarding so many PhDs, or until we create more senior posts. It’s simple arithmetic. So we shouldn’t expect reform of publishing, or alt-metrics, to save people from perishing. These reforms could certainly make the system fairer and better, but the fundamental problem is one of recruitment.

Now as we turn to the publishing, this is not terribly popular either. But what’s wrong with it? Surely, scientists have always had to publish, and surely, they always will?

Well, yes, in a broad sense, to publish simply means to make public, but I’ve never met a scientist who has a problem with that. What many scientists dislike is the process of getting our work into the public domain: writing it up, getting it through peer-review, etc.

I think there are two main reasons for disliking publishing: one, it’s hard work and takes time; two, we feel that the process of turning science into papers actually distorts the science e.g. it encourages only writing up positive results; splitting a coherent series of findings into multiple papers, etc. In other words, scientists resent publications, not publishing. What we don’t like is the need to produce ‘papers’.

So it’s encouraging that people are now talking about the end of the scientific paper, which I think could be a very good thing. However, we shouldn’t put all of our hopes into this. There will still be just as much perishing as before – that’s a separate challenge.

CATEGORIZED UNDER: science, select, Top Posts, Uncategorized
  • http://twitter.com/johnmyleswhite John Myles White

    Aren’t all the incentives faced by faculty pointed towards excessive recruitment? As everyone says, it’s an effective source of cheap labor that makes it easier to publish and acquire tenure. But it’s also an effective way to have enough students to justify acquiring lots of grants, which not only helps one acquire tenure, but also pleases the larger university administration.

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Yes, that’s a large part of the problem I think. I’m not sure how best to fix it. Possibly a quota system would be necessary – a maximum number of PhDs per researcher (or even institution)?

  • http://twitter.com/EricRietzschel Eric Rietzschel

    Another problem may be that so many parts of turning a paper into a publication lie outside of individual researchers’ control (or at least seem to do so) that it feels unfair to be judged by publications as the only measure of output. Whether a paper gets accepted for publication is only partly determined by the quality of the paper and the quality of the research. Journal policies, editors’ preferences, ‘luck of the draw’ in reviewer selection, papers submitted by other researchers, and so on, all contribute to a paper’s fate (or are believed to do so), and hence to a researcher’s career opportunities. In a sense, then, publication records as a measure of job performance may be seen by researchers as suffering from criterion contamination.

  • JonFrum

    I would argue that it’s less publish or perish (at least in many fields), than bring in grant money or perish. An independently wealthy associate professor who finances every penny of her own research and publishes in a constant stream wouldn’t bring in grant money – a good chunk of which goes not into the research directly, but into the university kitty to pay for overhead. So publishing is just a proxy for bringing home the bacon. Any faculty who brings in a large and constant stream of grant money isn’t going to perish, no matter how few publications they produce, or what journals they publish in.

    • http://twitter.com/jdottan Joseph Tan

      I think this is a sensible observation and it brings in the issue of requirements for funding. I’m curious, do you think bringing in grant money is still dependent on publishing? Because while your grant proposal has to be well written, your merit as a researcher able to carry out the grant is considered as well, correct? Then that would bring in how that merit is indexed, and it seems like publishing is how that happens.

  • http://twitter.com/davenuss79 Dave Nussbaum

    Just 2 more cents: the inflation of publications has been enormous the past few decades — it used to be that we published when we had something important to share. Now we publish because we have to, and usually without respect to whether it’s worth sharing. It’s a destructive race, but it’s one of those collective action problems that no individual can solve, and it’s hard to see what external forces are going to step in and call for fewer publications.

    Some departments focus on quality over quantity, but they’re relatively rare. Perhaps a metric that shows the “impact” of your average paper, so that every useless publication actually pulls down your average? Simmons et al suggested (jokingly, but…) that everyone gets one publication a year (max) and so you really have to stop and consider what you publish, and obviously inflation would be capped. In any case, a situation worth thinking about. As Dick Thaler noted in an interview on indecisionblog, the research that led to Kahneman’s (and Tversky) Nobel prize would not be enough to get you tenure many places these days — yet it’s obviously more deserving of tenure than a long list of publications each of which contributes little or nothing, as we see on many CVs these days.

    • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

      Yes. I’ve suggested (semi-seriously) that rather than # of citations it should be mean citations per paper.

      • linda518

        Matthew. true that Jacqueline`s st0ry is really cool, yesterday I got a great new Honda since getting a cheque for $5010 this – 4 weeks past and-just over, $10 thousand last munth. with-out any question its my favourite-work I’ve ever had. I started this 3 months ago and straight away started making a nice more than $78… per-hr. I follow the details on this straightforward website,, jump15.comCHECK IT OUT

  • http://twitter.com/practiCalfMRI practiCal fMRI

    I realize that the PhD evolved largely to provide a conduit to an academic career (not necessarily an academic research career, I hasten to add), but I would be loth to restrict the numbers pursuing PhDs just because of a restricted pool of academic jobs. There are many (more?) jobs outside of academia, lots of them with high research content, or with other duties that a PhD-trained person may excel at, and the training obtained in a PhD is still exceptionally useful. Thus, I concur, that the academic success/perish question should be well separated from the publishing issues.

    • http://twitter.com/infinidiv infinidiv

      Glad I kept scrolling before posting, this is exactly what I wanted to mention. There is almost no discussion in universities and among academics of the opportunities outside academia. And there are really many of those, in many different fields. Private research often gets a bad rap because people see themselves having to work for a pesticide or genetics conglomerate, but even in this topic we often forget that the value of a PhD goes far beyond the learned facts. In many NGOs and government agencies it is the general skills that are acquired and shown during a PhD that matter.

      PhD students or people interested in starting a PhD should remember that outside of academia that title means a lot more than within. Organizing your own project, dealing with roadblocks, an ability to teach yourself from the ground up, are all expected in the academic community, but they are not as widespread everywhere else (I am not saying they are non-existent! just that this is a very good way to figure out if someone has the ability, and it is made use of in many areas outside academia in that sense).

      • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

        Maybe the solution then would be to encourage people to see non-academic careers as viable options. Because there is certainly a problem in academia of too many PhDs; but as you & practiCal fMRI point out that doesn’t mean there are too many PhDs in society as a whole…

        • http://twitter.com/davenuss79 Dave Nussbaum

          Here’s an interesting recent interview with Paul Litvak (@paullitvak) about his experiences going back and forth from academia to non-academic jobs. http://indecisionblog.com/2013/03/25/into-the-wild-paul-litvak/

        • http://twitter.com/practiCalfMRI practiCal fMRI

          Several people from my institution have gone to work for social media & other online content delivery companies after getting a PhD. One even went into sports performance, starting his own company. And these are psych PhDs, too. So, yes, I think there are increasingly viable ways to have a great career outside of academia, even more so if you’re good at stats and/or programming. Indeed, around here it’s the academics who are starting to feel left out!

  • Pingback: What's Wrong With "Publish or Perish"? | WorldTopStories.com()

  • Audrey

    The inefficiencies of publication go beyond the peer-review and paywall issues. The structure of articles that require a literature review slows writing and adds bloat to each article. If I ran an experiment on, say, the verbal memory skills of kids with ADHD, the lit review should consist of citing a couple of review articles and perhaps an article or two that was published after the review articles came out. The rest just takes time, accomplishes little, and gets used by reviewers as a way to get themselves cited. But mostly, just a waste of time.

    • Dennis

      Sorry, but if you can’t motivate your work and show your thought process that lead you to do the experiments based on current scientific opinion, you are not doing science. You just measure shit.

      • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic

        Well, if you need the exhaustive detail seen in current papers to be able to understand those papers, you’re probably not doing science either.

      • JonFrum

        A lot of references I’ve seen were just filling in the blanks to make sure there were enough at the end of the paper to make it look like you knew what you were doing. This kind of referencing is good for students just getting into the literature, but does nothing to add to the paper itself. In other words, it’s make-work.

        • http://blogs.discovermagazine.com/neuroskeptic/ Neuroskeptic


  • hudasx

    When I first heard “publish or perish” as a graduate student in a developing country I thought it was the silliest thing ever. I’m supposed to do beg for research funds that could otherwise be used to build a new elementary school, read journals that cost as more than the library’s operating budget, write a paper that maybe 10 other people in the world will read, then pay the equivalent of our clerk’s annual salary to actually publish it (if it is worthy). Maybe this system works in the first world, but publishing shouldn’t be what the scanty few scholars in the third world should be aspiring for. And yet that’s how they’re measured simply because everyone else does.

    • http://twitter.com/infinidiv infinidiv

      Thanks for mentioning this. I feel like I am one of the few people in the “first world” who mentions this kind of thing. Its a lens that most people have a hard time looking past. We are annoyed by it all, but often we forget that there are many more people on this planet for whom it is even more difficult. The entire system is based on a very western view… and even that view is not even a century old. We need to remember that when we talk about “publish or perish” as such a static system. Considering how much animosity it has created in such a short time, I don’t think it will be around for long

      @hudasx:disqus on the note of the costs of publishing (only one of the points you made) have a look at the open access journals. Several of them offer free publishing for researchers who cannot pay the exorbitantly high fees. That might be an option for you. Otherwise, you could also have a look at “peerj”. Could be a more affordable option for you as well.

      • hudasx

        I should have mentioned I was a graduate student 30 years ago. Open source journals are a godsend, especially because even if I don’t believe in the importance of publishing it is critical to keep up-to-date in the science. The publish-or-perish system kept very ineffective but prolific professors in the classroom, while treating excellent lecturers dedicated only to teaching well as second-class academics. Unfortunately, teaching ability is hard to measure objectively; counting the number of publications is easy. That, sadly, is why the system will stay in place.

  • http://profiles.google.com/jan.moren Jan Moren

    I suspect that perhaps what people are really ranting against is less “publish or perish” but rather the “up or out” structure of the science career. When you think about it, the idea that you have to be either promoted or lose your job is both somewhat silly and rare in the work environment at large.

    Imagine an engineering division at a major manufacturer where, no matter how good an engineer you are, you must become group leader within ten years and department head before fifty or you will lose your job? Or a hospital where you will not be allowed to continue as a physician or nurse for more than a set number of years unless you get promoted?

    • Dennis

      Given postdoc salary, it is only responsible to kick them out when they don’t make it into a PI position.

      • http://profiles.google.com/jan.moren Jan Moren

        A post-doc salary is really not that bad. May not be as much as you’d make out of academia, but it’s still entirely comparable to that of a teacher or many other professions.

  • Pingback: Somewhere else, part 41 | Freakonometrics()

  • Scientific Editing

    Quality matters the most when it comes to scientific publishing and what matters is the reason for getting perished. Of course, stricter norms by journals do make researcher’s job painful. We at http://www.enago.com make this job easier for researchers.

  • Pingback: Friday Paper List – Saturday March 30th | Mo Papers Mo Problems()

  • Pingback: Impact Factor’s flaws, in 200 words - sMemo()

  • Pingback: Bird for Buck or ‘You left a well paid job to do what?!’ | DeanDoesResearch()

  • Pingback: Breaking the invisible boundaries of your research | Psychology and Neuroscience Research()

  • Pingback: Student Research on Voluntourism: Creating “Real-Time” Benefits for Practitioners | VolunTourism Institute()

  • Pingback: bjoern.brembs.blog » On the productivity of scientists()

  • Pingback: Bad Science and the case of ANOVA | Psychology and Neuroscience stuff()



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar